Refactor of caching & data storage part 17 #118

.idea/encodings.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="Encoding">
+    <file url="file://$PROJECT_DIR$/datafiles/MUNICIPIOS.csv" charset="ISO-8859-1" />
+  </component>
+</project>

.idea/inspectionProfiles/Project_Default.xml

@@ -1,6 +1,48 @@
 <component name="InspectionProjectProfileManager">
   <profile version="1.0">
     <option name="myName" value="Project Default" />
+    <inspection_tool class="Annotator" enabled="false" level="ERROR" enabled_by_default="false" />
+    <inspection_tool class="BadExpressionStatementJS" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
+    <inspection_tool class="CssOverwrittenProperties" enabled="false" level="WARNING" enabled_by_default="false" />
+    <inspection_tool class="CssUnresolvedCustomProperty" enabled="false" level="ERROR" enabled_by_default="false" />
+    <inspection_tool class="CssUnusedSymbol" enabled="false" level="WARNING" enabled_by_default="false" />
+    <inspection_tool class="GrazieInspection" enabled="false" level="GRAMMAR_ERROR" enabled_by_default="false" />
+    <inspection_tool class="GrazieStyle" enabled="false" level="STYLE_SUGGESTION" enabled_by_default="false" />
+    <inspection_tool class="HtmlFormInputWithoutLabel" enabled="true" level="WEAK WARNING" enabled_by_default="true" editorAttributes="INFO_ATTRIBUTES" />
+    <inspection_tool class="HtmlUnknownAttribute" enabled="false" level="WARNING" enabled_by_default="false">
+      <option name="myValues">
+        <value>
+          <list size="0" />
+        </value>
+      </option>
+      <option name="myCustomValuesEnabled" value="true" />
+    </inspection_tool>
+    <inspection_tool class="HtmlUnknownTag" enabled="false" level="WARNING" enabled_by_default="false">
+      <option name="myValues">
+        <value>
+          <list size="6">
+            <item index="0" class="java.lang.String" itemvalue="nobr" />
+            <item index="1" class="java.lang.String" itemvalue="noembed" />
+            <item index="2" class="java.lang.String" itemvalue="comment" />
+            <item index="3" class="java.lang.String" itemvalue="noscript" />
+            <item index="4" class="java.lang.String" itemvalue="embed" />
+            <item index="5" class="java.lang.String" itemvalue="script" />
+          </list>
+        </value>
+      </option>
+      <option name="myCustomValuesEnabled" value="true" />
+    </inspection_tool>
+    <inspection_tool class="HtmlUnknownTarget" enabled="false" level="WARNING" enabled_by_default="false" />
+    <inspection_tool class="HttpUrlsUsage" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
+    <inspection_tool class="JSDeprecatedSymbols" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
+    <inspection_tool class="JSIgnoredPromiseFromCall" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
+    <inspection_tool class="JSJQueryEfficiency" enabled="false" level="WARNING" enabled_by_default="false" />
+    <inspection_tool class="JSUnnecessarySemicolon" enabled="false" level="WARNING" enabled_by_default="false" />
+    <inspection_tool class="JSUnresolvedReference" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
+    <inspection_tool class="JSUnusedGlobalSymbols" enabled="true" level="WEAK WARNING" enabled_by_default="true" editorAttributes="INFO_ATTRIBUTES" />
+    <inspection_tool class="LanguageDetectionInspection" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
+    <inspection_tool class="OutdatedRequirementInspection" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
+    <inspection_tool class="PyBroadExceptionInspection" enabled="false" level="WEAK WARNING" enabled_by_default="false" />
     <inspection_tool class="SpellCheckingInspection" enabled="false" level="TYPO" enabled_by_default="false">
       <option name="processCode" value="true" />
       <option name="processLiterals" value="true" />

.idea/inspectionProfiles/profiles_settings.xml

@@ -1,6 +0,0 @@
-<component name="InspectionProjectProfileManager">
-  <settings>
-    <option name="USE_PROJECT_PROFILE" value="false" />
-    <version value="1.0" />
-  </settings>
-</component>

.idea/jsLibraryMappings.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="JavaScriptLibraryMappings">
+    <file url="file://$PROJECT_DIR$" libraries="{redoc.standalone}" />
+  </component>
+</project>

.idea/spothole.iml

@@ -2,9 +2,12 @@
 <module type="PYTHON_MODULE" version="4">
   <component name="NewModuleRootManager">
     <content url="file://$MODULE_DIR$">
+      <sourceFolder url="file://$MODULE_DIR$" isTestSource="false" />
       <excludeFolder url="file://$MODULE_DIR$/.venv" />
+      <excludeFolder url="file://$MODULE_DIR$/static/vendor" />
     </content>
     <orderEntry type="jdk" jdkName="Python 3.13 virtualenv at ~/code/spothole/.venv" jdkType="Python SDK" />
     <orderEntry type="sourceFolder" forTests="false" />
+    <orderEntry type="library" name="redoc.standalone" level="application" />
   </component>
 </module>

README.md

@@ -1,16 +1,25 @@
-# ![Spothole](/webassets/img/logo.png)
+# ![Spothole](/static/img/logo.png)
 
-Spothole is a utility to aggregate "spots" from amateur radio DX clusters and xOTA spotting sites, and provide an open JSON API as well as a website to browse the data.
+Spothole is a utility to aggregate "spots" from amateur radio DX clusters and xOTA spotting sites, and provide an open
+JSON API as well as a website to browse the data.
 
 ![Screenshot](/images/screenshot.png)
 
-While there are several other web-based interfaces to DX clusters, and sites that aggregate spots from various outdoor activity programmes for amateur radio, Spothole differentiates itself by supporting a large number of data sources, and by being "API first" rather than just providing a web front-end. This allows other software to be built on top of it.
+While there are several other web-based interfaces to DX clusters, and sites that aggregate spots from various outdoor
+activity programmes for amateur radio, Spothole differentiates itself by supporting a large number of data sources, and
+by being "API first" rather than just providing a web front-end. This allows other software to be built on top of it.
 
-The API is deliberately well-defined with an OpenAPI specification and auto-generated API documentation. The API delivers spots in a consistent format regardless of the data source, freeing developers from needing to know how each individual data source presents its data.
+The API is deliberately well-defined with an OpenAPI specification and auto-generated API documentation. The API
+delivers spots in a consistent format regardless of the data source, freeing developers from needing to know how each
+individual data source presents its data.
 
 Spothole itself is also open source, Public Domain licenced code that anyone can take and modify.
 
-Supported data sources include DX Clusters, the Reverse Beacon Network (RBN), the APRS Internet Service (APRS-IS), POTA, SOTA, WWFF, GMA, WWBOTA, HEMA, Parks 'n' Peaks, ZLOTA, WOTA, BOTA, LLOTA, WWTOTA, Tiles on the Air, the UK Packet Repeater Network, NG3K, and any site based on the xOTA software by nischu.
+Supported data sources include DX Clusters, the Reverse Beacon Network (RBN), the APRS Internet Service (APRS-IS), POTA,
+SOTA, WWFF, GMA, WWBOTA, HEMA, Parks 'n' Peaks, ZLOTA, WOTA, BOTA, LLOTA, WWTOTA, Tiles on the Air, the UK Packet
+Repeater Network, NG3K, and any site based on the xOTA software by nischu.
+
+Additional Special Interest Groups (SIGs) without their own specific data source include WAB, WAI and DME.
 
 ![Screenshot](/images/screenshot2.png)
 
@@ -18,24 +27,39 @@ Supported data sources include DX Clusters, the Reverse Beacon Network (RBN), th
 
 ## Accessing the public version
 
-You can access the public version's web interface at [https://spothole.app](https://spothole.app), and see [https://spothole.app/apidocs](https://spothole.app/apidocs) for the API details.
+You can access the public version's web interface at [https://spothole.app](https://spothole.app), and
+see [https://spothole.app/apidocs](https://spothole.app/apidocs) for the API details.
 
-This is a Progressive Web App, so you can also "install" it to your Android or iOS device by accessing it in Chrome or Safari respectively, and following the menu-driven process for installing PWAs.
+This is a Progressive Web App, so you can also "install" it to your Android or iOS device by accessing it in Chrome or
+Safari respectively, and following the menu-driven process for installing PWAs.
+
+You are more than welcome to use the data and the API that Spothole provides to power your own software. There are many
+ways to do this; see below.
 
 ## Embedding Spothole in another website
 
-You can embed Spothole in another website, e.g. for use as part of a ham radio custom dashboard.
+You can embed Spothole's web interface in another website, e.g. for use as part of a ham radio custom dashboard.
 
-URL parameters can be used to trigger an "embedded" mode which hides the headers, footers and settings. In this mode, you provide configuration for the various filter and display options via additional URL parameters. Any settings that the user has set for Spothole are ignored. This is so that the embedding site can select, for example, their choice of dark mode or SIG filters, which will not impact how Spothole appears when the user accesses it directly. Effectively, it becomes separate to their normal Spothole settings.
+URL parameters can be used to trigger an "embedded" mode which hides the headers, footers and settings. In this mode,
+you provide configuration for the various filter and display options via additional URL parameters. Any settings that
+the user has set for Spothole are ignored. This is so that the embedding site can select, for example, their choice of
+dark mode or SIG filters, which will not impact how Spothole appears when the user accesses it directly. Effectively, it
+becomes separate to their normal Spothole settings.
 
-Setting `embedded` to true is important for the rest of the settings to be applied; otherwise, the user's defaults will be used in preference to the URL params.
+Setting `embedded` to true is important for the rest of the settings to be applied; otherwise, the user's defaults will
+be used in preference to the URL params.
 
-These are supplied with the URL to the page you want to embed, for example for an embedded version of the band map in dark mode, use `https://spothole.app/bands?embedded=true&dark-mode=true`. For an embedded version of the main spots/home page in the system light/dark mode, use `https://spothole.app/?embedded=true`. For dark mode showing 70cm TOTA spots only, use `https://spothole.app/?embedded=true&dark-mode=true&sig=TOTA&band=70cm`. Providing no URL params causes the page to be loaded in the normal way it would when accessed directly in the user's browser.
+These are supplied with the URL to the page you want to embed, for example for an embedded version of the band map in
+dark mode, use `https://spothole.app/bands?embedded=true&dark-mode=true`. For an embedded version of the main spots/home
+page in the system light/dark mode, use `https://spothole.app/?embedded=true`. For dark mode showing 70cm TOTA spots
+only, use `https://spothole.app/?embedded=true&dark-mode=true&sig=TOTA&band=70cm`. Providing no URL params causes the
+page to be loaded in the normal way it would when accessed directly in the user's browser.
 
-The supported parameters are as follows. Generally these match the equivalent parameters in the real Spothole API, where a mapping exists.
+The supported parameters are as follows. Generally these match the equivalent parameters in the real Spothole API, where
+a mapping exists.
 
 | Name             | Allowed Values          | Default | Example                 | Description                                                                                                                                                           |
-|-------------------|-------------------------|---------|--------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+|------------------|-------------------------|---------|-------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------|
 | `embedded`       | `true`, `false`         | `false` | `?embedded=true`        | Enables embedded mode.                                                                                                                                                |
 | `color-scheme`   | `light`, `dark`, `auto` | `auto`  | `?color-scheme=dark`    | Forces light or dark mode in preference to the operating system default.                                                                                              |
 | `time-zone`      | `UTC`, `local`          | `UTC`   | `?time-zone=local`      | Sets times to be in UTC or local time.                                                                                                                                |
@@ -52,29 +76,58 @@ The supported parameters are as follows. Generally these match the equivalent pa
 | `map-center-lon` | Numeric (decimal)       | (auto)  | `?map-center-lon=-0.1`  | Sets the initial longitude of the map centre on the map page. If omitted, the map auto-fits to the loaded spots.                                                      |
 | `map-zoom`       | Numeric (integer)       | (auto)  | `?map-zoom=6`           | Sets the initial zoom level of the map on the map page. If omitted, the map auto-fits to the loaded spots.                                                            |
 
-More will be added soon to allow customisation of filters and other display properties.
+See the comment at the end of the next section regarding reliability and uptime of the "main" server.
 
 ## Writing your own client
 
-One of the key strengths of Spothole is that the API is well-defined and open to anyone to use. This means you can build your own software that uses data from Spothole.
+One of the key strengths of Spothole is that the API is well-defined and open to anyone to use. This means you can build
+your own software that uses data from Spothole.
 
-As well as the main API endpoints to fetch spots and alerts, with various possible query parameters, there are also Server-Sent Events (SSE) API endpoints to receive a live feed, plus various utility lookup endpoints for things like callsign and park data.
+As well as the main API endpoints to fetch spots and alerts, with various possible query parameters, there are also
+Server-Sent Events (SSE) API endpoints to receive a live feed, plus various utility lookup endpoints for things like
+callsign and park data.
 
 Various approaches exist to writing your own client, but in general:
 
-* Refer to the API docs. These are built on an OpenAPI definition file (`/webassets/apidocs/openapi.yml`), which you can automatically use to generate a client skeleton using various software.
-* Call the main "spots" or "alerts" API endpoints to get the data you want. Apply filters if necessary.
-* Call the "options" API to get an idea of which bands, modes etc. the server knows about. You might want to do that first before calling the spots/alerts APIs, to allow you to populate your filters correctly.
-* Refer to the provided HTML/JS interface for a reference on different approaches. For example, the "map" and "bands" pages simply query the main spot API on a timer, whereas the main/spots page combines this approach with using the Server-Sent Events (SSE) endpoint to update live.
-* Let me know if you get stuck, I'm happy to help!
+* Refer to the API docs. These are built on an OpenAPI definition file (`/static/apidocs/openapi.yml`), which you can
+  automatically use to generate a client skeleton using various software.
+* Call the main "spots" or "alerts" API endpoints to get the data you want. For example, your app could call
+  `https://spothole.app/api/v1/spots` once every few minutes. Apply filters if necessary.
+* Call the "options" API to get an idea of which bands, modes etc. the server knows about. You might want to do that
+  first before calling the spots/alerts APIs, to allow you to populate your filters correctly.
+* Refer to the provided HTML/JS interface for a reference on different approaches. For example, the "alerts"/"upcoming"
+  page simply query the main spot API on a timer, whereas the spots, map and bands pages combine this approach with
+  using the Server-Sent Events (SSE) endpoint to update live.
+* Let me know if you get stuck, I'm happy to help.
+
+Please don't hammer the API with an unnecessarily high request rate. For example, Spothole only queries the POTA API
+once every two minutes, so if your client is interested in POTA data there's no need to poll Spothole any more often
+than that.
+
+If you absolutely must be informed within seconds of a spot arriving in Spothole, please use the SSE endpoints instead,
+e.g. `https://spothole.app/api/v1/spots/stream`.
+
+If you want to handle different types of spot or alert differently within your client, please consider making a single
+request to the Spothole API to retrieve all the data, then filtering on your side. For example, call
+`https://spothole.app/api/v1/spots?sig=POTA,SOTA` rather than making two separate calls to
+`https://spothole.app/api/v1/spots?sig=POTA` and `https://spothole.app/api/v1/spots?sig=SOTA`.
+
+Remember, here at Spothole Inc. we offer an industry-standard "five nines" uptime on our server, with our own unique
+twist: we don't tell you which side of the decimal point the nines start! (Translation: This is a hobby project.
+`spothole.app` runs on the same server as my blog and other stuff. It might go down without warning. By all means base
+your own project on data from the main server if you like, but if you want any control over reliability and downtime,
+please run your own copy instead.)
 
 ## Running your own copy
 
-If you want to run a copy of Spothole with different configuration settings than the main instance, you can download it and run it on your own local machine or server.
+If you want to run a copy of Spothole with different configuration settings than the main instance, you can download it
+and run it on your own local machine or server.
 
-You will require Python version 3.8 or later. If you encounter an error about `gdal-config` during the following process, you will also need `libgdal-dev` installed.
+You will require Python version 3.8 or later. If you encounter an error about `gdal-config` during the following
+process, you will also need `libgdal-dev` installed.
 
-To download and set up Spothole on a Debian server, run the following commands. Other operating systems will likely be similar.
+To download and set up Spothole on a Debian server, run the following commands. Other operating systems will likely be
+similar.
 
 ```bash
 git clone ssh://git@git.ianrenton.com/ian/spothole.git
@@ -86,15 +139,23 @@ deactivate
 cp config-example.yml config.yml
 ```
 
-Then edit `config.yml` in your text editor of choice to set up the software as you like it. Mostly, this will involve enabling or disabling the various providers of spot and alert data.
+Then edit `config.yml` in your text editor of choice to set up the software as you like it. Mostly, this will involve
+enabling or disabling the various providers of spot, alert and other data.
 
-By default, all outdoor programme providers are enabled, as is one cluster node and the NG3K DXpedition data. The RBN spot providers are turned off by default due to the volume of traffic from CW/RTTY/FT8 skimmers, and the APRS and Packet spot providers are off by default on the assumption that Spothole users want a spot with a human at the other end of it, but all can be easily re-enabled.
+By default, all outdoor programme providers are enabled, as is one cluster node and the NG3K DXpedition data. The RBN
+spot providers are turned off by default due to the volume of traffic from CW/RTTY/FT8 skimmers, and the APRS and Packet
+spot providers are off by default on the assumption that Spothole users want a spot with a human at the other end of it,
+but all can be easily re-enabled.
 
-`config.yml` has some entries for QRZ.com username & password, and Clublog API keys. If provided, these allow Spothole to retrieve more information about DX spots, such as the country their callsign corresponds to. The software will work just fine without them, but you may find a few country flags etc. are less accurate or missing.
+Other parameters you will want to update include the base URL to your instance, and whether you want to serve a full
+web-based DX cluster interface or just the API endpoints for client software to use.
 
-Clublog API keys are free, but you'll need to get your own by submitting a helpdesk ticket and explaining what you'll use it for. The admin team are happy with the rate of requests made by my Spothole server, so unless you change the source code of yours to radically increase the rate of querying Clublog, I'm sure they will be fine with your server too.
-
-Free QRZ.com accounts offer only limited access to the site's data via their API. You'll have to sign up for one of their "XML Data Subscriber" plans to gain access to the full data, but if you're on a free account then the software will get what information it can.
+`config.yml` has an entry for a Clublog API key. If provided, this will allow Spothole to retrieve some more information
+about DX spots. The software will work just fine without it, but you may find a few country flags etc. are less accurate
+or missing. Clublog API keys are free, but you'll need to get your own by submitting a helpdesk ticket and explaining
+what you'll use it for. The admin team are happy with the rate of requests made by my Spothole server, so unless you
+change the source code of yours to radically increase the rate of querying Clublog, I'm sure they will be fine with your
+server too.
 
 Once you're happy with the content of `config.yml`, you can proceed to running the software.
 
@@ -105,13 +166,17 @@ source .venv/bin/activate
 python3 spothole.py
 ```
 
-The software can take a few seconds to start up, mostly because it is downloading an updated file to match callsigns to countries. This is normal, don't panic!
+The software can take a few seconds to start up, mostly because it is downloading updated files containing the various
+bits of reference data it uses. This is normal, don't panic! Once you see `You can access your copy of Spothole at
+http://localhost:8080` in the log, your server is good to go.
 
-If you see some errors on startup, check your configuration, e.g. in case you have specified a port for the web server that is already in use by something else.
+If you see some errors on startup, check your configuration, e.g. in case you have specified a port for the web server
+that is already in use by something else.
 
 ### Multiple cluster nodes with different settings
 
-Dan, S50U has written in with his Spothole cluster settings. He is using a cluster node which provides RBN spots, and uses different SSIDs on his callsign to get different settings when logged into the same cluster node. For example:
+Dan, S50U has written in with his Spothole cluster settings. He is using a cluster node which provides RBN spots, and
+uses different SSIDs on his callsign to get different settings when logged into the same cluster node. For example:
 
 ```
   -
@@ -207,9 +272,12 @@ For each callsign-SSID, we also specify our basic information with commands:
 
 ### systemd configuration
 
-If you want Spothole to run automatically on startup on a Linux distribution that uses `systemd`, follow the instructions here. For distros that don't use `systemd`, or Windows/OSX/etc., you can find generic instructions for your OS online.
+If you want Spothole to run automatically on startup on a Linux distribution that uses `systemd`, follow the
+instructions here. For distros that don't use `systemd`, or Windows/OSX/etc., you can find generic instructions for your
+OS online.
 
-Create a file at `/etc/systemd/system/spothole.service`. Give it the following content, adjusting for the user you want to run it as and the directory in which you have installed it:
+Create a file at `/etc/systemd/system/spothole.service`. Give it the following content, adjusting for the user you want
+to run it as and the directory in which you have installed it:
 
 ```
 [Unit]
@@ -239,43 +307,103 @@ Check the service has started up correctly with `sudo journalctl -u spothole -f`
 
 ### nginx Reverse Proxy configuration
 
-Web servers generally serve their pages from port 80. However, it's best not to serve Spothole's web interface directly on port 80, as that requires root privileges on a Linux system. It also and prevents us using HTTPS to serve a secure site, since Spothole itself doesn't directly support acting as an HTTPS server. The normal solution to this is to use a "reverse proxy" setup, where a general web server handles HTTP and HTTP requests (to port 80 & 443 respectively), then passes on the request to the back-end application (in this case Spothole). nginx is a common choice for this general web server.
+Web servers generally serve their pages from port 80. However, it's best not to serve Spothole's web interface directly
+on port 80, as that requires root privileges on a Linux system. It also and prevents us using HTTPS to serve a secure
+site, since Spothole itself doesn't directly support acting as an HTTPS server. The normal solution to this is to use a
+"reverse proxy" setup, where a general web server handles HTTP and HTTP requests (to port 80 & 443 respectively), then
+passes on the request to the back-end application (in this case Spothole). nginx is a common choice for this general web
+server.
 
-To set up nginx as a reverse proxy that sits in front of Spothole, first ensure it's installed e.g. `sudo apt install nginx`, and enabled e.g. `sudo systemd enable nginx`.
+To set up nginx as a reverse proxy that sits in front of Spothole, first ensure it's installed e.g.
+`sudo apt install nginx`, and enabled e.g. `sudo systemd enable nginx`.
 
-Create a file at `/etc/nginx/sites-available/` called `spothole`. Give it the following contents, replacing `spothole.app` with the domain name on which you want to run Spothole. If you changed the port on which Spothole runs, update that on the "proxy_pass" line too.
+Create a file at `/etc/nginx/sites-available/` called `spothole`. Give it the following contents, replacing
+`spothole.app` with the domain name on which you want to run Spothole. If you changed the port on which Spothole runs,
+update that on the "proxy_pass" line, and if you installed Spothole somewhere other than `/home/spothole/spothole`,
+adjust the alias location for serving static files.
+
+(The latter section, configuring the nginx server to serve static files directly, improves efficiency because it saves
+Spothole itself from serving JS, CSS etc. files. If you can't do this for some reason, e.g. your nginx and spothole are
+on different computers, you can omit the `location /static/ {}` block.)
 
 ```nginx
-map $request_uri $xssorigin {
-    ~^/api   *;
-}
-
 server {
     server_name spothole.app;
 
+    # Global proxy settings
+    proxy_http_version 1.1;
+    proxy_set_header Connection "";
+    proxy_connect_timeout 10s;
+    proxy_buffering on;
+    
+    # Pass on IP address and host information to Spothole, in case logging this information is required
+    proxy_set_header X-Real-IP $remote_addr;
+    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+    proxy_set_header Host $http_host;
+    proxy_set_header X-Forwarded-Proto $scheme;
+
     # Wellknown area for Lets Encrypt
     location /.well-known/ {
         alias /var/www/html/.well-known/;
     }
 
-    location / {
-        proxy_http_version 1.1;
-        proxy_set_header Connection "";
+    # Load static assets directly from the Spothole static directory
+    location /static/ {
+        alias /home/spothole/spothole/static/;
+        expires 1h;
+        add_header Cache-Control "public, max-age=3600, must-revalidate";
+    }
+    
+    # SSE endpoints
+    location ~ ^/api/v1/(spots|alerts)/stream/? {
         proxy_pass http://127.0.0.1:8080;
+        
+        # Remove buffering, remove caching, add suitable timeouts for SSE API calls
+        proxy_buffering off;
+        proxy_cache off;
+        proxy_read_timeout 24h;
+        proxy_send_timeout 24h;
+        proxy_set_header X-Accel-Buffering no;
+        add_header Cache-Control no-store always;
+        
+        # Allow cross-origin requests to API
         proxy_hide_header Access-Control-Allow-Origin;
-        add_header Access-Control-Allow-Origin $xssorigin;
+        add_header Access-Control-Allow-Origin * always; 
+    }
+
+    # Other API endpoints
+    location /api/ {
+        proxy_pass http://127.0.0.1:8080;
+        
+        # Remove buffering, remove caching, add suitable timeouts for API calls
+        proxy_buffering off;
+        proxy_cache off;
+        proxy_read_timeout 30s;
+        add_header Cache-Control no-store always;
+        
+        # Allow cross-origin requests to API
+        proxy_hide_header Access-Control-Allow-Origin;
+        add_header Access-Control-Allow-Origin * always;
+    }
+
+    # Templated pages
+    location / {
+        proxy_pass http://127.0.0.1:8080;
+        proxy_read_timeout 30s;
+        add_header Cache-Control "no-cache, must-revalidate" always;
     }
 }
 ```
 
-One further change you might want to make to the file above is the `add_header Access-Control-Allow-Origin` statement. This is what's used on
-my own Spothole server to make sure that other third-party web-based software can get the data from my instance, and applies to any endpoint underneath `/api`. If you want
-*your* Spothole instance to be set up the same way, so that others can write software in JavaScript that can access it,
-leave this intact. But if you want your Spothole instance to only be usable by scripts running on the web server you write,
-you can remove this line. (Note that this doesn't stop other people writing *non-web-based* software that accesses your
-Spothole API—the enforcement of cross-origin headers only happens within the user's browser. If you need to lock your
-instance down so that no-one else can access it with *any* software, that's an aspect of nginx or firewall config that you will need
-to find help with elsewhere.)
+One further change you might want to make to the file above is the `add_header Access-Control-Allow-Origin` statements.
+These are what's used on my own Spothole server to make sure that other third-party web-based software can get the data
+from my instance, and applies to any endpoint underneath `/api`. If you want *your* Spothole instance to be set up the
+same way, so that others can write software in JavaScript that can access it, leave this intact. But if you want your
+Spothole instance to only be usable by scripts running on the web server you write, you can remove these lines. (Note
+that this doesn't stop other people writing *non-web-based* software that accesses your Spothole API—the
+enforcement of cross-origin headers only happens within the user's browser. If you need to lock your instance down so
+that no-one else can access it with *any* software, that's an aspect of nginx or firewall config that you will need to
+find help with elsewhere.)
 
 Now, make a symbolic link to enable the site:
 
@@ -284,17 +412,22 @@ cd /etc/nginx/sites-enabled
 sudo ln -sf ../sites-available/spothole
 ```
 
-Test that your nginx config isn't broken using `nginx -t`. If it works, restart nginx with `sudo systemctl restart nginx`.
+Test that your nginx config isn't broken using `nginx -t`. If it works, restart nginx with
+`sudo systemctl restart nginx`.
 
-If you haven't already done so, set up a DNS entry to make sure requests for your domain name end up at the server that's running Spothole.
+If you haven't already done so, set up a DNS entry to make sure requests for your domain name end up at the server
+that's running Spothole.
 
 You should now be able to access the web interface by going to the domain from your browser.
 
-Once that's working, [install certbot](https://certbot.eff.org/instructions?ws=nginx&os=snap) onto your server. Run it as root, and when prompted pick your domain name from the list. After a few seconds, it should successfully provision a certificate and modify your nginx config files automatically. You should then be able to access the site via HTTPS.
+Once that's working, [install certbot](https://certbot.eff.org/instructions?ws=nginx&os=snap) onto your server. Run it
+as root, and when prompted pick your domain name from the list. After a few seconds, it should successfully provision a
+certificate and modify your nginx config files automatically. You should then be able to access the site via HTTPS.
 
 ## Modifying the source code
 
-Spothole is Public Domain licenced, so you can grab the source code and start modifying it for your own needs. Contributions of code back to the main repository are encouraged, but completely optional.
+Spothole is Public Domain licenced, so you can grab the source code and start modifying it for your own needs.
+Contributions of code back to the main repository are encouraged, but completely optional.
 
 ### Code structure
 
@@ -302,12 +435,17 @@ To navigate your way around the source code, this list may help.
 
 *Python back-end code*
 
-* `/core` - Core classes and scripts
+* `/core` - Core classes and utilities
 * `/data` - Data storage classes
-* `/spotproviders` - Classes providing spots by accessing the APIs of other services
-* `/alertproviders` - Classes providing alerts by accessing the APIs of other services
-* `/solarconditionsproviders` - Classes providing solar and propagation by accessing the APIs of other services
+* `/providers/spot` - Classes providing spots by accessing the APIs of other services
+* `/providers/alert` - Classes providing alerts by accessing the APIs of other services
+* `/providers/solarconditions` - Classes providing solar and propagation by accessing the APIs of other services
+* `/providers/staticdata` - Classes providing static lookup data by accessing bundled data files or the APIs of other
+  services
+* `/providers/sigrefdata` - Classes providing SIG reference lookup data by accessing bundled data files or the APIs of
+  other services
 * `/server` - Classes for running Spothole's own web server
+* `spothole.py` - Main application script
 
 *Templates*
 
@@ -315,52 +453,82 @@ To navigate your way around the source code, this list may help.
 
 *HTML/JS/CSS front-end code*
 
-* `/webassets` - Root for static files served by the web server
-* `/webassets/apidocs` - Contains the OpenAPI spec (`openapi.yml`)
-* `/webassets/css` - CSS files used by the web front-end
-* `/webassets/fa` - a copy of the FontAwesome library
-* `/webassets/img` - image files used by the web front-end
-* `/webassets/js` - JavaScript used by the web front-end
+* `/static` - Root for static files served by the web server. These are all served from a path starting `/static/`.
+* `/static/apidocs` - Contains the OpenAPI spec (`openapi.yml`)
+* `/static/audio` - Audio files used by the web front-end
+* `/static/css` - CSS files used by the web front-end
+* `/static/img` - image files used by the web front-end
+* `/static/js` - JavaScript used by the web front-end
+* `/static/vendor` - Third-party libraries (CSS, JS, fonts and images)
 
 *Miscellaneous*
 
-* `/` - Main script (`spothole.py`), pip `requirements.txt`, config, README, etc.
+* `/` - pip `requirements.txt`, config, README, etc.
 * `/images` - Image sources
-* `/datafiles` - Local data sources (differentiated from the majority of data files which are loaded from URLs and cached in `/cache`)
-* `/cache` - Directory where static-ish data downloaded from the internet is cached to avoid rapid re-requests, and where spot/alert data is cached so that it survives a software restart. Created on first run.
+* `/datafiles` - Local data files, used by some providers when the data will never change and/or is not easily available
+  online in a format Spothole can handle
+* `/cache` - Directory where Spothole stores all the data it uses that should be persisted to disk. Created on first
+ run.
 
 ### Extending the server
 
-Spothole is designed to be easily extensible. If you want to write your own spot provider, for example, simply add a module to the `spotproviders` package containing your class. (Currently, in order to be loaded correctly, the module (file) name should be the same as the class name, but lower case.)
+Spothole is designed to be easily extensible. If you want to write your own spot provider, for example, simply add a
+module to the `providers.spot` package containing your class. (Currently, in order to be loaded correctly, the module
+(file) name should be the same as the class name, but lower case.)
 
-Your class should extend "SpotProvider"; if it operates by polling an HTTP Server on a timer, it can instead extend "HTTPSpotProvider" where some of the work is done for you.
+Your class should extend "SpotProvider"; if it operates by polling an HTTP Server on a timer, it can instead extend "
+HTTPSpotProvider" where some of the work is done for you.
 
-The class will need to implement a constructor that takes in the `provider_config` and provides it to the superclass constructor, while also taking any other config parameters it needs.
+The class will need to implement a constructor that takes in the `provider_config` and provides it to the superclass
+constructor, while also taking any other config parameters it needs.
 
-If you're extending the base `SpotProvider` class, you will need to implement `start()` and `stop()` methods that start and stop a separate thread which handles the provider's processing needs. The thread should call `submit()` or `submit_batch()` when it has one or more spots to report.
+If you're extending the base `SpotProvider` class, you will need to implement `start()` and `stop()` methods that start
+and stop a separate thread which handles the provider's processing needs. The thread should call `submit()` or
+`submit_batch()` when it has one or more spots to report.
 
-If you're extending the `HTTPSpotProvider` class, you will need to provide a URI to query and an interval to the superclass constructor. You'll then need to implement the `http_response_to_spots()` method which is called when new data is retrieved. Your implementation should then call `submit()` or `submit_batch()` when it has one or more spots to report.
+If you're extending the `HTTPSpotProvider` class, you will need to provide a URI to query and an interval to the
+superclass constructor. You'll then need to implement the `http_response_to_spots()` method which is called when new
+data is retrieved. Your implementation should then call `submit()` or `submit_batch()` when it has one or more spots to
+report.
 
-When constructing spots, use the comments in the Spot class and the existing implementations as an example. All parameters are optional, but you will at least want to provide a `time` (which must be timezone-aware) and a `dx_call`.
+When constructing spots, use the comments in the Spot class and the existing implementations as an example. All
+parameters are optional, but you will at least want to provide a `time` (which must be timezone-aware) and a `dx_call`.
 
-Finally, simply add the appropriate config to the `spot_providers` section of `config.yml`, and your provider should be instantiated on startup.
+Finally, simply add the appropriate config to the `spot_providers` section of `config.yml`, and your provider should be
+instantiated on startup.
 
-The same approach as above is also used for alert providers.
+The same approach as above is also used for alerts, and other types of providers. Give me a shout if you need any
+advice.
 
 ## Thanks
 
 As well as being my work, I have also gratefully received feature patches from Steven, M1SDH.
 
-The project contains GeoJSON files for CQ and ITU zones, in the `/datafiles/` directory. These are MIT-licenced and, to my knowledge, created by HA8TKS for his CQ and ITU zone layers for Leaflet.
+The project contains GeoJSON files for CQ and ITU zones, in the `/datafiles/` directory. These are MIT-licenced and, to
+my knowledge, created by HA8TKS for his CQ and ITU zone layers for Leaflet. `/datafiles` also contains a
+`MUNICIPIOS.csv` file, from the "Nomenclátor Geográfico de Municipios y Entidades de Población" data set sourced from
+[el Centro Nacional de Información Geográfica](https://centrodedescargas.cnig.es/CentroDescargas/home).
+`didbase-stations.csv` and the TOTA file were created by me based on publicly available data from GIRO and from
+maps of conference centres.
 
-The project contains a self-hosted copy of Font Awesome's free library, in the `/webassets/fa/` directory. This is subject to Font Awesome's licence and is not covered by the overall licence declared in the `LICENSE` file. This approach was taken in preference to using their hosted kits due to the popularity of this project exceeding the page view limit for their free hosted offering.
+The project contains a set of flag icons generated using the "Noto Color Emoji" font on a Debian system, in the
+`/static/img/flags/` directory.
 
-The project contains a set of flag icons generated using the "Noto Color Emoji" font on a Debian system, in the `/webassets/img/flags/` directory.
+The software uses a number of Python libraries as listed in `requirements.txt`, and a number of JavaScript libraries.
+This project would not have been possible without these libraries, so many thanks to their developers.
 
-The software uses a number of Python libraries as listed in `requirements.txt`, and a number of JavaScript libraries such as jQuery, Leaflet and Bootstrap. This project would not have been possible without these libraries, so many thanks to their developers.
+### Third Party Libraries
 
-Particular thanks go to country-files.com for providing country lookup data for amateur radio, to K0SWE for [this JSON-formatted DXCC data](https://github.com/k0swe/dxcc-json/), and to the developers of `pyhamtools` for making it easy to use country-files.com data as well as QRZ.com and Clublog lookup.
+A number of third-party libraries are self-hosted in the `/static/vendor/` directory. These files are subject to their
+own licences and are not covered by the overall licence declared in the `LICENSE` file.
 
-Amateur radio clusters, outdoor programmes, propagation data providers etc. are almost all volunteer-run services that make no or little profit, and are done for the love of amateur radio. Services like Spothole, which build on top of them, are truly standing on the shoulders of giants. None of this would have been possible without the hard work and dedication of many other people within the amaetur radio community.
+Particular thanks go to country-files.com for providing country lookup data for amateur radio, to K0SWE
+for [this JSON-formatted DXCC data](https://github.com/k0swe/dxcc-json/), and to the developers of `pyhamtools` for
+making it easy to use country-files.com data as well as QRZ.com and Clublog lookup.
+
+Amateur radio clusters, outdoor programmes, propagation data providers etc. are almost all volunteer-run services that
+make no or little profit, and are done for the love of amateur radio. Services like Spothole, which build on top of
+them, are truly standing on the shoulders of giants. None of this would have been possible without the hard work and
+dedication of many other people within the amaetur radio community.
 
 The project's name was suggested by Harm, DK4HAA. Thanks!

config-example.yml

@@ -6,70 +6,68 @@
 # this as "N0CALL" and it shouldn't do any harm, as we're not sending anything to the various networks, only receiving.
 server-owner-callsign: "N0CALL"
 
+# Port to open the local web server on
+web-server-port: 8080
+
+# Run in API-only mode? When enabled, the web UI is not served, only the API endpoints and the OpenAPI documentation
+# page. If you are running your own Spothole instance purely to serve client software, and not wanting visitors to
+# discover a full web-based cluster UI, enable this flag.
+api-only-mode: false
+
 # The base URL at which the software runs.
 base-url: "http://localhost:8080"
 
 # Spot providers to use. This is an example set, tailor it to your liking by commenting and uncommenting.
 # RBN and APRS-IS are supported but have such a high data rate, you probably don't want them enabled.
-# Each provider needs a class, a name, and an enabled/disabled state. Some require more config such as hostnames/IP
+# Each provider needs a class and an enabled/disabled state. Some require more config such as hostnames/IP
 # addresses and ports. You can duplicate them if you like, e.g. to support several DX clusters. RBN uses two ports, 7000
 # for CW/RTTY and 7001 for FT8, so if you want both, you need two entries, as shown below.
 # Feel free to write your own provider classes! There are details in the README.
 spot-providers:
-  -
-    class: "POTA"
-    name: "POTA"
+  - class: "POTA"
     enabled: true
-  -
-    class: "SOTA"
-    name: "SOTA"
+
+  - class: "SOTA"
     enabled: true
-  -
-    class: "WWFF"
-    name: "WWFF"
+
+  - class: "WWFF"
     enabled: true
-  -
-    class: "WWBOTA"
-    name: "WWBOTA"
+
+  - class: "WWBOTA"
     enabled: true
-  -
-    class: "GMA"
-    name: "GMA"
+
+  - class: "GMA"
     enabled: true
-  -
-    class: "HEMA"
-    name: "HEMA"
+    # GMA requires an API key to fetch spots. After creating an account on cqgma.org, email support and request one.
+    api-key: ""
+
+  - class: "HEMA"
     enabled: true
-  -
-    class: "ParksNPeaks"
-    name: "ParksNPeaks"
+
+  - class: "ParksNPeaks"
     enabled: true
-  -
-    class: "ZLOTA"
-    name: "ZLOTA"
+
+  - class: "ZLOTA"
     enabled: true
-  -
-    class: "WOTA"
-    name: "WOTA"
+
+  - class: "WOTA"
     enabled: true
-  -
-    class: "LLOTA"
-    name: "LLOTA"
+
+  - class: "LLOTA"
     enabled: true
-  -
-    class: "WWTOTA"
-    name: "WWTOTA"
+
+  - class: "Towers"
     enabled: true
-  -
-    class: "Tiles"
-    name: "Tiles"
+
+  - class: "Tiles"
     enabled: true
-  -
-    class: "APRSIS"
-    name: "APRS-IS"
+
+  - class: "APRSIS"
     enabled: false
-  -
-    class: "DXCluster"
+
+  - class: "DXCluster"
+    # Clusters have a "name" property in case you want to say which cluster node the data is from, or connect to
+    # several clusters and name them separately.
     name: "HRD Cluster"
     enabled: true
     host: "hrd.wa9pie.net"
@@ -84,24 +82,18 @@ spot-providers:
     # sure you aren't also separately connecting to RBN directly, otherwise you may get duplicate spots.) Note that not
     # all clusters sent RBN spots anyway.
     allow_rbn_spots: false
-  -
-    class: "DXCluster"
+
+  - class: "DXCluster"
     name: "W3LPL Cluster"
     enabled: false
     host: "w3lpl.net"
     port: 7373
-    # Prompt the cluster node gives when asking for a callsign to log in. Varies between cluster node software.
     login_prompt: "Please enter your call:"
-    # Callsign Spothole will use to log into this cluster. Ensure the SSID (e.g. -99) is different to any personal
-    # connection you might make to this cluster node.
     login_callsign: "N0CALL-99"
-    # Whether to allow RBN spots that come via this cluster. If you don't want RBN spots or you are making a separate
-    # connection to RBN directly, leave this as False. If you want RBN spots from this cluster, set this to True. (Make
-    # sure you aren't also separately connecting to RBN directly, otherwise you may get duplicate spots.) Note that not
-    # all clusters sent RBN spots anyway.
     allow_rbn_spots: false
-  -
-    class: "RBN"
+
+  - class: "RBN"
+  - # RBN sources have a "name" property in case you want to differentiate between the CW/RTTY and FT8 sources.
     name: "RBN CW/RTTY"
     enabled: false
     port: 7000
@@ -110,99 +102,211 @@ spot-providers:
     # received by Spothole but not shown on the web UI unless the user explicitly turns it on. For that behaviour,
     # set enabled to true, but enabled-by-default-in-web-ui to false.
     enabled-by-default-in-web-ui: false
-  -
-    class: "RBN"
+
+  - class: "RBN"
     name: "RBN FT8"
     enabled: false
     port: 7001
     enabled-by-default-in-web-ui: false
-  -
-    class: "UKPacketNet"
-    name: "UK Packet Radio Net"
+
+  - class: "UKPacketNet"
     enabled: false
     enabled-by-default-in-web-ui: false
-  -
-    class: "XOTA"
-    name: "39C3 TOTA"
+
+  - class: "XOTA"
+    # xOTA sources have a "name" property so you can define what programme it is for, since xOTA is generic software
+    # for running "something-on-the-air" programmes.
+    name: "C3 TOTA"
     enabled: false
     url: "wss://39c3.totawatch.de/api/spot/live"
-    # Fixed SIG for all spots from a provider & location CSV are currently only a feature for the "XOTA" provider,
-    # the software found at https://github.com/nischu/xOTA/. This is because this is a generic backend for xOTA
-    # programmes and so different URLs provide different programmes.
-    sig: "TOTA"
-    locations-csv: "datafiles/39c3-tota.csv"
-  -
-    class: "XOTA"
+    # For the "XOTA" provider, a SIG must be set menually here because xOTA is a generic backend for xOTA
+    # programmes and so different URLs potentially provide different programmes.
+    sig: "Toilets"
+    # For Toilets on the Air, we prefix the SIG references (T-01 etc) with some characters that define the conference:
+    # C3, EH or HOPE - so we can look up the correct locations in our database, because each conference starts from T-01
+    # but refers to a toilet in a different building (or continent!)
+    sig-ref-prefix: "C3"
+
+  - class: "XOTA"
     name: "EH23 TOTA"
-    enabled: true
+    enabled: false
     url: "wss://eh23.totawatch.de/api/spot/live"
-    sig: "TOTA"
-    locations-csv: "datafiles/eh23-tota.csv"
+    sig: "Toilets"
+    sig-ref-prefix: "EH"
+
+  - class: "XOTA"
+    name: "HOPE26 TOTA"
+    enabled: false
+    url: "wss://hope-26.totawatch.de/api/spot/live"
+    sig: "Toilets"
+    sig-ref-prefix: "HOPE"
 
 
 # Alert providers to use. Same setup as the spot providers list above.
 alert-providers:
-  -
-    class: "POTA"
-    name: "POTA"
+  - class: "POTA"
     enabled: true
-  -
-    class: "SOTA"
-    name: "SOTA"
+
+  - class: "SOTA"
     enabled: true
-  -
-    class: "WWFF"
-    name: "WWFF"
+
+  - class: "WWFF"
     enabled: true
-  -
-    class: "ParksNPeaks"
-    name: "ParksNPeaks"
+
+  - class: "ParksNPeaks"
     enabled: true
-  -
-    class: "WOTA"
-    name: "WOTA"
+
+  - class: "WOTA"
     enabled: true
-  -
-    class: "BOTA"
-    name: "BOTA"
+
+  - class: "BOTA"
     enabled: true
-  -
-    class: "NG3K"
-    name: "NG3K"
+
+  - class: "NG3K"
     enabled: true
 
 
 # Solar condition providers to use. These poll external APIs for solar propagation data (SFI, A/K indices, band
 # conditions, etc.) and make it available via the /api/v1/solar endpoint.
 solar-condition-providers:
-  -
-    class: "HamQSL"
-    name: "HamQSL"
-    enabled: true
-  -
-    class: "NOAA3dayForecast"
-    name: "NOAA 3-day Forecast"
-    enabled: true
-  -
-    class: "GIROIonosonde"
-    name: "GIRO Ionosonde Data"
+  - class: "HamQSL"
+    enabled: true
+
+  - class: "NOAA3dayForecast"
+    enabled: true
+
+  - class: "GIROIonosonde"
+    enabled: true
+
+  - class: "KC2GProp"
+    enabled: true
+
+
+# Static reference data providers to use. These allow Spothole to download data such as mapping between callsign
+# prefixes and DXCC entities.
+static-data-providers:
+  - class: "K0SWE"
+    enabled: true
+
+  - class: "CQZoneData"
+    enabled: true
+
+  - class: "ITUZoneData"
+    enabled: true
+
+
+# SIG reference data providers to use. These allow Spothole to download, for example, the WWFF directory that maps WWFF
+# park IDs to their name and location.
+sig-ref-data-providers:
+  - class: "POTA"
+    enabled: true
+
+  - class: "SOTA"
+    enabled: true
+
+  - class: "WWFF"
+    enabled: true
+
+  - class: "WWBOTA"
+    enabled: true
+
+  - class: "GMA"
+    enabled: true
+
+  - class: "MOTA"
+    enabled: true
+
+  - class: "ILLW"
+    enabled: true
+
+  - class: "ARLHS"
+    enabled: true
+
+  - class: "WCA"
+    enabled: true
+
+  - class: "IOTA"
+    enabled: true
+
+  - class: "WOTA"
+    enabled: true
+
+  - class: "SIOTA"
+    enabled: true
+
+  - class: "ZLOTA"
+    enabled: true
+
+  - class: "LLOTA"
+    enabled: true
+
+  - class: "Towers"
+    enabled: true
+
+  - class: "DME"
+    enabled: true
+
+  - class: "Toilets"
+    enabled: true
+
+
+# Callsign data providers to use. These allow Spothole to provide information about callsigns, either from static
+# lookups or from web-based services such as QRZ. Note that use of QRZ and HamQTH is *on behalf of the user*, and
+# requires the user to enter their credentials into the website UI or provide them in an API call. Spothole will not
+# look up all calls via QRZ/HamQTH using the server owner's credentials, as this is against their usage policy.
+callsign-data-providers:
+  - class: "QRZ"
+    enabled: true
+    # Callsign data providers can often provide conflicting data, e.g. a home location from QRZ vs the centre of the
+    # DXCC entity from a country file lookup. The priority flag sets which source "wins" in the event of conflict.
+    # Lower numbers take priority over higher numbers. Generally then, QRZ/HamQTH should have low numbers as they are
+    # likely to have the most accurate data.
+    priority: 1
+    # No server-side credentials for QRZ. Users must provide their own as per QRZ policy.
+
+  - class: "HamQTH"
+    enabled: true
+    priority: 2
+    # No server-side credentials for HamQTH. Users must provide their own.
+
+  - class: "ClublogAPI"
+    # Querying the Clublog API directly doesn't provide any more data than the XML version, it just provides slightly
+    # more up-to-date information in the rare case that the prefix data changes, at a significant cost of looking up
+    # every callsign via an API call. Normally left disabled but it exists as an option.
+    enabled: false
+    priority: 3
+    # API key for Clublog to look up information. Required in order to enable this provider. Unlike QRZ and HamQTH,
+    # Clublog uses an API key issued to Spothole, not to the end user.
+    api-key: ""
+
+  - class: "ClublogXML"
+    enabled: true
+    priority: 4
+    # API key for Clublog to look up information. Required in order to enable this provider. You will need to request
+    # one via their helpdesk portal if you want to use callsign lookups from Clublog.
+    api-key: ""
+
+  - class: "CountryFiles"
+    priority: 5
     enabled: true
 
-# Port to open the local web server on
-web-server-port: 8080
 
 # Maximum time to keep spots and alerts in the system before deleting them. By default, one hour for spots and one week
 # for alerts.
 max-spot-age-sec: 3600
 max-alert-age-sec: 604800
 
-# API key for Clublog to look up information. Optional. You sill need to request one via their helpdesk portal if you
-# want to use callsign lookups from Clublog.
-clublog-api-key: ""
-
 # Allow submitting spots to the Spothole API?
 allow-spotting: true
 
+# Log web requests? Useful to see what your users are requesting in terms of pages and API endpoints, but will fill up
+# your log quickly on a popular server.
+log-web-requests: false
+
+# Minimum severity of log statements to print to the system log. Defaults to INFO, change to DEBUG if you need more
+# information.
+log-level: INFO
+
 # Options for the web UI.
 web-ui-options:
   spot-count: [ 10, 25, 50, 100 ]

core/cache_utils.py

@@ -1,10 +0,0 @@
-from datetime import timedelta
-
-from requests_cache import CachedSession
-
-# Cache for "semi-static" data such as the locations of parks, CSVs of reference lists, etc.
-# This has an expiry time of 30 days, so will re-request from the source after that amount
-# of time has passed. This is used throughout Spothole to cache data that does not change
-# rapidly.
-SEMI_STATIC_URL_DATA_CACHE = CachedSession("cache/semi_static_url_data_cache",
-                                           expire_after=timedelta(days=30))

core/call_lookup_helper.py

@@ -0,0 +1,30 @@
+from core.data_providers import DATA_PROVIDERS
+from data.callsign import Callsign
+
+
+def get_call_info(callsign, lookup_credentials):
+    """Utility method to get the best set of data for a callsign as we can, using all enabled providers.
+    lookup_credentials is an optional object that carries the user's QRZ.com/HamQTH credentials, if they provided them,
+    to enable lookup using those providers."""
+
+    callsign_data = Callsign(call=callsign)
+
+    if callsign:
+        # Sort callsign providers by priority order, so we query the highest priority (lowest numbers) first, and only
+        # query other providers for data we are missing as we go along.
+        for p in sorted(DATA_PROVIDERS.callsign_data_providers, key=lambda p: p.priority):
+            if p.enabled:
+                # Get new lookup data
+                data = p.lookup(callsign, lookup_credentials)
+                if data:
+                    # If we have new data for fields that were previously unpopulated, add them in
+                    for key, value in data.__dict__.items():
+                        if value is not None and callsign_data.__dict__.get(key) is None:
+                            callsign_data.__dict__[key] = value
+
+                    # If callsign data is fully populated, avoid looking up using other providers as their data will not
+                    # be used
+                    if callsign_data.fully_populated():
+                        break
+
+    return callsign_data

core/cleanup.py

@@ -1,73 +0,0 @@
-import logging
-from datetime import datetime
-from threading import Event, Thread
-
-import pytz
-
-
-class CleanupTimer:
-    """Provides a timed cleanup of the spot list."""
-
-    def __init__(self, spots, alerts, web_server, cleanup_interval):
-        """Constructor"""
-
-        self._spots = spots
-        self._alerts = alerts
-        self._web_server = web_server
-        self._cleanup_interval = cleanup_interval
-        self.last_cleanup_time = datetime.min.replace(tzinfo=pytz.UTC)
-        self.status = "Starting"
-        self._thread = None
-        self._stop_event = Event()
-
-    def start(self):
-        """Start the cleanup timer"""
-
-        self._thread = Thread(target=self._run, daemon=True)
-        self._thread.start()
-
-    def stop(self):
-        """Stop any threads and prepare for application shutdown"""
-
-        self._stop_event.set()
-
-    def _run(self):
-        while not self._stop_event.wait(timeout=self._cleanup_interval):
-            self._cleanup()
-
-    def _cleanup(self):
-        """Perform cleanup and reschedule next timer"""
-
-        try:
-            # Perform cleanup via letting the data expire
-            self._spots.expire()
-            self._alerts.expire()
-
-            # Explicitly clean up any spots and alerts that have expired
-            for i in list(self._spots.iterkeys()):
-                try:
-                    spot = self._spots[i]
-                    if spot.expired():
-                        self._spots.delete(i)
-                except KeyError:
-                    # Must have already been deleted, OK with that
-                    pass
-            for i in list(self._alerts.iterkeys()):
-                try:
-                    alert = self._alerts[i]
-                    if alert.expired():
-                        self._alerts.delete(i)
-                except KeyError:
-                    # Must have already been deleted, OK with that
-                    pass
-
-            # Clean up web server SSE spot/alert queues
-            self._web_server.clean_up_sse_queues()
-
-            self.status = "OK"
-            self.last_cleanup_time = datetime.now(pytz.UTC)
-
-        except Exception:
-            self.status = "Error"
-            logging.exception("Exception in Cleanup thread")
-            self._stop_event.wait(timeout=1)

core/config.py

@@ -1,3 +1,4 @@
+import importlib
 import logging
 import os
 
@@ -21,12 +22,27 @@ SERVER_OWNER_CALLSIGN = config["server-owner-callsign"]
 WEB_SERVER_PORT = config["web-server-port"]
 ALLOW_SPOTTING = config["allow-spotting"]
 WEB_UI_OPTIONS = config["web-ui-options"]
+API_ONLY_MODE = config.get("api-only-mode", False)
+LOG_LEVEL = config.get("log-level", "INFO")
+LOG_WEB_REQUESTS = config.get("log-web-requests", False)
 
 # For ease of config, each spot provider owns its own config about whether it should be enabled by default in the web UI
 # but for consistency we provide this to the front-end in web-ui-options because it has no impact outside of the web UI.
 WEB_UI_OPTIONS["spot-providers-enabled-by-default"] = [p["name"] for p in config["spot-providers"] if p["enabled"] and (
         "enabled-by-default-in-web-ui" not in p or p["enabled-by-default-in-web-ui"])]
+WEB_UI_OPTIONS["qrz-enabled"] = any(p["class"] == "QRZ" and p["enabled"] for p in config["callsign-data-providers"])
+WEB_UI_OPTIONS["hamqth-enabled"] = any(p["class"] == "HamQTH" and p["enabled"] for p in config["callsign-data-providers"])
 # If spotting to this server is enabled, "API" is another valid spot source even though it does not come from
 # one of our proviers. We set that to also be enabled by default.
 if ALLOW_SPOTTING:
     WEB_UI_OPTIONS["spot-providers-enabled-by-default"].append("API")
+
+
+def create_provider_from_config(package, config_providers_entry):
+    """Utility method to get a provider based on the class specified in its config entry. You must also provide the
+    package to look for it in, as there are several types of provider. e.g. package "providers.spot", where the config
+    entry is for a POTA spot provider."""
+
+    module = importlib.import_module(package + "." + config_providers_entry["class"].lower())
+    provider_class = getattr(module, config_providers_entry["class"])
+    return provider_class(config_providers_entry)

core/constants.py

@@ -3,37 +3,38 @@ from data.band import Band
 from data.sig import SIG
 
 # General software
-SOFTWARE_NAME = "Spothole by M0TRT"
-SOFTWARE_VERSION = "1.3-pre"
+SOFTWARE_VERSION = "2.0-pre"
 
 # HTTP headers used for spot providers that use HTTP
-HTTP_HEADERS = {"User-Agent": SOFTWARE_NAME + ", v" + SOFTWARE_VERSION + " (operated by " + SERVER_OWNER_CALLSIGN + ")"}
-HAMQTH_PRG = (SOFTWARE_NAME + " v" + SOFTWARE_VERSION + " operated by " + SERVER_OWNER_CALLSIGN).replace(" ", "_")
+HTTP_HEADERS = {"User-Agent": "Spothole v" + SOFTWARE_VERSION + " (operated by " + SERVER_OWNER_CALLSIGN + ")"}
+HAMQTH_PRG = ("Spothole v" + SOFTWARE_VERSION + " operated by " + SERVER_OWNER_CALLSIGN).replace(" ", "_")
 
 # Special Interest Groups
 SIGS = [
-    SIG(name="POTA", description="Parks on the Air", ref_regex=r"[A-Z]{2}\-\d{4,5}"),
-    SIG(name="SOTA", description="Summits on the Air", ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{2}\-\d{3}"),
-    SIG(name="WWFF", description="World Wide Flora & Fauna", ref_regex=r"[A-Z0-9]{1,3}FF\-\d{4}"),
-    SIG(name="GMA", description="Global Mountain Activity", ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{2}\-\d{3}"),
-    SIG(name="WWBOTA", description="Worldwide Bunkers on the Air", ref_regex=r"B\/[A-Z0-9]{1,3}\-\d{3,4}"),
-    SIG(name="HEMA", description="HuMPs Excluding Marilyns Award", ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{3}\-\d{3}"),
-    SIG(name="IOTA", description="Islands on the Air", ref_regex=r"[A-Z]{2}\-\d{3}"),
-    SIG(name="MOTA", description="Mills on the Air", ref_regex=r"X\d{4,6}"),
-    SIG(name="ARLHS", description="Amateur Radio Lighthouse Society", ref_regex=r"[A-Z]{3}\-\d{3,4}"),
-    SIG(name="ILLW", description="International Lighthouse & Lightship Weekend", ref_regex=r"[A-Z]{2}\d{4}"),
-    SIG(name="SIOTA", description="Silos on the Air", ref_regex=r"[A-Z]{2}\-[A-Z]{3}\d"),
-    SIG(name="WCA", description="World Castles Award", ref_regex=r"[A-Z0-9]{1,3}\-\d{5}"),
-    SIG(name="ZLOTA", description="New Zealand on the Air", ref_regex=r"ZL[A-Z]/[A-Z]{2}\-\d{3,4}"),
-    SIG(name="WOTA", description="Wainwrights on the Air", ref_regex=r"[A-Z]{3}-[0-9]{2}"),
-    SIG(name="BOTA", description="Beaches on the Air"),
-    SIG(name="KRMNPA", description="Keith Roget Memorial National Parks Award"),
-    SIG(name="LLOTA", description="Lagos y Lagunas on the Air", ref_regex=r"[A-Z]{2}\-\d{4}"),
-    SIG(name="WWTOTA", description="Towers on the Air", ref_regex=r"[A-Z]{2}R\-\d{4}"),
-    SIG(name="Tiles", description="Tiles on the Air", ref_regex=r"[A-Za-z]{2}[0-9]{2}[A-Za-z]{2}"),
-    SIG(name="WAB", description="Worked All Britain", ref_regex=r"[A-Z]{1,2}[0-9]{2}"),
-    SIG(name="WAI", description="Worked All Ireland", ref_regex=r"[A-Z][0-9]{2}"),
-    SIG(name="TOTA", description="Toilets on the Air", ref_regex=r"T\-[0-9]{2}")
+    SIG(name="POTA", comment_names=["POTA"], description="Parks on the Air", ref_regex=r"[A-Z]{2}\-\d{4,5}"),
+    SIG(name="SOTA", comment_names=["SOTA"], description="Summits on the Air", ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{2}\-\d{3}"),
+    SIG(name="WWFF", comment_names=["WWFF"], description="World Wide Flora & Fauna", ref_regex=r"[A-Z0-9]{1,3}FF\-\d{4}"),
+    SIG(name="GMA", comment_names=["GMA"], description="Global Mountain Activity", ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{2}\-\d{3}"),
+    SIG(name="WWBOTA", comment_names=["WWBOTA", "BOTA"], description="Worldwide Bunkers on the Air", ref_regex=r"B\/[A-Z0-9]{1,3}\-\d{3,4}"),
+    SIG(name="HEMA", comment_names=["HEMA"], description="HuMPs Excluding Marilyns Award", ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{3}\-\d{3}"),
+    SIG(name="IOTA", comment_names=["IOTA"], description="Islands on the Air", ref_regex=r"[A-Z]{2}\-\d{3}"),
+    SIG(name="MOTA", comment_names=["MOTA"], description="Mills on the Air", ref_regex=r"X\d{4,6}"),
+    SIG(name="ARLHS", comment_names=["ARLHS"], description="Amateur Radio Lighthouse Society", ref_regex=r"[A-Z]{3}[\- ]\d{3,4}"),
+    SIG(name="ILLW", comment_names=["ILLW"], description="International Lighthouse & Lightship Weekend", ref_regex=r"[A-Z]{2}\d{4}"),
+    SIG(name="SIOTA", comment_names=["SIOTA"], description="Silos on the Air", ref_regex=r"[A-Z]{2}\-[A-Z]{3}\d"),
+    SIG(name="WCA", comment_names=["WCA"], description="World Castles Award", ref_regex=r"[A-Z0-9]{1,3}\-\d{5}"),
+    SIG(name="ZLOTA", comment_names=["ZLOTA"], description="New Zealand on the Air", ref_regex=r"ZL[A-Z]/[A-Z]{2}\-\d{3,4}"),
+    SIG(name="WOTA", comment_names=["WOTA"], description="Wainwrights on the Air", ref_regex=r"[A-Z]{3}-[0-9]{2}"),
+    SIG(name="BOTA", comment_names=[], description="Beaches on the Air"),
+    SIG(name="KRMNPA", comment_names=["KRMNPA"], description="Keith Roget Memorial National Parks Award", ref_regex=r"VKFF\-\d{4}"),
+    SIG(name="SANPCPA", comment_names=["SANPCPA"], description="South Australian National Parks and Conservation Parks Award", ref_regex=r"VKFF\-\d{4}"),
+    SIG(name="LLOTA", comment_names=["LLOTA"], description="Lagos y Lagunas on the Air", ref_regex=r"LL[A-Z]{2}\-\d{4}"),
+    SIG(name="Towers", comment_names=["TOTA"], description="Towers on the Air", ref_regex=r"[A-Z]{2,3}R\-\d{4}"),
+    SIG(name="Tiles", comment_names=[], description="Tiles on the Air", ref_regex=r"[A-Za-z]{2}[0-9]{2}[A-Za-z]{2}"),
+    SIG(name="WAB", comment_names=["WAB"], description="Worked All Britain", ref_regex=r"[A-Z]{1,2}[0-9]{2}"),
+    SIG(name="WAI", comment_names=["WAI"], description="Worked All Ireland", ref_regex=r"[A-Z][0-9]{2}"),
+    SIG(name="DME", comment_names=["DME"], description="Diplomas de Municipios Españoles", ref_regex=r"\d{4,5}"),
+    SIG(name="Toilets", comment_names=[], description="Toilets on the Air", ref_regex=r"T\-[0-9]{2}")
 ]
 
 # Modes. Note "DIGI" and "DIGITAL" are also supported but are normalised into "DATA".
@@ -62,17 +63,17 @@ MODE_ALIASES = {
 BANDS = [
     Band(name="2200m", start_freq=135700, end_freq=137800),
     Band(name="600m", start_freq=472000, end_freq=479000),
-    Band(name="160m", start_freq=1800000, end_freq=2000000),
-    Band(name="80m", start_freq=3500000, end_freq=4000000),
-    Band(name="60m", start_freq=5250000, end_freq=5410000),
-    Band(name="40m", start_freq=7000000, end_freq=7300000),
-    Band(name="30m", start_freq=10100000, end_freq=10150000),
-    Band(name="20m", start_freq=14000000, end_freq=14350000),
-    Band(name="17m", start_freq=18068000, end_freq=18168000),
-    Band(name="15m", start_freq=21000000, end_freq=21450000),
-    Band(name="12m", start_freq=24890000, end_freq=24990000),
+    Band(name="160m", start_freq=1800000, end_freq=2000000, is_ham_hf=True),
+    Band(name="80m", start_freq=3500000, end_freq=4000000, is_ham_hf=True),
+    Band(name="60m", start_freq=5250000, end_freq=5410000, is_ham_hf=True),
+    Band(name="40m", start_freq=7000000, end_freq=7300000, is_ham_hf=True),
+    Band(name="30m", start_freq=10100000, end_freq=10150000, is_ham_hf=True),
+    Band(name="20m", start_freq=14000000, end_freq=14350000, is_ham_hf=True),
+    Band(name="17m", start_freq=18068000, end_freq=18168000, is_ham_hf=True),
+    Band(name="15m", start_freq=21000000, end_freq=21450000, is_ham_hf=True),
+    Band(name="12m", start_freq=24890000, end_freq=24990000, is_ham_hf=True),
     Band(name="11m", start_freq=26965000, end_freq=27405000),
-    Band(name="10m", start_freq=28000000, end_freq=29700000),
+    Band(name="10m", start_freq=28000000, end_freq=29700000, is_ham_hf=True),
     Band(name="6m", start_freq=50000000, end_freq=54000000),
     Band(name="5m", start_freq=56000000, end_freq=60500000),
     Band(name="4m", start_freq=70000000, end_freq=70500000),
@@ -90,3 +91,18 @@ UNKNOWN_BAND = Band(name="Unknown", start_freq=0, end_freq=0)
 
 # Continents
 CONTINENTS = ["EU", "NA", "SA", "AS", "AF", "OC", "AN"]
+
+# Propagation modes used in VHF/UHF DX cluster comments, e.g. "JN61ES<ES>JM56XT". I don't think there's an official list
+# of these anywhere, but here are some I've seen or seen reference to
+PROPAGATION_MODES = {
+    "F2": "F2 layer ionospheric",
+    "ES": "Sporadic-E",
+    "TR": "Tropospheric ducting",
+    "TEP": "Trans-Equatorial Propagation",
+    "EME": "Earth-Moon-Earth",
+    "AU": "Aurora",
+    "MS": "Meteor scatter",
+    "RS": "Rain scatter",
+    "AS": "Aircraft scatter",
+    "SAT": "Satellite"
+}

core/data_providers.py

@@ -0,0 +1,73 @@
+from core.config import config, create_provider_from_config
+
+
+class DataProviders:
+    """Global object for storing data providers."""
+
+    def __init__(self):
+        self.spot_providers = []
+        self.alert_providers = []
+        self.solar_condition_providers = []
+        self.static_data_providers = []
+        self.sig_ref_data_providers = []
+        self.callsign_data_providers = []
+
+
+    def setup(self):
+        for entry in config["spot-providers"]:
+            self.spot_providers.append(create_provider_from_config("providers.spot", entry))
+        for entry in config["alert-providers"]:
+            self.alert_providers.append(create_provider_from_config("providers.alert", entry))
+        for entry in config.get("solar-condition-providers", []):
+            self.solar_condition_providers.append(create_provider_from_config("providers.solarconditions", entry))
+        for entry in config.get("static-data-providers", []):
+            self.static_data_providers.append(create_provider_from_config("providers.staticdata", entry))
+        for entry in config.get("sig-ref-data-providers", []):
+            self.sig_ref_data_providers.append(create_provider_from_config("providers.sigrefdata", entry))
+        for entry in config.get("callsign-data-providers", []):
+            self.callsign_data_providers.append(create_provider_from_config("providers.callsigndata", entry))
+
+
+    def start(self):
+        # Start data providers before spot/alert providers so the lookup data is there already for incoming spots
+        for p in self.static_data_providers:
+            if p.enabled:
+                p.start()
+        for p in self.sig_ref_data_providers:
+            if p.enabled:
+                p.start()
+        for p in self.callsign_data_providers:
+            if p.enabled:
+                p.start()
+        for p in self.spot_providers:
+            if p.enabled:
+                p.start()
+        for p in self.alert_providers:
+            if p.enabled:
+                p.start()
+        for p in self.solar_condition_providers:
+            if p.enabled:
+                p.start()
+
+    def stop(self):
+        for sp in self.spot_providers:
+            if sp.enabled:
+                sp.stop()
+        for ap in self.alert_providers:
+            if ap.enabled:
+                ap.stop()
+        for scp in self.solar_condition_providers:
+            if scp.enabled:
+                scp.stop()
+        for srdp in self.sig_ref_data_providers:
+            if srdp.enabled:
+                srdp.stop()
+        for sdp in self.static_data_providers:
+            if sdp.enabled:
+                sdp.stop()
+        for cdp in self.callsign_data_providers:
+            if cdp.enabled:
+                cdp.stop()
+
+# Global object
+DATA_PROVIDERS = DataProviders()

core/data_store.py

@@ -0,0 +1,119 @@
+import logging
+import re
+from pathlib import Path
+
+import diskcache
+
+from core.config import MAX_SPOT_AGE, MAX_ALERT_AGE
+from core.live_data_cache import LiveDataCache
+from data.solar_conditions import SolarConditions
+
+CACHE_DIR = "./cache/"
+
+
+class DataStore:
+    """Data caching/storage object. Handles storage of spots, alerts, solar conditions, SIG reference data, and callsign
+    lookup data using different caching strategies for each."""
+
+    def __init__(self):
+        # Constants
+        self._MAX_SPOT_COUNT = 100000
+        self._MAX_ALERT_COUNT = 100000
+        self._SPOT_ALERT_SNAPSHOT_INTERVAL_SEC = 300
+        self.CALLSIGN_DATA_TTL_SEC = 30 * 24 * 60 * 60
+        # Caches
+        self.alerts = None
+        self.spots = None
+        self.callsign_data_countryfiles = None
+        self.callsign_data_clublogxml = None
+        self.callsign_data_clublogapi = None
+        self.callsign_data_qrz = None
+        self.callsign_data_hamqth = None
+        self.dxcc_data = None
+        self.dxcc_lookup_by_call_regex = []
+        self.sigrefs = None
+        self.status_data = {}
+        self._status = None
+        self.solar_conditions = {}
+        self._solar = None
+        # ITU/CQ zone GeoJSON data is only ever loaded statically from a local file so these don't even need to be
+        # caches, they can just be straight objects
+        self.cq_zone_data = None
+        self.itu_zone_data = None
+
+    def setup(self):
+        Path(CACHE_DIR).mkdir(parents=True, exist_ok=True)
+
+        # Standard disk cache for solar data and status data, but each cache contains only a single object which we
+        # expose to the wider application
+        self._solar = diskcache.Cache(CACHE_DIR + "solar")
+        if "solar_conditions" not in self._solar:
+            self._solar.add("solar_conditions", SolarConditions())
+        self.solar_conditions = self._solar.get("solar_conditions")
+        self._status = diskcache.Cache(CACHE_DIR + "status")
+        if "status_data" not in self._status:
+            self._status.add("status_data", {})
+        self.status_data = self._status.get("status_data")
+
+        # Standard disk cache for static reference and SIG ref data. Separate provider threads will repopulate these on
+        # a regular basis but there's no need for a TTL since old data is better than no data.
+        self.dxcc_data = diskcache.Cache(CACHE_DIR + "dxcc_data")
+        self.regenerate_call_regex_to_dxcc_entity_map()
+
+        # For SIG reference data specifically, we need to key on both SIG *and* reference, and trying to do two layers
+        # of dict in diskcache absolutely destroys performance with unpickling huge dicts, so we have an ugly "SIG:ref"
+        # syntax for keys to keep it a single level.
+        self.sigrefs = diskcache.Cache(CACHE_DIR + "sigrefs")
+        logging.info(f"Loaded data for %d SIG references.", len(self.sigrefs))
+
+        # Standard disk cache for callsign data. This data does have a TTL to trigger an occasional re-lookup.
+        # Old data *is* better than no data, but we can't have a background thread re-looking-up every callsign
+        # we've seen, so we rely on them timing out and this triggering another lookup.
+        self.callsign_data_countryfiles = diskcache.Cache(CACHE_DIR + "callsign_data_countryfiles")
+        self.callsign_data_clublogxml = diskcache.Cache(CACHE_DIR + "callsign_data_clublogxml")
+        self.callsign_data_clublogapi = diskcache.Cache(CACHE_DIR + "callsign_data_clublogapi")
+        self.callsign_data_qrz = diskcache.Cache(CACHE_DIR + "callsign_data_qrz")
+        self.callsign_data_hamqth = diskcache.Cache(CACHE_DIR + "callsign_data_hamqth")
+        unique_keys = set()
+        for c in [self.callsign_data_countryfiles, self.callsign_data_clublogxml, self.callsign_data_clublogapi,
+                  self.callsign_data_qrz, self.callsign_data_hamqth]:
+            unique_keys.update(c)
+        logging.info(f"Loaded data for %d callsigns.", len(unique_keys))
+
+        # Special caches for spots and alerts, which have TTL and write snapshots to disk at an interval. We
+        # specifically load these caches *last* so that any sigref and callsign data is already loaded from disk cache
+        # before the spots and alerts are live in the system.
+        self.spots = LiveDataCache(maxsize=self._MAX_SPOT_COUNT, ttl=MAX_SPOT_AGE,
+                                   snapshot_dir=CACHE_DIR + "spots",
+                                   snapshot_interval_sec=self._SPOT_ALERT_SNAPSHOT_INTERVAL_SEC)
+        logging.info(f"Loaded %d spots from a previous run.", len(self.spots.keys()))
+
+        self.alerts = LiveDataCache(maxsize=self._MAX_ALERT_COUNT, ttl=MAX_ALERT_AGE,
+                                    snapshot_dir=CACHE_DIR + "alerts",
+                                    snapshot_interval_sec=self._SPOT_ALERT_SNAPSHOT_INTERVAL_SEC)
+        logging.info(f"Loaded %d alerts from a previous run.", len(self.alerts.keys()))
+
+    def regenerate_call_regex_to_dxcc_entity_map(self):
+        """DXCC entity data from K0SWE includes a regex which we can use to match a callsign, and determine which DXCC
+        entity it belongs to. But getting every DXCC entity data object out of DiskCache, iterating, compiling its regex
+        and testing the callsign every time is expensive. So instead we build a separate in-memory lookup of compiled
+        regex against DXCC entity code, as a list of tuples we can iterate through."""
+
+        for entry in [DATA_STORE.dxcc_data[key] for key in DATA_STORE.dxcc_data]:
+            self.dxcc_lookup_by_call_regex.append((re.compile(entry["prefixRegex"]), entry["entityCode"]))
+
+    def close(self):
+        self.spots.close()
+        self.alerts.close()
+        self._solar.close()
+        self._status.close()
+        self.dxcc_data.close()
+        self.sigrefs.close()
+        self.callsign_data_countryfiles.close()
+        self.callsign_data_clublogxml.close()
+        self.callsign_data_clublogapi.close()
+        self.callsign_data_qrz.close()
+        self.callsign_data_hamqth.close()
+
+# Global object
+DATA_STORE = DataStore()

core/geo_utils.py

@@ -1,32 +1,23 @@
-import json
 import logging
 import re
 from math import floor
 
-import geopandas
 from pyproj import Transformer
-from shapely import prepare
 from shapely.geometry import Point, Polygon
 
+from core.data_store import DATA_STORE
+
 TRANSFORMER_OS_GRID_TO_WGS84 = Transformer.from_crs("EPSG:27700", "EPSG:4326")
 TRANSFORMER_IRISH_GRID_TO_WGS84 = Transformer.from_crs("EPSG:29903", "EPSG:4326")
 TRANSFORMER_CI_UTM_GRID_TO_WGS84 = Transformer.from_crs("+proj=utm +zone=30 +ellps=WGS84", "EPSG:4326")
 
-with open("datafiles/cqzones.geojson") as f:
-    cq_zone_data = geopandas.GeoDataFrame.from_features(json.load(f)["features"])
-with open("datafiles/ituzones.geojson") as f:
-    itu_zone_data = geopandas.GeoDataFrame.from_features(json.load(f)["features"])
-for idx in cq_zone_data.index:
-    prepare(cq_zone_data.at[idx, 'geometry'])
-for idx in itu_zone_data.index:
-    prepare(itu_zone_data.at[idx, 'geometry'])
-
 
 def lat_lon_to_cq_zone(lat, lon):
     """Finds out which CQ zone a lat/lon point is in."""
 
+    if DATA_STORE.cq_zone_data is not None:
         lon = ((lon + 180) % 360) - 180
-    for index, row in cq_zone_data.iterrows():
+        for index, row in DATA_STORE.cq_zone_data.iterrows():
             polygon = Polygon(row["geometry"])
             test_point = Point(lon, lat)
             if polygon.contains(test_point):
@@ -46,8 +37,9 @@ def lat_lon_to_cq_zone(lat, lon):
 def lat_lon_to_itu_zone(lat, lon):
     """Finds out which ITU zone a lat/lon point is in."""
 
+    if DATA_STORE.itu_zone_data is not None:
         lon = ((lon + 180) % 360) - 180
-    for index, row in itu_zone_data.iterrows():
+        for index, row in DATA_STORE.itu_zone_data.iterrows():
             polygon = Polygon(row["geometry"])
             test_point = Point(lon, lat)
             if polygon.contains(test_point):

core/live_data_cache.py

@@ -0,0 +1,100 @@
+import logging
+import threading
+import time
+
+import diskcache
+from cachetools import TTLCache
+
+
+class LiveDataCache:
+    """Cache for spots and alerts. Uses the fast in-memory TTLCache for normal data I/O, including the TTL to enforce
+    maximum lifetime, and adds a separate diskcache to which we can save and load the TTLCache to provide persistence.
+    Also adds thread safety so spots and alerts can come from any thread, and a listener mechanism so the web server
+    can get a callback when new spots/alerts are added, and send them to any SSE clients."""
+
+    def __init__(self, maxsize, ttl, snapshot_dir, snapshot_interval_sec):
+        self._cache = TTLCache(maxsize=maxsize, ttl=ttl)
+        self._lock = threading.Lock()
+        self._ttl = ttl
+        self._listeners = []
+        self._listeners_lock = threading.Lock()
+        self._snapshot_dir = snapshot_dir
+        self._disk_cache = diskcache.Cache(str(snapshot_dir))
+        self._load_snapshot()
+        self._start_periodic_snapshot(snapshot_interval_sec)
+
+    def set(self, key, value):
+        with self._lock:
+            self._cache[key] = value
+
+        # Notify listeners
+        with self._listeners_lock:
+            listeners = list(self._listeners)
+        for callback in listeners:
+            try:
+                callback(value)
+            except Exception:
+                logging.error("Listener raised an exception for key %s", key, exc_info=True)
+
+
+    def get(self, key, default=None):
+        with self._lock:
+            return self._cache.get(key, default)
+
+    def delete(self, key):
+        with self._lock:
+            self._cache.pop(key, None)
+
+    def keys(self):
+        with self._lock:
+            return list(self._cache.keys())
+
+    def values(self):
+        with self._lock:
+            return list(self._cache.values())
+
+    def add_listener(self, callback):
+        """Register callback(value) which will be called whenever a new spot/alert item is added via set(). Used by the
+        web server (via SSEBroadcaster) to send SSE clients an update on every new spot."""
+
+        with self._listeners_lock:
+            self._listeners.append(callback)
+
+    def remove_listener(self, callback):
+        with self._listeners_lock:
+            self._listeners.remove(callback)
+
+    def save_snapshot(self):
+        with self._lock:
+            # Store the time with the data so we can avoid loading anything nxt time that's older than TTL
+            data = [(k, v, time.time()) for k, v in self._cache.items()]
+        try:
+            self._disk_cache.set("snapshot", data)
+        except Exception as e:
+            logging.error("Failed to write snapshot to %s", self._snapshot_dir, e, exc_info=True)
+
+    def _load_snapshot(self):
+        data = self._disk_cache.get("snapshot")
+        if not data:
+            return
+
+        now = time.time()
+        with self._lock:
+            for key, value, saved_at in data:
+                # Only restore entries that would still be within TTL
+                if now - saved_at < self._ttl:
+                    self._cache[key] = value
+        logging.info("Loaded snapshot from %s", self._snapshot_dir)
+
+    def _start_periodic_snapshot(self, interval):
+        def loop():
+            while True:
+                time.sleep(interval)
+                self.save_snapshot()
+
+        t = threading.Thread(target=loop, name=f"LiveDataCache-Snapshot-{self._snapshot_dir}")
+        t.start()
+
+    def close(self):
+        self.save_snapshot()
+        self._disk_cache.close()

core/lookup_helper.py

@@ -1,682 +0,0 @@
-import gzip
-import json
-import logging
-import re
-import urllib.parse
-from datetime import timedelta
-
-import requests
-import xmltodict
-from diskcache import Cache
-from pyhamtools import LookupLib, Callinfo, callinfo
-from pyhamtools.exceptions import APIKeyMissingError
-from pyhamtools.frequency import freq_to_band
-from pyhamtools.locator import latlong_to_locator
-from requests_cache import CachedSession
-
-from core.cache_utils import SEMI_STATIC_URL_DATA_CACHE
-from core.config import config
-from core.constants import BANDS, UNKNOWN_BAND, CW_MODES, PHONE_MODES, DATA_MODES, ALL_MODES, \
-    HTTP_HEADERS, HAMQTH_PRG, MODE_ALIASES
-from data.lookup_credentials import LookupCredentials
-
-# QRZ XML field names differ from pyhamtools' normalised names; map them here.
-_QRZ_FIELD_MAP = {
-    "lat": "latitude",
-    "lon": "longitude",
-    "grid": "locator",
-    "ituzone": "ituz",
-    "cqzone": "cqz",
-}
-_QRZ_INT_FIELDS = {"adif", "cqz", "ituz"}
-_QRZ_FLOAT_FIELDS = {"latitude", "longitude"}
-
-
-def _normalize_qrz_data(raw):
-    data = {}
-    for k, v in raw.items():
-        if v is None:
-            continue
-        mapped_key = _QRZ_FIELD_MAP.get(k, k)
-        if mapped_key in _QRZ_INT_FIELDS:
-            try:
-                v = int(v)
-            except (ValueError, TypeError):
-                pass
-        elif mapped_key in _QRZ_FLOAT_FIELDS:
-            try:
-                v = float(v)
-            except (ValueError, TypeError):
-                pass
-        data[mapped_key] = v
-    return data
-
-
-class LookupHelper:
-    """Singleton class that provides lookup functionality."""
-
-    def __init__(self):
-        """Create the lookup helper. Note that nothing actually happens until the start() method is called, and that all
-        lookup methods will fail if start() has not yet been called. This therefore needs starting before any spot or
-        alert handlers are created."""
-
-        self._clublog_callsign_data_cache = None
-        self._lookup_lib_clublog_xml = None
-        self._clublog_xml_available = None
-        self._lookup_lib_clublog_api = None
-        self._clublog_xml_download_location = None
-        self._clublog_api_available = None
-        self._clublog_cty_xml_cache = None
-        self._clublog_api_key = None
-        self._qrz_callsign_data_cache = None
-        self._qrz_base_url = "https://xmldata.qrz.com/xml/current/"
-        # QRZ session keys expire after an hour; cache the login response for 55 minutes.
-        self._qrz_session_cache = CachedSession("cache/qrz_session_cache",
-                                                expire_after=timedelta(minutes=55))
-        self._hamqth_callsign_data_cache = None
-        self._hamqth_base_url = "https://www.hamqth.com/xml.php"
-        # HamQTH session keys expire after an hour. Rather than working out how much time has passed manually, we cheat
-        # and cache the HTTP response for 55 minutes, so when the login URL is queried within 55 minutes of the previous
-        # time, you just get the cached response.
-        self._hamqth_session_lookup_cache = CachedSession("cache/hamqth_session_cache",
-                                                          expire_after=timedelta(minutes=55))
-        self._call_info_basic = None
-        self._lookup_lib_basic = None
-        self._country_files_cty_plist_download_location = None
-        self._dxcc_json_download_location = None
-        self._dxcc_data = None
-
-    def start(self):
-        # Lookup helpers from pyhamtools. We use five (!) of these. The simplest is country-files.com, which downloads
-        # the data once on startup, and requires no login/key, but does not have the best coverage.
-        # If the user provides login details/API keys, we also set up helpers for QRZ.com, HamQTH, Clublog (live API
-        # request), and Clublog (XML download). The lookup functions iterate through these in a sensible order, looking
-        # for suitable data.
-        self._country_files_cty_plist_download_location = "cache/cty.plist"
-        success = self._download_country_files_cty_plist()
-        if success:
-            self._lookup_lib_basic = LookupLib(lookuptype="countryfile",
-                                               filename=self._country_files_cty_plist_download_location)
-        else:
-            self._lookup_lib_basic = LookupLib(lookuptype="countryfile")
-        self._call_info_basic = Callinfo(self._lookup_lib_basic)
-
-        self._qrz_callsign_data_cache = Cache('cache/qrz_callsign_lookup_cache')
-
-        self._hamqth_callsign_data_cache = Cache('cache/hamqth_callsign_lookup_cache')
-
-        self._clublog_api_key = config["clublog-api-key"]
-        self._clublog_cty_xml_cache = CachedSession("cache/clublog_cty_xml_cache", expire_after=timedelta(days=10))
-        self._clublog_api_available = self._clublog_api_key != ""
-        self._clublog_xml_download_location = "cache/cty.xml"
-        if self._clublog_api_available:
-            self._lookup_lib_clublog_api = LookupLib(lookuptype="clublogapi", apikey=self._clublog_api_key)
-            success = self._download_clublog_ctyxml()
-            self._clublog_xml_available = success
-            if success:
-                self._lookup_lib_clublog_xml = LookupLib(lookuptype="clublogxml",
-                                                         filename=self._clublog_xml_download_location)
-        self._clublog_callsign_data_cache = Cache('cache/clublog_callsign_lookup_cache')
-
-        # We also get a lookup of DXCC data from K0SWE to use for additional lookups of e.g. flags.
-        self._dxcc_json_download_location = "cache/dxcc.json"
-        success = self._download_dxcc_json()
-        if success:
-            with open(self._dxcc_json_download_location) as f:
-                tmp_dxcc_data = json.load(f)["dxcc"]
-                # Reformat as a map for faster lookup
-                self._dxcc_data = {}
-                for dxcc in tmp_dxcc_data:
-                    self._dxcc_data[dxcc["entityCode"]] = dxcc
-        else:
-            logging.error("Could not download DXCC data, flags and similar data may be missing!")
-
-        # Precompile regex matches for DXCCs to improve efficiency when iterating through them
-        for dxcc in (self._dxcc_data.values() if self._dxcc_data else []):
-            dxcc["_prefixRegexCompiled"] = re.compile(dxcc["prefixRegex"])
-
-    def _download_country_files_cty_plist(self):
-        """Download the cty.plist file from country-files.com on first startup. The pyhamtools lib can actually download and use
-        this itself, but it's occasionally offline which causes it to throw an error. By downloading it separately, we can
-        catch errors and handle them, falling back to a previous copy of the file in the cache, and we can use the
-        requests_cache library to prevent re-downloading too quickly if the software keeps restarting."""
-
-        try:
-            logging.info("Downloading Country-files.com cty.plist...")
-            response = SEMI_STATIC_URL_DATA_CACHE.get("https://www.country-files.com/cty/cty.plist",
-                                                      headers=HTTP_HEADERS).text
-
-            with open(self._country_files_cty_plist_download_location, "w") as f:
-                f.write(response)
-                f.flush()
-                return True
-
-        except Exception as e:
-            logging.error("Exception when downloading Clublog cty.xml", e)
-            return False
-
-    def _download_dxcc_json(self):
-        """Download the dxcc.json file on first startup."""
-
-        try:
-            logging.info("Downloading dxcc.json...")
-            response = SEMI_STATIC_URL_DATA_CACHE.get(
-                "https://raw.githubusercontent.com/k0swe/dxcc-json/refs/heads/main/dxcc.json",
-                headers=HTTP_HEADERS).text
-
-            with open(self._dxcc_json_download_location, "w") as f:
-                f.write(response)
-                f.flush()
-                return True
-
-        except Exception as e:
-            logging.error("Exception when downloading dxcc.json", e)
-            return False
-
-    def _download_clublog_ctyxml(self):
-        """Download the cty.xml (gzipped) file from Clublog on first startup, so we can use it in preference to querying the
-        database live if possible."""
-
-        try:
-            logging.info("Downloading Clublog cty.xml.gz...")
-            response = self._clublog_cty_xml_cache.get("https://cdn.clublog.org/cty.php?api=" + self._clublog_api_key,
-                                                       headers=HTTP_HEADERS)
-            logging.info("Caching Clublog cty.xml.gz...")
-            open(self._clublog_xml_download_location + ".gz", 'wb').write(response.content)
-            with gzip.open(self._clublog_xml_download_location + ".gz", "rb") as uncompressed:
-                file_content = uncompressed.read()
-            logging.info("Caching Clublog cty.xml...")
-            with open(self._clublog_xml_download_location, "wb") as f:
-                f.write(file_content)
-                f.flush()
-                return True
-
-        except Exception as e:
-            logging.error("Exception when downloading Clublog cty.xml", e)
-            return False
-
-    def infer_country_from_callsign(self, call, credentials=None):
-        """Infer a country name from a callsign"""
-
-        try:
-            # Start with the basic country-files.com-based decoder.
-            country = self._call_info_basic.get_country_name(call)
-        except (KeyError, ValueError):
-            country = None
-        # Couldn't get anything from basic call info database, try QRZ.com
-        if not country:
-            qrz_data = self._get_qrz_data_for_callsign(call, credentials)
-            if qrz_data and "country" in qrz_data:
-                country = qrz_data["country"]
-        # Couldn't get anything from QRZ.com database, try HamQTH
-        if not country:
-            hamqth_data = self._get_hamqth_data_for_callsign(call, credentials)
-            if hamqth_data and "country" in hamqth_data:
-                country = hamqth_data["country"]
-        # Couldn't get anything from HamQTH database, try Clublog data
-        if not country:
-            clublog_data = self._get_clublog_xml_data_for_callsign(call)
-            if clublog_data and "Name" in clublog_data:
-                country = clublog_data["Name"]
-        if not country:
-            clublog_data = self._get_clublog_api_data_for_callsign(call)
-            if clublog_data and "Name" in clublog_data:
-                country = clublog_data["Name"]
-        # Couldn't get anything from Clublog database, try DXCC data
-        if not country:
-            dxcc_data = self._get_dxcc_data_for_callsign(call)
-            if dxcc_data and "name" in dxcc_data:
-                country = dxcc_data["name"]
-        return country
-
-    def infer_dxcc_id_from_callsign(self, call, credentials=None):
-        """Infer a DXCC ID from a callsign"""
-
-        try:
-            # Start with the basic country-files.com-based decoder.
-            dxcc = self._call_info_basic.get_adif_id(call)
-        except (KeyError, ValueError):
-            dxcc = None
-        # Couldn't get anything from basic call info database, try QRZ.com
-        if not dxcc:
-            qrz_data = self._get_qrz_data_for_callsign(call, credentials)
-            if qrz_data and "adif" in qrz_data:
-                dxcc = qrz_data["adif"]
-        # Couldn't get anything from QRZ.com database, try HamQTH
-        if not dxcc:
-            hamqth_data = self._get_hamqth_data_for_callsign(call, credentials)
-            if hamqth_data and "adif" in hamqth_data:
-                dxcc = hamqth_data["adif"]
-        # Couldn't get anything from HamQTH database, try Clublog data
-        if not dxcc:
-            clublog_data = self._get_clublog_xml_data_for_callsign(call)
-            if clublog_data and "DXCC" in clublog_data:
-                dxcc = clublog_data["DXCC"]
-        if not dxcc:
-            clublog_data = self._get_clublog_api_data_for_callsign(call)
-            if clublog_data and "DXCC" in clublog_data:
-                dxcc = clublog_data["DXCC"]
-        # Couldn't get anything from Clublog database, try DXCC data
-        if not dxcc:
-            dxcc_data = self._get_dxcc_data_for_callsign(call)
-            if dxcc_data and "entityCode" in dxcc_data:
-                dxcc = dxcc_data["entityCode"]
-        return dxcc
-
-    def infer_continent_from_callsign(self, call, credentials=None):
-        """Infer a continent shortcode from a callsign"""
-
-        try:
-            # Start with the basic country-files.com-based decoder.
-            continent = self._call_info_basic.get_continent(call)
-        except (KeyError, ValueError):
-            continent = None
-        # Couldn't get anything from basic call info database, try HamQTH
-        if not continent:
-            hamqth_data = self._get_hamqth_data_for_callsign(call, credentials)
-            if hamqth_data and "continent" in hamqth_data:
-                continent = hamqth_data["continent"]
-        # Couldn't get anything from HamQTH database, try Clublog data
-        if not continent:
-            clublog_data = self._get_clublog_xml_data_for_callsign(call)
-            if clublog_data and "Continent" in clublog_data:
-                continent = clublog_data["Continent"]
-        if not continent:
-            clublog_data = self._get_clublog_api_data_for_callsign(call)
-            if clublog_data and "Continent" in clublog_data:
-                continent = clublog_data["Continent"]
-        # Couldn't get anything from Clublog database, try DXCC data
-        if not continent:
-            dxcc_data = self._get_dxcc_data_for_callsign(call)
-            # Some DXCCs are in two continents, if so don't use the continent data as we can't be sure
-            if dxcc_data and "continent" in dxcc_data and len(dxcc_data["continent"]) == 1:
-                continent = dxcc_data["continent"][0]
-        return continent
-
-    def infer_cq_zone_from_callsign(self, call, credentials=None):
-        """Infer a CQ zone from a callsign"""
-
-        try:
-            # Start with the basic country-files.com-based decoder.
-            cqz = self._call_info_basic.get_cqz(call)
-        except (KeyError, ValueError):
-            cqz = None
-        # Couldn't get anything from basic call info database, try QRZ.com
-        if not cqz:
-            qrz_data = self._get_qrz_data_for_callsign(call, credentials)
-            if qrz_data and "cqz" in qrz_data:
-                cqz = qrz_data["cqz"]
-        # Couldn't get anything from QRZ.com database, try HamQTH
-        if not cqz:
-            hamqth_data = self._get_hamqth_data_for_callsign(call, credentials)
-            if hamqth_data and "cq" in hamqth_data:
-                cqz = hamqth_data["cq"]
-        # Couldn't get anything from HamQTH database, try Clublog data
-        if not cqz:
-            clublog_data = self._get_clublog_xml_data_for_callsign(call)
-            if clublog_data and "CQZ" in clublog_data:
-                cqz = clublog_data["CQZ"]
-        if not cqz:
-            clublog_data = self._get_clublog_api_data_for_callsign(call)
-            if clublog_data and "CQZ" in clublog_data:
-                cqz = clublog_data["CQZ"]
-        # Couldn't get anything from Clublog database, try DXCC data
-        if not cqz:
-            dxcc_data = self._get_dxcc_data_for_callsign(call)
-            # Some DXCCs are in multiple zones, if so don't use the zone data as we can't be sure
-            if dxcc_data and "cq" in dxcc_data and len(dxcc_data["cq"]) == 1:
-                cqz = dxcc_data["cq"][0]
-        return cqz
-
-    def infer_itu_zone_from_callsign(self, call, credentials=None):
-        """Infer a ITU zone from a callsign"""
-
-        try:
-            # Start with the basic country-files.com-based decoder.
-            ituz = self._call_info_basic.get_ituz(call)
-        except (KeyError, ValueError):
-            ituz = None
-        # Couldn't get anything from basic call info database, try QRZ.com
-        if not ituz:
-            qrz_data = self._get_qrz_data_for_callsign(call, credentials)
-            if qrz_data and "ituz" in qrz_data:
-                ituz = qrz_data["ituz"]
-        # Couldn't get anything from QRZ.com database, try HamQTH
-        if not ituz:
-            hamqth_data = self._get_hamqth_data_for_callsign(call, credentials)
-            if hamqth_data and "itu" in hamqth_data:
-                ituz = hamqth_data["itu"]
-        # Couldn't get anything from HamQTH database, Clublog doesn't provide this, so try DXCC data
-        if not ituz:
-            dxcc_data = self._get_dxcc_data_for_callsign(call)
-            # Some DXCCs are in multiple zones, if so don't use the zone data as we can't be sure
-            if dxcc_data and "itu" in dxcc_data and len(dxcc_data["itu"]) == 1:
-                ituz = dxcc_data["itu"]
-        return ituz
-
-    def get_flag_for_dxcc(self, dxcc):
-        """Get an emoji flag for a given DXCC entity ID"""
-
-        return self._dxcc_data[dxcc]["flag"] if dxcc in self._dxcc_data else None
-
-    def infer_name_from_callsign_online_lookup(self, call, credentials=None):
-        """Infer an operator name from a callsign (requires QRZ.com/HamQTH)"""
-
-        data = self._get_qrz_data_for_callsign(call, credentials)
-        if data and "fname" in data:
-            name = data["fname"]
-            if "name" in data:
-                name = name + " " + data["name"]
-            return name
-        data = self._get_hamqth_data_for_callsign(call, credentials)
-        if data and "nick" in data:
-            return data["nick"]
-        else:
-            return None
-
-    def infer_latlon_from_callsign_online_lookup(self, call, credentials=None):
-        """Infer a latitude and longitude from a callsign (requires QRZ.com/HamQTH)
-        Coordinates that look default are rejected (apologies if your position really is 0,0, enjoy your voyage)"""
-
-        data = self._get_qrz_data_for_callsign(call, credentials)
-        if data and "latitude" in data and "longitude" in data and (
-                float(data["latitude"]) != 0 or float(data["longitude"]) != 0) and -89.9 < float(
-                data["latitude"]) < 89.9:
-            return [float(data["latitude"]), float(data["longitude"])]
-        data = self._get_hamqth_data_for_callsign(call, credentials)
-        if data and "latitude" in data and "longitude" in data and (
-                float(data["latitude"]) != 0 or float(data["longitude"]) != 0) and -89.9 < float(
-                data["latitude"]) < 89.9:
-            return [float(data["latitude"]), float(data["longitude"])]
-        else:
-            return None
-
-    def infer_grid_from_callsign_online_lookup(self, call, credentials=None):
-        """Infer a grid locator from a callsign (requires QRZ.com/HamQTH).
-        Grids that look default are rejected (apologies if your grid really is AA00aa, enjoy your research)"""
-
-        data = self._get_qrz_data_for_callsign(call, credentials)
-        if data and "locator" in data and data["locator"].upper() != "AA00" and data["locator"].upper() != "AA00AA" and \
-                data["locator"].upper() != "AA00AA00":
-            return data["locator"]
-        data = self._get_hamqth_data_for_callsign(call, credentials)
-        if data and "grid" in data and data["grid"].upper() != "AA00" and data["grid"].upper() != "AA00AA" and data[
-            "grid"].upper() != "AA00AA00":
-            return data["grid"]
-        else:
-            return None
-
-    def infer_qth_from_callsign_online_lookup(self, call, credentials=None):
-        """Infer a textual QTH from a callsign (requires QRZ.com/HamQTH)"""
-
-        data = self._get_qrz_data_for_callsign(call, credentials)
-        if data and "addr2" in data:
-            return data["addr2"]
-        data = self._get_hamqth_data_for_callsign(call, credentials)
-        if data and "qth" in data:
-            return data["qth"]
-        else:
-            return None
-
-    def infer_latlon_from_callsign_dxcc(self, call):
-        """Infer a latitude and longitude from a callsign (using DXCC, probably very inaccurate)"""
-
-        try:
-            data = self._call_info_basic.get_lat_long(call)
-            if data and "latitude" in data and "longitude" in data:
-                loc = [float(data["latitude"]), float(data["longitude"])]
-            else:
-                loc = None
-        except KeyError:
-            loc = None
-        # Couldn't get anything from basic call info database, try Clublog data
-        if not loc:
-            data = self._get_clublog_xml_data_for_callsign(call)
-            if data and "Lat" in data and "Lon" in data:
-                loc = [float(data["Lat"]), float(data["Lon"])]
-        if not loc:
-            data = self._get_clublog_api_data_for_callsign(call)
-            if data and "Lat" in data and "Lon" in data:
-                loc = [float(data["Lat"]), float(data["Lon"])]
-        return loc
-
-    def infer_grid_from_callsign_dxcc(self, call):
-        """Infer a grid locator from a callsign (using DXCC, probably very inaccurate)"""
-
-        latlon = self.infer_latlon_from_callsign_dxcc(call)
-        grid = None
-        try:
-            grid = latlong_to_locator(latlon[0], latlon[1], 8)
-        except:
-            logging.debug("Invalid lat/lon received for DXCC")
-        return grid
-
-    def _get_qrz_data_for_callsign(self, call, credentials):
-        """Utility method to get QRZ.com data from cache if possible, if not get it from the API and cache it.
-        Returns None immediately if no credentials are provided."""
-
-        # Return from cache if available (a cached None means 'not found in QRZ')
-        if call in self._qrz_callsign_data_cache:
-            return self._qrz_callsign_data_cache.get(call)
-
-        # Obtain session key from credentials
-        session_key = None
-        if credentials and credentials.qrz_session_key:
-            session_key = credentials.qrz_session_key
-        elif credentials and credentials.qrz_username and credentials.qrz_password:
-            try:
-                login_response = self._qrz_session_cache.get(
-                    self._qrz_base_url + "?username=" + urllib.parse.quote_plus(credentials.qrz_username) +
-                    "&password=" + urllib.parse.quote_plus(credentials.qrz_password) + "&agent=spothole",
-                    headers=HTTP_HEADERS).content
-                login_data = xmltodict.parse(login_response)
-                session = login_data.get("QRZDatabase", {}).get("Session", {})
-                if "Key" in session:
-                    session_key = session["Key"]
-                else:
-                    logging.warning("QRZ.com login details incorrect, failed to look up with QRZ.")
-                    return None
-            except Exception:
-                logging.error("Exception when getting QRZ.com session key")
-                return None
-
-        if not session_key:
-            return None
-
-        # Try the call as given, then fall back to the base call (strips /P, /M etc.)
-        calls_to_try = [call]
-        home_call = callinfo.Callinfo.get_homecall(call)
-        if home_call != call:
-            calls_to_try.append(home_call)
-
-        for lookup_call in calls_to_try:
-            try:
-                lookup_response = requests.get(
-                    self._qrz_base_url + "?s=" + session_key + "&callsign=" + urllib.parse.quote_plus(lookup_call),
-                    headers=HTTP_HEADERS, timeout=10).content
-                raw = xmltodict.parse(lookup_response).get("QRZDatabase", {}).get("Callsign")
-                if raw:
-                    data = _normalize_qrz_data(raw)
-                    self._qrz_callsign_data_cache.add(call, data, expire=604800)  # 1 week in seconds
-                    return data
-            except (KeyError, ValueError):
-                continue
-            except Exception:
-                logging.error("Exception when looking up QRZ data")
-                return None
-
-        # Not found in QRZ; cache None so we don't keep retrying
-        self._qrz_callsign_data_cache.add(call, None, expire=604800)  # 1 week in seconds
-        return None
-
-    def _get_hamqth_data_for_callsign(self, call, credentials):
-        """Utility method to get HamQTH data from cache if possible, if not get it from the API and cache it.
-        Returns None immediately if no credentials are provided."""
-
-        # Return from cache if available
-        if call in self._hamqth_callsign_data_cache:
-            return self._hamqth_callsign_data_cache.get(call)
-
-        # Obtain session ID from credentials
-        session_id = None
-        if credentials and credentials.hamqth_session_id:
-            session_id = credentials.hamqth_session_id
-        elif credentials and credentials.hamqth_username and credentials.hamqth_password:
-            try:
-                session_data = self._hamqth_session_lookup_cache.get(
-                    self._hamqth_base_url + "?u=" + urllib.parse.quote_plus(credentials.hamqth_username) +
-                    "&p=" + urllib.parse.quote_plus(credentials.hamqth_password), headers=HTTP_HEADERS).content
-                dict_data = xmltodict.parse(session_data)
-                if "session_id" in dict_data["HamQTH"]["session"]:
-                    session_id = dict_data["HamQTH"]["session"]["session_id"]
-                else:
-                    logging.warning("HamQTH login details incorrect, failed to look up with HamQTH.")
-                    return None
-            except Exception:
-                logging.error("Exception when getting HamQTH session ID")
-                return None
-
-        if not session_id:
-            return None
-
-        # Try the call as given, then fall back to the base call (strips /P, /M etc.)
-        calls_to_try = [call]
-        home_call = callinfo.Callinfo.get_homecall(call)
-        if home_call != call:
-            calls_to_try.append(home_call)
-
-        for lookup_call in calls_to_try:
-            try:
-                lookup_data = SEMI_STATIC_URL_DATA_CACHE.get(
-                    self._hamqth_base_url + "?id=" + session_id + "&callsign=" + urllib.parse.quote_plus(
-                        lookup_call) + "&prg=" + HAMQTH_PRG, headers=HTTP_HEADERS).content
-                data = xmltodict.parse(lookup_data)["HamQTH"]["search"]
-                self._hamqth_callsign_data_cache.add(call, data, expire=604800)  # 1 week in seconds
-                return data
-            except (KeyError, ValueError):
-                continue
-            except Exception:
-                logging.error("Exception when looking up HamQTH data")
-                return None
-
-        # Not found in HamQTH; cache None so we don't keep retrying
-        self._hamqth_callsign_data_cache.add(call, None, expire=604800)  # 1 week in seconds
-        return None
-
-    def _get_clublog_api_data_for_callsign(self, call):
-        """Utility method to get Clublog API data from cache if possible, if not get it from the API and cache it"""
-
-        # Fetch from cache if we can, otherwise fetch from the API and cache it
-        if call in self._clublog_callsign_data_cache:
-            return self._clublog_callsign_data_cache.get(call)
-        elif self._clublog_api_available:
-            try:
-                data = self._lookup_lib_clublog_api.lookup_callsign(callsign=call)
-                self._clublog_callsign_data_cache.add(call, data, expire=604800)  # 1 week in seconds
-                return data
-            except (KeyError, ValueError):
-                # Clublog had no info for the call, but maybe it had prefixes or suffixes. Try again with the base call.
-                try:
-                    data = self._lookup_lib_clublog_api.lookup_callsign(callsign=callinfo.Callinfo.get_homecall(call))
-                    self._clublog_callsign_data_cache.add(call, data, expire=604800)  # 1 week in seconds
-                    return data
-                except (KeyError, ValueError):
-                    # Clublog had no info for the call, that's OK. Cache a None so we don't try to look this up again
-                    self._clublog_callsign_data_cache.add(call, None, expire=604800)  # 1 week in seconds
-                    return None
-            except APIKeyMissingError:
-                # User API key was wrong, warn
-                logging.error("Could not look up via Clublog API, key " + self._clublog_api_key + " was rejected.")
-                return None
-        else:
-            return None
-
-    def _get_clublog_xml_data_for_callsign(self, call):
-        """Utility method to get Clublog XML data from file"""
-
-        if self._clublog_xml_available:
-            try:
-                data = self._lookup_lib_clublog_xml.lookup_callsign(callsign=call)
-                return data
-            except (KeyError, ValueError):
-                # Clublog had no info for the call, that's OK. Cache a None so we don't try to look this up again
-                self._clublog_callsign_data_cache.add(call, None, expire=604800)  # 1 week in seconds
-                return None
-        else:
-            return None
-
-    def _get_dxcc_data_for_callsign(self, call):
-        """Utility method to get generic DXCC data from our lookup table, if we can find it"""
-
-        for entry in self._dxcc_data.values():
-            if entry["_prefixRegexCompiled"].match(call):
-                return entry
-        return None
-
-    def stop(self):
-        """Shutdown method to close down any caches neatly."""
-
-        self._qrz_callsign_data_cache.close()
-        self._hamqth_callsign_data_cache.close()
-        self._clublog_callsign_data_cache.close()
-
-
-# Singleton object
-lookup_helper = LookupHelper()
-
-def infer_mode_from_comment(comment):
-    """Infer a mode from the comment"""
-
-    for mode in ALL_MODES:
-        if mode in comment.upper():
-            return mode
-    for mode in MODE_ALIASES.keys():
-        if mode in comment.upper():
-            return MODE_ALIASES[mode]
-    return None
-
-
-def infer_mode_type_from_mode(mode):
-    """Infer a "mode family" from a mode."""
-
-    if mode.upper() in CW_MODES:
-        return "CW"
-    elif mode.upper() in PHONE_MODES:
-        return "PHONE"
-    elif mode.upper() in DATA_MODES:
-        return "DATA"
-    else:
-        if mode.upper() != "OTHER":
-            logging.warning("Found an unrecognised mode: " + mode + ". Developer should categorise this.")
-        return None
-
-
-def infer_band_from_freq(freq):
-    """Infer a band from a frequency in Hz"""
-
-    for b in BANDS:
-        if b.start_freq <= freq <= b.end_freq:
-            return b
-    return UNKNOWN_BAND
-
-
-def infer_mode_from_frequency(freq):
-    """Infer a mode from the frequency (in Hz) according to the band plan. Just a guess really."""
-
-    try:
-        khz = freq / 1000.0
-        mode = freq_to_band(khz)["mode"]
-        # Some additional common digimode ranges in addition to what the 3rd-party freq_to_band function returns.
-        # This is mostly here just because freq_to_band is very specific about things like FT8 frequencies, and e.g.
-        # a spot at 7074.5 kHz will be indicated as LSB, even though it's clearly in the FT8 range. Future updates
-        # might include other common digimode centres of activity here, but this achieves the main goal of keeping
-        # large numbers of clearly-FT* spots off the list of people filtering out digimodes.
-        if (7074 <= khz < 7077) or (10136 <= khz < 10139) or (14074 <= khz < 14077) or (18100 <= khz < 18103) or (
-                21074 <= khz < 21077) or (24915 <= khz < 24918) or (28074 <= khz < 28077):
-            mode = "FT8"
-        if (7047.5 <= khz < 7050.5) or (10140 <= khz < 10143) or (14080 <= khz < 14083) or (
-                18104 <= khz < 18107) or (21140 <= khz < 21143) or (24919 <= khz < 24922) or (28180 <= khz < 28183):
-            mode = "FT4"
-        return mode
-    except KeyError:
-        return None

core/sig_lookup_helper.py

@@ -0,0 +1,108 @@
+import logging
+
+from pyhamtools.locator import locator_to_latlong, latlong_to_locator
+
+from core.data_store import DATA_STORE
+from core.geo_utils import wab_wai_square_to_lat_lon
+from data.sig_ref import SIGRef
+
+
+def get_sig_ref_info(sig, ref_id):
+    """Look up details of a SIG reference (e.g. POTA park) such as name, lat/lon, and grid. Takes in a sig name and
+    a reference ID (both strings) and returns a SigRef object populated with as much data as we can find. This makes
+    use of SIG ref data in the data store, live lookups from the web, or just automatic calculation depending on which
+    SIG we are getting data for."""
+
+    if sig is None or sig == "" or ref_id is None or ref_id == "":
+        logging.debug("Failed to look up sig_ref info, sig or ref were not set.")
+        return None
+
+    sig_ref = SIGRef(sig=sig, id=ref_id)
+
+    try:
+        ### FUDGES ###
+        #
+        # DME fudge. Our database has leading zeros padding to 5 digits which is the expected format, but not all
+        # activators add leading zeros.
+        if sig.upper() == "DME":
+            ref_id = ref_id.zfill(5)
+
+        # KRMNPA & SANPCPA fudge. These don't have their own reference system, they just use VKFF references, so pretend
+        # the sig is WWFF and carry on
+        if sig.upper() == "KRMNPA" or sig.upper() == "SANPCPA":
+            sig = "WWFF"
+
+        ### SKIPS ###
+        #
+        # If the SIG is HEMA, we have no current lookup for this so just skip the lookup here.
+        if sig.upper() == "HEMA":
+            return sig_ref
+
+        ### PROGRAMMATIC DATA GENERATION INSTEAD OF LOOKUPS ###
+        #
+        # If the SIG is Tiles, WAB, WAI or BOTA (Beaches), we don't have anything to look up from the data store, we can
+        # calculate all the information we are going to get directly.
+        if sig.upper() == "TILES":
+            # Tiles on the Air just uses Maidenhead 6-digit squares, so ID, Name and Grid are all the same
+            if not sig_ref.name:
+                sig_ref.name = sig_ref.id
+            if not sig_ref.grid:
+                sig_ref.grid = sig_ref.id
+            if sig_ref.grid and not sig_ref.latitude:
+                ll = locator_to_latlong(str(sig_ref.grid))
+                sig_ref.latitude = ll[0]
+                sig_ref.longitude = ll[1]
+            return sig_ref
+
+        elif sig.upper() == "WAB" or sig.upper() == "WAI":
+            ll = wab_wai_square_to_lat_lon(ref_id)
+            if ll:
+                sig_ref.name = ref_id
+                try:
+                    sig_ref.grid = latlong_to_locator(ll[0], ll[1], 6)
+                    sig_ref.latitude = ll[0]
+                    sig_ref.longitude = ll[1]
+                except:
+                    logging.warning("Invalid lat/lon received for WAB/WAI reference")
+            return sig_ref
+
+        elif sig.upper() == "BOTA":
+            # For BOTA all we can ever generate is the URL, there is no data file or lookup for lat/longs
+            if not sig_ref.name:
+                sig_ref.name = sig_ref.id
+            sig_ref.url = "https://www.beachesontheair.com/beaches/" + sig_ref.name.lower().replace(" ", "-")
+            return sig_ref
+
+        ### ACTUAL LOOKUP ###
+        #
+        # OK, this is something we have to look up. Now check to see if our data store contains SIG ref information for
+        # this SIG. If so, check for the reference data and use that.
+        key = sig + ":" + ref_id
+        lookup_data = DATA_STORE.sigrefs.get(key) if key in DATA_STORE.sigrefs else None
+        if lookup_data:
+            return lookup_data
+
+        else:
+            # Maybe a super new reference we don't know about yet, but more likely a typo or a test reference,
+            # just silently ignore it.
+            logging.debug("%s database did not contain data for ref %s", sig, ref_id)
+
+    except Exception:
+        logging.error("Exception when looking up sig_ref info for " + sig + " ref " + ref_id, exc_info=True)
+    return sig_ref
+
+
+def populate_missing_sig_ref_info(sig_ref):
+    """Look up details of a SIG reference (e.g. POTA park) such as name, lat/lon, and grid. Takes in a sig_ref object
+    which must at minimum have a "sig" and an "id". The rest of the object will be populated and returned. Any data
+    currently in the object will be kept, only missing data in the object will be populated if it can be determined."""
+
+    lookup_data = get_sig_ref_info(sig_ref.sig, sig_ref.id)
+
+    if lookup_data:
+        # Copy new sig ref data into existing object where data was previously missing
+        for key, value in lookup_data.__dict__.items():
+            if value is not None and sig_ref.__dict__.get(key) is None:
+                sig_ref.__dict__[key] = value
+
+    return sig_ref

core/sig_utils.py

@@ -1,11 +1,4 @@
-import csv
-import logging
-
-from pyhamtools.locator import latlong_to_locator, locator_to_latlong
-
-from core.cache_utils import SEMI_STATIC_URL_DATA_CACHE
-from core.constants import SIGS, HTTP_HEADERS
-from core.geo_utils import wab_wai_square_to_lat_lon
+from core.constants import SIGS
 
 
 def get_ref_regex_for_sig(sig):
@@ -17,156 +10,15 @@ def get_ref_regex_for_sig(sig):
     return None
 
 
-def populate_sig_ref_info(sig_ref):
-    """Look up details of a SIG reference (e.g. POTA park) such as name, lat/lon, and grid. Takes in a sig_ref object which
-    must at minimum have a "sig" and an "id". The rest of the object will be populated and returned.
-    Note there is currently no support for KRMNPA location lookup, see issue #61."""
+def get_sig_name_from_comment_name(sig):
+    """Utility function to get the name of a SIG from its "comment name". Generally these will be the same but there are
+    some cases (e.g. is "TOTA" Towers, Tiles or Toilets?) where we need to transform one to the other."""
 
-    if sig_ref.sig is None or sig_ref.id is None:
-        logging.warning("Failed to look up sig_ref info, sig or id were not set.")
-
-    sig = sig_ref.sig
-    ref_id = sig_ref.id
-    try:
-        if sig.upper() == "POTA":
-            data = SEMI_STATIC_URL_DATA_CACHE.get("https://api.pota.app/park/" + ref_id, headers=HTTP_HEADERS).json()
-            if data:
-                fullname = data["name"] if "name" in data else None
-                if fullname and "parktypeDesc" in data and data["parktypeDesc"] != "":
-                    fullname = fullname + " " + data["parktypeDesc"]
-                sig_ref.name = fullname
-                sig_ref.url = "https://pota.app/#/park/" + ref_id
-                sig_ref.grid = data["grid6"] if "grid6" in data else None
-                sig_ref.latitude = data["latitude"] if "latitude" in data else None
-                sig_ref.longitude = data["longitude"] if "longitude" in data else None
-        elif sig.upper() == "SOTA":
-            data = SEMI_STATIC_URL_DATA_CACHE.get("https://api-db2.sota.org.uk/api/summits/" + ref_id,
-                                                  headers=HTTP_HEADERS).json()
-            if data:
-                sig_ref.name = data["name"] if "name" in data else None
-                sig_ref.url = "https://www.sotadata.org.uk/en/summit/" + ref_id
-                sig_ref.grid = data["locator"] if "locator" in data else None
-                sig_ref.latitude = data["latitude"] if "latitude" in data else None
-                sig_ref.longitude = data["longitude"] if "longitude" in data else None
-                sig_ref.activation_score = data["points"] if "points" in data else None
-        elif sig.upper() == "WWBOTA":
-            data = SEMI_STATIC_URL_DATA_CACHE.get("https://api.wwbota.org/bunkers/" + ref_id,
-                                                  headers=HTTP_HEADERS).json()
-            if data:
-                sig_ref.name = data["name"] if "name" in data else None
-                sig_ref.url = "https://bunkerwiki.org/?s=" + ref_id if ref_id.startswith("B/G") else None
-                sig_ref.grid = data["locator"] if "locator" in data else None
-                sig_ref.latitude = data["lat"] if "lat" in data else None
-                sig_ref.longitude = data["long"] if "long" in data else None
-        elif sig.upper() == "GMA" or sig.upper() == "ARLHS" or sig.upper() == "ILLW" or sig.upper() == "WCA" or sig.upper() == "MOTA" or sig.upper() == "IOTA":
-            data = SEMI_STATIC_URL_DATA_CACHE.get("https://www.cqgma.org/api/ref/?" + ref_id,
-                                                  headers=HTTP_HEADERS).json()
-            if data:
-                sig_ref.name = data["name"] if "name" in data else None
-                sig_ref.url = "https://www.cqgma.org/zinfo.php?ref=" + ref_id
-                sig_ref.grid = data["locator"] if "locator" in data else None
-                sig_ref.latitude = data["latitude"] if "latitude" in data else None
-                sig_ref.longitude = data["longitude"] if "longitude" in data else None
-        elif sig.upper() == "WWFF":
-            wwff_csv_data = SEMI_STATIC_URL_DATA_CACHE.get("https://wwff.co/wwff-data/wwff_directory.csv",
-                                                           headers=HTTP_HEADERS)
-            wwff_index = {row["reference"]: row for row in csv.DictReader(wwff_csv_data.content.decode().splitlines())}
-            row = wwff_index.get(ref_id)
-            if row:
-                sig_ref.name = row["name"] if "name" in row else None
-                sig_ref.url = "https://wwff.co/directory/?showRef=" + ref_id
-                sig_ref.grid = row["iaruLocator"] if "iaruLocator" in row and row["iaruLocator"] != "-" else None
-                sig_ref.latitude = float(row["latitude"]) if "latitude" in row and row["latitude"] != "-" else None
-                sig_ref.longitude = float(row["longitude"]) if "longitude" in row and row["longitude"] != "-" else None
-        elif sig.upper() == "SIOTA":
-            siota_csv_data = SEMI_STATIC_URL_DATA_CACHE.get("https://www.silosontheair.com/data/silos.csv",
-                                                            headers=HTTP_HEADERS)
-            siota_index = {row["SILO_CODE"]: row for row in csv.DictReader(siota_csv_data.content.decode().splitlines())}
-            row = siota_index.get(ref_id)
-            if row:
-                sig_ref.name = row["NAME"] if "NAME" in row else None
-                sig_ref.grid = row["LOCATOR"] if "LOCATOR" in row else None
-                sig_ref.latitude = float(row["LAT"]) if "LAT" in row else None
-                sig_ref.longitude = float(row["LNG"]) if "LNG" in row else None
-        elif sig.upper() == "WOTA":
-            data = SEMI_STATIC_URL_DATA_CACHE.get("https://www.wota.org.uk/mapping/data/summits.json",
-                                                  headers=HTTP_HEADERS).json()
-            if data:
-                for feature in data["features"]:
-                    if feature["properties"]["wotaId"] == ref_id:
-                        sig_ref.name = feature["properties"]["title"]
-                        # Fudge WOTA URLs. Outlying fell (LDO) URLs don't match their ID numbers but require 214 to be
-                        # added to them
-                        sig_ref.url = "https://www.wota.org.uk/MM_" + ref_id
-                        if ref_id.upper().startswith("LDO-"):
-                            number = int(ref_id.upper().replace("LDO-", ""))
-                            sig_ref.url = "https://www.wota.org.uk/MM_LDO-" + str(number + 214)
-                        sig_ref.grid = feature["properties"]["qthLocator"]
-                        sig_ref.latitude = feature["geometry"]["coordinates"][1]
-                        sig_ref.longitude = feature["geometry"]["coordinates"][0]
-                        break
-        elif sig.upper() == "ZLOTA":
-            data = SEMI_STATIC_URL_DATA_CACHE.get("https://ontheair.nz/assets/assets.json", headers=HTTP_HEADERS).json()
-            if data:
-                for asset in data:
-                    if asset["code"] == ref_id:
-                        sig_ref.name = asset["name"]
-                        sig_ref.url = "https://ontheair.nz/assets/ZLI_OT-030" + ref_id.replace("/", "_")
-                        try:
-                            sig_ref.grid = latlong_to_locator(asset["y"], asset["x"], 6)
-                        except:
-                            logging.debug("Invalid lat/lon received for reference")
-                            sig_ref.latitude = asset["y"]
-                            sig_ref.longitude = asset["x"]
-                        break
-        elif sig.upper() == "BOTA":
-            if not sig_ref.name:
-                sig_ref.name = sig_ref.id
-            sig_ref.url = "https://www.beachesontheair.com/beaches/" + sig_ref.name.lower().replace(" ", "-")
-        elif sig.upper() == "LLOTA":
-            data = SEMI_STATIC_URL_DATA_CACHE.get("https://llota.app/api/public/references",
-                                                  headers=HTTP_HEADERS).json()
-            if data:
-                for ref in data:
-                    if ref["reference_code"] == ref_id:
-                        sig_ref.name = ref["name"]
-                        sig_ref.url = "https://llota.app/list/ref/" + ref_id
-                        sig_ref.grid = ref["grid_locator"]
-                        ll = locator_to_latlong(sig_ref.grid)
-                        sig_ref.latitude = ll[0]
-                        sig_ref.longitude = ll[1]
-                        break
-        elif sig.upper() == "WWTOTA":
-            if not sig_ref.name:
-                sig_ref.name = sig_ref.id
-            sig_ref.url = "https://wwtota.com/seznam/karta_rozhledny.php?ref=" + sig_ref.name
-        elif sig.upper() == "TILES":
-            # Tiles on the Air just uses Maidenhead 6-digit squares, so ID, Name and Grid are all the same
-            if not sig_ref.name:
-                sig_ref.name = sig_ref.id
-            if not sig_ref.grid:
-                sig_ref.grid = sig_ref.id
-            if sig_ref.grid and not sig_ref.latitude:
-                ll = locator_to_latlong(sig_ref.grid)
-                sig_ref.latitude = ll[0]
-                sig_ref.longitude = ll[1]
-        elif sig.upper() == "WAB" or sig.upper() == "WAI":
-            ll = wab_wai_square_to_lat_lon(ref_id)
-            if ll:
-                sig_ref.name = ref_id
-                try:
-                    sig_ref.grid = latlong_to_locator(ll[0], ll[1], 6)
-                    sig_ref.latitude = ll[0]
-                    sig_ref.longitude = ll[1]
-                except:
-                    logging.debug("Invalid lat/lon received for reference")
-    except:
-        logging.warning("Failed to look up sig_ref info for " + sig + " ref " + ref_id + ".")
-    return sig_ref
+    for s in SIGS:
+        if any(n.upper() == sig.upper() for n in s.comment_names):
+            return s.name
+    return None
 
 
-# Regex matching any SIG
-ANY_SIG_REGEX = r"(" + r"|".join(list(map(lambda p: p.name, SIGS))) + r")"
-
-# Regex matching any SIG reference
-ANY_XOTA_SIG_REF_REGEX = r"[\w\/]+\-\d+"
+# Regex matching any SIG's "comment name", i.e. how it may be referred to in spot comments
+ANY_SIG_REGEX = r"(" + r"|".join(n for s in SIGS for n in s.comment_names) + r")"

core/status_reporter.py

@@ -7,36 +7,30 @@ import pytz
 
 from core.config import SERVER_OWNER_CALLSIGN
 from core.constants import SOFTWARE_VERSION
+from core.data_providers import DATA_PROVIDERS
+from core.data_store import DATA_STORE
 from core.prometheus_metrics_handler import memory_use_gauge, spots_gauge, alerts_gauge
+from server.webserver import WEB_SERVER
 
 
 class StatusReporter:
     """Provides a timed update of the application's status data."""
 
-    def __init__(self, status_data, run_interval, web_server, cleanup_timer, spots, spot_providers, alerts,
-                 alert_providers, solar_condition_providers):
+    def __init__(self, run_interval):
         """Constructor"""
 
-        self._status_data = status_data
         self._run_interval = run_interval
-        self._web_server = web_server
-        self._cleanup_timer = cleanup_timer
-        self._spots = spots
-        self._spot_providers = spot_providers
-        self._alerts = alerts
-        self._alert_providers = alert_providers
-        self._solar_condition_providers = solar_condition_providers
         self._thread = None
         self._stop_event = Event()
         self._startup_time = datetime.now(pytz.UTC)
 
-        self._status_data["software-version"] = SOFTWARE_VERSION
-        self._status_data["server-owner-callsign"] = SERVER_OWNER_CALLSIGN
+        DATA_STORE.status_data["software-version"] = SOFTWARE_VERSION
+        DATA_STORE.status_data["server-owner-callsign"] = SERVER_OWNER_CALLSIGN
 
     def start(self):
         """Start the reporter thread"""
 
-        self._thread = Thread(target=self._run, daemon=True)
+        self._thread = Thread(target=self._run, name="StatusReporter")
         self._thread.start()
 
     def stop(self):
@@ -55,43 +49,58 @@ class StatusReporter:
     def _report(self):
         """Write status information"""
 
-        self._status_data["uptime"] = (datetime.now(pytz.UTC) - self._startup_time).total_seconds()
-        self._status_data["mem_use_mb"] = round(psutil.Process(os.getpid()).memory_info().rss / (1024 * 1024), 3)
-        self._status_data["num_spots"] = len(self._spots)
-        self._status_data["num_alerts"] = len(self._alerts)
-        self._status_data["spot_providers"] = list(
+        DATA_STORE.status_data["uptime"] = (datetime.now(pytz.UTC) - self._startup_time).total_seconds()
+        DATA_STORE.status_data["mem_use_mb"] = round(psutil.Process(os.getpid()).memory_info().rss / (1024 * 1024), 3)
+        DATA_STORE.status_data["num_spots"] = len(DATA_STORE.spots.values())
+        DATA_STORE.status_data["num_alerts"] = len(DATA_STORE.alerts.values())
+        DATA_STORE.status_data["spot_providers"] = list(
             map(lambda p: {"name": p.name, "enabled": p.enabled, "status": p.status,
                            "last_updated": p.last_update_time.replace(
                                tzinfo=pytz.UTC).timestamp() if p.last_update_time.year > 2000 else 0,
                            "last_spot": p.last_spot_time.replace(
                                tzinfo=pytz.UTC).timestamp() if p.last_spot_time.year > 2000 else 0},
-                self._spot_providers))
-        self._status_data["alert_providers"] = list(
+                DATA_PROVIDERS.spot_providers))
+        DATA_STORE.status_data["alert_providers"] = list(
             map(lambda p: {"name": p.name, "enabled": p.enabled, "status": p.status,
                            "last_updated": p.last_update_time.replace(
                                tzinfo=pytz.UTC).timestamp() if p.last_update_time.year > 2000 else 0},
-                self._alert_providers))
-        self._status_data["solar_condition_providers"] = list(
+                DATA_PROVIDERS.alert_providers))
+        DATA_STORE.status_data["solar_condition_providers"] = list(
             map(lambda p: {"name": p.name, "enabled": p.enabled, "status": p.status,
                            "last_updated": p.last_update_time.replace(
                                tzinfo=pytz.UTC).timestamp() if p.last_update_time.year > 2000 else 0},
-                self._solar_condition_providers))
-        self._status_data["cleanup"] = {"status": self._cleanup_timer.status,
-                                        "last_ran": self._cleanup_timer.last_cleanup_time.replace(
-                                            tzinfo=pytz.UTC).timestamp() if self._cleanup_timer.last_cleanup_time else 0}
-        self._status_data["webserver"] = {"status": self._web_server.web_server_metrics["status"],
-                                          "last_api_access": self._web_server.web_server_metrics[
+                DATA_PROVIDERS.solar_condition_providers))
+        DATA_STORE.status_data["static_data_providers"] = list(
+            map(lambda p: {"name": p.name, "enabled": p.enabled, "status": p.status,
+                           "last_updated": p.last_update_time.replace(
+                               tzinfo=pytz.UTC).timestamp() if p.last_update_time.year > 2000 else 0},
+                DATA_PROVIDERS.static_data_providers))
+        DATA_STORE.status_data["sig_ref_data_providers"] = list(
+            map(lambda p: {"sig_name": p.sig_name, "enabled": p.enabled, "status": p.status,
+                           "last_updated": p.last_update_time.replace(
+                               tzinfo=pytz.UTC).timestamp() if p.last_update_time.year > 2000 else 0,
+                           "reference_count": p.reference_count},
+                DATA_PROVIDERS.sig_ref_data_providers))
+        DATA_STORE.status_data["callsign_data_providers"] = list(
+            map(lambda p: {"name": p.name, "enabled": p.enabled, "status": p.status,
+                           "last_updated": p.last_update_time.replace(
+                               tzinfo=pytz.UTC).timestamp() if p.last_update_time.year > 2000 else 0,
+                           "lookup_count": p.lookup_count},
+                DATA_PROVIDERS.callsign_data_providers))
+        DATA_STORE.status_data["webserver"] = {"status": WEB_SERVER.web_server_metrics["status"],
+                                               "last_api_access": WEB_SERVER.web_server_metrics[
                                                    "last_api_access_time"].replace(
-                                              tzinfo=pytz.UTC).timestamp() if self._web_server.web_server_metrics[
+                                                   tzinfo=pytz.UTC).timestamp() if WEB_SERVER.web_server_metrics[
                                                    "last_api_access_time"] else 0,
-                                          "api_access_count": self._web_server.web_server_metrics["api_access_counter"],
-                                          "last_page_access": self._web_server.web_server_metrics[
+                                               "api_access_count": WEB_SERVER.web_server_metrics["api_access_counter"],
+                                               "last_page_access": WEB_SERVER.web_server_metrics[
                                                    "last_page_access_time"].replace(
-                                              tzinfo=pytz.UTC).timestamp() if self._web_server.web_server_metrics[
+                                                   tzinfo=pytz.UTC).timestamp() if WEB_SERVER.web_server_metrics[
                                                    "last_page_access_time"] else 0,
-                                          "page_access_count": self._web_server.web_server_metrics["page_access_counter"]}
+                                               "page_access_count": WEB_SERVER.web_server_metrics[
+                                                   "page_access_counter"]}
 
         # Update Prometheus metrics
         memory_use_gauge.set(psutil.Process(os.getpid()).memory_info().rss)
-        spots_gauge.set(len(self._spots))
-        alerts_gauge.set(len(self._alerts))
+        spots_gauge.set(len(DATA_STORE.spots.values()))
+        alerts_gauge.set(len(DATA_STORE.alerts.values()))

core/url_data_cache.py

@@ -0,0 +1,25 @@
+import threading
+from datetime import timedelta
+
+from requests_cache import CachedSession
+
+from core.data_store import CACHE_DIR
+
+
+class URLDataCache(CachedSession):
+    """Cache for "semi-static" data retrieved from a URL. This is an extra layer of caching in addition to the normal
+    caching of spots, alerts and other data in the DataStore class. Its purpose is to avoid hitting remote endpoints
+    frequently when e.g. restarting Spothole many times during testing. It should only be used for data that isn't
+    expected to change on a sub-daily basis, so never for spots & alerts. It also implements a lock to allow it to be
+    used across multiple threads, though note that URL lookups will block each other this way, so it is still better to
+    create one of these objects per thread if possible."""
+
+    _lock = threading.Lock()
+
+    def __init__(self, name):
+        super().__init__(CACHE_DIR + "urls/" + name, expire_after=timedelta(days=1),
+                         allowable_codes=(200, 400, 401, 403, 404))
+
+    def get(self, *args, **kwargs):
+        with self._lock:
+            return super().get(*args, **kwargs)

core/utils.py

@@ -1,15 +1,117 @@
-def serialize_everything(obj):
-    """Convert objects to serialisable things. Used by JSON serialiser as a default when it encounters unserializable things.
-    Just converts objects to dict. Try to avoid doing anything clever here when serialising spots, because we also need
-    to receive spots without complex handling."""
-    return obj.__dict__
+import logging
+
+import simplejson
+from pyhamtools.frequency import freq_to_band
+from pyhamtools.locator import latlong_to_locator
+
+from core.constants import UNKNOWN_BAND, BANDS, CW_MODES, PHONE_MODES, DATA_MODES, MODE_ALIASES, ALL_MODES
+from core.data_store import DATA_STORE
+from data.callsign import Callsign
 
 
-def empty_queue(q):
-    """Empty a queue"""
+def safe_json_dumps(obj):
+    """Safe version of json.dumps that also converts objects to dicts so they can be output, and ignores NaN floats
+    which are invalid in JSON."""
+
+    return simplejson.dumps(obj, ensure_ascii=False, ignore_nan=True, default=lambda o: o.__dict__)
+
+
+def infer_mode_from_comment(comment):
+    """Infer a mode from the comment"""
+
+    for mode in ALL_MODES:
+        if mode in comment.upper():
+            return mode
+    for mode in MODE_ALIASES.keys():
+        if mode in comment.upper():
+            return MODE_ALIASES[mode]
+    return None
+
+
+def infer_mode_type_from_mode(mode):
+    """Infer a "mode family" from a mode."""
+
+    if mode.upper() in CW_MODES:
+        return "CW"
+    elif mode.upper() in PHONE_MODES:
+        return "PHONE"
+    elif mode.upper() in DATA_MODES:
+        return "DATA"
+    else:
+        if mode.upper() != "OTHER":
+            logging.warning("Found an unrecognised mode: " + mode + ". Developer should categorise this.")
+        return None
+
+
+def infer_band_from_freq(freq):
+    """Infer a band from a frequency in Hz"""
+
+    for b in BANDS:
+        if b.start_freq <= freq <= b.end_freq:
+            return b
+    return UNKNOWN_BAND
+
+
+def infer_mode_from_frequency(freq):
+    """Infer a mode from the frequency (in Hz) according to the band plan. Just a guess really."""
 
-    while not q.empty():
     try:
-            q.get_nowait()
-        except:
-            break
+        khz = freq / 1000.0
+        mode = freq_to_band(khz)["mode"]
+        # Some additional common digimode ranges in addition to what the 3rd-party freq_to_band function returns.
+        # This is mostly here just because freq_to_band is very specific about things like FT8 frequencies, and e.g.
+        # a spot at 7074.5 kHz will be indicated as LSB, even though it's clearly in the FT8 range. Future updates
+        # might include other common digimode centres of activity here, but this achieves the main goal of keeping
+        # large numbers of clearly-FT* spots off the list of people filtering out digimodes.
+        if (7074 <= khz < 7077) or (10136 <= khz < 10139) or (14074 <= khz < 14077) or (18100 <= khz < 18103) or (
+                21074 <= khz < 21077) or (24915 <= khz < 24918) or (28074 <= khz < 28077):
+            mode = "FT8"
+        if (7047.5 <= khz < 7050.5) or (10140 <= khz < 10143) or (14080 <= khz < 14083) or (
+                18104 <= khz < 18107) or (21140 <= khz < 21143) or (24919 <= khz < 24922) or (28180 <= khz < 28183):
+            mode = "FT4"
+        return mode
+    except KeyError:
+        return None
+
+
+def get_flag_for_dxcc(dxcc):
+    """Get an emoji flag for a given DXCC entity ID"""
+
+    dxcc_data = DATA_STORE.dxcc_data[dxcc] if dxcc in DATA_STORE.dxcc_data else None
+    return dxcc_data["flag"] if dxcc_data else None
+
+
+def get_callsign_object_from_pyhamtools_callinfo(callsign, callinfo):
+    """Utility function to take the data provided by a PyHamTools CallInfo object and populate our own Callsign data
+    object from it"""
+
+    try:
+        home_call = callinfo.get_homecall(callsign)
+        data = callinfo.get_all(callsign)
+
+        country = data["country"] if "country" in data else None
+        dxcc_id = data["adif"] if "adif" in data else None
+        continent = data["continent"] if "continent" in data else None
+        cq_zone = data["cqz"] if "cqz" in data else None
+        itu_zone = data["ituz"] if "ituz" in data else None
+        lat = float(data["latitude"]) if "latitude" in data else None
+        lon = float(data["longitude"]) if "longitude" in data else None
+        grid = None
+        if lat and lon:
+            grid = latlong_to_locator(lat, lon)
+
+        return Callsign(call=callsign,
+                        home_call=home_call,
+                        country=country,
+                        dxcc_id=dxcc_id,
+                        continent=continent,
+                        cq_zone=cq_zone,
+                        itu_zone=itu_zone,
+                        latitude=lat,
+                        longitude=lon,
+                        grid=grid,
+                        location_source="DXCC")
+
+    except (KeyError, ValueError):
+        # Unknown callsign, can't look anything up, return a Callsign object with basic data so that gets cached
+        return Callsign(call=callsign)

data/alert.py

@@ -1,13 +1,15 @@
 import copy
 import hashlib
 import json
+import logging
 from dataclasses import dataclass
 from datetime import datetime, timedelta
 
 import pytz
 
-from core.lookup_helper import lookup_helper
-from core.sig_utils import populate_sig_ref_info
+from core.call_lookup_helper import get_call_info
+from core.sig_lookup_helper import populate_missing_sig_ref_info
+from core.utils import get_flag_for_dxcc
 
 
 @dataclass
@@ -15,55 +17,56 @@ class Alert:
     """Data class that defines an alert."""
 
     # Unique identifier for the alert
-    id: str = None
+    id: str | None = None
     # Callsigns of the operators that has been alerted
-    dx_calls: list = None
+    dx_calls: list | None = None
     # Names of the operators that has been alerted
-    dx_names: list = None
+    dx_names: list | None = None
     # Country of the DX operator
-    dx_country: str = None
+    dx_country: str | None = None
     # Country flag of the DX operator
-    dx_flag: str = None
+    dx_flag: str | None = None
     # Continent of the DX operator
-    dx_continent: str = None
+    dx_continent: str | None = None
     # DXCC ID of the DX operator
-    dx_dxcc_id: int = None
+    dx_dxcc_id: int | None = None
     # CQ zone of the DX operator
-    dx_cq_zone: int = None
+    dx_cq_zone: int | None = None
     # ITU zone of the DX operator
-    dx_itu_zone: int = None
+    dx_itu_zone: int | None = None
     # Intended frequencies & modes of operation. Essentially just a different kind of comment field.
-    freqs_modes: str = None
+    freqs_modes: str | None = None
     # Start time of the activation, UTC seconds since UNIX epoch
-    start_time: float = None
+    start_time: float | None = None
     # Start time of the activation of the alert, ISO 8601
-    start_time_iso: str = None
+    start_time_iso: str | None = None
     # End time of the activation, UTC seconds since UNIX epoch. Optional
-    end_time: float = None
+    end_time: float | None = None
     # End time of the activation of the alert, ISO 8601
-    end_time_iso: str = None
+    end_time_iso: str | None = None
     # Time that this software received the alert, UTC seconds since UNIX epoch. This is used with the "since_received"
     # call to our API to receive all data that is new to us, even if by a quirk of the API it might be older than the
     # list time the client polled the API.
-    received_time: float = None
+    received_time: float | None = None
     # Time that this software received the alert, ISO 8601
-    received_time_iso: str = None
+    received_time_iso: str | None = None
     # Comment made by the alerter, if any
-    comment: str = None
+    comment: str | None = None
     # Special Interest Group (SIG), e.g. outdoor activity programme such as POTA
-    sig: str = None
+    sig: str | None = None
     # SIG references. We allow multiple here for e.g. n-fer activations, unlike ADIF SIG_INFO
-    sig_refs: list = None
+    sig_refs: list | None = None
     # Whether this alert is for a DXpedition, as opposed to e.g. an xOTA programme.
     is_dxpedition: bool = False
     # Where we got the alert from, e.g. "POTA", "SOTA"...
-    source: str = None
+    source: str | None = None
     # The ID the source gave it, if any.
-    source_id: str = None
+    source_id: str | None = None
 
     def infer_missing(self, credentials=None):
         """Infer missing parameters where possible"""
 
+        try:
             # If we somehow don't have a start time, set it to zero so it sorts off the bottom of any list but
             # clients can still reliably parse it as a number.
             if not self.start_time:
@@ -83,25 +86,26 @@ class Alert:
 
             # DX country, continent, zones etc. from callsign. CQ/ITU zone are better looked up with a location but we don't
             # have a real location for alerts.
+            call_info = get_call_info(self.dx_calls[0], credentials)
             if self.dx_calls and self.dx_calls[0] and not self.dx_country:
-            self.dx_country = lookup_helper.infer_country_from_callsign(self.dx_calls[0], credentials)
+                self.dx_country = call_info.country
             if self.dx_calls and self.dx_calls[0] and not self.dx_continent:
-            self.dx_continent = lookup_helper.infer_continent_from_callsign(self.dx_calls[0], credentials)
+                self.dx_continent = call_info.continent
             if self.dx_calls and self.dx_calls[0] and not self.dx_cq_zone:
-            self.dx_cq_zone = lookup_helper.infer_cq_zone_from_callsign(self.dx_calls[0], credentials)
+                self.dx_cq_zone = call_info.cq_zone
             if self.dx_calls and self.dx_calls[0] and not self.dx_itu_zone:
-            self.dx_itu_zone = lookup_helper.infer_itu_zone_from_callsign(self.dx_calls[0], credentials)
+                self.dx_itu_zone = call_info.itu_zone
             if self.dx_calls and self.dx_calls[0] and not self.dx_dxcc_id:
-            self.dx_dxcc_id = lookup_helper.infer_dxcc_id_from_callsign(self.dx_calls[0], credentials)
+                self.dx_dxcc_id = call_info.dxcc_id
             if self.dx_dxcc_id and not self.dx_flag:
-            self.dx_flag = lookup_helper.get_flag_for_dxcc(self.dx_dxcc_id)
+                self.dx_flag = get_flag_for_dxcc(self.dx_dxcc_id)
 
             # Fetch SIG data. In case a particular API doesn't provide a full set of name, lat, lon & grid for a reference
             # in its initial call, we use this code to populate the rest of the data. This includes working out grid refs
             # from WAB and WAI, which count as a SIG even though there's no real lookup, just maths
             if self.sig_refs and len(self.sig_refs) > 0:
                 for sig_ref in self.sig_refs:
-                populate_sig_ref_info(sig_ref)
+                    populate_missing_sig_ref_info(sig_ref)
 
             # If the spot itself doesn't have a SIG yet, but we have at least one SIG reference, take that reference's SIG
             # and apply it to the whole spot.
@@ -121,12 +125,13 @@ class Alert:
                 self_copy.received_time_iso = ""
                 self.id = hashlib.sha256(str(self_copy).encode("utf-8")).hexdigest()
 
-        # DX operator details lookup, using QRZ.com/HamQTH. This should be the last resort compared to taking the data
-        # from the actual alerting service, e.g. we don't want to accidentally use a user's QRZ.com home lat/lon
-        # instead of the one from the park reference they're at.
+            # DX operator name lookup, using QRZ.com/HamQTH.
             if self.dx_calls and not self.dx_names:
                 self.dx_names = list(
-                map(lambda c: lookup_helper.infer_name_from_callsign_online_lookup(c, credentials), self.dx_calls))
+                    map(lambda c: get_call_info(c, credentials).name, self.dx_calls))
+
+        except Exception as e:
+            logging.error("Exception while inferring missing data from spot", e, exc_info=True)
 
     def to_json(self):
         """JSON serialise"""

data/band.py

@@ -11,3 +11,5 @@ class Band:
     start_freq: float
     # Stop frequency, in Hz
     end_freq: float
+    # Whether this is an HF amateur radio band
+    is_ham_hf: bool = False

data/callsign.py

@@ -0,0 +1,47 @@
+from dataclasses import dataclass
+
+
+@dataclass
+class Callsign:
+    """Data class that defines a callsign and the data associated with it. This will have been retrieved by a callsign
+    lookup provider using data files or online lookup. This can be used to infer missing data for a spot, though if the
+    spot has a SIG (e.g. POTA) reference this data for their home location (or even just their country) will be less
+    accurate and should not be used in preference to that."""
+
+    # Callsign as spotted
+    call: str
+    # "Home" call, i.e. with any prefixes and suffixes stripped off
+    home_call: str | None = None
+    # Operator name
+    name : str | None = None
+    # QTH (location), free text
+    qth : str | None = None
+    # Maidenhead grid locator. Depending on the source of lookup this could be a home location from QRZ/HamQTH or just
+    # the centre of the country they're operating in if no other data is available.
+    grid : str | None = None
+    # Latitude. Depending on the source of lookup this could be a home location from QRZ/HamQTH or just
+    # the centre of the country they're operating in if no other data is available.
+    latitude : float | None = None
+    # Longitude. Depending on the source of lookup this could be a home location from QRZ/HamQTH or just
+    # the centre of the country they're operating in if no other data is available.
+    longitude : float | None = None
+    # Country in which the callsign indicates they are operating
+    country: str | None = None
+    # Continent in which the callsign indicates they are operating
+    continent: str | None = None
+    # DXCC ID in which the callsign indicates they are operating
+    dxcc_id: int | None = None
+    # CQ zone in which the callsign indicates they are operating
+    cq_zone: int | None = None
+    # ITU zone in which the callsign indicates they are operating
+    itu_zone: int | None = None
+    # Location source. This can be "HOME QTH" or "DXCC" depending on which provider gave us a location
+    location_source: str | None = None
+
+    def fully_populated(self):
+        """Utility method to indicate that the callsign data is fully populated. Multiple providers can return data for
+        a callsign, and we try them in sequence until we have all the data we can, in which case there's no point
+        querying any other providers."""
+        return self.home_call is not None and self.name is not None and self.qth is not None and self.grid is not None\
+            and self.latitude is not None and self.longitude is not None and self.country is not None and self.continent\
+            is not None and self.dxcc_id is not None and self.cq_zone is not None and self.itu_zone is not None

data/sig.py

@@ -1,13 +1,21 @@
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 
 
 @dataclass
 class SIG:
-    """Data class that defines a Special Interest Group."""
+    """Data class that defines a Special Interest Group. Each contains a name and a longer form description.
+     They also contain comment_names which attempts to separate out the way people might refer to it in
+     cluster comments from how it is referred to in the UI & API. (For example, "TOTA" in cluster spot comments
+     almost always means Towers on the Air, but no single programme is referred to in the UI as "TOTA" as
+     it's ambiguous between Towers, Toilets and Tiles. And while Beaches got the name "BOTA" first, "BOTA" spots
+     are much more likely to be bunkers.) Finally, there is a ref_regex which provides a regular expression to
+     match what references (such as parks and summits) look like for that programme."""
 
-    # SIG name, e.g. "POTA"
+    # SIG name as used in the UI and API, e.g. "Towers"
     name: str
-    # Description, e.g. "Parks on the Air"
+    # Description, e.g. "Towers on the Air"
     description: str
+    # SIG names as they might appear in cluster spot comments, e.g. ["TOTA"]
+    comment_names: list[str] = field(default_factory=list)
     # Regex matcher for references, e.g. for POTA r"[A-Z]{2}\-\d+".
-    ref_regex: str = None
+    ref_regex: str | None = None

data/sig_ref.py

@@ -11,14 +11,14 @@ class SIGRef:
     # SIG that this reference is in, e.g. "POTA".
     sig: str
     # Name of the reference, e.g. "Null Country Park", if known.
-    name: str = None
+    name: str | None = None
     # URL to look up more information about the reference, if known.
-    url: str = None
+    url: str | None = None
     # Latitude of the reference, if known.
-    latitude: float = None
+    latitude: float | None = None
     # Longitude of the reference, if known.
-    longitude: float = None
+    longitude: float | None = None
     # Maidenhead grid reference of the reference, if known.
-    grid: str = None
+    grid: str | None = None
     # Activation score. SOTA only
-    activation_score: int = None
+    activation_score: int | None = None

data/solar_conditions.py

@@ -110,11 +110,11 @@ class HFBandCondition:
     """Data class representing HF propagation conditions for certain bands and time of day."""
 
     # Band name, e.g. "80m-40m", "20m-17m", "10m-6m"
-    band: str = None
+    band: str | None = None
     # Time of day: "day" or "night"
-    time: str = None
+    time: str | None = None
     # Propagation condition: "Good", "Fair", or "Poor"
-    condition: str = None
+    condition: str | None = None
 
 
 @dataclass
@@ -122,65 +122,66 @@ class SolarConditions:
     """Data class representing current solar and propagation conditions."""
 
     # Time the data was last updated at the source, UTC seconds since UNIX epoch
-    updated: float = None
+    updated: float | None = None
     # Solar Flux Index (SFI)
-    sfi: int = None
+    sfi: int | None = None
     # A-index (daily geomagnetic activity)
-    a_index: int = None
+    a_index: int | None = None
     # K-index (3-hour geomagnetic activity)
-    k_index: int = None
+    k_index: int | None = None
     # X-ray flux class, e.g. "B2.3", "C1.0"
-    xray: str = None
+    xray: str | None = None
     # Proton flux
-    proton_flux: int = None
+    proton_flux: int | None = None
     # Electron flux
-    electron_flux: int = None
+    electron_flux: int | None = None
     # Aurora activity level
-    aurora: int = None
+    aurora: int | None = None
     # Latitude in degrees of the aurora boundary
-    aurora_latitude: float = None
+    aurora_latitude: float | None = None
     # Sunspot count
-    sunspots: int = None
+    sunspots: int | None = None
     # Solar wind speed in km/s
-    solar_wind: float = None
+    solar_wind: float | None = None
     # Interplanetary magnetic field strength in nT
-    magnetic_field: float = None
+    magnetic_field: float | None = None
     # Geomagnetic field condition, e.g. "Quiet", "Unsettled", "Active", "Storm"
-    geomag_field: str = None
+    geomag_field: str | None = None
     # Geomagnetic background noise level, e.g. "S0", "S1", "S2"
-    geomag_noise: str = None
+    geomag_noise: str | None = None
     # HF band propagation conditions, keyed by "{band}-{time}" e.g. "80m-40m-day"
-    hf_conditions: dict = None
+    hf_conditions: dict | None = None
     # VHF propagation conditions, keyed by condition name
-    vhf_conditions: dict = None
+    vhf_conditions: dict | None = None
     # NOAA Kp index 3-day forecast, keyed by UNIX timestamp of the start of each 3-hour UTC period
-    k_index_forecast: dict = None
+    k_index_forecast: dict | None = None
     # NOAA Solar Radiation Storm (S1 or greater) probability forecast, keyed by UNIX timestamp of start of day UTC
-    solar_storm_forecast: dict = None
+    solar_storm_forecast: dict | None = None
     # NOAA Radio Blackout (R1-R2) probability forecast, keyed by UNIX timestamp of start of day UTC
-    blackout_forecast_r1r2: dict = None
+    blackout_forecast_r1r2: dict | None = None
     # NOAA Radio Blackout (R3 or greater) probability forecast, keyed by UNIX timestamp of start of day UTC
-    blackout_forecast_r3_or_greater: dict = None
-    # Ionosonde measurements from LGDC, list of dicts with keys: ursi, name, fof2, muf
-    ionosonde_data: list = None
+    blackout_forecast_r3_or_greater: dict | None = None
+    # Ionosonde measurements, dict keyed by URSI code, values are dicts with keys: ursi, name, fof2, muf, luf,
+    # band_states. Populated by GIROIonosonde or KC2GProp providers.
+    ionosonde_data: dict | None = None
 
     # Derived values (populated by infer_descriptions())
     # HF radio blackout risk description, derived from xray
-    xray_desc: str = None
+    xray_desc: str | None = None
     # HF radio blackout scale number (R0-R5), derived from xray
-    radio_blackout_scale: int = None
+    radio_blackout_scale: int | None = None
     # Solar radiation storm level description, derived from proton_flux
-    proton_flux_desc: str = None
+    proton_flux_desc: str | None = None
     # Solar radiation storm scale number (S0-S5), derived from proton_flux
-    solar_storm_scale: int = None
+    solar_storm_scale: int | None = None
     # Geomagnetic storm level description, derived from k_index
-    geomag_storm_desc: str = None
+    geomag_storm_desc: str | None = None
     # Geomagnetic storm scale number (G0-G5), derived from k_index
-    geomag_storm_scale: int = None
+    geomag_storm_scale: int | None = None
     # Overall HF band conditions summary, derived from sfi
-    band_conditions_desc: str = None
+    band_conditions_desc: str | None = None
     # Electron flux description, derived from electron_flux
-    electron_flux_desc: str = None
+    electron_flux_desc: str | None = None
 
     def infer_descriptions(self):
         """Populate derived text description fields from the current numeric/raw field values."""
@@ -196,6 +197,7 @@ class SolarConditions:
         self.electron_flux_desc = _lookup_by_threshold(self.electron_flux, ELECTRON_FLUX_DESCRIPTIONS)
 
     def to_json(self):
-        """JSON serialise"""
+        """JSON serialise. Dict key order is insertion order (Python 3.7+ guarantee), so callers receive
+        fields in a predictable, logical sequence without relying on sort_keys."""
 
-        return json.dumps(self, default=lambda o: o.__dict__, sort_keys=True)
+        return json.dumps(self, default=lambda o: o.__dict__)

data/spot.py

@@ -10,12 +10,13 @@ import pytz
 from pyhamtools.locator import locator_to_latlong, latlong_to_locator
 
 from core.config import MAX_SPOT_AGE
-from core.constants import MODE_ALIASES
+from core.constants import MODE_ALIASES, PROPAGATION_MODES
 from core.geo_utils import lat_lon_to_cq_zone, lat_lon_to_itu_zone
-from core.lookup_helper import lookup_helper, infer_band_from_freq, infer_mode_from_comment, \
-    infer_mode_from_frequency, infer_mode_type_from_mode
-from data.lookup_credentials import LookupCredentials
-from core.sig_utils import populate_sig_ref_info, ANY_SIG_REGEX, get_ref_regex_for_sig
+from core.call_lookup_helper import get_call_info
+from core.sig_utils import ANY_SIG_REGEX, get_ref_regex_for_sig, get_sig_name_from_comment_name
+from core.sig_lookup_helper import populate_missing_sig_ref_info
+from core.utils import infer_band_from_freq, infer_mode_from_comment, \
+    infer_mode_from_frequency, infer_mode_type_from_mode, get_flag_for_dxcc
 from data.sig_ref import SIGRef
 
 
@@ -24,38 +25,38 @@ class Spot:
     """Data class that defines a spot."""
 
     # Unique identifier for the spot
-    id: str = None
+    id: str | None = None
 
     # DX (spotted) operator info
 
     # Callsign of the operator that has been spotted
-    dx_call: str = None
+    dx_call: str | None = None
     # Name of the operator that has been spotted
-    dx_name: str = None
+    dx_name: str | None = None
     # QTH of the operator that has been spotted. This could be from any SIG refs or could be from online lookup of their
     # home QTH.
-    dx_qth: str = None
+    dx_qth: str | None = None
     # Country of the DX operator
-    dx_country: str = None
+    dx_country: str | None = None
     # Country flag of the DX operator
-    dx_flag: str = None
+    dx_flag: str | None = None
     # Continent of the DX operator
-    dx_continent: str = None
+    dx_continent: str | None = None
     # DXCC ID of the DX operator
-    dx_dxcc_id: int = None
+    dx_dxcc_id: int | None = None
     # CQ zone of the DX operator
-    dx_cq_zone: int = None
+    dx_cq_zone: int | None = None
     # ITU zone of the DX operator
-    dx_itu_zone: int = None
+    dx_itu_zone: int | None = None
     # If this is an APRS/Packet/etc spot, what SSID was the DX operator using?
-    dx_ssid: str = None
+    dx_ssid: str | None = None
     # Maidenhead grid locator for the DX.  This could be from a geographical reference e.g. POTA, or just from the
     # country
-    dx_grid: str = None
+    dx_grid: str | None = None
     # Latitude & longitude of the DX, in degrees. This could be from a geographical reference e.g. POTA, or from a QRZ
     # lookup
-    dx_latitude: float = None
-    dx_longitude: float = None
+    dx_latitude: float | None = None
+    dx_longitude: float | None = None
     # DX Location source. Indicates how accurate the location might be. Values: "SPOT", "WAB/WAI GRID", "HOME QTH",
     # "DXCC", "NONE"
     dx_location_source: str = "NONE"
@@ -67,74 +68,88 @@ class Spot:
     # DE (Spotter) info
 
     # Callsign of the spotter
-    de_call: str = None
+    de_call: str | None = None
     # Country of the spotter
-    de_country: str = None
+    de_country: str | None = None
     # Country flag of the spotter
-    de_flag: str = None
+    de_flag: str | None = None
     # Continent of the spotter
-    de_continent: str = None
+    de_continent: str | None = None
     # DXCC ID of the spotter
-    de_dxcc_id: int = None
+    de_dxcc_id: int | None = None
     # If this is an APRS/Packet/etc spot, what SSID was the spotter/receiver using?
-    de_ssid: str = None
+    de_ssid: str | None = None
     # Maidenhead grid locator for the spotter. This is not going to be from a xOTA reference so it will likely just be
     # a QRZ or DXCC lookup. If the spotter is also portable, this is probably wrong, but it's good enough for some
     # simple mapping.
-    de_grid: str = None
+    de_grid: str | None = None
     # Latitude & longitude of the DX, in degrees. This is not going to be from a xOTA reference so it will likely just
     # be a QRZ or DXCC lookup. If the spotter is also portable, this is probably wrong, but it's good enough for some
     # simple mapping.
-    de_latitude: float = None
-    de_longitude: float = None
+    de_latitude: float | None = None
+    de_longitude: float | None = None
 
     # General QSO info
 
     # Reported mode, such as SSB, PHONE, CW, FT8...
-    mode: str = None
+    mode: str | None = None
     # Inferred mode "family". One of "CW", "PHONE" or "DIGI".
-    mode_type: str = None
+    mode_type: str | None = None
     # Source of the mode information. "SPOT", "COMMENT", "BANDPLAN" or "NONE"
     mode_source: str = "NONE"
     # Frequency, in Hz
-    freq: float = None
+    freq: float | None = None
     # Band, defined by the frequency, e.g. "40m" or "70cm"
-    band: str = None
+    band: str | None = None
+    # Propagation mode, if known
+    propagation_mode: str | None = None
     # Comment left by the spotter, if any
-    comment: str = None
+    comment: str | None = None
     # QRT state. Some APIs return spots marked as QRT. Otherwise we can check the comments.
     qrt: bool = False
 
     # Special Interest Group info
 
     # Special Interest Group (SIG), e.g. outdoor activity programme such as POTA
-    sig: str = None
+    sig: str | None = None
     # SIG references. We allow multiple here for e.g. n-fer activations, unlike ADIF SIG_INFO
-    sig_refs: list = None
+    sig_refs: list | None = None
 
     # Timing info
 
     # Time of the spot, UTC seconds since UNIX epoch
-    time: float = None
+    time: float | None = None
     # Time of the spot, ISO 8601
-    time_iso: str = None
+    time_iso: str | None = None
     # Time that this software received the spot, UTC seconds since UNIX epoch. This is used with the "since_received"
     # call to our API to receive all data that is new to us, even if by a quirk of the API it might be older than the
     # list time the client polled the API.
-    received_time: float = None
+    received_time: float | None = None
     # Time that this software received the spot, ISO 8601
-    received_time_iso: str = None
+    received_time_iso: str | None = None
 
     # Source info
 
     # Where we got the spot from, e.g. "POTA", "Cluster"...
-    source: str = None
+    source: str | None = None
     # The ID the source gave it, if any.
-    source_id: str = None
+    source_id: str | None = None
+
+    def __post_init__(self):
+        """Normalise fields that don't survive a plain dict to Spot conversion. This is used in the "add spot" API
+        endpoint where the client is submitting JSON, and we want to recreate a full Spot object, including nested
+        objects such as the sig_refs list.."""
+
+        if self.sig_refs:
+            self.sig_refs = [
+                sig_ref if isinstance(sig_ref, SIGRef) else SIGRef(**sig_ref)
+                for sig_ref in self.sig_refs
+            ]
 
     def infer_missing(self, credentials=None):
         """Infer missing parameters where possible"""
 
+        try:
             # If we somehow don't have a spot time, set it to zero so it sorts off the bottom of any list but
             # clients can still reliably parse it as a number.
             if not self.time:
@@ -158,14 +173,15 @@ class Spot:
                     self.dx_ssid = split[1]
 
             # DX country, continent etc. from callsign
+            dx_call_info = get_call_info(self.dx_call, credentials)
             if self.dx_call and not self.dx_country:
-            self.dx_country = lookup_helper.infer_country_from_callsign(self.dx_call, credentials)
+                self.dx_country = dx_call_info.country
             if self.dx_call and not self.dx_continent:
-            self.dx_continent = lookup_helper.infer_continent_from_callsign(self.dx_call, credentials)
+                self.dx_continent = dx_call_info.continent
             if self.dx_call and not self.dx_dxcc_id:
-            self.dx_dxcc_id = lookup_helper.infer_dxcc_id_from_callsign(self.dx_call, credentials)
+                self.dx_dxcc_id = dx_call_info.dxcc_id
             if self.dx_dxcc_id and not self.dx_flag:
-            self.dx_flag = lookup_helper.get_flag_for_dxcc(self.dx_dxcc_id)
+                self.dx_flag = get_flag_for_dxcc(self.dx_dxcc_id)
 
             # Clean up spotter call if it has an SSID or -# from RBN
             if self.de_call and "-" in self.de_call:
@@ -190,16 +206,17 @@ class Spot:
 
             # Spotter country, continent, zones etc. from callsign.
             # DE call with no digits, or APRS servers starting "T2" are not things we can look up location for
+            de_call_info = get_call_info(self.de_call, credentials)
             if self.de_call and any(char.isdigit() for char in self.de_call) and not (
                     self.de_call.startswith("T2") and self.source == "APRS-IS"):
                 if not self.de_country:
-                self.de_country = lookup_helper.infer_country_from_callsign(self.de_call, credentials)
+                    self.de_country = de_call_info.country
                 if not self.de_continent:
-                self.de_continent = lookup_helper.infer_continent_from_callsign(self.de_call, credentials)
+                    self.de_continent = de_call_info.continent
                 if not self.de_dxcc_id:
-                self.de_dxcc_id = lookup_helper.infer_dxcc_id_from_callsign(self.de_call, credentials)
+                    self.de_dxcc_id = de_call_info.dxcc_id
                 if self.de_dxcc_id and not self.de_flag:
-                self.de_flag = lookup_helper.get_flag_for_dxcc(self.de_dxcc_id)
+                    self.de_flag = get_flag_for_dxcc(self.de_dxcc_id)
 
             # Remove NaNs in frequency
             if self.freq and self.freq == float("nan"):
@@ -254,7 +271,7 @@ class Spot:
                 for sig_match in sig_matches:
                     # First of all, if we haven't got a SIG for this spot set yet, now we have. This covers things like cluster
                     # spots where the comment is just "POTA".
-                found_sig = sig_match.group(2).upper()
+                    found_sig = get_sig_name_from_comment_name(sig_match.group(2))
                     if not self.sig:
                         self.sig = found_sig
 
@@ -262,7 +279,7 @@ class Spot:
                     # If so, add that to the sig_refs list for this spot.
                     ref_regex = get_ref_regex_for_sig(found_sig)
                     if ref_regex:
-                    ref_matches = re.finditer(r"(^|\W)" + found_sig + r"($|\W)(" + ref_regex + r")($|\W)", self.comment,
+                        ref_matches = re.finditer(r"(^|\W)" + found_sig + r"([ -])(" + ref_regex + r")($|\W)", self.comment,
                                                   re.IGNORECASE)
                         for ref_match in ref_matches:
                             self._append_sig_ref_if_missing(SIGRef(id=ref_match.group(3).upper(), sig=found_sig))
@@ -272,7 +289,7 @@ class Spot:
             # from WAB and WAI, which count as a SIG even though there's no real lookup, just maths
             if self.sig_refs and len(self.sig_refs) > 0:
                 for sig_ref in self.sig_refs:
-                sig_ref = populate_sig_ref_info(sig_ref)
+                    sig_ref = populate_missing_sig_ref_info(sig_ref)
                     # If the spot itself doesn't have location yet, but the SIG ref does, extract it
                     if sig_ref.grid and not self.dx_grid:
                         self.dx_grid = sig_ref.grid
@@ -289,6 +306,30 @@ class Spot:
             if self.sig_refs and len(self.sig_refs) > 0 and not self.sig:
                 self.sig = self.sig_refs[0].sig
 
+            # Parse "de_grid<prop_mode>dx_grid" structures from the comment, e.g. "JN61ES(ES)JM56XT" or "JO02GQ<>KN17LG".
+            # These are common on cluster spots and can provide grid references in preference to e.g. QRZ lookup, as well as
+            # being the only source we have for propagation mode. Brace for nightmare regex from hell.
+            if self.comment:
+                grid_mode_grid_match = re.search(
+                    r'\b([A-Ra-r]{2}\d{2}(?:[A-Xa-x]{2}(?:\d{2})?)?)(?:<([^>]*)>|\(([^)]*)\))([A-Ra-r]{2}\d{2}(?:[A-Xa-x]{2}(?:\d{2})?)?)\b',
+                    self.comment)
+                if grid_mode_grid_match:
+                    # regex matches, so extract grids:
+                    if not self.de_grid:
+                        self.de_grid = grid_mode_grid_match.group(1).upper()
+                    if not self.dx_grid:
+                        self.dx_grid = grid_mode_grid_match.group(4).upper()
+                        self.dx_location_source = "SPOT"
+
+                    # And extract propagation mode (group 2 for <...>, group 3 for (...)):
+                    mode_tag = (grid_mode_grid_match.group(2) or grid_mode_grid_match.group(3) or "").upper()
+                    if mode_tag and not self.propagation_mode:
+                        if mode_tag in PROPAGATION_MODES:
+                            self.propagation_mode = PROPAGATION_MODES[mode_tag]
+                        else:
+                            self.propagation_mode = mode_tag
+                            logging.info("Seen a new propagation mode tag not yet in the system: %s", mode_tag)
+
             # DX Grid to lat/lon and vice versa in case one is missing
             if self.dx_grid and not self.dx_latitude:
                 try:
@@ -320,84 +361,57 @@ class Spot:
                 self_copy.received_time_iso = ""
                 self.id = hashlib.sha256(str(self_copy).encode("utf-8")).hexdigest()
 
-        # DX operator details lookup, using QRZ.com/HamQTH. This should be the last resort compared to taking the data
-        # from the actual spotting service, e.g. we don't want to accidentally use a user's QRZ.com home lat/lon
+            # DX operator details lookup. This should be the last resort compared to taking the data from the actual
+            # spotting service, e.g. we don't want to accidentally use a user's QRZ.com home lat/lon or DXCC lat/lon
             # instead of the one from the park reference they're at.
             if self.dx_call and not self.dx_name:
-            self.dx_name = lookup_helper.infer_name_from_callsign_online_lookup(self.dx_call, credentials)
+                self.dx_name = dx_call_info.name
             if self.dx_call and not self.dx_latitude:
-            latlon = lookup_helper.infer_latlon_from_callsign_online_lookup(self.dx_call, credentials)
-            if latlon:
-                self.dx_latitude = latlon[0]
-                self.dx_longitude = latlon[1]
-            self.dx_grid = lookup_helper.infer_grid_from_callsign_online_lookup(self.dx_call, credentials)
-            self.dx_location_source = "HOME QTH"
+                self.dx_latitude = dx_call_info.latitude
+                self.dx_longitude = dx_call_info.longitude
+                self.dx_grid = dx_call_info.grid
+                self.dx_location_source = dx_call_info.location_source
 
             # Determine a "QTH" string. If we have a SIG ref, pick the first one and turn it into a suitable string,
             # otherwise see what they have set on an online lookup service.
             if self.sig_refs and len(self.sig_refs) > 0:
-            self.dx_qth = self.sig_refs[0].id
+                qth = self.sig_refs[0].id
                 if self.sig_refs[0].name:
-                self.dx_qth = self.dx_qth + " " + self.sig_refs[0].name
+                    qth += " " + self.sig_refs[0].name
+                self.dx_qth = qth
             else:
-            self.dx_qth = lookup_helper.infer_qth_from_callsign_online_lookup(self.dx_call, credentials)
-
-        # Last resort for getting a DX position, use the DXCC entity.
-        if self.dx_call and not self.dx_latitude:
-            latlon = lookup_helper.infer_latlon_from_callsign_dxcc(self.dx_call)
-            if latlon:
-                self.dx_latitude = latlon[0]
-                self.dx_longitude = latlon[1]
-                self.dx_grid = lookup_helper.infer_grid_from_callsign_dxcc(self.dx_call)
-                self.dx_location_source = "DXCC"
-
-        # It looks like we can sometimes get a string into lat/lon, so try to parse as float, reject if not valid
-        if isinstance(self.dx_latitude, str) or isinstance(self.dx_longitude, str):
-            try:
-                self.dx_latitude = float(self.dx_latitude)
-                self.dx_longitude = float(self.dx_longitude)
-            except (TypeError, ValueError):
-                logging.warning("Received non-numeric strings in lat/lon (" + str(self.dx_latitude) + ", " + str(self.dx_longitude) + ") for call " + self.dx_call + ", rejecting it")
-                self.dx_latitude = None
-                self.dx_longitude = None
+                self.dx_qth = dx_call_info.qth
 
             # CQ and ITU zone lookup, preferably from location but failing that, from callsign
             if not self.dx_cq_zone:
                 if self.dx_latitude:
                     self.dx_cq_zone = lat_lon_to_cq_zone(self.dx_latitude, self.dx_longitude)
                 elif self.dx_call:
-                self.dx_cq_zone = lookup_helper.infer_cq_zone_from_callsign(self.dx_call, credentials)
+                    self.dx_cq_zone = dx_call_info.cq_zone
             if not self.dx_itu_zone:
                 if self.dx_latitude:
                     self.dx_itu_zone = lat_lon_to_itu_zone(self.dx_latitude, self.dx_longitude)
                 elif self.dx_call:
-                self.dx_itu_zone = lookup_helper.infer_itu_zone_from_callsign(self.dx_call, credentials)
+                    self.dx_itu_zone = dx_call_info.itu_zone
 
             # DX Location is "good" if it is from a spot, or from QRZ if the callsign doesn't contain a slash, so the operator
             # is likely at home.
-        self.dx_location_good = self.dx_latitude and self.dx_longitude and (
+            self.dx_location_good = bool(self.dx_latitude and self.dx_longitude and (
                     self.dx_location_source == "SPOT" or self.dx_location_source == "SIG REF LOOKUP"
                     or self.dx_location_source == "WAB/WAI GRID"
-                or (self.dx_location_source == "HOME QTH" and not "/" in self.dx_call))
+                    or (self.dx_location_source == "HOME QTH" and "/" not in (self.dx_call or ""))))
 
             # DE with no digits and APRS servers starting "T2" are not things we can look up location for
             if self.de_call and any(char.isdigit() for char in self.de_call) and not (
                     self.de_call.startswith("T2") and self.source == "APRS-IS"):
-            # DE operator position lookup, using QRZ.com/HamQTH.
+                # DE operator location lookup
                 if not self.de_latitude:
-                latlon = lookup_helper.infer_latlon_from_callsign_online_lookup(self.de_call, credentials)
-                if latlon:
-                    self.de_latitude = latlon[0]
-                    self.de_longitude = latlon[1]
-                self.de_grid = lookup_helper.infer_grid_from_callsign_online_lookup(self.de_call, credentials)
+                    self.de_latitude = de_call_info.latitude
+                    self.de_longitude = de_call_info.longitude
+                    self.de_grid = de_call_info.grid
 
-            # Last resort for getting a DE position, use the DXCC entity.
-            if not self.de_latitude:
-                latlon = lookup_helper.infer_latlon_from_callsign_dxcc(self.de_call)
-                if latlon:
-                    self.de_latitude = latlon[0]
-                    self.de_longitude = latlon[1]
-                    self.de_grid = lookup_helper.infer_grid_from_callsign_dxcc(self.de_call)
+        except Exception as e:
+            logging.error("Exception while inferring missing data from spot", e, exc_info=True)
 
     def to_json(self):
         """JSON serialise"""
@@ -407,16 +421,16 @@ class Spot:
     def _append_sig_ref_if_missing(self, new_sig_ref):
         """Append a sig_ref to the list, so long as it's not already there."""
 
-        if not self.sig_refs:
-            self.sig_refs = []
+        sig_refs = self.sig_refs or []
+        self.sig_refs = sig_refs
         new_sig_ref.id = new_sig_ref.id.strip().upper()
         new_sig_ref.sig = new_sig_ref.sig.strip().upper()
         if new_sig_ref.id == "":
             return
-        for sig_ref in self.sig_refs:
+        for sig_ref in sig_refs:
             if sig_ref.id == new_sig_ref.id and sig_ref.sig == new_sig_ref.sig:
                 return
-        self.sig_refs.append(new_sig_ref)
+        sig_refs.append(new_sig_ref)
 
     def expired(self):
         """Decide if this spot has expired (in which case it should not be added to the system in the first place, and not

datafiles/39c3-tota.csv

@@ -1,18 +0,0 @@
-ref,lat,lon
-T-01,53.56278090617755,9.984341869295505
-T-02,53.562383404176416,9.98551893027115
-T-03,53.56170184391514,9.985416035619778
-T-04,53.562026534393176,9.986372919078974
-T-11,53.56284641242506,9.98475590239655
-T-12,53.562431705517035,9.98551675702443
-T-13,53.56223704898424,9.985774520335664
-T-14,53.5617893512591,9.986344302837976
-T-21,53.56284641242506,9.98475590239655
-T-22,53.56245816412497,9.985456089490567
-T-23,53.56199560857136,9.985636761412673
-T-24,53.5617893512591,9.986344302837976
-T-31,53.56247470064887,9.985611427551902
-T-32,53.5617893512591,9.986344302837976
-T-41,53.56245039134992,9.985486136112701
-T-91,53.56147934973529,9.984626806439744
-T-92,53.561396810300735,9.987553052152899

datafiles/eh23-tota.csv

@@ -1,13 +0,0 @@
-ref,lat,lon
-T-01,50.3636495,7.5584857
-T-02,50.3636495,7.5584857
-T-03,50.3636495,7.5584857
-T-11,50.3636495,7.5584857
-T-13,50.3636495,7.5584857
-T-14,50.3636495,7.5584857
-T-21,50.3636495,7.5584857
-T-31,50.3636495,7.5584857
-T-33,50.3636495,7.5584857
-T-34,50.3636495,7.5584857
-T-41,50.3636495,7.5584857
-T-51,50.3636495,7.5584857

datafiles/toilets.csv

@@ -0,0 +1,30 @@
+ref,lat,lon
+C3 T-01,53.56278090617755,9.984341869295505
+C3 T-02,53.562383404176416,9.98551893027115
+C3 T-03,53.56170184391514,9.985416035619778
+C3 T-04,53.562026534393176,9.986372919078974
+C3 T-11,53.56284641242506,9.98475590239655
+C3 T-12,53.562431705517035,9.98551675702443
+C3 T-13,53.56223704898424,9.985774520335664
+C3 T-14,53.5617893512591,9.986344302837976
+C3 T-21,53.56284641242506,9.98475590239655
+C3 T-22,53.56245816412497,9.985456089490567
+C3 T-23,53.56199560857136,9.985636761412673
+C3 T-24,53.5617893512591,9.986344302837976
+C3 T-31,53.56247470064887,9.985611427551902
+C3 T-32,53.5617893512591,9.986344302837976
+C3 T-41,53.56245039134992,9.985486136112701
+C3 T-91,53.56147934973529,9.984626806439744
+C3 T-92,53.561396810300735,9.987553052152899
+EH T-01,50.3636495,7.5584857
+EH T-02,50.3636495,7.5584857
+EH T-03,50.3636495,7.5584857
+EH T-11,50.3636495,7.5584857
+EH T-13,50.3636495,7.5584857
+EH T-14,50.3636495,7.5584857
+EH T-21,50.3636495,7.5584857
+EH T-31,50.3636495,7.5584857
+EH T-33,50.3636495,7.5584857
+EH T-34,50.3636495,7.5584857
+EH T-41,50.3636495,7.5584857
+EH T-51,50.3636495,7.5584857

providers/alert/alert_provider.py

@@ -2,27 +2,20 @@ from datetime import datetime
 
 import pytz
 
-from core.config import MAX_ALERT_AGE
+from core.data_store import DATA_STORE
 
 
 class AlertProvider:
     """Generic alert provider class. Subclasses of this query the individual APIs for alerts."""
 
-    def __init__(self, provider_config):
+    def __init__(self, name, provider_config):
         """Constructor"""
 
-        self.name = provider_config["name"]
+        self.name = name
         self.enabled = provider_config["enabled"]
         self.last_update_time = datetime.min.replace(tzinfo=pytz.UTC)
         self.status = "Not Started" if self.enabled else "Disabled"
-        self._alerts = None
-        self._web_server = None
-
-    def setup(self, alerts, web_server):
-        """Set up the provider, e.g. giving it the alert list to work from"""
-
-        self._alerts = alerts
-        self._web_server = web_server
+        self._alerts = DATA_STORE.alerts
 
     def start(self):
         """Start the provider. This should return immediately after spawning threads to access the remote resources"""
@@ -44,10 +37,7 @@ class AlertProvider:
 
     def _add_alert(self, alert):
         if not alert.expired():
-            self._alerts.add(alert.id, alert, expire=MAX_ALERT_AGE)
-            # Ping the web server in case we have any SSE connections that need to see this immediately
-            if self._web_server:
-                self._web_server.notify_new_alert(alert)
+            self._alerts.set(alert.id, alert)
 
     def stop(self):
         """Stop any threads and prepare for application shutdown"""

providers/alert/bota.py

@@ -3,7 +3,7 @@ from datetime import datetime, timedelta
 import pytz
 from bs4 import BeautifulSoup
 
-from alertproviders.http_alert_provider import HTTPAlertProvider
+from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 from data.sig_ref import SIGRef
 
@@ -15,25 +15,36 @@ class BOTA(HTTPAlertProvider):
     ALERTS_URL = "https://www.beachesontheair.com/"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("BOTA", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []
         # Find the table of upcoming alerts
-        bs = BeautifulSoup(http_response.content.decode(), features="lxml")
+        bs = BeautifulSoup(http_response.content.decode("utf-8-sig"), features="lxml")
+        if not bs.body:
+            return new_alerts
         div = bs.body.find('div', attrs={'class': 'view-activations-public'})
         if div:
             table = div.find('table', attrs={'class': 'views-table'})
             if table:
                 tbody = table.find('tbody')
+                if not tbody:
+                    return new_alerts
                 for row in tbody.find_all('tr'):
                     cells = row.find_all('td')
-                    first_cell_text = str(cells[0].find('a').contents[0]).strip()
+                    first_cell_anchor = cells[0].find('a') if len(cells) > 0 else None
+                    second_cell_anchor = cells[1].find('a') if len(cells) > 1 else None
+                    if not first_cell_anchor or not second_cell_anchor:
+                        continue
+                    first_cell_text = first_cell_anchor.get_text().strip()
                     ref_name = first_cell_text.split(" by ")[0]
-                    dx_call = str(cells[1].find('a').contents[0]).strip().upper()
+                    dx_call = second_cell_anchor.get_text().strip().upper()
 
                     # Get the date, dealing with the fact we get no year so have to figure out if it's last year or next year
-                    date_text = str(cells[2].find('span').contents[0]).strip()
+                    date_span = cells[2].find('span') if len(cells) > 2 else None
+                    if not date_span:
+                        continue
+                    date_text = date_span.get_text().strip()
                     date_time = datetime.strptime(date_text, "%d %b - %H:%M UTC").replace(tzinfo=pytz.UTC)
                     date_time = date_time.replace(year=datetime.now(pytz.UTC).year)
                     # If this was more than a day ago, activation is actually next year

providers/alert/http_alert_provider.py

@@ -4,8 +4,9 @@ from threading import Thread, Event
 
 import pytz
 import requests
+from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
 
-from alertproviders.alert_provider import AlertProvider
+from providers.alert.alert_provider import AlertProvider
 from core.constants import HTTP_HEADERS
 
 
@@ -13,8 +14,8 @@ class HTTPAlertProvider(AlertProvider):
     """Generic alert provider class for providers that request data via HTTP(S). Just for convenience to avoid code
     duplication. Subclasses of this query the individual APIs for data."""
 
-    def __init__(self, provider_config, url, poll_interval):
-        super().__init__(provider_config)
+    def __init__(self, name, provider_config, url, poll_interval):
+        super().__init__(name, provider_config)
         self._url = url
         self._poll_interval = poll_interval
         self._thread = None
@@ -24,7 +25,7 @@ class HTTPAlertProvider(AlertProvider):
         # Fire off the polling thread. It will poll immediately on startup, then sleep for poll_interval between
         # subsequent polls, so start() returns immediately and the application can continue starting.
         logging.info("Set up query of " + self.name + " alert API every " + str(self._poll_interval) + " seconds.")
-        self._thread = Thread(target=self._run, daemon=True)
+        self._thread = Thread(target=self._run, name=f"HTTPAlertProvider-{self.name}")
         self._thread.start()
 
     def stop(self):
@@ -41,6 +42,8 @@ class HTTPAlertProvider(AlertProvider):
             # Request data from API
             logging.debug("Polling " + self.name + " alert API...")
             http_response = requests.get(self._url, headers=HTTP_HEADERS, timeout=(5, 30))
+            # Check response code was good
+            if http_response.ok:
                 # Pass off to the subclass for processing
                 new_alerts = self._http_response_to_alerts(http_response)
                 # Submit the new alerts for processing. There might not be any alerts for the less popular programs.
@@ -50,7 +53,14 @@ class HTTPAlertProvider(AlertProvider):
                 self.status = "OK"
                 self.last_update_time = datetime.now(pytz.UTC)
                 logging.debug("Received data from " + self.name + " alert API.")
+            else:
+                self.status = "Error"
+                logging.warning(f"HTTP {http_response.status_code} when calling {self.name} alerts API.")
 
+        except ConnectionError:
+            logging.warning(f"Connection error when accessing {self.name} alerts API.")
+        except (ConnectTimeout, ReadTimeout):
+            logging.warning(f"Timeout when accessing {self.name} alerts API.")
         except Exception:
             self.status = "Error"
             logging.exception("Exception in HTTP JSON Alert Provider (" + self.name + ")")

providers/alert/ng3k.py

@@ -1,10 +1,12 @@
 import re
 from datetime import datetime
+from typing import cast
 
 import pytz
 from rss_parser import Parser
+from rss_parser.models.rss import RSS
 
-from alertproviders.http_alert_provider import HTTPAlertProvider
+from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 
 
@@ -16,11 +18,11 @@ class NG3K(HTTPAlertProvider):
     AS_CALL_PATTERN = re.compile("as ([a-z0-9/]+)", re.IGNORECASE)
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("NG3K", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []
-        rss = Parser.parse(http_response.content.decode())
+        rss = cast(RSS, Parser.parse(http_response.content.decode("utf-8-sig")))
         # Iterate through source data
         for source_alert in rss.channel.items:
             # Deal with "the format"...

providers/alert/parksnpeaks.py

@@ -3,7 +3,7 @@ from datetime import datetime
 
 import pytz
 
-from alertproviders.http_alert_provider import HTTPAlertProvider
+from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 from data.sig_ref import SIGRef
 
@@ -12,17 +12,17 @@ class ParksNPeaks(HTTPAlertProvider):
     """Alert provider for Parks n Peaks"""
 
     POLL_INTERVAL_SEC = 1800
-    ALERTS_URL = "http://parksnpeaks.org/api/ALERTS/"
+    ALERTS_URL = "https://parksnpeaks.org/api/ALERTS/"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("ParksNPeaks", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []
         # Iterate through source data
         for source_alert in http_response.json():
             # Calculate some things
-            sig = source_alert["Class"]
+            sig = source_alert["Class"].upper()
             if " - " in source_alert["Location"]:
                 split = source_alert["Location"].split(" - ")
                 sig_ref = split[0]
@@ -33,18 +33,24 @@ class ParksNPeaks(HTTPAlertProvider):
             start_time = datetime.strptime(source_alert["alTime"], "%Y-%m-%d %H:%M:%S").replace(
                 tzinfo=pytz.UTC).timestamp()
 
+            sigrefs = []
+            # PnP can give us an alert of class "QRP" which is the only one that's not a real SIG in Spothole's list,
+            # so mask this out if we got it.
+            if sig != "QRP":
+                sigrefs = [SIGRef(id=sig_ref, sig=sig, name=sig_ref_name)]
+
             # Convert to our alert format
             alert = Alert(source=self.name,
                           source_id=source_alert["alID"],
                           dx_calls=[source_alert["CallSign"].upper()],
                           freqs_modes=source_alert["Freq"] + " " + source_alert["MODE"],
                           comment=source_alert["Comments"],
-                          sig_refs=[SIGRef(id=sig_ref, sig=sig, name=sig_ref_name)],
+                          sig_refs=sigrefs,
                           start_time=start_time,
                           is_dxpedition=False)
 
             # Log a warning for the developer if PnP gives us an unknown programme we've never seen before
-            if sig and sig not in ["POTA", "SOTA", "WWFF", "SiOTA", "ZLOTA", "KRMNPA"]:
+            if sig and sig not in ["POTA", "SOTA", "WWFF", "SIOTA", "ZLOTA", "KRMNPA", "LLOTA", "QRP"]:
                 logging.warning("PNP alert found with sig " + sig + ", developer needs to add support for this!")
 
             # If this is POTA, SOTA or WWFF data we already have it through other means, so ignore. Otherwise, add to

providers/alert/pota.py

@@ -2,7 +2,7 @@ from datetime import datetime
 
 import pytz
 
-from alertproviders.http_alert_provider import HTTPAlertProvider
+from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 from data.sig_ref import SIGRef
 
@@ -14,7 +14,7 @@ class POTA(HTTPAlertProvider):
     ALERTS_URL = "https://api.pota.app/activation"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("POTA", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []
@@ -37,6 +37,6 @@ class POTA(HTTPAlertProvider):
             # Add to our list, but exclude any old spots that POTA can sometimes give us where even the end time is
             # in the past. Don't worry about de-duping, removing old alerts etc. at this point; other code will do
             # that for us.
-            if alert.end_time > datetime.now(pytz.UTC).timestamp():
+            if alert.end_time and alert.end_time > datetime.now(pytz.UTC).timestamp():
                 new_alerts.append(alert)
         return new_alerts

providers/alert/sota.py

@@ -2,7 +2,7 @@ from datetime import datetime
 
 import pytz
 
-from alertproviders.http_alert_provider import HTTPAlertProvider
+from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 from data.sig_ref import SIGRef
 
@@ -14,7 +14,7 @@ class SOTA(HTTPAlertProvider):
     ALERTS_URL = "https://api-db2.sota.org.uk/api/alerts/365/all/all"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("SOTA", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []

providers/alert/wota.py

@@ -1,9 +1,11 @@
 from datetime import datetime
+from typing import cast
 
 import pytz
 from rss_parser import Parser as RSSParser
+from rss_parser.models.rss import RSS
 
-from alertproviders.http_alert_provider import HTTPAlertProvider
+from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 from data.sig_ref import SIGRef
 
@@ -16,11 +18,11 @@ class WOTA(HTTPAlertProvider):
     RSS_DATE_TIME_FORMAT = "%a, %d %b %Y %H:%M:%S %z"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("WOTA", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []
-        rss = RSSParser.parse(http_response.content.decode())
+        rss = cast(RSS, RSSParser.parse(http_response.content.decode("utf-8-sig")))
         # Iterate through source data
         for source_alert in rss.channel.items:
 
@@ -35,9 +37,9 @@ class WOTA(HTTPAlertProvider):
             ref_name = None
             if len(title_split) > 1:
                 ref_split = title_split[1].split(" - ")
-                ref = ref_split[0]
+                ref = str(ref_split[0])
                 if len(ref_split) > 1:
-                    ref_name = ref_split[1]
+                    ref_name = str(ref_split[1])
 
             # Pick apart the description
             desc_split = source_alert.description.split(". ")

providers/alert/wwff.py

@@ -2,7 +2,7 @@ from datetime import datetime
 
 import pytz
 
-from alertproviders.http_alert_provider import HTTPAlertProvider
+from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 from data.sig_ref import SIGRef
 
@@ -14,7 +14,7 @@ class WWFF(HTTPAlertProvider):
     ALERTS_URL = "https://spots.wwff.co/static/agendas.json"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("WWFF", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []

providers/callsigndata/api_query_callsign_data_provider.py

@@ -0,0 +1,28 @@
+import logging
+from datetime import datetime
+from threading import Thread, Event
+
+import pytz
+from requests import ReadTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from core.url_data_cache import URLDataCache
+from providers.callsigndata.callsign_data_provider import CallsignDataProvider
+
+
+class APIQueryCallsignDataProvider(CallsignDataProvider):
+    """Generic callsign data provider class for providers that fetch their data from the web on-demand using an API."""
+
+    def __init__(self, name, provider_config, storage):
+        """ Set up the provider."""
+        super().__init__(name, provider_config, storage)
+
+        if self.enabled:
+            self.status = "Ready"
+
+    def start(self):
+        pass
+
+    def stop(self):
+        pass

providers/callsigndata/callsign_data_provider.py

@@ -0,0 +1,59 @@
+import logging
+from datetime import datetime
+
+import pytz
+
+from core.data_store import DATA_STORE
+
+
+class CallsignDataProvider:
+    """Generic callsign reference data provider class. Subclasses of this set up the various mechanisms via which
+    Spothole can look up data for callsigns."""
+
+    def __init__(self, name, provider_config, storage):
+        """Constructor. As well as name and config, provide the storage object from DATA_STORE that will be used to
+        store the result of lookups to speed up future access."""
+
+        self.name = name
+        self.enabled = provider_config["enabled"]
+        self.priority = int(provider_config["priority"])
+        self.last_update_time = datetime.min.replace(tzinfo=pytz.UTC)
+        self.status = "Not Started" if self.enabled else "Disabled"
+        self.lookup_count = 0
+        self._storage = storage
+
+    def start(self):
+        """Start the provider. This should return immediately after spawning threads to access remote resources, if
+        needed."""
+
+        raise NotImplementedError("Subclasses must implement this method")
+
+    def stop(self):
+        """Stop any threads and prepare for application shutdown"""
+
+        raise NotImplementedError("Subclasses must implement this method")
+
+    def lookup(self, callsign, lookup_credentials):
+        """Looks up data for the provided callsign. Takes a LookupCredentials object, which provides any credentials
+        that have been provided by the user for this session (QRZ.com/HamQTH) to allow us to look up using those
+        services on the user's behalf. (Clublog is looked up using an API key owned by the server and provided in its
+        config file, so users need not provide their own.) Returns a Callsign object with as much data populated as
+        possible. Data is cached internally for a set period of 30 days to avoid the need to request data from servers
+        each time."""
+
+        if self.enabled:
+            if callsign in self._storage:
+                return self._storage[callsign]
+            else:
+                c = self._perform_new_lookup(callsign, lookup_credentials)
+                if c:
+                    self._storage.set(callsign, c, expire=DATA_STORE.CALLSIGN_DATA_TTL_SEC)
+                return c
+        else:
+            return None
+
+
+    def _perform_new_lookup(self, callsign, lookup_credentials):
+        """Makes a new request to the data source for callsign data."""
+
+        raise NotImplementedError("Subclasses must implement this method")

providers/callsigndata/clublogapi.py

@@ -0,0 +1,48 @@
+import logging
+from datetime import datetime
+
+import pytz
+from pyhamtools import LookupLib, Callinfo
+
+from core.data_store import DATA_STORE
+from core.utils import get_callsign_object_from_pyhamtools_callinfo
+from data.callsign import Callsign
+from providers.callsigndata.api_query_callsign_data_provider import APIQueryCallsignDataProvider
+
+
+class ClublogAPI(APIQueryCallsignDataProvider):
+    """Callsign data provider for Clublog's API."""
+
+    _callinfo = None
+
+    def __init__(self, provider_config):
+        # API key required for this provider
+        self._api_key = provider_config.get("api-key", "")
+        if self._api_key != "":
+            lookuplib = LookupLib(lookuptype="clublogapi", apikey=self._api_key)
+            self._callinfo = Callinfo(lookuplib)
+        else:
+            provider_config["enabled"] = False
+            logging.warning(
+                "Clublog XML callsign data provider configured but no api key was provided, this has been disabled.")
+
+        super().__init__("Clublog API", provider_config, DATA_STORE.callsign_data_clublogapi)
+
+
+    def _perform_new_lookup(self, callsign, lookup_credentials):
+        callsign_data = Callsign(call=callsign)
+
+        try:
+            if self._callinfo:
+                callsign_data = get_callsign_object_from_pyhamtools_callinfo(callsign, self._callinfo)
+                self.status = "OK"
+                self.last_update_time = datetime.now(pytz.UTC)
+                self.lookup_count += 1
+            else:
+                return None
+
+        except Exception as e:
+            self.status = "Error"
+            logging.error("Exception when looking up data from Clublog API", e, exc_info=True)
+
+        return callsign_data

providers/callsigndata/clublogxml.py

@@ -0,0 +1,69 @@
+import gzip
+import logging
+from datetime import datetime
+
+import pytz
+from pyhamtools import LookupLib, Callinfo
+
+from core.data_store import DATA_STORE
+from core.utils import get_callsign_object_from_pyhamtools_callinfo
+from data.callsign import Callsign
+from providers.callsigndata.file_download_callsign_data_provider import FileDownloadCallsignDataProvider
+
+
+class ClublogXML(FileDownloadCallsignDataProvider):
+    """Callsign data provider for ClubLog's Country File, which provides basic callsign to DXCC entity mapping."""
+
+    POLL_INTERVAL_DAYS = 30
+    DATA_URL = "https://cdn.clublog.org/cty.php"
+    CACHE_PATH_ZIPPED = "cache/cty.xml.gz"
+    CACHE_PATH_UNZIPPED = "cache/cty.xml"
+    _callinfo = None
+
+    def __init__(self, provider_config):
+        # API key required for this provider
+        self._api_key = provider_config.get("api-key", "")
+        if self._api_key == "":
+            provider_config["enabled"] = False
+            logging.warning(
+                "Clublog XML callsign data provider configured but no api key was provided, this has been disabled.")
+
+        super().__init__("Clublog XML", provider_config, self.DATA_URL + "?api=" + self._api_key,
+                         self.CACHE_PATH_ZIPPED, self.POLL_INTERVAL_DAYS, DATA_STORE.callsign_data_clublogxml)
+
+    def _handle_file(self, path):
+        try:
+            # The download from Clublog is gzipped so we need to uncompress that and re-save as a separate file that
+            # the LookupLib can actually use.
+            with gzip.open(path, "rb") as uncompressed:
+                file_content = uncompressed.read()
+            assert isinstance(file_content, bytes)
+            with open(self.CACHE_PATH_UNZIPPED, "wb") as f:
+                f.write(file_content)
+                f.flush()
+
+            # Now load the data
+            lookuplib = LookupLib(lookuptype="clublogxml", filename=self.CACHE_PATH_UNZIPPED)
+            self._callinfo = Callinfo(lookuplib)
+            return True
+
+        except Exception as e:
+            logging.error("Exception when loading Clublog XML.", e, exc_info=True)
+            return False
+
+    def _perform_new_lookup(self, callsign, lookup_credentials):
+        callsign_data = Callsign(call=callsign)
+
+        try:
+            if self._callinfo:
+                callsign_data = get_callsign_object_from_pyhamtools_callinfo(callsign, self._callinfo)
+                self.status = "OK"
+                self.lookup_count += 1
+            else:
+                return None
+
+        except Exception as e:
+            self.status = "Error"
+            logging.error("Exception when looking up data from Clublog XML data", e, exc_info=True)
+
+        return callsign_data

providers/callsigndata/countryfiles.py

@@ -0,0 +1,48 @@
+import logging
+
+from pyhamtools import LookupLib, Callinfo
+
+from core.data_store import DATA_STORE
+from core.utils import get_callsign_object_from_pyhamtools_callinfo
+from data.callsign import Callsign
+from providers.callsigndata.file_download_callsign_data_provider import FileDownloadCallsignDataProvider
+
+
+class CountryFiles(FileDownloadCallsignDataProvider):
+    """Callsign data provider for Country-files.com, which provides basic callsign to DXCC entity mapping."""
+
+    POLL_INTERVAL_DAYS = 30
+    DATA_URL = "https://www.country-files.com/cty/cty.plist"
+    CACHE_PATH = "cache/cty.plist"
+    _callinfo = None
+
+    def __init__(self, provider_config):
+        super().__init__("CountryFiles.com", provider_config, self.DATA_URL, self.CACHE_PATH, self.POLL_INTERVAL_DAYS,
+                         DATA_STORE.callsign_data_countryfiles)
+
+    def _handle_file(self, path):
+        try:
+            lookuplib = LookupLib(lookuptype="countryfile", filename=path)
+            self._callinfo = Callinfo(lookuplib)
+            return True
+
+        except Exception as e:
+            logging.error("Exception when loading Country Files cty.plist.", e, exc_info=True)
+            return False
+
+    def _perform_new_lookup(self, callsign, lookup_credentials):
+        callsign_data = Callsign(call=callsign)
+
+        try:
+            if self._callinfo:
+                callsign_data = get_callsign_object_from_pyhamtools_callinfo(callsign, self._callinfo)
+                self.status = "OK"
+                self.lookup_count += 1
+            else:
+                return None
+
+        except Exception as e:
+            self.status = "Error"
+            logging.error("Exception when looking up data from Country file", e, exc_info=True)
+
+        return callsign_data

providers/callsigndata/file_download_callsign_data_provider.py

@@ -0,0 +1,87 @@
+import logging
+from datetime import datetime
+from threading import Thread, Event
+
+import pytz
+from requests import ReadTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from core.url_data_cache import URLDataCache
+from providers.callsigndata.callsign_data_provider import CallsignDataProvider
+
+
+class FileDownloadCallsignDataProvider(CallsignDataProvider):
+    """Generic callsign data provider class for providers that fetch their data from the web by downloading a file."""
+
+    def __init__(self, name, provider_config, url, cache_file_path, poll_interval, storage):
+        """ Set up the provider, note poll_interval is in *days*."""
+        super().__init__(name, provider_config, storage)
+        self._url = url
+        self._cache_file_path = cache_file_path
+        self._poll_interval = poll_interval
+        self._thread = None
+        self._stop_event = Event()
+        self._url_data_cache = URLDataCache("callsigndata_" + name)
+
+        if self.enabled:
+            self.status = "Ready"
+
+    def start(self):
+        # Fire off the polling thread. It will poll immediately on startup, then sleep for poll_interval between
+        # subsequent polls, so start() returns immediately and the application can continue starting.
+        logging.info(
+            "Set up query of " + self.name + " callsign reference data every " + str(self._poll_interval) + " days.")
+        self._thread = Thread(target=self._run, name=f"FileDownloadCallsignDataProvider-{self.name}")
+        self._thread.start()
+
+    def stop(self):
+        self._stop_event.set()
+
+    def _run(self):
+        while True:
+            self._poll()
+            if self._stop_event.wait(timeout=self._poll_interval * 60 * 60 * 24):
+                break
+
+    def _poll(self):
+        try:
+            # Request the file. Use the data cache (with a TTL of 1 day) here, not as the main mechanism for
+            # caching, but just so continual restarts of the software during testing don't hammer the servers.
+            logging.debug("Downloading " + self.name + " callsign reference data...")
+            http_response = self._url_data_cache.get(self._url, headers=HTTP_HEADERS)
+            # Check response code was good
+            if http_response.ok:
+                # Save the data to a local file
+                with open(self._cache_file_path, "wb") as f:
+                    f.write(http_response.content)
+                    f.flush()
+                    # Pass off to the subclass for processing
+                    ok = self._handle_file(self._cache_file_path)
+                    if ok:
+                        self.status = "OK"
+                        self.last_update_time = datetime.now(pytz.UTC)
+                        logging.info("Updated callsign reference data from " + self.name)
+                    else:
+                        self.status = "Error"
+                        logging.warning(f"Error updating callsign reference data from {self.name}.")
+
+            else:
+                self.status = "Error"
+                logging.warning(f"HTTP {http_response.status_code} when downloading callsign reference data from {self.name}.")
+
+        except ConnectionError:
+            self.status = "Error"
+            logging.warning(f"Connection error when downloading callsign reference data from {self.name}.")
+        except (ConnectTimeout, ReadTimeout):
+            self.status = "Error"
+            logging.warning(f"Timeout when downloading callsign reference data from {self.name}.")
+        except Exception:
+            self.status = "Error"
+            logging.exception("Exception in callsign reference data provider (" + self.name + ")")
+            self._stop_event.wait(timeout=1)
+
+    def _handle_file(self, path):
+        """Handle an updated file downloaded from the server. Return true if successful, false otherwise."""
+
+        raise NotImplementedError("Subclasses must implement this method")

providers/callsigndata/hamqth.py

@@ -0,0 +1,142 @@
+import logging
+import urllib.parse
+from datetime import timedelta, datetime
+
+import pytz
+import xmltodict
+from pyhamtools import callinfo
+from requests import ConnectTimeout, ReadTimeout
+from requests_cache import CachedSession
+
+from core.config import SERVER_OWNER_CALLSIGN
+from core.constants import HTTP_HEADERS, SOFTWARE_VERSION
+from core.data_store import DATA_STORE, CACHE_DIR
+from core.url_data_cache import URLDataCache
+from data.callsign import Callsign
+from providers.callsigndata.api_query_callsign_data_provider import APIQueryCallsignDataProvider
+
+
+class HamQTH(APIQueryCallsignDataProvider):
+    """Callsign data provider for HamQTH."""
+
+    def __init__(self, provider_config):
+        super().__init__("HamQTH", provider_config, DATA_STORE.callsign_data_hamqth)
+        self._HAMQTH_BASE_URL = "https://www.hamqth.com/xml.php"
+        self._PRG = ("Spothole v" + SOFTWARE_VERSION + " operated by " + SERVER_OWNER_CALLSIGN).replace(" ", "_")
+        self._URL_DATA_CACHE = URLDataCache("hamqth")
+        # Separate URL cache for session key lookups. Once a session key is returned from logging in with a username
+        # and password, this is valid for an hour, so our cache stores this specifically for 55 minutes.
+        self._CREDENTIALS_CACHE = CachedSession(CACHE_DIR + "/urls/hamqth-creds",
+                                                expire_after=timedelta(minutes=55))
+
+    def _perform_new_lookup(self, callsign, lookup_credentials):
+        # If we don't have HamQTH credentials, skip this lookup Return None so we don't *cache* the lack of data, because
+        #         # someone might provide credentials next time around.
+        if not lookup_credentials or not ((lookup_credentials.hamqth_username and lookup_credentials.hamqth_password)
+                or lookup_credentials.hamqth_session_id):
+            return None
+
+        try:
+            # Obtain session key from credentials, by looking it up from username & password if necessary.
+            session_id = None
+            if lookup_credentials.hamqth_session_id:
+                session_id = lookup_credentials.hamqth_session_id
+            elif lookup_credentials.hamqth_username and lookup_credentials.hamqth_password:
+                try:
+                    session_data = self._CREDENTIALS_CACHE.get(
+                        self._HAMQTH_BASE_URL + "?u=" + urllib.parse.quote_plus(lookup_credentials.hamqth_username) +
+                        "&p=" + urllib.parse.quote_plus(lookup_credentials.hamqth_password),
+                        headers=HTTP_HEADERS).content
+                    dict_data = xmltodict.parse(session_data)
+                    if "session_id" in dict_data["HamQTH"]["session"]:
+                        session_id = str(dict_data["HamQTH"]["session"]["session_id"])
+                    else:
+                        # Log this failure at debug level only, not our problem if user entered the wrong password.
+                        logging.debug("HamQTH login details incorrect, failed to look up with HamQTH.")
+                        return None
+                except Exception:
+                    logging.error("Exception when getting HamQTH session key")
+                    return None
+
+            if not session_id:
+                return None
+
+            # Try the call as given, then fall back to the base call (strips /P, /M etc.)
+            calls_to_try = [callsign]
+            try:
+                home_call = callinfo.Callinfo.get_homecall(callsign)
+                if home_call != callsign:
+                    calls_to_try.append(home_call)
+            except ValueError:
+                logging.debug("Could not look up home call for callsign %s", callsign)
+
+            # Try looking up each call using the API
+            for lookup_call in calls_to_try:
+                try:
+                    response = self._URL_DATA_CACHE.get(
+                        self._HAMQTH_BASE_URL + "?id=" + session_id + "&callsign=" + urllib.parse.quote_plus(
+                            lookup_call) + "&prg=" + self._PRG, headers=HTTP_HEADERS, timeout=10)
+                    if response.ok:
+                        # Found data, convert it to our object and return it
+                        data = xmltodict.parse(response.content)["HamQTH"]["search"]
+                        self.status = "OK"
+                        self.last_update_time = datetime.now(pytz.UTC)
+                        self.lookup_count += 1
+                        return self.hamqth_response_to_callsign(callsign, data)
+
+                    elif not response.from_cache:
+                        logging.warning("HTTP %d looking up callsign %s using HamQTH", response.status_code,
+                                        lookup_call)
+
+                except (KeyError, ValueError):
+                    continue
+                except ConnectionError:
+                    logging.warning(f"Connection error when looking up callsign %s using HamQTH", lookup_call)
+                    continue
+                except (ConnectTimeout, ReadTimeout):
+                    logging.warning(f"Timeout when looking up callsign %s using HamQTH", lookup_call)
+                    continue
+                except Exception:
+                    logging.error("Exception when looking up callsign %s using HamQTH", lookup_call, exc_info=True)
+                    continue
+
+            # Not found in HamQTH; return a Callsign object with no data so we cache that and don't keep retrying
+            return Callsign(call=callsign)
+
+        except Exception as e:
+            self.status = "Error"
+            logging.error("Exception when looking up data from HamQTH", e, exc_info=True)
+            # Return None, this won't be cached so we will be asked to query data again for this call next time.
+            return None
+
+    @staticmethod
+    def hamqth_response_to_callsign(callsign, data):
+        """Convert the "Callsign" block in HamQTH's API response to our own Callsign object."""
+
+        # Check for sensible latitudes
+        lat = None
+        lon = None
+        if "latitude" in data and "longitude" in data and (
+                float(data["latitude"]) != 0 or float(data["longitude"]) != 0) and -89.9 < float(
+                data["latitude"]) < 89.9:
+            lat = float(data["latitude"])
+            lon = float(data["longitude"])
+
+        # Check for sensible grids
+        grid = None
+        if "grid" in data and not data["grid"].startswith("AA00"):
+            grid = data["grid"]
+
+        return Callsign(call=callsign,
+                        home_call=callinfo.Callinfo.get_homecall(callsign),
+                        name=data["nick"] if "nick" in data else None,
+                        qth=data["qth"] if "qth" in data else None,
+                        country=data["country"] if "country" in data else None,
+                        continent=data["continent"] if "continent" in data else None,
+                        latitude=lat,
+                        longitude=lon,
+                        grid=grid,
+                        dxcc_id=int(data["adif"]) if "adif" in data else None,
+                        cq_zone=int(data["cq"]) if "cq" in data else None,
+                        itu_zone=int(data["itu"]) if "itu" in data else None,
+                        location_source="HOME QTH")

providers/callsigndata/qrz.py

@@ -0,0 +1,162 @@
+import logging
+import urllib.parse
+from datetime import timedelta, datetime
+
+import pytz
+import xmltodict
+from pyhamtools import callinfo
+from requests import ConnectTimeout, ReadTimeout
+from requests_cache import CachedSession
+
+from core.constants import HTTP_HEADERS
+from core.data_store import DATA_STORE, CACHE_DIR
+from core.url_data_cache import URLDataCache
+from data.callsign import Callsign
+from providers.callsigndata.api_query_callsign_data_provider import APIQueryCallsignDataProvider
+
+
+class QRZ(APIQueryCallsignDataProvider):
+    """Callsign data provider for QRZ.com."""
+
+    def __init__(self, provider_config):
+        super().__init__("QRZ.com", provider_config, DATA_STORE.callsign_data_qrz)
+        self._QRZ_BASE_URL = "https://xmldata.qrz.com/xml/current/"
+        self._URL_DATA_CACHE = URLDataCache("qrz")
+        # Separate URL cache for session key lookups. Once a session key is returned from logging in with a username
+        # and password, this is valid for an hour, so our cache stores this specifically for 55 minutes.
+        self._CREDENTIALS_CACHE = CachedSession(CACHE_DIR + "/urls/qrz-creds",
+                                                expire_after=timedelta(minutes=55))
+
+    def _perform_new_lookup(self, callsign, lookup_credentials):
+        # If we don't have QRZ credentials, skip this lookup. Return None so we don't *cache* the lack of data, because
+        # someone might provide credentials next time around.
+        if not lookup_credentials or not ((lookup_credentials.qrz_username and lookup_credentials.qrz_password)
+                or lookup_credentials.qrz_session_key):
+            return None
+
+        try:
+            # Obtain session key from credentials, by looking it up from username & password if necessary.
+            session_key = None
+            if lookup_credentials.qrz_session_key:
+                session_key = lookup_credentials.qrz_session_key
+            elif lookup_credentials.qrz_username and lookup_credentials.qrz_password:
+                try:
+                    login_response = self._CREDENTIALS_CACHE.get(
+                        self._QRZ_BASE_URL + "?username=" + urllib.parse.quote_plus(lookup_credentials.qrz_username) +
+                        "&password=" + urllib.parse.quote_plus(lookup_credentials.qrz_password) + "&agent=spothole",
+                        headers=HTTP_HEADERS).content
+                    login_data = xmltodict.parse(login_response)
+                    session = login_data.get("QRZDatabase", {}).get("Session", {})
+                    if "Key" in session:
+                        session_key = str(session["Key"])
+                    else:
+                        # Log this failure at debug level only, not our problem if user entered the wrong password.
+                        logging.debug("QRZ.com login details incorrect, failed to look up with QRZ.")
+                        return None
+                except Exception:
+                    logging.error("Exception when getting QRZ.com session key")
+                    return None
+
+            if not session_key:
+                return None
+
+            # Try the call as given, then fall back to the base call (strips /P, /M etc.)
+            calls_to_try = [callsign]
+            try:
+                home_call = callinfo.Callinfo.get_homecall(callsign)
+                if home_call != callsign:
+                    calls_to_try.append(home_call)
+            except ValueError:
+                logging.debug("Could not look up home call for callsign %s", callsign)
+
+            # Try looking up each call using the API
+            for lookup_call in calls_to_try:
+                try:
+                    response = self._URL_DATA_CACHE.get(
+                        self._QRZ_BASE_URL + "?s=" + session_key + "&callsign=" + urllib.parse.quote_plus(lookup_call),
+                        headers=HTTP_HEADERS, timeout=10)
+                    if response.ok:
+                        qrz_response = xmltodict.parse(response.content).get("QRZDatabase", {})
+                        if qrz_response:
+                            if "Callsign" in qrz_response:
+                                qrz_data = qrz_response.get("Callsign")
+                                self.status = "OK"
+                                self.last_update_time = datetime.now(pytz.UTC)
+                                self.lookup_count += 1
+                                # Found data, convert it to our object and return it
+                                return self.qrz_response_to_callsign(callsign, qrz_data)
+
+                            elif "Session" in qrz_response and "Error" in qrz_response.get("Session"):
+                                # Errors here are normally just "callsign not in database", no need to log that ourselves
+                                # above debug level.
+                                logging.debug("QRZ returned an error looking up callsign %s: %s", lookup_call,
+                                              qrz_response.get("Session").get("Error"))
+
+                        elif not response.from_cache:
+                            logging.warning("QRZ returned a malformed response looking up callsign %s", lookup_call)
+                    elif not response.from_cache:
+                        logging.warning("HTTP %d looking up callsign %s using QRZ", lookup_call)
+
+                except (KeyError, ValueError):
+                    continue
+                except ConnectionError:
+                    logging.warning(f"Connection error when looking up callsign %s using QRZ", lookup_call)
+                    continue
+                except (ConnectTimeout, ReadTimeout):
+                    logging.warning(f"Timeout when looking up callsign %s using QRZ.", lookup_call)
+                    continue
+                except Exception:
+                    logging.error("Exception when looking up callsign %s using QRZ", lookup_call, exc_info=True)
+                    continue
+
+            # Not found in QRZ; return a Callsign object with no data so we cache that and don't keep retrying
+            return Callsign(call=callsign)
+
+        except Exception as e:
+            self.status = "Error"
+            logging.error("Exception when looking up data from QRZ.com", e, exc_info=True)
+            # Return None, this won't be cached so we will be asked to query data again for this call next time.
+            return None
+
+    @staticmethod
+    def qrz_response_to_callsign(callsign, data):
+        """Convert the "Callsign" block in QRZ's API response to our own Callsign object."""
+
+        # Get a name
+        name = None
+        if "name_fmt" in data:
+            name = data["name_fmt"]
+        if "fname" in data:
+            name = data["fname"]
+            if "nick" in data:
+                name = name + " \"" + data["nick"] + "\""
+            if "name" in data:
+                name = name + " " + data["name"]
+
+        # Check for sensible latitudes
+        lat = None
+        lon = None
+        if "latitude" in data and "longitude" in data and (
+                float(data["latitude"]) != 0 or float(data["longitude"]) != 0) and -89.9 < float(
+                data["latitude"]) < 89.9:
+            lat = float(data["latitude"])
+            lon = float(data["longitude"])
+
+        # Check for sensible grids
+        grid = None
+        if "grid" in data and not data["grid"].startswith("AA00"):
+            grid = data["grid"]
+
+        return Callsign(call=callsign,
+                        home_call=callinfo.Callinfo.get_homecall(callsign),
+                        name=name,
+                        qth=data["addr2"] if "addr2" in data else None,
+                        country=data["country"] if "country" in data else None,
+                        continent=data["continent"] if "continent" in data else None,
+                        latitude=lat,
+                        longitude=lon,
+                        grid=grid,
+                        dxcc_id=int(data["adif"]) if "adif" in data else None,
+                        cq_zone=int(data["cqzone"]) if "cqzone" in data else None,
+                        itu_zone=int(data["ituzone"]) if "ituzone" in data else None,
+                        location_source="HOME QTH")

providers/sigrefdata/arlhs.py

@@ -0,0 +1,35 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class ARLHS(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Amateur Radio Light House Society"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "ARLHS"
+    DATA_URL = "https://www.gma.rocks/download/lighthouse.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()[1:]):
+            if "ARLHS" in row and row["ARLHS"] != "":
+                ref_id = row["ARLHS"]
+                new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["Name"] if "Name" in row else None,
+                                          url="https://www.cqgma.org/zinfo.php?ref=" + ref_id,
+                                          latitude=float(row["Latitude"]) if "Latitude" in row and row[
+                                              "Latitude"] != "" else None,
+                                          longitude=float(row["Longitude"]) if "Longitude" in row and row[
+                                              "Longitude"] != "" else None,
+                                          grid=row["Maidenhead Locator"]))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/dme.py

@@ -0,0 +1,41 @@
+import csv
+
+from pyhamtools.locator import latlong_to_locator
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.local_file_sig_ref_data_provider import LocalFileSIGRefDataProvider
+
+
+class DME(LocalFileSIGRefDataProvider):
+    """SIG ref data provider for Diploma Municipios de Espana"""
+
+    SIG = "DME"
+    PATH = "datafiles/MUNICIPIOS.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.PATH)
+
+    def _file_to_data(self, path):
+        new_data = []
+        with open(path, encoding="latin-1") as _f:
+            for row in csv.DictReader(_f, delimiter=";"):
+                ref_id = row["COD_INE"][:5]
+                latitude = float(row["LATITUD_ETRS89_REGCAN95"].replace(",", ".")) if row.get(
+                    "LATITUD_ETRS89_REGCAN95") else None
+                longitude = float(row["LONGITUD_ETRS89_REGCAN95"].replace(",", ".")) if row.get(
+                    "LONGITUD_ETRS89_REGCAN95") else None
+
+                ref = SIGRef(sig=self.SIG, id=ref_id,
+                                          name=row["NOMBRE_ACTUAL"] + ", " + row["PROVINCIA"],
+                                          latitude=latitude,
+                                          longitude=longitude)
+                if latitude and longitude:
+                    ref.grid = latlong_to_locator(latitude, longitude, 6)
+                new_data.append(ref)
+
+                # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest
+                # of the data in this case
+                if self._stop:
+                    break
+
+        return new_data

providers/sigrefdata/file_download_sig_ref_data_provider.py

@@ -0,0 +1,80 @@
+import logging
+from datetime import datetime
+from threading import Thread, Event
+
+import pytz
+from requests import ReadTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from core.url_data_cache import URLDataCache
+from providers.sigrefdata.sig_ref_data_provider import SIGRefDataProvider
+
+
+class FileDownloadSIGRefDataProvider(SIGRefDataProvider):
+    """Generic SIG ref data provider class for providers that fetch their data from the web by downloading a file."""
+
+    def __init__(self, sig_name, provider_config, url, poll_interval):
+        """ Set up the provider, note poll_interval is in *days*."""
+        super().__init__(sig_name, provider_config)
+        self._url = url
+        self._poll_interval = poll_interval
+        self._thread = None
+        self._stop_event = Event()
+        self._url_data_cache = URLDataCache("sigrefdata_" + sig_name)
+
+    def start(self):
+        # Fire off the polling thread. It will poll immediately on startup, then sleep for poll_interval between
+        # subsequent polls, so start() returns immediately and the application can continue starting.
+        logging.info(
+            "Set up query of " + self.sig_name + " SIG ref data every " + str(self._poll_interval) + " days.")
+        self._thread = Thread(target=self._run, name=f"FileDownloadSIGRefDataProvider-{self.sig_name}")
+        self._thread.start()
+
+    def stop(self):
+        self._stop_event.set()
+
+    def _run(self):
+        while True:
+            self._poll()
+            if self._stop_event.wait(timeout=self._poll_interval * 60 * 60 * 24):
+                break
+
+    def _poll(self):
+        try:
+            # Request data from API. Use the data cache (with a TTL of 1 day) here, not as the main mechanism for
+            # caching, but just so continual restarts of the software during testing don't hammer the servers.
+            logging.debug("Downloading " + self.sig_name + " SIG ref data...")
+            http_response = self._url_data_cache.get(self._url, headers=HTTP_HEADERS)
+            # Check response code was good
+            if http_response.ok:
+                # Pass off to the subclass for processing
+                new_data = self._http_response_to_data(http_response)
+                # Add the new data to the SIG Ref data store
+                if new_data:
+                    self._add_data(new_data)
+
+                self.status = "OK"
+                self.last_update_time = datetime.now(pytz.UTC)
+                logging.debug("Received SIG ref data for " + self.sig_name)
+            else:
+                self.status = "Error"
+                logging.warning(f"HTTP {http_response.status_code} when downloading SIG ref data for {self.sig_name}.")
+
+        except ConnectionError:
+            self.status = "Error"
+            logging.warning(f"Connection error when downloading SIG ref data for {self.sig_name}.")
+        except (ConnectTimeout, ReadTimeout):
+            self.status = "Error"
+            logging.warning(f"Timeout when downloading SIG ref data for {self.sig_name}.")
+        except Exception:
+            self.status = "Error"
+            logging.exception("Exception in HTTP SIG Ref Data Provider (" + self.sig_name + ")")
+            self._stop_event.wait(timeout=1)
+
+    def _http_response_to_data(self, http_response):
+        """Convert an HTTP response returned by the server into SIG Ref data. The whole response is provided here so the
+        subclass implementations can check for HTTP status codes if necessary, and handle the response as JSON, CSV,
+        whatever the remote file actually is."""
+
+        raise NotImplementedError("Subclasses must implement this method")

providers/sigrefdata/gma.py

@@ -0,0 +1,34 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class GMA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Global Mountain Activity"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "GMA"
+    DATA_URL = "https://www.gma.rocks/download/summits.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()[1:]):
+            ref_id = row["Reference"]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["Name"] if "Name" in row else None,
+                                      url="https://www.cqgma.org/zinfo.php?ref=" + ref_id,
+                                      latitude=float(row["Latitude"]) if "Latitude" in row and row[
+                                          "Latitude"] != "" else None,
+                                      longitude=float(row["Longitude"]) if "Longitude" in row and row[
+                                          "Longitude"] != "" else None,
+                                      grid=row["Maidenhead Locator"]))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/illw.py

@@ -0,0 +1,35 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class ILLW(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for International Lighthouse & Lightship Weekend"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "ILLW"
+    DATA_URL = "https://www.gma.rocks/download/lighthouse.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()[1:]):
+            if "ILLW" in row and row["ILLW"] != "":
+                ref_id = row["ILLW"]
+                new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["Name"] if "Name" in row else None,
+                                          url="https://www.cqgma.org/zinfo.php?ref=" + ref_id,
+                                          latitude=float(row["Latitude"]) if "Latitude" in row and row[
+                                              "Latitude"] != "" else None,
+                                          longitude=float(row["Longitude"]) if "Longitude" in row and row[
+                                              "Longitude"] != "" else None,
+                                          grid=row["Maidenhead Locator"]))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/iota.py

@@ -0,0 +1,41 @@
+import logging
+
+from pyhamtools.locator import latlong_to_locator
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class IOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Islands on the Air"""
+
+    POLL_INTERVAL_DAYS = 365
+    SIG = "IOTA"
+    DATA_URL = "https://www.iota-world.org/islands-on-the-air/downloads/download-file.html?path=groups.json"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        data = http_response.json()
+        if isinstance(data, list):
+            for ref in data:
+                ref_id = ref["refno"]
+                latitude = float(ref["latitude_min"]) + float(ref["latitude_max"]) / 2.0
+                longitude = float(ref["longitude_min"]) + float(ref["longitude_max"]) / 2.0
+                grid = None
+                try:
+                    grid = latlong_to_locator(latitude, longitude, 6)
+                except ValueError:
+                    logging.debug(f"Error converting lat/lon to locator for an IOTA reference %f %f", latitude, longitude)
+
+                new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=ref["name"],
+                                          grid=grid, latitude=latitude, longitude=longitude))
+
+                # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest
+                # of the data in this case
+                if self._stop_event.is_set():
+                    break
+
+        return new_data

providers/sigrefdata/llota.py

@@ -0,0 +1,37 @@
+from pyhamtools.locator import locator_to_latlong
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class LLOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Lagos y Lagunas on the Air"""
+
+    POLL_INTERVAL_DAYS = 7
+    SIG = "LLOTA"
+    DATA_URL = "https://llota.app/api/public/references"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        data = http_response.json()
+        if isinstance(data, list):
+            for ref in data:
+                ref_id = ref["reference_code"]
+                grid = str(ref["grid_locator"])
+                ll = locator_to_latlong(grid)
+
+                new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=str(ref["name"]),
+                                          url="https://llota.app/list/ref/" + ref_id,
+                                          grid=grid,
+                                          latitude=ll[0],
+                                          longitude=ll[1]))
+
+                # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest
+                # of the data in this case
+                if self._stop_event.is_set():
+                    break
+
+        return new_data

providers/sigrefdata/local_file_sig_ref_data_provider.py

@@ -0,0 +1,38 @@
+import logging
+from datetime import datetime
+
+import pytz
+
+from providers.sigrefdata.sig_ref_data_provider import SIGRefDataProvider
+
+
+class LocalFileSIGRefDataProvider(SIGRefDataProvider):
+    """Generic SIG ref data provider class for providers that fetch their data from a local file on startup."""
+
+    def __init__(self, sig, provider_config, path):
+        super().__init__(sig, provider_config)
+        self._path = path
+        self._stop = False
+
+    def start(self):
+        logging.debug("Loading " + self.sig_name + " SIG ref data from file.")
+        try:
+            new_data = self._file_to_data(self._path)
+            if new_data:
+                self._add_data(new_data)
+                self.status = "OK"
+                self.last_update_time = datetime.now(pytz.UTC)
+            else:
+                self.status = "Error"
+                logging.info("Failed to load SIG ref data for " + self.sig_name)
+        except Exception as e:
+            self.status = "Error"
+            logging.error("Exception in local file SIG Ref Data Provider (" + self.sig_name + ")", e, exc_info=True)
+
+    def stop(self):
+        self._stop = True
+
+    def _file_to_data(self, path):
+        """Load a file on the given path and turn it into SIG Ref data."""
+
+        raise NotImplementedError("Subclasses must implement this method")

providers/sigrefdata/mota.py

@@ -0,0 +1,34 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class MOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Mills on the Air"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "MOTA"
+    DATA_URL = "https://www.gma.rocks/download/mills.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()[1:]):
+            ref_id = row["Reference"]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["Name"] if "Name" in row else None,
+                                      url="https://www.cqgma.org/zinfo.php?ref=" + ref_id,
+                                      latitude=float(row["Latitude"]) if "Latitude" in row and row[
+                                          "Latitude"] != "" else None,
+                                      longitude=float(row["Longitude"]) if "Longitude" in row and row[
+                                          "Longitude"] != "" else None,
+                                      grid=row["Maidenhead Locator"]))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/pota.py

@@ -0,0 +1,34 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class POTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Parks on the Air"""
+
+    POLL_INTERVAL_DAYS = 7
+    SIG = "POTA"
+    DATA_URL = "https://pota.app/all_parks_ext.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()):
+            ref_id = row["reference"]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["name"] if "name" in row else None,
+                                      url="https://pota.app/#/park/" + ref_id,
+                                      grid=row["grid"] if "grid" in row else None,
+                                      latitude=float(row["latitude"]) if "latitude" in row and row[
+                                          "latitude"] != "" else None,
+                                      longitude=float(row["longitude"]) if "longitude" in row and row[
+                                          "longitude"] != "" else None))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/sig_ref_data_provider.py

@@ -0,0 +1,39 @@
+import logging
+from datetime import datetime
+
+import pytz
+
+from core.data_store import DATA_STORE
+
+
+class SIGRefDataProvider:
+    """Generic SIG reference data provider class. Subclasses of this query the individual URLs or files for data."""
+
+    def __init__(self, sig_name, provider_config):
+        """Constructor"""
+
+        self.sig_name = sig_name
+        self.enabled = provider_config["enabled"]
+        self.last_update_time = datetime.min.replace(tzinfo=pytz.UTC)
+        self.status = "Not Started" if self.enabled else "Disabled"
+        self.reference_count = 0
+
+
+    def start(self):
+        """Start the provider. This should return immediately after spawning threads to access the remote resources"""
+
+        raise NotImplementedError("Subclasses must implement this method")
+
+
+    def stop(self):
+        """Stop any threads and prepare for application shutdown"""
+
+        raise NotImplementedError("Subclasses must implement this method")
+
+    def _add_data(self, new_data):
+        """Add all the provided reference data objects to the data store."""
+
+        for d in new_data:
+            DATA_STORE.sigrefs[self.sig_name + ":" + d.id] = d
+        self.reference_count = len(new_data)
+        logging.info(f"Loaded %d references for %s into the data store.", self.reference_count, self.sig_name)

providers/sigrefdata/siota.py

@@ -0,0 +1,31 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class SIOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Silos on the Air"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "SIOTA"
+    DATA_URL = "https://www.silosontheair.com/data/silos.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()):
+            ref_id = row["SILO_CODE"]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["NAME"] if "NAME" in row else None,
+                                      grid=row["LOCATOR"] if "LOCATOR" in row else None,
+                                      latitude=float(row["LAT"]) if "LAT" in row else None,
+                                      longitude=float(row["LNG"]) if "LNG" in row else None))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/sota.py

@@ -0,0 +1,39 @@
+import csv
+
+from pyhamtools.locator import latlong_to_locator
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class SOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Summits on the Air"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "SOTA"
+    DATA_URL = "https://storage.sota.org.uk/summitslist.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()[1:]):
+            ref_id = row["SummitCode"]
+            latitude = float(row["Latitude"]) if "Latitude" in row and row["Latitude"] != "" else None
+            longitude = float(row["Longitude"]) if "Longitude" in row and row["Longitude"] != "" else None
+            ref = SIGRef(sig=self.SIG, id=ref_id, name=row["SummitName"] if "SummitName" in row else None,
+                                      url="https://www.sotadata.org.uk/en/summit/" + ref_id,
+                                      latitude=latitude,
+                                      longitude=longitude,
+                                      activation_score=int(row["Points"]) if "Points" in row else None)
+            if latitude and longitude:
+                ref.grid = latlong_to_locator(latitude, longitude, 6)
+            new_data.append(ref)
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/toilets.py

@@ -0,0 +1,30 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.local_file_sig_ref_data_provider import LocalFileSIGRefDataProvider
+
+
+class Toilets(LocalFileSIGRefDataProvider):
+    """SIG ref data provider for Toilets on the Air"""
+
+    SIG = "Toilets"
+    PATH = "datafiles/toilets.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.PATH)
+
+    def _file_to_data(self, path):
+        new_data = []
+        with open(path) as _f:
+            csv_data = _f.read()
+            dr = csv.DictReader(csv_data.splitlines())
+            for row in dr:
+                new_data.append(SIGRef(sig=self.SIG, id=row["ref"], name=row["ref"], latitude=float(row["lat"]),
+                                              longitude=float(row["lon"])))
+
+                # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest
+                # of the data in this case
+                if self._stop:
+                    break
+
+        return new_data

providers/sigrefdata/towers.py

@@ -0,0 +1,32 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class Towers(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Towers on the Air"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "Towers"
+    DATA_URL = "https://wwtota.com/servis/generate_csv.php?ref=&filter=all"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines(), delimiter=";"):
+            ref_id = row["Ref"]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["Nazev"] if "Nazev" in row else None,
+                                      url="https://wwtota.com/seznam/karta_rozhledny.php?ref=" + ref_id,
+                                      grid=row["Lokator"] if "Lokator" in row and row["Lokator"] != "" else None,
+                                      latitude=float(row["Lat"]) if "Lat" in row and row["Lat"] != "" else None,
+                                      longitude=float(row["Lon"]) if "Lon" in row and row["Lon"] != "" else None))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/wca.py

@@ -0,0 +1,49 @@
+import csv
+import logging
+
+from pyhamtools.locator import latlong_to_locator
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class WCA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for World Castles Award"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "WCA"
+    DATA_URL = "https://polo.ham2k.com/data/activities/wca/all-castles.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()):
+            ref_id = row["REF"]
+
+            coords_str = row["COORDINATES"]
+            latitude = None
+            longitude = None
+            grid = None
+            try:
+                if coords_str:
+                    split = coords_str.split(", ")
+                    latitude = float(split[0])
+                    longitude = float(split[1])
+                    grid = latlong_to_locator(latitude, longitude)
+            except ValueError:
+                logging.debug(f"Encountered dodgy formatting in WCA CSV, skipping location data for %s", ref_id)
+
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["CLEAN NAME"] if "CLEAN NAME" in row else None,
+                                      url="https://www.cqgma.org/zinfo.php?ref=" + ref_id,
+                                      latitude=latitude,
+                                      longitude=longitude,
+                                      grid=grid))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/wota.py

@@ -0,0 +1,36 @@
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class WOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Wainwrights on the Air"""
+
+    POLL_INTERVAL_DAYS = 365
+    SIG = "WOTA"
+    DATA_URL = "https://www.wota.org.uk/mapping/data/summits.json"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for feature in http_response.json().get("features", []):
+            ref_id = feature["properties"]["wotaId"]
+            # Fudge WOTA URLs. Outlying fell (LDO) URLs don't match their ID numbers but require 214 to be
+            # added to them
+            url = "https://www.wota.org.uk/MM_" + ref_id
+            if ref_id.upper().startswith("LDO-"):
+                number = int(ref_id.upper().replace("LDO-", ""))
+                url = "https://www.wota.org.uk/MM_LDO-" + str(number + 214)
+
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=feature["properties"]["title"], url=url,
+                                      grid=feature["properties"]["qthLocator"],
+                                      latitude=feature["geometry"]["coordinates"][1],
+                                      longitude=feature["geometry"]["coordinates"][0]))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/wwbota.py

@@ -0,0 +1,32 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class WWBOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Worldwide Bunkers on the Air"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "WWBOTA"
+    DATA_URL = "https://api.wwbota.org/bunkers/?format=CSV"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()):
+            ref_id = row["Reference"]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["Name"] if "Name" in row else None,
+                                      url="https://bunkerwiki.org/?s=" + ref_id if ref_id.startswith("B/G") else None,
+                                      grid=row["Locator"] if "Locator" in row and row["Locator"] != "" else None,
+                                      latitude=float(row["Lat"]) if "Lat" in row and row["Lat"] != "" else None,
+                                      longitude=float(row["Long"]) if "Long" in row and row["Long"] != "" else None))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/wwff.py

@@ -0,0 +1,35 @@
+import csv
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class WWFF(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Worldwide Flora & Fauna"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "WWFF"
+    DATA_URL = "https://wwff.co/wwff-data/wwff_directory.csv"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        for row in csv.DictReader(http_response.content.decode("utf-8-sig").splitlines()):
+            ref_id = row["reference"]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=row["name"] if "name" in row else None,
+                                      url="https://wwff.co/directory/?showRef=" + ref_id,
+                                      grid=row["iaruLocator"] if "iaruLocator" in row and row[
+                                          "iaruLocator"] != "-" else None,
+                                      latitude=float(row["latitude"]) if "latitude" in row and row[
+                                          "latitude"] != "" and row["latitude"] != "-" else None,
+                                      longitude=float(row["longitude"]) if "longitude" in row and row[
+                                          "longitude"] != "" and row["longitude"] != "-" else None))
+
+            # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest of
+            # the data in this case
+            if self._stop_event.is_set():
+                break
+
+        return new_data

providers/sigrefdata/zlota.py

@@ -0,0 +1,39 @@
+from pyhamtools.locator import latlong_to_locator
+
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class ZLOTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for New Zealand on the Air"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "ZLOTA"
+    DATA_URL = "https://ontheair.nz/assets/assets.json"
+
+    def __init__(self, provider_config):
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+
+    def _http_response_to_data(self, http_response):
+        new_data = []
+        data = http_response.json()
+        if isinstance(data, list):
+            for ref in data:
+                ref_id = ref["code"]
+                latitude = ref["y"]
+                longitude = ref["x"]
+
+                ref = SIGRef(sig=self.SIG, id=ref_id, name=ref["name"],
+                                          url="https://ontheair.nz/assets/" + ref_id.replace("/", "_"),
+                                          latitude=latitude,
+                                          longitude=longitude)
+                if latitude and longitude:
+                    ref.grid = latlong_to_locator(latitude, longitude, 6)
+                new_data.append(ref)
+
+                # Bail out if a stop has been requested, i.e. the program is shutting down - no need to parse the rest
+                # of the data in this case
+                if self._stop_event.is_set():
+                    break
+
+        return new_data

providers/solarconditions/giroionosonde.py

@@ -0,0 +1,173 @@
+import csv
+import logging
+from datetime import datetime, timezone, timedelta
+from threading import Thread, Event
+
+import pytz
+import requests
+from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from providers.solarconditions.ionosonde_utils import compute_band_states
+from providers.solarconditions.solar_conditions_provider import SolarConditionsProvider
+
+# Each station gets polled roughly once every hour (3600 seconds). Note that to avoid a burst of requests to the server
+# every hour, the requests for data from each station are spaced out throughout the hour, leading to one request being
+# sent every 1-2 minutes.
+POLL_INTERVAL = 3600
+# To avoid looking up all stations in the GIRO system and working out which ones are providing live data, this has been
+# manually determined and a CSV provided of all the stations that we can query for live data.
+STATIONS_INDEX = "datafiles/didbase-stations.csv"
+LGDC_URL = "https://lgdc.uml.edu/common/DIDBGetValues"
+HISTORY_HOURS = 24
+
+
+class GIROIonosonde(SolarConditionsProvider):
+    """Solar conditions provider using ionosonde data from the GIRO Data Center.
+    Queries foF2, MUF, and LUF measurements for all stations in datafiles/didbase-stations.csv.
+
+    Designed to run alongside KC2GProp even though they produce similar data. GIRO has more stations and includes LUF
+    data, but is less reliable and often offline."""
+
+    def __init__(self, provider_config):
+        super().__init__("GIRO Ionosonde Data", provider_config)
+        self._stations = self._load_stations()
+        self._thread = None
+        self._stop_event = Event()
+
+        # Pre-populate ionosonde_data with known station names for stations not already present,
+        # so the station dropdown is available before the first poll. Does not overwrite existing
+        # entries so KC2G cache data is preserved.
+        existing = self._solar_conditions.ionosonde_data or {}
+        new_entries = {
+            s["ursi"]: {"ursi": s["ursi"], "name": s["name"], "fof2": None, "muf": None,
+                        "luf": None, "band_states": None}
+            for s in self._stations if s["ursi"] not in existing
+        }
+        if new_entries:
+            self.update_data({"ionosonde_data": {**existing, **new_entries}})
+
+    @staticmethod
+    def _load_stations():
+        stations = []
+        with open(STATIONS_INDEX, newline='') as f:
+            for row in csv.reader(f):
+                if len(row) >= 2:
+                    stations.append({"ursi": row[0].strip(), "name": row[1].strip()})
+        return stations
+
+    def start(self):
+        logging.info(f"Set up query of GIRO ionosonde data API every {POLL_INTERVAL} seconds.")
+        self._thread = Thread(target=self._run, name="GIROIonosondeDataProvider")
+        self._thread.start()
+
+    def stop(self):
+        self._stop_event.set()
+
+    def _run(self):
+        # Real interval at which we poll is the "once per hour" divided by the number of stations, so each one gets
+        # polled once per hour, just not all at once
+        interval = POLL_INTERVAL / len(self._stations)
+        station_index = 0
+        while True:
+            self._poll_station(self._stations[station_index])
+            station_index = (station_index + 1) % len(self._stations)
+            if self._stop_event.wait(timeout=interval):
+                break
+
+    def _poll_station(self, station):
+        ursi = station["ursi"]
+        name = station["name"]
+        try:
+            logging.debug(f"Polling GIRO ionosonde data for {ursi} ({name})...")
+            now = datetime.now(timezone.utc)
+            from_time = now - timedelta(hours=HISTORY_HOURS)
+            cutoff_ts = from_time.timestamp()
+
+            fof2, muf, luf = self._fetch_station_data(ursi, from_time, now)
+            if not fof2 or not muf:
+                return
+
+            # Start from the existing ionosonde_data so stations provided by other providers
+            # (e.g. KC2GProp) are preserved for stations GIRO does not cover.
+            ionosonde_data = dict(self._solar_conditions.ionosonde_data or {})
+
+            # Merge GIRO's readings into any existing data for this station.
+            existing = ionosonde_data.get(ursi, {})
+            merged_fof2 = {**{float(t): v for t, v in (existing.get("fof2") or {}).items()}, **fof2}
+            merged_muf = {**{float(t): v for t, v in (existing.get("muf") or {}).items()}, **muf}
+            merged_luf = dict(luf) if luf else {}
+
+            merged_fof2 = {t: v for t, v in merged_fof2.items() if t >= cutoff_ts}
+            merged_muf = {t: v for t, v in merged_muf.items() if t >= cutoff_ts}
+            merged_luf = {t: v for t, v in merged_luf.items() if t >= cutoff_ts}
+
+            band_states = compute_band_states(merged_fof2, merged_muf, merged_luf)
+            ionosonde_data[ursi] = {
+                "ursi": ursi, "name": name,
+                "fof2": merged_fof2 or None,
+                "muf": merged_muf or None,
+                "luf": merged_luf or None,
+                "band_states": band_states,
+            }
+            self.update_data({"ionosonde_data": ionosonde_data})
+            self.status = "OK"
+            self.last_update_time = datetime.now(pytz.UTC)
+            logging.debug(f"Updated ionosonde data for {ursi} ({name}).")
+
+        except Exception:
+            self.status = "Error"
+            logging.exception(f"Exception fetching GIRO ionosonde data for {ursi} ({name})")
+
+    def _fetch_station_data(self, ursi, from_time, to_time):
+        """Fetch foF2, MUF and LUF readings for a station. Returns (fof2_dict, muf_dict, luf_dict) keyed by UNIX timestamp."""
+
+        from_str = from_time.strftime("%Y.%m.%d+%H:%M:%S")
+        to_str = to_time.strftime("%Y.%m.%d+%H:%M:%S")
+        url = f"{LGDC_URL}?ursiCode={ursi}&charName=foF2,MUFD,fmin&DMUF=3000&fromDate={from_str}&toDate={to_str}"
+        try:
+            http_response = requests.get(url, headers=HTTP_HEADERS, timeout=(5, 15))
+            if not http_response.ok:
+                logging.warning(f"HTTP {http_response.status_code} when calling Giro ionosonde API.")
+                return None, None, None
+            return self._parse_all(http_response.text)
+        except (ConnectTimeout, ReadTimeout):
+            logging.warning(f"Timeout when accessing Giro ionosonde API.")
+            return None, None, None
+        except ConnectionError:
+            logging.warning("Connection error when accessing Giro ionosonde API.")
+            return None, None, None
+
+    @staticmethod
+    def _parse_all(text):
+        """Parse web server response and return (fof2_dict, muf_dict, luf_dict) keyed by UNIX timestamp."""
+
+        fof2_data = {}
+        muf_data = {}
+        luf_data = {}
+        for line in text.splitlines():
+            line = line.strip()
+            if not line or line.startswith('#'):
+                continue
+            # Data rows have the following format: timestamp CS foF2 QD MUFD QD fmin QD
+            parts = line.split()
+            if len(parts) >= 5:
+                try:
+                    # Python 3.8 TZ parsing fudge
+                    ts = datetime.fromisoformat(parts[0].replace('Z', '+00:00')).timestamp()
+                except ValueError:
+                    continue
+                try:
+                    fof2_data[ts] = float(parts[2])
+                except ValueError:
+                    pass
+                try:
+                    muf_data[ts] = float(parts[4])
+                except ValueError:
+                    pass
+                if len(parts) >= 7:
+                    try:
+                        luf_data[ts] = float(parts[6])
+                    except ValueError:
+                        pass
+        return fof2_data, muf_data, luf_data

providers/solarconditions/hamqsl.py

@@ -4,8 +4,7 @@ from xml.etree import ElementTree
 import pytz
 from dateutil import parser as dateutil_parser, tz as dateutil_tz
 
-
-from solarconditionsproviders.http_solar_conditions_provider import HTTPSolarConditionsProvider
+from providers.solarconditions.http_solar_conditions_provider import HTTPSolarConditionsProvider
 
 POLL_INTERVAL = 3600  # 1 hour
 URL = "https://www.hamqsl.com/solarxml.php"
@@ -16,13 +15,9 @@ class HamQSL(HTTPSolarConditionsProvider):
     Provides solar flux index, geomagnetic indices, and HF/VHF propagation condition summaries."""
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, URL, POLL_INTERVAL)
+        super().__init__("HamQSL", provider_config, URL, POLL_INTERVAL)
 
     def _http_response_to_solar_conditions(self, http_response):
-        if http_response.status_code != 200:
-            logging.warning("HamQSL solar conditions API returned HTTP " + str(http_response.status_code))
-            return None
-
         root = ElementTree.fromstring(http_response.text)
         sd = root.find("solardata")
         if sd is None:
@@ -32,6 +27,9 @@ class HamQSL(HTTPSolarConditionsProvider):
         # Some error checking functions in case the data is janky.
 
         def text(tag, default=None):
+            if sd is None:
+                logging.warning("HamQSL solar conditions API returned unexpected XML structure")
+                return default
             el = sd.find(tag)
             return el.text.strip() if el is not None and el.text else default
 
@@ -104,7 +102,8 @@ class HamQSL(HTTPSolarConditionsProvider):
             "geomag_noise": text("signalnoise"),
             "hf_conditions": hf_conditions,
             "vhf_conditions": {
-                "vhf_aurora_northern_hemi": vhf_map.get(("vhf-aurora", "northern_hemi")).title().replace("Lat Aur", "Latitude"),
+                "vhf_aurora_northern_hemi": (vhf_map.get(("vhf-aurora", "northern_hemi")) or "").title().replace(
+                    "Lat Aur", "Latitude") or None,
                 "es_2m_europe": vhf_map.get(("E-Skip", "europe")),
                 "es_4m_europe": vhf_map.get(("E-Skip", "europe_4m")),
                 "es_6m_europe": vhf_map.get(("E-Skip", "europe_6m")),

providers/solarconditions/http_solar_conditions_provider.py

@@ -4,17 +4,18 @@ from threading import Thread, Event
 
 import pytz
 import requests
+from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
 
 from core.constants import HTTP_HEADERS
-from solarconditionsproviders.solar_conditions_provider import SolarConditionsProvider
+from providers.solarconditions.solar_conditions_provider import SolarConditionsProvider
 
 
 class HTTPSolarConditionsProvider(SolarConditionsProvider):
     """Generic solar conditions provider for providers that request data via HTTP(S). Subclasses implement
     _http_response_to_solar_conditions() to parse the specific API response format."""
 
-    def __init__(self, provider_config, url, poll_interval):
-        super().__init__(provider_config)
+    def __init__(self, name, provider_config, url, poll_interval):
+        super().__init__(name, provider_config)
         self._url = url
         self._poll_interval = poll_interval
         self._thread = None
@@ -23,7 +24,7 @@ class HTTPSolarConditionsProvider(SolarConditionsProvider):
     def start(self):
         logging.info(
             "Set up query of " + self.name + " solar conditions API every " + str(self._poll_interval) + " seconds.")
-        self._thread = Thread(target=self._run, daemon=True)
+        self._thread = Thread(target=self._run, name=f"HTTPSolarConditionsProvider-{self.name}")
         self._thread.start()
 
     def stop(self):
@@ -39,13 +40,22 @@ class HTTPSolarConditionsProvider(SolarConditionsProvider):
         try:
             logging.debug("Polling " + self.name + " solar conditions API...")
             http_response = requests.get(self._url, headers=HTTP_HEADERS, timeout=(5, 30))
+            # Check response code was good
+            if http_response.ok:
                 new_data = self._http_response_to_solar_conditions(http_response)
                 self.update_data(new_data)
 
                 self.status = "OK"
                 self.last_update_time = datetime.now(pytz.UTC)
                 logging.debug("Received data from " + self.name + " solar conditions API.")
+            else:
+                self.status = "Error"
+                logging.warning(f"HTTP {http_response.status_code} when calling {self.name} solar conditions API.")
 
+        except ConnectionError:
+            logging.warning(f"Connection error when accessing {self.name} solar conditions API.")
+        except (ConnectTimeout, ReadTimeout):
+            logging.warning(f"Timeout when accessing {self.name} solar conditions API.")
         except Exception:
             self.status = "Error"
             logging.exception("Exception in HTTP Solar Conditions Provider (" + self.name + ")")

providers/solarconditions/ionosonde_utils.py

@@ -0,0 +1,37 @@
+from core.constants import BANDS
+
+HF_BANDS = [b for b in BANDS if b.is_ham_hf]
+
+
+def _latest(d) -> float | None:
+    """Given a map where the key is a timestamp and the value is a number represented as a string, find the latest
+    timestamp and return the corresponding value as a float."""
+
+    val = str(d[max(d.keys())]) if d else None
+    return float(val) if (val is not None and val != "None") else None
+
+
+def compute_band_states(fof2_dict, muf_dict, luf_dict):
+    """Compute HF band states from the latest foF2, MUF and LUF values.
+
+    Returns a map where the keys are HF bands and the values are as follows:
+      "Closed" if band frequency is above MUF or below LUF (if known)
+      "Short" if band frequency is >= LUF and < foF2 (good for NVIS)
+      "Long" if band frequency is >= foF2 and < MUF (good for DX)
+    """
+
+    fof2 = _latest(fof2_dict)
+    muf = _latest(muf_dict)
+    luf = _latest(luf_dict) if luf_dict else None
+    if fof2 is None or muf is None:
+        return {}
+    band_states = {}
+    for band in HF_BANDS:
+        freq = band.start_freq / 1_000_000
+        if freq > muf or (luf is not None and freq < luf):
+            band_states[band.name] = "Closed"
+        elif freq < fof2:
+            band_states[band.name] = "Short"
+        else:
+            band_states[band.name] = "Long"
+    return band_states

providers/solarconditions/kc2gprop.py

@@ -0,0 +1,126 @@
+import logging
+from datetime import datetime, timezone, timedelta
+from threading import Thread, Event
+
+import pytz
+import requests
+from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from providers.solarconditions.ionosonde_utils import compute_band_states
+from providers.solarconditions.solar_conditions_provider import SolarConditionsProvider
+
+POLL_INTERVAL = 900  # 15 minutes
+KC2G_URL = "https://prop.kc2g.com/api/stations.json"
+HISTORY_HOURS = 24
+
+
+class KC2GProp(SolarConditionsProvider):
+    """Solar conditions provider using ionosonde data from prop.kc2g.com. The API returns only the latest reading per
+    station, so this provider polls every 15 minutes and accumulates a 24-hour time series by merging each new reading
+    into the persisted ionosonde_data, producing the same data structure as GIROIonosonde.
+
+    Designed to run alongside GIROIonosonde even though they produce similar data. KC2G is more reliable and is always
+    online, but has fewer stations and does not provide LUF data."""
+
+    def __init__(self, provider_config):
+        super().__init__("KC2G Propagation Data", provider_config)
+        self._thread = None
+        self._stop_event = Event()
+
+    def start(self):
+        logging.info(f"Set up query of KC2G ionosonde data API every {POLL_INTERVAL} seconds.")
+        self._thread = Thread(target=self._run, name="KC2GPropProvider")
+        self._thread.start()
+
+    def stop(self):
+        self._stop_event.set()
+
+    def _run(self):
+        while True:
+            self._poll()
+            if self._stop_event.wait(timeout=POLL_INTERVAL):
+                break
+
+    def _poll(self):
+        try:
+            logging.debug("Polling KC2G ionosonde data...")
+            http_response = requests.get(KC2G_URL, headers=HTTP_HEADERS, timeout=(5, 30))
+            if not http_response.ok:
+                logging.warning(f"HTTP {http_response.status_code} when calling KG2G ionosonde API.")
+                return
+
+            now = datetime.now(timezone.utc)
+            cutoff_ts = (now - timedelta(hours=HISTORY_HOURS)).timestamp()
+
+            # Start from existing ionosonde_data so the accumulated time series survives across polls and restarts and
+            # stations provided only by GIROIonosonde are not discarded
+            ionosonde_data = dict(self._solar_conditions.ionosonde_data or {})
+            updated_count = 0
+
+            for reading in http_response.json():
+                station = reading.get("station", {})
+                ursi = station.get("code")
+                name = station.get("name")
+                if not ursi or not name:
+                    continue
+
+                time_str = reading.get("time")
+                if not time_str:
+                    continue
+                try:
+                    ts = datetime.fromisoformat(time_str)
+                    if ts.tzinfo is None:
+                        ts = ts.replace(tzinfo=timezone.utc)
+                    ts_float = ts.timestamp()
+                except ValueError:
+                    continue
+
+                # Skip readings outside our history window (some stations have months-old data)
+                if ts_float < cutoff_ts:
+                    continue
+
+                fof2_val = reading.get("fof2")
+                muf_val = reading.get("mufd")
+                if fof2_val is None and muf_val is None:
+                    continue
+
+                # Merge this reading into the existing time series for the station.
+                existing = ionosonde_data.get(ursi, {})
+                fof2_dict = dict(existing.get("fof2") or {})
+                muf_dict = dict(existing.get("muf") or {})
+                # LUF is not available from KC2G; carry forward whatever GIRO has written.
+                luf_dict = {float(t): v for t, v in (existing.get("luf") or {}).items()}
+
+                fof2_dict[ts_float] = fof2_val
+                muf_dict[ts_float] = muf_val
+
+                # Trim all series to the 24-hour window.
+                fof2_dict = {t: v for t, v in fof2_dict.items() if t >= cutoff_ts}
+                muf_dict = {t: v for t, v in muf_dict.items() if t >= cutoff_ts}
+                luf_dict = {t: v for t, v in luf_dict.items() if t >= cutoff_ts}
+
+                band_states = compute_band_states(fof2_dict, muf_dict, luf_dict)
+                ionosonde_data[ursi] = {
+                    "ursi": ursi,
+                    "name": name,
+                    "fof2": fof2_dict or None,
+                    "muf": muf_dict or None,
+                    "luf": luf_dict or None,
+                    "band_states": band_states,
+                }
+                updated_count += 1
+
+            self.update_data({"ionosonde_data": ionosonde_data})
+            self.status = "OK"
+            self.last_update_time = datetime.now(pytz.UTC)
+            logging.debug(f"Updated KC2G ionosonde data for {updated_count} stations.")
+
+        except ConnectionError:
+            logging.warning("Connection error when accessing KC2G ionosonde API.")
+        except (ConnectTimeout, ReadTimeout):
+            logging.warning(f"Timeout when accessing KC2G ionosonde API.")
+        except Exception:
+            self.status = "Error"
+            logging.exception("Exception in KC2G ionosonde data provider")
+            self._stop_event.wait(timeout=1)

providers/solarconditions/noaa3dayforecast.py

@@ -2,7 +2,7 @@ import logging
 import re
 from datetime import datetime, timezone
 
-from solarconditionsproviders.http_solar_conditions_provider import HTTPSolarConditionsProvider
+from providers.solarconditions.http_solar_conditions_provider import HTTPSolarConditionsProvider
 
 POLL_INTERVAL = 10800  # Every 3 hours
 URL = "https://services.swpc.noaa.gov/text/3-day-forecast.txt"
@@ -13,12 +13,13 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
     corresponding fields in the solar conditions object.."""
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, URL, POLL_INTERVAL)
+        super().__init__("NOAA 3-day Forecast", provider_config, URL, POLL_INTERVAL)
 
     @staticmethod
     def _parse_percentage_table(lines, section_header, year):
         """Find and parse a forecast table using percentages, identified by section_header. This is common to the lookup
         of the solar storm and radio blackout forecast parsing."""
+
         start_idx = None
         for i, line in enumerate(lines):
             if section_header in line:
@@ -28,7 +29,7 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
             logging.warning(f"NOAA 3-day forecast: could not find '{section_header}' section")
             return None
 
-        # Find the date header line — the first line within the next few that contains month+day patterns
+        # Find the date header line by scanning the next few lines for month & day patterns
         date_header_idx = None
         for j in range(start_idx + 1, min(start_idx + 6, len(lines))):
             if re.search(r'[A-Za-z]{3}\s+\d{2}', lines[j]):
@@ -37,12 +38,12 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
         if date_header_idx is None:
             logging.warning(f"NOAA 3-day forecast: could not find date header after '{section_header}'")
             return None
-
         date_matches = re.findall(r'([A-Za-z]{3})\s+(\d{2})', lines[date_header_idx])
         if not date_matches:
             logging.warning(f"NOAA 3-day forecast: no dates in header: {lines[date_header_idx]}")
             return None
 
+        # Figure out the date based on the line found
         column_timestamps = []
         for month_str, day_str in date_matches:
             try:
@@ -52,7 +53,7 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
                 logging.warning(f"NOAA 3-day forecast: could not parse date: {month_str} {day_str} {year}")
                 return None
 
-        # Parse data rows: each non-empty line should have a text label and percentage values
+        # Parse data rows. Each non-empty line should have a text label followed by percentage values
         result = {}
         for line in lines[date_header_idx + 1:]:
             line_stripped = line.strip()
@@ -65,6 +66,7 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
                 if result:
                     break
                 continue
+
             # Row label is everything before the first percentage value
             row_label = line_stripped[:line_stripped.index(pct_matches[0].group())].strip()
             row_data = {}
@@ -78,10 +80,6 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
         return result if result else None
 
     def _http_response_to_solar_conditions(self, http_response):
-        if http_response.status_code != 200:
-            logging.warning("NOAA K-index forecast API returned HTTP " + str(http_response.status_code))
-            return None
-
         lines = http_response.text.splitlines()
 
         # Find the "NOAA Kp index breakdown" section header
@@ -90,7 +88,6 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
             if "NOAA Kp index breakdown" in line:
                 start_idx = i
                 break
-
         if start_idx is None:
             logging.warning("NOAA K-index forecast: could not find 'NOAA Kp index breakdown' section")
             return None

providers/solarconditions/solar_conditions_provider.py

@@ -2,24 +2,21 @@ from datetime import datetime
 
 import pytz
 
+from core.data_store import DATA_STORE
+
 
 class SolarConditionsProvider:
     """Generic solar conditions provider class. Subclasses of this query individual APIs for space weather and
      propagation data."""
 
-    def __init__(self, provider_config):
+    def __init__(self, name, provider_config):
         """Constructor"""
 
-        self.name = provider_config["name"]
+        self.name = name
         self.enabled = provider_config["enabled"]
         self.last_update_time = datetime.min.replace(tzinfo=pytz.UTC)
         self.status = "Not Started" if self.enabled else "Disabled"
-        self._solar_conditions = None
-
-    def setup(self, solar_conditions):
-        """Set up the provider, giving it the solar conditions dict to update"""
-
-        self._solar_conditions = solar_conditions
+        self._solar_conditions = DATA_STORE.solar_conditions
 
     def start(self):
         """Start the provider. This should return immediately after spawning threads to access the remote resources"""

providers/spot/aprsis.py

@@ -7,15 +7,15 @@ import pytz
 
 from core.config import SERVER_OWNER_CALLSIGN
 from data.spot import Spot
-from spotproviders.spot_provider import SpotProvider
+from providers.spot.spot_provider import SpotProvider
 
 
 class APRSIS(SpotProvider):
     """Spot provider for the APRS-IS."""
 
     def __init__(self, provider_config):
-        super().__init__(provider_config)
-        self._thread = Thread(target=self._connect)
+        super().__init__("APRS-IS", provider_config)
+        self._thread = Thread(target=self._connect, name="APRSISSpotProvider")
         self._thread.daemon = True
         self._aprsis = None
 
@@ -37,20 +37,20 @@ class APRSIS(SpotProvider):
 
     def _handle(self, data):
         # Split SSID in "from" call and store separately
-        from_parts = data["from"].split("-").upper()
-        dx_call = from_parts[0]
-        dx_ssid = from_parts[1] if len(from_parts) > 1 else None
-        via_parts = data["via"].split("-").upper()
-        de_call = via_parts[0]
-        de_ssid = via_parts[1] if len(via_parts) > 1 else None
+        from_parts = str(data["from"]).split("-")
+        dx_call = from_parts[0].upper()
+        dx_ssid = from_parts[1].upper() if len(from_parts) > 1 else None
+        via_parts = str(data["via"]).split("-")
+        de_call = via_parts[0].upper()
+        de_ssid = via_parts[1].upper() if len(via_parts) > 1 else None
         spot = Spot(source="APRS-IS",
                     dx_call=dx_call,
                     dx_ssid=dx_ssid,
                     de_call=de_call,
                     de_ssid=de_ssid,
-                    comment=data["comment"] if "comment" in data else None,
-                    dx_latitude=data["latitude"] if "latitude" in data else None,
-                    dx_longitude=data["longitude"] if "longitude" in data else None,
+                    comment=str(data["comment"]) if "comment" in data else None,
+                    dx_latitude=float(data["latitude"]) if "latitude" in data else None,
+                    dx_longitude=float(data["longitude"]) if "longitude" in data else None,
                     time=datetime.now(
                         pytz.UTC).timestamp())  # APRS-IS spots are live so we can assume spot time is "now"
 

providers/spot/dxcluster.py

@@ -9,7 +9,7 @@ import telnetlib3
 
 from core.config import SERVER_OWNER_CALLSIGN
 from data.spot import Spot
-from spotproviders.spot_provider import SpotProvider
+from providers.spot.spot_provider import SpotProvider
 
 
 class DXCluster(SpotProvider):
@@ -26,7 +26,8 @@ class DXCluster(SpotProvider):
     def __init__(self, provider_config):
         """Constructor requires hostname and port"""
 
-        super().__init__(provider_config)
+        name = provider_config["name"] if "name" in provider_config else "Cluster"
+        super().__init__(name, provider_config)
         self._hostname = provider_config["host"]
         self._port = provider_config["port"]
         self._login_prompt = provider_config["login_prompt"] if "login_prompt" in provider_config else "login:"
@@ -35,7 +36,7 @@ class DXCluster(SpotProvider):
         self._allow_rbn_spots = provider_config["allow_rbn_spots"] if "allow_rbn_spots" in provider_config else False
         self._spot_line_pattern = self._LINE_PATTERN_ALLOW_RBN if self._allow_rbn_spots else self._LINE_PATTERN_EXCLUDE_RBN
         self._telnet = None
-        self._thread = Thread(target=self._handle)
+        self._thread = Thread(target=self._handle, name=f"DXClusterSpotProvider-{self.name}")
         self._thread.daemon = True
         self._running = True
 
@@ -59,6 +60,10 @@ class DXCluster(SpotProvider):
                     self._telnet.write((self._login_callsign + "\n").encode("latin-1"))
                     connected = True
                     logging.info("DX Cluster " + self._hostname + " connected.")
+                except ConnectionRefusedError:
+                    self.status = "Error"
+                    logging.warning("Connection refused to DX cluster " + self._hostname)
+                    sleep(300)
                 except Exception:
                     self.status = "Error"
                     logging.exception("Exception while connecting to DX Cluster Provider (" + self._hostname + ").")
@@ -72,7 +77,8 @@ class DXCluster(SpotProvider):
                     match = self._spot_line_pattern.match(telnet_output.decode("latin-1"))
                     if match:
                         spot_time = datetime.strptime(match.group(5), "%H%MZ")
-                        spot_datetime = datetime.combine(datetime.now(pytz.UTC).date(), spot_time.time(), tzinfo=pytz.UTC)
+                        spot_datetime = datetime.combine(datetime.now(pytz.UTC).date(), spot_time.time(),
+                                                         tzinfo=pytz.UTC)
                         spot = Spot(source=self.name,
                                     dx_call=match.group(3),
                                     de_call=match.group(1),

providers/spot/gma.py

@@ -0,0 +1,115 @@
+import logging
+from datetime import datetime
+
+import pytz
+
+from core.constants import HTTP_HEADERS
+from core.url_data_cache import URLDataCache
+from data.sig_ref import SIGRef
+from data.spot import Spot
+from providers.spot.http_spot_provider import HTTPSpotProvider
+
+
+class GMA(HTTPSpotProvider):
+    """Spot provider for General Mountain Activity"""
+
+    POLL_INTERVAL_SEC = 120
+    SPOTS_URL = "https://www.gma.rocks/api/spots/25/"
+    # GMA spots don't contain the details of the programme they are for, we need a separate lookup for that
+    REF_INFO_URL_ROOT = "https://www.gma.rocks/api/ref/?"
+
+    def __init__(self, provider_config):
+        # Ensure there is an API key in our config, and set up the query URL using it. If no key is provided,
+        # disable this spot provider.
+        self._api_key = provider_config.get("api-key", "")
+        if self._api_key == "":
+            provider_config["enabled"] = False
+            logging.warning("GMA spot provider configured but no api key was provided, this API will not be queried.")
+        self._url_data_cache = URLDataCache("GMA")
+
+        super().__init__("GMA", provider_config, self.SPOTS_URL + "?key=" + self._api_key, self.POLL_INTERVAL_SEC)
+
+    def _http_response_to_spots(self, http_response):
+        new_spots = []
+        # Iterate through source data
+        if "RCD" in http_response.json():
+            for source_spot in http_response.json()["RCD"]:
+                # Convert to our spot format
+                spot = Spot(source=self.name,
+                            dx_call=source_spot["ACTIVATOR"].upper(),
+                            de_call=source_spot["SPOTTER"].upper(),
+                            # Seen GMA spots with no frequency or with "QRT" in this field
+                            freq=float(source_spot["QRG"]) * 1000 if (
+                                    source_spot["QRG"] != "" and source_spot["QRG"] != "QRT") else None,
+                            # Filter out some weird mode strings
+                            mode=source_spot["MODE"].upper() if "<>" not in source_spot["MODE"] else None,
+                            comment=source_spot["TEXT"],
+                            sig_refs=[SIGRef(id=source_spot["REF"], sig="", name=source_spot["NAME"])],
+                            time=datetime.strptime(source_spot["DATE"] + source_spot["TIME"], "%Y%m%d%H%M").replace(
+                                tzinfo=pytz.UTC).timestamp(),
+                            # Seen GMA spots with no (or empty) lat/lon
+                            dx_latitude=float(source_spot["LAT"]) if (
+                                    source_spot["LAT"] and source_spot["LAT"] != "") else None,
+                            dx_longitude=float(source_spot["LON"]) if (
+                                    source_spot["LON"] and source_spot["LON"] != "") else None,
+                            qrt=source_spot["QRG"] == "QRT")
+
+                # GMA doesn't give what programme (SIG) the reference is for until we separately look it up.
+                if "REF" in source_spot:
+                    try:
+                        ref_response = self._url_data_cache.get(self.REF_INFO_URL_ROOT + source_spot["REF"],
+                                                          headers=HTTP_HEADERS)
+                        # Sometimes this is blank even if it's a 200 response, so handle that
+                        if ref_response.ok and ref_response.text is not None and ref_response.text != "":
+                            ref_info = ref_response.json()
+                            # If this is POTA, SOTA or WWFF data we already have it through other means, so ignore. POTA and WWFF
+                            # spots come through with reftype=POTA or reftype=WWFF. SOTA is harder to figure out because both SOTA
+                            # and GMA summits come through with reftype=Summit, so we must check for the presence of a "sota" entry
+                            # to determine if it's a SOTA summit.
+                            if spot.sig_refs and "reftype" in ref_info and ref_info["reftype"] not in ["POTA",
+                                                                                                       "WWFF"] and (
+                                    ref_info["reftype"] != "Summit" or "sota" not in ref_info or ref_info[
+                                "sota"] == ""):
+                                match ref_info["reftype"]:
+                                    case "Summit":
+                                        spot.sig_refs[0].sig = "GMA"
+                                        spot.sig = "GMA"
+                                    case "IOTA Island":
+                                        spot.sig_refs[0].sig = "IOTA"
+                                        spot.sig = "IOTA"
+                                    case "Lighthouse (ILLW)":
+                                        spot.sig_refs[0].sig = "ILLW"
+                                        spot.sig = "ILLW"
+                                    case "Lighthouse (ARLHS)":
+                                        spot.sig_refs[0].sig = "ARLHS"
+                                        spot.sig = "ARLHS"
+                                    case "Castle":
+                                        spot.sig_refs[0].sig = "WCA"
+                                        spot.sig = "WCA"
+                                    case "Mill":
+                                        spot.sig_refs[0].sig = "MOTA"
+                                        spot.sig = "MOTA"
+                                    case _:
+                                        logging.warning("GMA spot found with ref type " + ref_info[
+                                            "reftype"] + ", developer needs to add support for this!")
+                                        spot.sig_refs[0].sig = ref_info["reftype"]
+                                        spot.sig = ref_info["reftype"]
+
+                                # Add to our list. Don't worry about de-duping, removing old spots etc. at this point;
+                                # other code will do that for us.
+                                new_spots.append(spot)
+
+                        elif not ref_response.from_cache:
+                            if not ref_response.ok:
+                                logging.warning(
+                                    f"HTTP {ref_response.status_code} when looking up GMA ref {source_spot["REF"]}")
+                            else:
+                                logging.warning(
+                                    f"GMA API returned a malformed response when looking up ref {source_spot["REF"]}")
+                    except:
+                        logging.warning("Exception when looking up " + self.REF_INFO_URL_ROOT + source_spot[
+                            "REF"] + ", ignoring this spot for now")
+        else:
+            logging.warning(f"The GMA API returned an unexpected response (HTTP {http_response.status_code}).")
+
+        return new_spots

providers/spot/hema.py

@@ -0,0 +1,75 @@
+import logging
+import re
+from datetime import datetime
+
+import pytz
+import requests
+from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from data.sig_ref import SIGRef
+from data.spot import Spot
+from providers.spot.http_spot_provider import HTTPSpotProvider
+
+
+class HEMA(HTTPSpotProvider):
+    """Spot provider for HuMPs Excluding Marilyns Award"""
+
+    POLL_INTERVAL_SEC = 300
+    # HEMA wants us to check for a "spot seed" from the API and see if it's actually changed before querying the main
+    # data API. So it's actually the SPOT_SEED_URL that we pass into the constructor and get the superclass to call on a
+    # timer. The actual data lookup all happens after parsing and checking the seed.
+    SPOT_SEED_URL = "http://www.hema.org.uk/spotSeed.jsp"
+    SPOTS_URL = "http://www.hema.org.uk/spotsMobile.jsp"
+    FREQ_MODE_PATTERN = re.compile("^([\\d.]*) \\((.*)\\)$")
+    SPOTTER_COMMENT_PATTERN = re.compile("^\\((.*)\\) (.*)$")
+
+    def __init__(self, provider_config):
+        super().__init__("HEMA", provider_config, self.SPOT_SEED_URL, self.POLL_INTERVAL_SEC)
+        self._spot_seed = ""
+
+    def _http_response_to_spots(self, http_response):
+        # OK, source data is actually just the spot seed at this point. We'll then go on to fetch real data if we know
+        # this has changed.
+        spot_seed_changed = http_response.text != self._spot_seed
+        self._spot_seed = http_response.text
+
+        new_spots = []
+        # OK, if the spot seed actually changed, now we make the real request for data.
+        if spot_seed_changed:
+            try:
+                source_data = requests.get(self.SPOTS_URL, headers=HTTP_HEADERS, timeout=(5, 30))
+                source_data_items = source_data.text.split("=")
+                # Iterate through source data items.
+                for source_spot in source_data_items:
+                    spot_items = source_spot.split(";")
+                    # Any line with less than 9 items is not a proper spot line
+                    if len(spot_items) >= 9:
+                        # Fiddle with some data to extract bits we need. Freq/mode and spotter/comment come in combined fields.
+                        freq_mode_match = re.search(self.FREQ_MODE_PATTERN, spot_items[5])
+                        spotter_comment_match = re.search(self.SPOTTER_COMMENT_PATTERN, spot_items[6])
+                        if not freq_mode_match or not spotter_comment_match:
+                            continue
+
+                        # Convert to our spot format
+                        spot = Spot(source=self.name,
+                                    dx_call=spot_items[2].upper(),
+                                    de_call=spotter_comment_match.group(1).upper(),
+                                    freq=float(freq_mode_match.group(1)) * 1000000,
+                                    mode=freq_mode_match.group(2).upper(),
+                                    comment=spotter_comment_match.group(2),
+                                    sig="HEMA",
+                                    sig_refs=[SIGRef(id=spot_items[3].upper(), sig="HEMA", name=spot_items[4])],
+                                    time=datetime.strptime(spot_items[0], "%d/%m/%Y %H:%M").replace(
+                                        tzinfo=pytz.UTC).timestamp(),
+                                    dx_latitude=float(spot_items[7]),
+                                    dx_longitude=float(spot_items[8]))
+
+                        # Add to our list. Don't worry about de-duping, removing old spots etc. at this point; other
+                        # code will do that for us.
+                        new_spots.append(spot)
+            except (ConnectTimeout, ReadTimeout):
+                logging.warning(f"Timeout when accessing HEMA spots API.")
+            except ConnectionError:
+                logging.warning("Connection error when accessing HEMA spots API.")
+        return new_spots

providers/spot/http_spot_provider.py

@@ -4,17 +4,19 @@ from threading import Thread, Event
 
 import pytz
 import requests
+from requests import ReadTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout
 
 from core.constants import HTTP_HEADERS
-from spotproviders.spot_provider import SpotProvider
+from providers.spot.spot_provider import SpotProvider
 
 
 class HTTPSpotProvider(SpotProvider):
     """Generic spot provider class for providers that request data via HTTP(S). Just for convenience to avoid code
     duplication. Subclasses of this query the individual APIs for data."""
 
-    def __init__(self, provider_config, url, poll_interval):
-        super().__init__(provider_config)
+    def __init__(self, name, provider_config, url, poll_interval):
+        super().__init__(name, provider_config)
         self._url = url
         self._poll_interval = poll_interval
         self._thread = None
@@ -24,7 +26,7 @@ class HTTPSpotProvider(SpotProvider):
         # Fire off the polling thread. It will poll immediately on startup, then sleep for poll_interval between
         # subsequent polls, so start() returns immediately and the application can continue starting.
         logging.info("Set up query of " + self.name + " spot API every " + str(self._poll_interval) + " seconds.")
-        self._thread = Thread(target=self._run, daemon=True)
+        self._thread = Thread(target=self._run, name=f"HTTPSpotProvider-{self.name}")
         self._thread.start()
 
     def stop(self):
@@ -41,6 +43,8 @@ class HTTPSpotProvider(SpotProvider):
             # Request data from API
             logging.debug("Polling " + self.name + " spot API...")
             http_response = requests.get(self._url, headers=HTTP_HEADERS, timeout=(5, 30))
+            # Check response code was good
+            if http_response.ok:
                 # Pass off to the subclass for processing
                 new_spots = self._http_response_to_spots(http_response)
                 # Submit the new spots for processing. There might not be any spots for the less popular programs.
@@ -50,10 +54,17 @@ class HTTPSpotProvider(SpotProvider):
                 self.status = "OK"
                 self.last_update_time = datetime.now(pytz.UTC)
                 logging.debug("Received data from " + self.name + " spot API.")
+            else:
+                self.status = "Error"
+                logging.warning(f"HTTP {http_response.status_code} when calling {self.name} spot API.")
 
+        except ConnectionError:
+            logging.warning(f"Connection error when accessing {self.name} spots API.")
+        except (ConnectTimeout, ReadTimeout):
+            logging.warning(f"Timeout when accessing {self.name} spots API.")
         except Exception:
             self.status = "Error"
-            logging.exception("Exception in HTTP JSON Spot Provider (" + self.name + ")")
+            logging.exception("Exception in HTTP Spot Provider (" + self.name + ")")
             self._stop_event.wait(timeout=1)
 
     def _http_response_to_spots(self, http_response):

providers/spot/llota.py

@@ -2,7 +2,7 @@ from datetime import datetime
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class LLOTA(HTTPSpotProvider):
@@ -12,7 +12,7 @@ class LLOTA(HTTPSpotProvider):
     SPOTS_URL = "https://llota.app/api/public/spots"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("LLOTA", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []
@@ -22,8 +22,8 @@ class LLOTA(HTTPSpotProvider):
             comment = None
             spotter = None
             if "history" in source_spot and len(source_spot["history"]) > 0:
-                comment = source_spot["history"][-1]["comment"]
-                spotter = source_spot["history"][-1]["spotter_callsign"]
+                comment = str(source_spot["history"][-1]["comment"])
+                spotter = str(source_spot["history"][-1]["spotter_callsign"])
             # Convert to our spot format
             spot = Spot(source=self.name,
                         source_id=source_spot["id"],

providers/spot/parksnpeaks.py

@@ -6,7 +6,7 @@ import pytz
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class ParksNPeaks(HTTPSpotProvider):
@@ -17,7 +17,7 @@ class ParksNPeaks(HTTPSpotProvider):
     SIOTA_LIST_URL = "https://www.silosontheair.com/data/silos.csv"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("ParksNPeaks", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []
@@ -30,8 +30,8 @@ class ParksNPeaks(HTTPSpotProvider):
                             dx_call=source_spot["actCallsign"].upper(),
                             de_call=source_spot["actSpoter"].upper() if source_spot["actSpoter"] != "" else None,
                             # typo exists in API
-                            freq=float(source_spot["actFreq"].replace(",", "")) * 1000000 if (
-                                    source_spot["actFreq"] != "") else None,
+                            freq=float(source_spot["actFreq"].replace(",", "").replace("+-", "")
+                                       .replace("+/-", "").strip()) * 1000000 if (source_spot["actFreq"] != "") else None,
                             # Seen PNP spots with empty frequency, and with comma-separated thousands digits
                             mode=source_spot["actMode"].upper(),
                             comment=source_spot["actComments"],
@@ -39,9 +39,9 @@ class ParksNPeaks(HTTPSpotProvider):
                                 tzinfo=pytz.UTC).timestamp())
 
                 # Extract a de_call if it's in the comment but not in the "actSpoter" field
-                m = re.search(r"\(de ([A-Za-z0-9]*)\)", spot.comment)
+                m = re.search(r"\(de ([A-Za-z0-9]*)\)", spot.comment or "")
                 if not spot.de_call and m:
-                    spot.de_call = m.group(1)
+                    spot.de_call = str(m.group(1))
 
                 # Record SIG information. Sometimes we get a "SIG" of "QRP", which we ignore as it's not a programme with a
                 # defined set of references
@@ -49,14 +49,15 @@ class ParksNPeaks(HTTPSpotProvider):
                 sig_ref = source_spot["actSiteID"]
                 if sig and sig != "" and sig != "QRP" and sig_ref and sig_ref != "":
                     spot.sig = sig
-                    spot.sig_refs = [SIGRef(id=source_spot["actSiteID"], sig=source_spot["actClass"].upper())]
+                    sig_refs = [SIGRef(id=source_spot["actSiteID"], sig=source_spot["actClass"].upper())]
+                    spot.sig_refs = sig_refs
 
                     # Free text location is not present in all spots, so only add it if it's set
                     if "actLocation" in source_spot and source_spot["actLocation"] != "":
-                        spot.sig_refs[0].name = source_spot["actLocation"]
+                        sig_refs[0].name = source_spot["actLocation"]
 
                     # Log a warning for the developer if PnP gives us an unknown programme we've never seen before
-                    if sig not in ["POTA", "SOTA", "WWFF", "SIOTA", "ZLOTA", "KRMNPA"]:
+                    if sig not in ["POTA", "SOTA", "WWFF", "SIOTA", "ZLOTA", "KRMNPA", "LLOTA"]:
                         logging.warning("PNP spot found with sig " + sig + ", developer needs to add support for this!")
 
                 # Add new spot to the list

providers/spot/pota.py

@@ -4,7 +4,7 @@ import pytz
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class POTA(HTTPSpotProvider):
@@ -14,7 +14,7 @@ class POTA(HTTPSpotProvider):
     SPOTS_URL = "https://api.pota.app/spot/activator"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("POTA", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []

providers/spot/rbn.py

@@ -9,7 +9,7 @@ import telnetlib3
 
 from core.config import SERVER_OWNER_CALLSIGN
 from data.spot import Spot
-from spotproviders.spot_provider import SpotProvider
+from providers.spot.spot_provider import SpotProvider
 
 
 class RBN(SpotProvider):
@@ -23,10 +23,11 @@ class RBN(SpotProvider):
     def __init__(self, provider_config):
         """Constructor requires port number."""
 
-        super().__init__(provider_config)
+        name = provider_config["name"] if "name" in provider_config else "RBN"
+        super().__init__(name, provider_config)
         self._port = provider_config["port"]
         self._telnet = None
-        self._thread = Thread(target=self._handle)
+        self._thread = Thread(target=self._handle, name=f"RBNSpotProvider-{self.name}")
         self._thread.daemon = True
         self._running = True
 
@@ -46,7 +47,7 @@ class RBN(SpotProvider):
                     self.status = "Connecting"
                     logging.info("RBN port " + str(self._port) + " connecting...")
                     self._telnet = telnetlib3.Telnet("telnet.reversebeacon.net", self._port)
-                    telnet_output = self._telnet.read_until("Please enter your call: ".encode("latin-1"))
+                    self._telnet.read_until("Please enter your call: ".encode("latin-1"))
                     self._telnet.write((SERVER_OWNER_CALLSIGN + "\n").encode("latin-1"))
                     connected = True
                     logging.info("RBN port " + str(self._port) + " connected.")
@@ -63,7 +64,8 @@ class RBN(SpotProvider):
                     match = self._LINE_PATTERN.match(telnet_output.decode("latin-1"))
                     if match:
                         spot_time = datetime.strptime(match.group(5), "%H%MZ")
-                        spot_datetime = datetime.combine(datetime.now(pytz.UTC).date(), spot_time.time(), tzinfo=pytz.UTC)
+                        spot_datetime = datetime.combine(datetime.now(pytz.UTC).date(), spot_time.time(),
+                                                         tzinfo=pytz.UTC)
                         spot = Spot(source=self.name,
                                     dx_call=match.group(3),
                                     de_call=match.group(1),

providers/spot/sota.py

@@ -0,0 +1,63 @@
+import logging
+from datetime import datetime
+
+import requests
+from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from data.sig_ref import SIGRef
+from data.spot import Spot
+from providers.spot.http_spot_provider import HTTPSpotProvider
+
+
+class SOTA(HTTPSpotProvider):
+    """Spot provider for Summits on the Air"""
+
+    POLL_INTERVAL_SEC = 120
+    # SOTA wants us to check for an "epoch" from the API and see if it's actually changed before querying the main data
+    # APIs. So it's actually the EPOCH_URL that we pass into the constructor and get the superclass to call on a timer.
+    # The actual data lookup all happens after parsing and checking the epoch.
+    EPOCH_URL = "https://api-db2.sota.org.uk/api/spots/epoch"
+    SPOTS_URL = "https://api-db2.sota.org.uk/api/spots/60/all/all"
+
+    def __init__(self, provider_config):
+        super().__init__("SOTA", provider_config, self.EPOCH_URL, self.POLL_INTERVAL_SEC)
+        self._api_epoch = ""
+
+    def _http_response_to_spots(self, http_response):
+        # OK, source data is actually just the epoch at this point. We'll then go on to fetch real data if we know this
+        # has changed.
+        epoch_changed = http_response.text != self._api_epoch
+        self._api_epoch = http_response.text
+
+        new_spots = []
+        # OK, if the epoch actually changed, now we make the real request for data.
+        if epoch_changed:
+            try:
+                source_data = requests.get(self.SPOTS_URL, headers=HTTP_HEADERS, timeout=(5, 30)).json()
+                # Iterate through source data
+                for source_spot in source_data:
+                    # Convert to our spot format
+                    spot = Spot(source=self.name,
+                                source_id=source_spot["id"],
+                                dx_call=source_spot["activatorCallsign"].upper(),
+                                dx_name=source_spot["activatorName"],
+                                de_call=source_spot["callsign"].upper(),
+                                freq=(float(source_spot["frequency"]) * 1000000) if (
+                                        source_spot["frequency"] is not None) else None,
+                                # Seen SOTA spots with no frequency!
+                                mode=source_spot["mode"].upper(),
+                                comment=source_spot["comments"],
+                                sig="SOTA",
+                                sig_refs=[SIGRef(id=source_spot["summitCode"], sig="SOTA", name=source_spot["summitName"],
+                                                 activation_score=source_spot["points"])],
+                                time=datetime.fromisoformat(source_spot["timeStamp"].replace("Z", "+00:00")).timestamp())
+
+                    # Add to our list. Don't worry about de-duping, removing old spots etc. at this point; other code will do
+                    # that for us.
+                    new_spots.append(spot)
+            except ConnectionError:
+                logging.warning("Connection error when accessing SOTA spots API")
+            except (ConnectTimeout, ReadTimeout):
+                logging.warning(f"Timeout when accessing SOTA spots API.")
+        return new_spots

providers/spot/spot_provider.py

@@ -2,28 +2,21 @@ from datetime import datetime
 
 import pytz
 
-from core.config import MAX_SPOT_AGE
+from core.data_store import DATA_STORE
 
 
 class SpotProvider:
     """Generic spot provider class. Subclasses of this query the individual APIs for data."""
 
-    def __init__(self, provider_config):
+    def __init__(self, name, provider_config):
         """Constructor"""
 
-        self.name = provider_config["name"]
+        self.name = name
         self.enabled = provider_config["enabled"]
         self.last_update_time = datetime.min.replace(tzinfo=pytz.UTC)
         self.last_spot_time = datetime.min.replace(tzinfo=pytz.UTC)
         self.status = "Not Started" if self.enabled else "Disabled"
-        self._spots = None
-        self._web_server = None
-
-    def setup(self, spots, web_server):
-        """Set up the provider, e.g. giving it the spot list to work from"""
-
-        self._spots = spots
-        self._web_server = web_server
+        self._spots = DATA_STORE.spots
 
     def start(self):
         """Start the provider. This should return immediately after spawning threads to access the remote resources"""
@@ -59,10 +52,7 @@ class SpotProvider:
 
     def _add_spot(self, spot):
         if not spot.expired():
-            self._spots.add(spot.id, spot, expire=MAX_SPOT_AGE)
-            # Ping the web server in case we have any SSE connections that need to see this immediately
-            if self._web_server:
-                self._web_server.notify_new_spot(spot)
+            self._spots.set(spot.id, spot)
 
     def stop(self):
         """Stop any threads and prepare for application shutdown"""

providers/spot/sse_spot_provider.py

@@ -0,0 +1,98 @@
+import logging
+from datetime import datetime
+from threading import Event, Lock, Thread
+
+import pytz
+from requests_sse import EventSource
+
+from core.constants import HTTP_HEADERS
+from providers.spot.spot_provider import SpotProvider
+
+
+class SSESpotProvider(SpotProvider):
+    """Spot provider using Server-Sent Events."""
+
+    def __init__(self, name, provider_config, url):
+        super().__init__(name, provider_config)
+        self._url = url
+        self._thread = None
+        self._last_event_id = None
+        self._stop_event = Event()
+        self._event_source_lock = Lock()
+        self._event_source = None
+
+    def start(self):
+        logging.info("Set up SSE connection to " + self.name + " spot API.")
+        self._stop_event.clear()
+        self._thread = Thread(target=self._run, name=f"SSESpotProvider-{self.name}")
+        self._thread.daemon = True
+        self._thread.start()
+
+    def stop(self):
+        self._stop_event.set()
+
+        with self._event_source_lock:
+            event_source = self._event_source
+        if event_source:
+            try:
+                event_source.close()
+            except Exception:
+                logging.exception(
+                    "Exception closing SSE connection for " + self.name + " during stop()")
+
+        if self._thread:
+            self._thread.join(timeout=15)
+            if self._thread.is_alive():
+                logging.warning(self.name + " SSE worker thread did not exit on time and will be killed.")
+
+    def _on_open(self):
+        self.status = "Waiting for Data"
+
+    def _on_error(self):
+        self.status = "Connecting"
+
+    def _set_event_source(self, event_source):
+        with self._event_source_lock:
+            self._event_source = event_source
+
+    def _run(self):
+        while not self._stop_event.is_set():
+            try:
+                logging.debug("Connecting to " + self.name + " spot API...")
+                self.status = "Connecting"
+                with EventSource(self._url, headers=HTTP_HEADERS, latest_event_id=self._last_event_id, timeout=10,
+                                 on_open=self._on_open, on_error=self._on_error) as event_source:
+                    self._set_event_source(event_source)
+                    try:
+                        for event in event_source:
+                            if self._stop_event.is_set():
+                                break
+                            if event.type == 'message':
+                                try:
+                                    self._last_event_id = event.last_event_id
+                                    new_spot = self._sse_message_to_spot(event.data)
+                                    if new_spot:
+                                        self._submit(new_spot)
+
+                                    self.status = "OK"
+                                    self.last_update_time = datetime.now(pytz.UTC)
+                                    logging.debug("Received data from " + self.name + " spot API.")
+
+                                except Exception:
+                                    logging.exception(
+                                        "Exception processing message from SSE Spot Provider (" + self.name + ")")
+                    finally:
+                        self._set_event_source(None)
+
+            except Exception:
+                self.status = "Error"
+                logging.exception("Exception in SSE Spot Provider (" + self.name + ")")
+            else:
+                self.status = "Disconnected"
+            self._stop_event.wait(timeout=5) # Wait before trying to reconnect
+
+    def _sse_message_to_spot(self, message_data):
+        """Convert an SSE message received from the API into a spot. The whole message data is provided here so the subclass
+        implementations can handle the message as JSON, XML, text, whatever the API actually provides."""
+
+        raise NotImplementedError("Subclasses must implement this method")

providers/spot/tiles.py

@@ -2,7 +2,7 @@ from datetime import datetime
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class Tiles(HTTPSpotProvider):
@@ -12,7 +12,7 @@ class Tiles(HTTPSpotProvider):
     SPOTS_URL = "https://icneuzxitdqtofutxbla.supabase.co/functions/v1/spots?active_hours=24"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("Tiles", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []
@@ -30,7 +30,8 @@ class Tiles(HTTPSpotProvider):
                         sig="Tiles",
                         # Tiles spots can include POTA & SOTA references, but ignore those on the basis that we will get them separately from the POTA/SOTA providers anyway.
                         # Just take the grid reference itself as the single Tiles SIG reference.
-                        sig_refs=[SIGRef(id=source_spot["maidenhead_grid"], sig="Tiles", name=source_spot["maidenhead_grid"])],
+                        sig_refs=[SIGRef(id=source_spot["maidenhead_grid"], sig="Tiles",
+                                         name=source_spot["maidenhead_grid"])],
                         time=datetime.fromisoformat(source_spot["created_at"].replace("Z", "+00:00")).timestamp(),
                         dx_grid=source_spot["maidenhead_grid"],
                         dx_latitude=source_spot["latitude"],
@@ -41,6 +42,7 @@ class Tiles(HTTPSpotProvider):
             new_spots.append(spot)
         return new_spots
 
+
 # Utility function to keep the first decimal point in a given string but remove any others. Used to parse Tiles'
 # strange frequency format where we can sometimes have e.g. "14.123.5".
 def strip_extra_decimal_points(s):

providers/spot/towers.py

@@ -1,20 +1,19 @@
-from datetime import datetime
-
 import json
+from datetime import datetime
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
-class WWTOTA(HTTPSpotProvider):
+class Towers(HTTPSpotProvider):
     """Spot provider for Towers on the Air"""
 
     POLL_INTERVAL_SEC = 120
     SPOTS_URL = "https://wwtota.com/api/cluster_live.php"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("Towers", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []
@@ -31,8 +30,8 @@ class WWTOTA(HTTPSpotProvider):
                         dx_call=source_spot["call"].upper(),
                         freq=likely_freq,
                         comment=source_spot["comment"],
-                        sig="WWTOTA",
-                        sig_refs=[SIGRef(id=source_spot["ref"], sig="WWTOTA")],
+                        sig="Towers",
+                        sig_refs=[SIGRef(id=source_spot["ref"], sig="Towers")],
                         time=datetime.strptime(response_json["updated"][:10] + source_spot["time"],
                                                "%Y-%m-%d%H:%M").timestamp())
 

providers/spot/ukpacketnet.py

@@ -4,7 +4,7 @@ from datetime import datetime
 import pytz
 
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class UKPacketNet(HTTPSpotProvider):
@@ -14,7 +14,7 @@ class UKPacketNet(HTTPSpotProvider):
     SPOTS_URL = "https://nodes.ukpacketradio.network/api/nodedata"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("UK Packet Net", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []

providers/spot/websocket_spot_provider.py

@@ -7,14 +7,14 @@ import pytz
 from websocket import create_connection
 
 from core.constants import HTTP_HEADERS
-from spotproviders.spot_provider import SpotProvider
+from providers.spot.spot_provider import SpotProvider
 
 
 class WebsocketSpotProvider(SpotProvider):
     """Spot provider using websockets."""
 
-    def __init__(self, provider_config, url):
-        super().__init__(provider_config)
+    def __init__(self, name, provider_config, url):
+        super().__init__(name, provider_config)
         self._url = url
         self._ws = None
         self._thread = None
@@ -24,7 +24,7 @@ class WebsocketSpotProvider(SpotProvider):
     def start(self):
         logging.info("Set up websocket connection to " + self.name + " spot API.")
         self._stopped = False
-        self._thread = Thread(target=self._run)
+        self._thread = Thread(target=self._run, name=f"WebsocketSpotProvider-{self.name}")
         self._thread.daemon = True
         self._thread.start()
 

providers/spot/wota.py

@@ -1,13 +1,15 @@
 import logging
 import re
 from datetime import datetime
+from typing import cast
 
 import pytz
 from rss_parser import Parser
+from rss_parser.models.rss import RSS
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class WOTA(HTTPSpotProvider):
@@ -19,11 +21,11 @@ class WOTA(HTTPSpotProvider):
     RSS_DATE_TIME_FORMAT = "%a, %d %b %Y %H:%M:%S %z"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("WOTA", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []
-        rss = Parser.parse(http_response.content.decode())
+        rss = cast(RSS, Parser.parse(http_response.content.decode("utf-8-sig")))
         # Iterate through source data
         for source_spot in rss.channel.items:
 
@@ -39,9 +41,9 @@ class WOTA(HTTPSpotProvider):
                 ref_name = None
                 if len(title_split) > 1:
                     ref_split = title_split[1].split(" - ")
-                    ref = ref_split[0]
+                    ref = str(ref_split[0])
                     if len(ref_split) > 1:
-                        ref_name = ref_split[1]
+                        ref_name = str(ref_split[1])
 
                 # Pick apart the description
                 desc_split = source_spot.description.split(". ")

providers/spot/wwbota.py

@@ -3,7 +3,7 @@ from datetime import datetime
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.sse_spot_provider import SSESpotProvider
+from providers.spot.sse_spot_provider import SSESpotProvider
 
 
 class WWBOTA(SSESpotProvider):
@@ -12,10 +12,10 @@ class WWBOTA(SSESpotProvider):
     SPOTS_URL = "https://api.wwbota.net/spots/"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL)
+        super().__init__("WWBOTA", provider_config, self.SPOTS_URL)
 
-    def _sse_message_to_spot(self, message):
-        source_spot = json.loads(message)
+    def _sse_message_to_spot(self, message_data):
+        source_spot = json.loads(message_data)
         # Convert to our spot format. First we unpack references, because WWBOTA spots can have more than one for
         # n-fer activations.
         refs = []

providers/spot/wwff.py

@@ -4,7 +4,7 @@ import pytz
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class WWFF(HTTPSpotProvider):
@@ -14,7 +14,7 @@ class WWFF(HTTPSpotProvider):
     SPOTS_URL = "https://spots.wwff.co/static/spots.json"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("WWFF", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []

providers/spot/xota.py

@@ -0,0 +1,42 @@
+import json
+from datetime import datetime
+
+import pytz
+
+from data.sig_ref import SIGRef
+from data.spot import Spot
+from providers.spot.websocket_spot_provider import WebsocketSpotProvider
+
+
+class XOTA(WebsocketSpotProvider):
+    """Spot provider for servers based on the "xOTA" software at https://github.com/nischu/xOTA/
+    The provider typically doesn't give us a lat/lon or SIG explicitly, so our own config provides a SIG which we can
+    then use for lookups. This functionality is implemented for Toilets on the Air events, of which there are
+    several - so a plain lookup of a "TOTA reference" doesn't make sense, it depends on which TOTA, which is why we also
+    provide a sig_ref_prefix in our config. This is applied to the reference ID, so e.g. "T-01" at C3 might become
+    "C3 T-01". This allows us to provide location lookups for TOTA at several conferences."""
+
+    LOCATION_DATA = {}
+    SIG = None
+
+    def __init__(self, provider_config):
+        name = provider_config["name"] if "name" in provider_config else "xOTA"
+        super().__init__(name, provider_config, provider_config["url"])
+        self.SIG = str(provider_config["sig"]) if "sig" in provider_config else None
+        self._sig_ref_prefix = str(provider_config["sig-ref-prefix"]) if "sig-ref-prefix" in provider_config else ""
+
+    def _ws_message_to_spot(self, b):
+        string = b.decode("utf-8")
+        source_spot = json.loads(string)
+        ref_id = self._sig_ref_prefix + " " + source_spot["reference"]["title"]
+        spot = Spot(source=self.name,
+                    source_id=source_spot["id"],
+                    dx_call=source_spot["stationCallSign"].upper(),
+                    freq=float(source_spot["freq"]) * 1000,
+                    mode=source_spot["mode"].upper(),
+                    sig=self.SIG,
+                    sig_refs=[
+                        SIGRef(id=ref_id, sig=self.SIG or "", url=source_spot["reference"]["website"])],
+                    time=datetime.now(pytz.UTC).timestamp(),
+                    qrt=source_spot["state"] != "active")
+        return spot

providers/spot/zlota.py

@@ -4,7 +4,7 @@ import pytz
 
 from data.sig_ref import SIGRef
 from data.spot import Spot
-from spotproviders.http_spot_provider import HTTPSpotProvider
+from providers.spot.http_spot_provider import HTTPSpotProvider
 
 
 class ZLOTA(HTTPSpotProvider):
@@ -15,7 +15,7 @@ class ZLOTA(HTTPSpotProvider):
     LIST_URL = "https://ontheair.nz/assets/assets.json"
 
     def __init__(self, provider_config):
-        super().__init__(provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("ZLOTA", provider_config, self.SPOTS_URL, self.POLL_INTERVAL_SEC)
 
     def _http_response_to_spots(self, http_response):
         new_spots = []

providers/staticdata/cqzonedata.py

@@ -0,0 +1,36 @@
+import json
+import logging
+
+import geopandas
+from shapely import prepare
+
+from core.data_store import DATA_STORE
+from providers.staticdata.local_file_static_data_provider import LocalFileStaticDataProvider
+
+
+class CQZoneData(LocalFileStaticDataProvider):
+    """Static data provider for CQ zone geodata."""
+
+    PATH = "datafiles/cqzones.geojson"
+
+    def __init__(self, provider_config):
+        super().__init__("CQ Zone Data", provider_config, self.PATH)
+
+    def _load_data(self, path):
+        try:
+            with open(path) as f:
+                cq_zone_data = geopandas.GeoDataFrame.from_features(json.load(f)["features"])
+            for idx in cq_zone_data.index:
+                prepare(cq_zone_data.at[idx, 'geometry'])
+
+                # Bail out if a stop has been requested, i.e. the program is shutting down - no need to prepare the rest
+                # of the data in this case
+                if self.stop:
+                    break
+
+            DATA_STORE.cq_zone_data = cq_zone_data
+            return True
+
+        except Exception as e:
+            logging.error("Exception when loading CQ zone data.", e, exc_info=True)
+            return False

providers/staticdata/file_download_static_data_provider.py

@@ -0,0 +1,77 @@
+import logging
+from datetime import datetime
+from threading import Thread, Event
+
+import pytz
+from requests import ReadTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout
+
+from core.constants import HTTP_HEADERS
+from core.url_data_cache import URLDataCache
+from providers.staticdata.static_data_provider import StaticDataProvider
+
+
+class FileDownloadStaticDataProvider(StaticDataProvider):
+    """Generic static reference data provider class for providers that fetch their data from the web by downloading a
+     file."""
+
+    def __init__(self, name, provider_config, url, poll_interval):
+        """ Set up the provider, note poll_interval is in *days*."""
+        super().__init__(name, provider_config)
+        self._url = url
+        self._poll_interval = poll_interval
+        self._thread = None
+        self._stop_event = Event()
+        self._url_data_cache = URLDataCache("staticdata_" + name)
+
+    def start(self):
+        # Fire off the polling thread. It will poll immediately on startup, then sleep for poll_interval between
+        # subsequent polls, so start() returns immediately and the application can continue starting.
+        logging.info(
+            "Set up query of " + self.name + " static reference data every " + str(self._poll_interval) + " days.")
+        self._thread = Thread(target=self._run, name=f"FileDownloadStaticDataProvider-{self.name}")
+        self._thread.start()
+
+    def stop(self):
+        self._stop_event.set()
+
+    def _run(self):
+        while True:
+            self._poll()
+            if self._stop_event.wait(timeout=self._poll_interval * 60 * 60 * 24):
+                break
+
+    def _poll(self):
+        try:
+            # Request data from API. Use the data cache (with a TTL of 1 day) here, not as the main mechanism for
+            # caching, but just so continual restarts of the software during testing don't hammer the servers.
+            logging.debug("Downloading " + self.name + " static reference data...")
+            http_response = self._url_data_cache.get(self._url, headers=HTTP_HEADERS)
+            # Check response code was good
+            if http_response.ok:
+                # Pass off to the subclass for processing
+                ok = self._handle_http_response(http_response)
+                if ok:
+                    self.status = "OK"
+                    self.last_update_time = datetime.now(pytz.UTC)
+                    logging.info("Updated static reference data for " + self.name)
+            else:
+                self.status = "Error"
+                logging.warning(f"HTTP {http_response.status_code} when downloading static reference data for {self.name}.")
+
+        except ConnectionError:
+            self.status = "Error"
+            logging.warning(f"Connection error when downloading static reference data for {self.name}.")
+        except (ConnectTimeout, ReadTimeout):
+            self.status = "Error"
+            logging.warning(f"Timeout when downloading static reference data for {self.name}.")
+        except Exception:
+            self.status = "Error"
+            logging.exception("Exception in HTTP static reference data provider (" + self.name + ")")
+            self._stop_event.wait(timeout=1)
+
+    def _handle_http_response(self, http_response):
+        """Handle an HTTP response returned by the server and load the data from it. Return true if successful,
+         false otherwise."""
+
+        raise NotImplementedError("Subclasses must implement this method")