Merge remote-tracking branch 'origin/main'

# Conflicts:
#	templates/add_spot.html
#	templates/alerts.html
#	templates/bands.html
#	templates/base.html
#	templates/conditions.html
#	templates/map.html
#	templates/spots.html
#	templates/status.html

.idea/codeStyles/codeStyleConfig.xml

@@ -0,0 +1,5 @@
+<component name="ProjectCodeStyleConfiguration">
+  <state>
+    <option name="PREFERRED_PROJECT_CODE_STYLE" value="Default" />
+  </state>
+</component>

.idea/inspectionProfiles/Project_Default.xml

@@ -3,6 +3,7 @@
     <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="CheckImageSize" enabled="false" level="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" />
@@ -43,6 +44,34 @@
     <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="PyCompatibilityInspection" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ourVersions">
+        <value>
+          <list size="6">
+            <item index="0" class="java.lang.String" itemvalue="3.15" />
+            <item index="1" class="java.lang.String" itemvalue="3.14" />
+            <item index="2" class="java.lang.String" itemvalue="3.13" />
+            <item index="3" class="java.lang.String" itemvalue="3.12" />
+            <item index="4" class="java.lang.String" itemvalue="3.11" />
+            <item index="5" class="java.lang.String" itemvalue="3.10" />
+          </list>
+        </value>
+      </option>
+    </inspection_tool>
+    <inspection_tool class="PyStringConversionWithoutDunderMethodInspection" enabled="true" level="WEAK WARNING" enabled_by_default="true">
+      <option name="ignoredTypes">
+        <list>
+          <option value="types.NoneType" />
+        </list>
+      </option>
+    </inspection_tool>
+    <inspection_tool class="PyStubPackagesAdvertiser" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ignoredPackages">
+        <list>
+          <option value="pandas" />
+        </list>
+      </option>
+    </inspection_tool>
     <inspection_tool class="SpellCheckingInspection" enabled="false" level="TYPO" enabled_by_default="false">
       <option name="processCode" value="true" />
       <option name="processLiterals" value="true" />

.idea/pyLspTools.xml

@@ -0,0 +1,16 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="PyToolsState">
+    <option name="tools">
+      <map>
+        <entry key="ruff">
+          <value>
+            <ToolEntry>
+              <option name="enabled" value="true" />
+            </ToolEntry>
+          </value>
+        </entry>
+      </map>
+    </option>
+  </component>
+</project>

.idea/runConfigurations/Run.xml

@@ -12,6 +12,8 @@
     <option name="IS_MODULE_SDK" value="true" />
     <option name="ADD_CONTENT_ROOTS" value="true" />
     <option name="ADD_SOURCE_ROOTS" value="true" />
+    <option name="DEBUG_JUST_MY_CODE" value="true" />
+    <option name="RUN_TOOL" value="" />
     <option name="SCRIPT_NAME" value="$PROJECT_DIR$/spothole.py" />
     <option name="PARAMETERS" value="" />
     <option name="SHOW_COMMAND_LINE" value="false" />

.idea/spothole.iml

@@ -1,5 +1,5 @@
 <?xml version="1.0" encoding="UTF-8"?>
-<module type="PYTHON_MODULE" version="4">
+<module external.system.id="pyproject.toml" type="PYTHON_MODULE" version="4">
   <component name="NewModuleRootManager">
     <content url="file://$MODULE_DIR$">
       <sourceFolder url="file://$MODULE_DIR$" isTestSource="false" />

Dockerfile

@@ -4,5 +4,6 @@ WORKDIR /app
 COPY . .
 RUN pip install --no-cache-dir -r requirements.txt
 EXPOSE 8080
+EXPOSE 7373
 
 CMD ["python3", "spothole.py"]

README.md

@@ -1,7 +1,7 @@
 # ![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.
+JSON API as well as a website to browse the data, and its own telnet server for integration with desktop loggers.
 
 ![Screenshot](/images/screenshot.png)
 
@@ -17,10 +17,8 @@ Spothole itself is also open source, Public Domain licenced code that anyone can
 
 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 KRMNPA, SANPCPA, WAB, WAI and
-DME.
+Repeater Network, NG3K, and any site based on the xOTA software by nischu. It also integrates with QRZ.com and HamQTH,
+retrieves solar data from various sources, provides information about upcoming contests, and more.
 
 ![Screenshot](/images/screenshot2.png)
 
@@ -34,18 +32,20 @@ see [https://spothole.app/apidocs](https://spothole.app/apidocs) for the API det
 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, to run your
-own Spothole server, or to modify it in any way. More details can be found in the following sections:
+You are more than welcome to use the data and the API that Spothole provides to power your own software, to run your own
+Spothole server, or to modify it in any way. More details can be found in the following sections:
 
 * [Embedding Spothole in another website](docs/embedding.md)
 * [Writing your own client](docs/clients.md)
 * [Running your own copy](docs/running.md)
-  * [nginx reverse proxy configuration](docs/nginx.md)
-  * [systemd service file](docs/systemd.md)
-  * [Running in Docker](docs/docker.md)
-  * [Use of multiple cluster logins](docs/multicluster.md)
+    * [nginx reverse proxy configuration](docs/nginx.md)
+    * [systemd service file](docs/systemd.md)
+    * [Running in Docker](docs/docker.md)
+    * [Use of multiple cluster logins](docs/multicluster.md)
 * [Modifying the source code](docs/modifying.md)
 
+![Screenshot](/images/screenshot4.png)
+
 ## Thanks
 
 As well as being my work, I have also gratefully received feature patches from Steven, M1SDH.
@@ -54,8 +54,8 @@ The project contains GeoJSON files for CQ and ITU zones, in the `/datafiles/` di
 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.
+`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 set of flag icons generated using the "Noto Color Emoji" font on a Debian system, in the
 `/static/img/flags/` directory.
@@ -65,12 +65,14 @@ This project would not have been possible without these libraries, so many thank
 
 ### Third Party Libraries
 
-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.
+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.
 
 A number of third-party libraries are self-hosted in the `/webassets/vendor/` directory. These files are subject to
 their own licences and are not covered by the overall licence declared in the `LICENSE` file.
 
+Sound effects are CC0-licenced and sourced from [Pixabay](https://pixabay.com).
+
 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.

config-example.yml

@@ -17,6 +17,15 @@ api_only_mode: false
 # The base URL at which the software runs.
 base_url: "http://localhost:8080"
 
+# Whether to run a telnet spot server as well as the web interface
+telnet_server_enabled: false
+
+# Telnet server address. This is not really needed by the software itself, just displayed in documentation.
+telnet_server_address: "localhost"
+
+# Port to run the telnet server on
+telnet_server_port: 7373
+
 # 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 and an enabled/disabled state. Some require more config such as hostnames/IP
@@ -67,8 +76,8 @@ spot_providers:
 
   - 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"
+    # several clusters and name them separately. The name will be used as the "source" field for the spots received.
+    name: "Cluster"
     enabled: true
     host: "hrd.wa9pie.net"
     port: 8000
@@ -84,11 +93,11 @@ spot_providers:
     allow_rbn_spots: false
 
   - class: "DXCluster"
-    name: "W3LPL Cluster"
+    name: "Cluster"
     enabled: false
     host: "w3lpl.net"
     port: 7373
-    login_prompt: "Please enter your call:"
+    login_prompt: "login:"
     login_callsign: "N0CALL-99"
     allow_rbn_spots: false
 
@@ -162,9 +171,21 @@ alert_providers:
   - class: "BOTA"
     enabled: true
 
+  - class: "Hamsat"
+    enabled: true
+
   - class: "NG3K"
     enabled: true
 
+  - class: "WA7BNM"
+    enabled: true
+
+  - class: "RSGBHFContests"
+    enabled: true
+
+  - class: "RSGBVHFContests"
+    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/v2/solar endpoint.
@@ -246,12 +267,36 @@ sig_ref_data_providers:
   - class: "DME"
     enabled: true
 
+  - class: "FEA"
+    enabled: true
+
+  - class: "DMVE"
+    enabled: true
+
+  - class: "DMUE"
+    enabled: true
+
+  - class: "DCE"
+    enabled: true
+
+  - class: "DEFE"
+    enabled: true
+
   - class: "KRMNPA"
     enabled: true
 
   - class: "SANPCPA"
     enabled: true
 
+  - class: "DTMBA"
+    enabled: true
+
+  - class: "COTA"
+    enabled: true
+
+  - class: "PGA"
+    enabled: true
+
   - class: "Toilets"
     enabled: true
 
@@ -275,27 +320,27 @@ callsign_data_providers:
     priority: 2
     # No server-side credentials for HamQTH. Users must provide their own.
 
+  - class: "CountryFiles"
+    priority: 3
+    enabled: true
+
   - 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
+    priority: 4
     # 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
+    priority: 5
     # 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
-
 
 # Maximum time to keep spots and alerts in the system before deleting them. By default, one hour for spots and one week
 # for alerts.
@@ -345,4 +390,6 @@ web_ui_options:
   # (home) page, although being arbitrary HTML you can also use a div with absolute, relative, float placement etc. This
   # is designed for a "donate/support the server" type button, though you are free to do whatever you want with it.
   # As the server owner you are responsible for the safe usage of this option!
-  support_button_html: ""
+  support_button_html: ""
+  # CartoDB API key. Required to use the CartoDB layers without watermark. Get one from https://carto.com/basemaps/apikey/.
+  cartodb_api_key: ""

core/call_lookup_helper.py

@@ -1,3 +1,5 @@
+import re
+
 from core.data_providers import DATA_PROVIDERS
 from data.callsign import Callsign
 
@@ -9,10 +11,11 @@ def get_call_info(callsign, lookup_credentials):
 
     callsign_data = Callsign(call=callsign)
 
-    if callsign:
+    # First check our input looks like a real callsign
+    if callsign and re.match(r"^[A-Za-z0-9/\-]*$", 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):
+        for p in sorted(DATA_PROVIDERS.callsign_data_providers, key=lambda p2: p2.priority):
             if p.enabled:
                 # Get new lookup data
                 data = p.lookup(callsign, lookup_credentials)
@@ -27,4 +30,4 @@ def get_call_info(callsign, lookup_credentials):
                     if callsign_data.fully_populated():
                         break
 
-    return callsign_data
+    return callsign_data

core/cleanup.py

@@ -0,0 +1,77 @@
+import logging
+from datetime import datetime
+from threading import Event, Thread
+
+import pytz
+
+from core.data_store import DATA_STORE
+
+logger = logging.getLogger(__name__)
+
+
+class CleanupTimer:
+    """Provides a timed cleanup of the spot list."""
+
+    def __init__(self):
+        """Constructor"""
+
+        self._cleanup_interval = None
+        self.last_cleanup_time = datetime.min.replace(tzinfo=pytz.UTC)
+        self.status = "Starting"
+        self._thread = None
+        self._stop_event = Event()
+
+    def setup(self, cleanup_interval):
+        self._cleanup_interval = cleanup_interval
+
+    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()
+        if self._thread:
+            self._thread.join(timeout=15)
+            if self._thread.is_alive():
+                logger.warning("Cleanup worker thread did not exit on time and will be killed.")
+
+    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:
+            for i in list(DATA_STORE.spots.keys()):
+                try:
+                    spot = DATA_STORE.spots.get(i)
+                    if spot and spot.expired():
+                        DATA_STORE.spots.delete(i)
+                except KeyError:
+                    # Must have already been deleted, OK with that
+                    pass
+            for i in list(DATA_STORE.alerts.keys()):
+                try:
+                    alert = DATA_STORE.alerts.get(i)
+                    if alert and alert.expired():
+                        DATA_STORE.alerts.delete(i)
+                except KeyError:
+                    # Must have already been deleted, OK with that
+                    pass
+
+            self.status = "OK"
+            self.last_cleanup_time = datetime.now(pytz.UTC)
+
+        except Exception:
+            self.status = "Error"
+            logger.exception("Exception in Cleanup thread")
+            self._stop_event.wait(timeout=1)
+
+
+# Global object
+CLEANUP_TIMER = CleanupTimer()

core/config.py

@@ -1,25 +1,32 @@
 import importlib
 import logging
 import os
+import sys
 
 import yaml
 
+logger = logging.getLogger(__name__)
+
 # Check you have a config file
 if not os.path.isfile("config.yml"):
-    logging.error(
-        "Your config file is missing. Ensure you have copied config-example.yml to config.yml and updated it according to your needs.")
-    exit()
+    logger.error(
+        "Your config file is missing. Ensure you have copied config-example.yml to config.yml and updated it according to your needs."
+    )
+    sys.exit()
 
 # Load config
 with open("config.yml") as f:
     config = yaml.safe_load(f)
-logging.info("Loaded config.")
+logger.info("Loaded config.")
 
 BASE_URL = config.get("base_url", "http://localhost:8080")
 MAX_SPOT_AGE = config.get("max_spot_age_sec", 3600)
 MAX_ALERT_AGE = config.get("max_alert_age_sec", 604800)
 SERVER_OWNER_CALLSIGN = config.get("server_owner_callsign", "N0CALL")
 WEB_SERVER_PORT = config.get("web_server_port", 8080)
+TELNET_SERVER_ENABLED = config.get("telnet_server_enabled", False)
+TELNET_SERVER_ADDRESS = config.get("telnet_server_address", "localhost")
+TELNET_SERVER_PORT = config.get("telnet_server_port", 7373)
 ALLOW_SPOTTING = config.get("allow_spotting", True)
 ALLOW_UPSTREAM_SPOTTING = config.get("allow_upstream_spotting", True)
 WEB_UI_OPTIONS = config.get("web_ui_options", {})
@@ -30,7 +37,9 @@ LOG_LEVEL = config.get("log_level", "INFO")
 LOG_WEB_REQUESTS = config.get("log_web_requests", False)
 
 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"])
+WEB_UI_OPTIONS["hamqth_enabled"] = any(
+    p["class"] == "HamQTH" and p["enabled"] for p in config["callsign_data_providers"]
+)
 WEB_UI_OPTIONS["recaptcha_site_key"] = RECAPTCHA_SITE_KEY
 WEB_UI_OPTIONS["allow_upstream_spotting"] = ALLOW_SPOTTING and ALLOW_UPSTREAM_SPOTTING
 
@@ -40,6 +49,6 @@ def create_provider_from_config(package, config_providers_entry):
     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())
+    module = importlib.import_module(f"{package}.{config_providers_entry['class'].lower()}")
     provider_class = getattr(module, config_providers_entry["class"])
     return provider_class(config_providers_entry)

core/constants.py

@@ -1,66 +1,403 @@
 from core.config import SERVER_OWNER_CALLSIGN
+from core.enums import SIGRefType, SIGType
 from data.band import Band
 from data.sig import SIG
 
 # General software
-SOFTWARE_VERSION = "2.0-pre"
+SOFTWARE_VERSION = "2.1.2"
 
 # HTTP headers used for spot providers that use HTTP
-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(" ", "_")
+HTTP_HEADERS = {"User-Agent": f"Spothole v{SOFTWARE_VERSION} (operated by {SERVER_OWNER_CALLSIGN})"}
+HAMQTH_PRG = f"Spothole v{SOFTWARE_VERSION} operated by {SERVER_OWNER_CALLSIGN}".replace(" ", "_")
 
 # Special Interest Groups
 SIGS = [
-    SIG(name="POTA", comment_names=["POTA"], description="Parks on the Air", ref_regex=r"[A-Z]{2}\-\d{4,5}|K\-TEST"),
-    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}")
+    SIG(
+        name="AMSAT",
+        comment_names=[],
+        description="Amateur Radio Satellites",
+        sig_type=SIGType.WORLDWIDE,
+        icon="fa-satellite",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="EME",
+        comment_names=[],
+        description="Moonbounce",
+        sig_type=SIGType.WORLDWIDE,
+        icon="fa-moon",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="POTA",
+        comment_names=["POTA"],
+        description="Parks on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.PARK,
+        ref_regex=r"[A-Z]{2}\-\d{4,5}|K\-TEST",
+        icon="fa-tree",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="SOTA",
+        comment_names=["SOTA"],
+        description="Summits on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.SUMMIT,
+        ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{2}\-\d{3}",
+        icon="fa-mountain-sun",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="WWFF",
+        comment_names=["WWFF"],
+        description="World Wide Flora & Fauna",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.PARK,
+        ref_regex=r"[A-Z0-9]{1,3}FF\-\d{4}",
+        icon="fa-seedling",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="GMA",
+        comment_names=["GMA"],
+        description="Global Mountain Activity",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.SUMMIT,
+        ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{2}\-\d{3}",
+        icon="fa-person-hiking",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="WWBOTA",
+        comment_names=["WWBOTA", "BOTA"],
+        description="Worldwide Bunkers on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.BUNKER,
+        ref_regex=r"B\/[A-Z0-9]{1,3}\-\d{3,4}",
+        icon="fa-radiation",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="HEMA",
+        comment_names=["HEMA"],
+        description="HuMPs Excluding Marilyns Award",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.SUMMIT,
+        ref_regex=r"[A-Z0-9]{1,3}\/[A-Z]{3}\-\d{3}",
+        icon="fa-mound",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="IOTA",
+        comment_names=["IOTA"],
+        description="Islands on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.ISLAND,
+        ref_regex=r"[A-Z]{2}\-\d{3}",
+        icon="fa-book-atlas",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="GMA Islands",
+        comment_names=[],
+        description="Global Mountain Activity - Islands",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.ISLAND,
+        ref_regex=r"(([A-Z]{2}\-\d{3})|([A-Z0-9]{1,3}\/[A-Z]{2}\-\d{3}))",
+        icon="fa-person-hiking",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="ARLHS",
+        comment_names=["ARLHS"],
+        description="Amateur Radio Lighthouse Society",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.LIGHTHOUSE,
+        ref_regex=r"[A-Z]{3}[\- ]\d{3,4}",
+        icon="fa-house-flood-water",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="ILLW",
+        comment_names=["ILLW"],
+        description="International Lighthouse & Lightship Weekend",
+        sig_type=SIGType.EVENT,
+        ref_type=SIGRefType.LIGHTHOUSE,
+        ref_regex=r"[A-Z]{2}\d{4}",
+        icon="fa-house-flood-water",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="MOTA",
+        comment_names=["MOTA"],
+        description="Mills on the Air",
+        sig_type=SIGType.EVENT,
+        ref_type=SIGRefType.MILL,
+        ref_regex=r"X\d{4,6}",
+        icon="fa-fan",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="SIOTA",
+        comment_names=["SIOTA"],
+        description="Silos on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.SILO,
+        ref_regex=r"[A-Z]{2}\-[A-Z]{3}\d",
+        icon="fa-wheat-awn",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="WCA",
+        comment_names=["WCA"],
+        description="World Castles Award",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.CASTLE,
+        ref_regex=r"[A-Z0-9]{1,3}\-\d{5}",
+        icon="fa-chess-rook",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="ZLOTA",
+        comment_names=["ZLOTA"],
+        description="New Zealand on the Air",
+        sig_type=SIGType.REGIONAL,
+        ref_type=None,
+        ref_regex=r"ZL[A-Z]/[A-Z]{2}\-\d{3,4}",
+        icon="fa-kiwi-bird",
+        region_flag="🇳🇿",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="WOTA",
+        comment_names=["WOTA"],
+        description="Wainwrights on the Air",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.SUMMIT,
+        ref_regex=r"[A-Z]{3}-[0-9]{2}",
+        icon="fa-w",
+        region_flag="🇬🇧",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="BOTA",
+        comment_names=[],
+        description="Beaches on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.BEACH,
+        icon="fa-umbrella-beach",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="KRMNPA",
+        comment_names=["KRMNPA"],
+        description="Keith Roget Memorial National Parks Award",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.PARK,
+        ref_regex=r"VKFF\-\d{4}",
+        icon="fa-earth-oceania",
+        region_flag="🇦🇺",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="SANPCPA",
+        comment_names=["SANPCPA"],
+        description="South Australian National Parks and Conservation Parks Award",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.PARK,
+        ref_regex=r"VKFF\-\d{4}",
+        icon="fa-earth-oceania",
+        region_flag="🇦🇺",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="LLOTA",
+        comment_names=["LLOTA"],
+        description="Lagos y Lagunas on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.LAKE,
+        ref_regex=r"LL[A-Z]{2}\-\d{4}",
+        icon="fa-water",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="Towers",
+        comment_names=["TOTA"],
+        description="Towers on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.TOWER,
+        ref_regex=r"[A-Z]{2,3}R\-\d{4}",
+        icon="fa-tower-observation",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="Tiles",
+        comment_names=[],
+        description="Tiles on the Air",
+        sig_type=SIGType.WORLDWIDE,
+        ref_type=SIGRefType.GRID,
+        ref_regex=r"[A-Za-z]{2}[0-9]{2}[A-Za-z]{2}",
+        icon="fa-square",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="WAB",
+        comment_names=["WAB"],
+        description="Worked All Britain",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.GRID,
+        ref_regex=r"[A-Z]{1,2}[0-9]{2}",
+        icon="fa-table-cells-large",
+        region_flag="🇬🇧",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="WAI",
+        comment_names=["WAI"],
+        description="Worked All Ireland",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.GRID,
+        ref_regex=r"[A-Z][0-9]{2}",
+        icon="fa-table-cells-large",
+        region_flag="🇮🇪",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="DMF",
+        comment_names=["DMF"],
+        description="Diplôme des Moulins de France",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.MILL,
+        icon="fa-fan",
+        region_flag="🇫🇷",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="DME",
+        comment_names=["DME"],
+        description="Diploma Municipios de España",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.TOWN,
+        ref_regex=r"DME[\- ]\d{3,5}",
+        icon="fa-building",
+        region_flag="🇪🇸",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="FEA",
+        comment_names=["FEA"],
+        description="Diploma Faros de España",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.LIGHTHOUSE,
+        # FEA references are technically [DE]\-\d{4}(\.\d)? but spotters always seem to miss out the D- or E-
+        # prefix and just use FEA-1234 or FEA 1234, so allow for that. The FEA sigref data provider adds both
+        # forms to the database.
+        ref_regex=r"([DE]|FEA)[\- ]\d{4}(\.\d)?",
+        icon="fa-house-flood-water",
+        region_flag="🇪🇸",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="DMUE",
+        comment_names=["DMUE"],
+        description="Diploma Museos de España",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.BUILDING,
+        ref_regex=r"MUE[A-Z]{2}-\d{3}",
+        icon="fa-landmark",
+        region_flag="🇪🇸",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="DMVE",
+        comment_names=["DMVE"],
+        description="Diploma Monumentos y Vestigios de España",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.BUILDING,
+        ref_regex=r"MV[A-Z]{1,2}-\d{4}",
+        icon="fa-monument",
+        region_flag="🇪🇸",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="DCE",
+        comment_names=["DCE"],
+        description="Diploma Castillos de España",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.CASTLE,
+        ref_regex=r"C[A-Z]{1,2}-\d{3}",
+        icon="fa-chess-rook",
+        region_flag="🇪🇸",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="DEFE",
+        comment_names=["DEFE"],
+        description="Diploma Estaciones de Ferrocarril de España",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.BUILDING,
+        ref_regex=r"EF[A-Z]{1,2}-\d{3}",
+        icon="fa-train",
+        region_flag="🇪🇸",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="DTMBA",
+        comment_names=["DTMBA"],
+        description="Diploma Teatri Musei e Belle Arti",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.BUILDING,
+        ref_regex=r"I-?[0-9]{3,4}\s?[A-Z]{2}",
+        icon="fa-landmark",
+        region_flag="🇮🇹",
+        refs_globally_unique=True,
+    ),
+    SIG(
+        name="BIWOTA",
+        comment_names=["BIWOTA"],
+        description="British Inland Waterways on the Air",
+        sig_type=SIGType.EVENT,
+        ref_type=SIGRefType.WATERWAY,
+        icon="fa-ship",
+        region_flag="🇬🇧",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="COTA",
+        comment_names=["COTA"],
+        description="Castles on the Air",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.CASTLE,
+        ref_regex=r"[A-Z]{3}\-[0-9]{3,5}",
+        icon="fa-chess-rook",
+        region_flag="🇩🇪",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="PGA",
+        comment_names=["PGA"],
+        description="Polish Gmina Award",
+        sig_type=SIGType.REGIONAL,
+        ref_type=SIGRefType.REGION,
+        ref_regex=r"[A-Z]{2}[0-9]{2}",
+        icon="fa-g",
+        region_flag="🇵🇱",
+        refs_globally_unique=False,
+    ),
+    SIG(
+        name="Toilets",
+        comment_names=[],
+        description="Toilets on the Air",
+        sig_type=SIGType.EVENT,
+        ref_type=SIGRefType.TOILET,
+        ref_regex=r"T\-[0-9]{2}",
+        icon="fa-toilet",
+        region_flag="🏴‍☠️",
+        refs_globally_unique=True,
+    ),
 ]
 
-# Modes. Note "DIGI" and "DIGITAL" are also supported but are normalised into "DATA".
-CW_MODES = ["CW"]
-PHONE_MODES = ["PHONE", "SSB", "USB", "LSB", "AM", "FM", "DV", "DMR", "DSTAR", "C4FM", "FUSION", "M17"]
-DATA_MODES = ["DATA", "FT8", "FT4", "RTTY", "SSTV", "JS8", "HELL", "PSK", "OLIVIA", "PKT", "MSK144"]
-ALL_MODES = CW_MODES + PHONE_MODES + DATA_MODES
-MODE_TYPES = ["CW", "PHONE", "DATA"]
-SSB_SUB_MODES = ["USB", "LSB"]
-DV_SUB_MODES = ["DMR", "DSTAR", "C4FM", "FUSION", "M17"]
-
-# Mode aliases. Sometimes we get spots with a mode described in a different way that is effectively the same as a mode
-# we already know, or we want to normalise things for consistency. The lookup table for this is here. Incoming spots
-# that match a key in this table will be converted to the corresponding value, so only the modes above will actually be
-# present in the spots.
-MODE_ALIASES = {
-    "RTT": "RTTY",
-    "BPSK": "PSK",
-    "PSK31": "PSK",
-    "BPSK31": "PSK",
-    "MFSK": "FSK",
-    "MFSK32": "FSK",
-    "DIGI": "DATA",
-    "DIGITAL": "DATA"
-}
-
 # Band definitions
 BANDS = [
     Band(name="2200m", start_freq=135700, end_freq=137800),
@@ -76,7 +413,7 @@ BANDS = [
     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, is_ham_hf=True),
-    Band(name="6m", start_freq=50000000, end_freq=54000000),
+    Band(name="6m", start_freq=50000000, end_freq=54000000, is_ham_hf=True),
     Band(name="5m", start_freq=56000000, end_freq=60500000),
     Band(name="4m", start_freq=70000000, end_freq=70500000),
     Band(name="2m", start_freq=144000000, end_freq=148000000),
@@ -88,12 +425,10 @@ BANDS = [
     Band(name="10GHz", start_freq=10000000000, end_freq=10500000000),
     Band(name="24GHz", start_freq=24000000000, end_freq=24050000000),
     Band(name="47GHz", start_freq=47000000000, end_freq=47200000000),
-    Band(name="76GHz", start_freq=75500000000, end_freq=81500000000)]
+    Band(name="76GHz", start_freq=75500000000, end_freq=81500000000),
+]
 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 = {
@@ -106,5 +441,6 @@ PROPAGATION_MODES = {
     "MS": "Meteor scatter",
     "RS": "Rain scatter",
     "AS": "Aircraft scatter",
-    "SAT": "Satellite"
+    "ACS": "Aircraft scatter",
+    "SAT": "Satellite",
 }

core/data_providers.py

@@ -1,8 +1,11 @@
 import logging
 import threading
+import time
 
 from core.config import config, create_provider_from_config
 
+logger = logging.getLogger(__name__)
+
 
 class DataProviders:
     """Global object for storing data providers."""
@@ -14,7 +17,7 @@ class DataProviders:
         self.static_data_providers = []
         self.sig_ref_data_providers = []
         self.callsign_data_providers = []
-
+        self._startup_timers = []
 
     def setup(self):
         for entry in config["spot_providers"]:
@@ -31,10 +34,10 @@ class DataProviders:
             self.callsign_data_providers.append(create_provider_from_config("providers.callsigndata", entry))
 
     @staticmethod
-    def start_providers(providers, type):
+    def start_providers(providers, provider_type):
         """Helper method to activate enabled providers in the list."""
 
-        logging.info(f"Starting %s providers...", type)
+        logger.info(f"Starting {provider_type} providers...")
         for p in providers:
             if p.enabled:
                 p.start()
@@ -42,32 +45,59 @@ class DataProviders:
     def start(self):
         # Start data providers before spot/alert providers so the lookup data is there already for incoming spots.
         # Each category is fired off after a small delay to give the rest of Spothole chance to start up.
-        threading.Timer(5.0, lambda: self.start_providers(self.static_data_providers, "static data")).start()
-        threading.Timer(10.0, lambda: self.start_providers(self.callsign_data_providers, "callsign data")).start()
-        threading.Timer(15.0, lambda: self.start_providers(self.spot_providers, "spot")).start()
-        threading.Timer(20.0, lambda: self.start_providers(self.alert_providers, "alert")).start()
-        threading.Timer(25.0, lambda: self.start_providers(self.solar_condition_providers, "solar condition")).start()
-        threading.Timer(30.0, lambda: self.start_providers(self.sig_ref_data_providers, "SIG ref data")).start()
+        self._startup_timers = [
+            threading.Timer(5.0, lambda: self.start_providers(self.static_data_providers, "static data")),
+            threading.Timer(10.0, lambda: self.start_providers(self.callsign_data_providers, "callsign data")),
+            threading.Timer(15.0, lambda: self.start_providers(self.spot_providers, "spot")),
+            threading.Timer(20.0, lambda: self.start_providers(self.alert_providers, "alert")),
+            threading.Timer(
+                25.0,
+                lambda: self.start_providers(self.solar_condition_providers, "solar condition"),
+            ),
+            threading.Timer(30.0, lambda: self.start_providers(self.sig_ref_data_providers, "SIG ref data")),
+        ]
+        for t in self._startup_timers:
+            t.daemon = True
+            t.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()
+        # Cancel any startup timers that haven't fired yet
+        for t in self._startup_timers:
+            t.cancel()
+
+        # Stop all providers
+        all_providers = [
+            p
+            for p in (
+                self.spot_providers
+                + self.alert_providers
+                + self.solar_condition_providers
+                + self.sig_ref_data_providers
+                + self.static_data_providers
+                + self.callsign_data_providers
+            )
+            if p.enabled
+        ]
+        if not all_providers:
+            return
+
+        def stop_provider(p):
+            try:
+                p.stop()
+            except Exception:
+                logger.exception("Exception stopping provider")
+
+        threads = [threading.Thread(target=stop_provider, args=(p,), daemon=True) for p in all_providers]
+        for t in threads:
+            t.start()
+
+        deadline = time.monotonic() + 40
+        for t in threads:
+            t.join(timeout=max(0.0, deadline - time.monotonic()))
+        still_running = [t for t in threads if t.is_alive()]
+        if still_running:
+            logger.warning("Some threads did not stop in time!")
+
 
 # Global object
-DATA_PROVIDERS = DataProviders()
+DATA_PROVIDERS = DataProviders()

core/data_store.py

@@ -4,10 +4,13 @@ from pathlib import Path
 
 import diskcache
 
-from core.config import MAX_SPOT_AGE, MAX_ALERT_AGE
+from core.config import MAX_ALERT_AGE, MAX_SPOT_AGE
 from core.live_data_cache import LiveDataCache
+from core.single_object_data_cache import SingleObjectDataCache
 from data.solar_conditions import SolarConditions
 
+logger = logging.getLogger(__name__)
+
 CACHE_DIR = "./cache/"
 
 
@@ -32,10 +35,8 @@ class DataStore:
         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
+        self.status = None
+        self.solar_conditions = 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
@@ -44,54 +45,60 @@ class DataStore:
     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")
+        # For solar data and status data, we use a wrapper around disk cache where each cache contains only a single
+        # object exposed to the wider application, and provides a store() method for callers to notify diskcache that
+        # the object has changed and needs to be re-cached.
+        self.solar_conditions = SingleObjectDataCache(f"{CACHE_DIR}solar", SolarConditions())
+        self.status = SingleObjectDataCache(f"{CACHE_DIR}status", {})
 
         # 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.dxcc_data = diskcache.Cache(f"{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))
+        self.sigrefs = diskcache.Cache(f"{CACHE_DIR}sigrefs")
+        logger.info(f"Loaded data for {len(self.sigrefs)} SIG references.")
 
         # 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")
+        self.callsign_data_countryfiles = diskcache.Cache(f"{CACHE_DIR}callsign_data_countryfiles")
+        self.callsign_data_clublogxml = diskcache.Cache(f"{CACHE_DIR}callsign_data_clublogxml")
+        self.callsign_data_clublogapi = diskcache.Cache(f"{CACHE_DIR}callsign_data_clublogapi")
+        self.callsign_data_qrz = diskcache.Cache(f"{CACHE_DIR}callsign_data_qrz")
+        self.callsign_data_hamqth = diskcache.Cache(f"{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]:
+        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))
+        logger.info(f"Loaded data for {len(unique_keys)} callsigns.")
 
         # 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.spots = LiveDataCache(
+            maxsize=self._MAX_SPOT_COUNT,
+            ttl=MAX_SPOT_AGE,
+            snapshot_dir=f"{CACHE_DIR}spots",
+            snapshot_interval_sec=self._SPOT_ALERT_SNAPSHOT_INTERVAL_SEC,
+        )
+        logger.info(f"Loaded {len(self.spots.keys())} spots from a previous run.")
 
-        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()))
+        self.alerts = LiveDataCache(
+            maxsize=self._MAX_ALERT_COUNT,
+            ttl=MAX_ALERT_AGE,
+            snapshot_dir=f"{CACHE_DIR}alerts",
+            snapshot_interval_sec=self._SPOT_ALERT_SNAPSHOT_INTERVAL_SEC,
+        )
+        logger.info(f"Loaded {len(self.alerts.keys())} alerts from a previous run.")
 
     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
@@ -99,14 +106,15 @@ class DataStore:
         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."""
 
+        self.dxcc_lookup_by_call_regex = []
         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.solar_conditions.close()
+        self.status.close()
         self.dxcc_data.close()
         self.sigrefs.close()
         self.callsign_data_countryfiles.close()
@@ -115,5 +123,6 @@ class DataStore:
         self.callsign_data_qrz.close()
         self.callsign_data_hamqth.close()
 
+
 # Global object
-DATA_STORE = DataStore()
+DATA_STORE = DataStore()

core/enums.py

@@ -0,0 +1,154 @@
+from enum import Enum
+
+
+class Continent(str, Enum):
+    EU = "EU"
+    NA = "NA"
+    SA = "SA"
+    AS = "AS"
+    AF = "AF"
+    OC = "OC"
+    AN = "AN"
+
+
+class Mode(str, Enum):
+    CW = "CW"
+    PHONE = "PHONE"
+    SSB = "SSB"
+    AM = "AM"
+    FM = "FM"
+    DV = "DV"
+    DATA = "DATA"
+    FT8 = "FT8"
+    FT4 = "FT4"
+    RTTY = "RTTY"
+    SSTV = "SSTV"
+    JS8 = "JS8"
+    HELL = "HELL"
+    PSK = "PSK"
+    FSK = "FSK"
+    OLIVIA = "OLIVIA"
+    PKT = "PKT"
+    MSK144 = "MSK144"
+
+    @property
+    def is_cw(self) -> bool:
+        return self == Mode.CW
+
+    @property
+    def is_phone(self) -> bool:
+        return self in {Mode.PHONE, Mode.SSB, Mode.AM, Mode.FM, Mode.DV}
+
+    @property
+    def is_data(self) -> bool:
+        return not (self.is_cw or self.is_phone)
+
+    @staticmethod
+    def from_name(name):
+        """Convert a string to an enum mode using the alias table."""
+
+        if not name:
+            return None
+
+        try:
+            return Mode(name.upper())
+        except (KeyError, ValueError):
+            try:
+                return Mode(MODE_ALIASES[name.upper()])
+            except (KeyError, ValueError):
+                return None
+
+
+class ModeType(str, Enum):
+    PHONE = "PHONE"
+    CW = "CW"
+    DATA = "DATA"
+
+
+class ModeSource(str, Enum):
+    """Where the mode data came from in a spot."""
+
+    SPOT = "SPOT"
+    COMMENT = "COMMENT"
+    BANDPLAN = "BANDPLAN"
+
+
+class LocationSourceForSpot(str, Enum):
+    """Where the location data came from in a spot."""
+
+    SPOT = "SPOT"
+    SIG_REF_LOOKUP = "SIG REF LOOKUP"
+    GRID = "GRID"
+    HOME_QTH = "HOME QTH"
+    DXCC = "DXCC"
+
+
+class LocationSourceForCallsign(str, Enum):
+    """Where the location data came from in a callsign data object."""
+
+    HOME_QTH = "HOME QTH"
+    DXCC = "DXCC"
+
+
+class SIGRefType(str, Enum):
+    """Type of a Special Interest Group reference."""
+
+    PARK = "PARK"
+    SUMMIT = "SUMMIT"
+    CASTLE = "CASTLE"
+    LIGHTHOUSE = "LIGHTHOUSE"
+    ISLAND = "ISLAND"
+    BUNKER = "BUNKER"
+    MILL = "MILL"
+    SILO = "SILO"
+    BEACH = "BEACH"
+    LAKE = "LAKE"
+    TOWER = "TOWER"
+    WATERWAY = "WATERWAY"
+    TOWN = "TOWN"
+    BUILDING = "BUILDING"
+    REGION = "REGION"
+    GRID = "GRID"
+    TOILET = "TOILET"
+
+
+class AlertType(str, Enum):
+    """Type of an alert."""
+
+    XOTA = "XOTA"
+    SATELLITE = "SATELLITE"
+    DXPEDITION = "DXPEDITION"
+    CONTEST = "CONTEST"
+
+
+class SIGType(str, Enum):
+    """Type of a Special Interest Group. Used to group them in the web UI."""
+
+    WORLDWIDE = "WORLDWIDE"
+    REGIONAL = "REGIONAL"
+    EVENT = "EVENT"
+
+
+# Mode aliases. Sometimes we get spots with a mode described in a different way that is effectively the same as a mode
+# we already know, or we want to normalise things for consistency. The lookup table for this is here. Incoming spots
+# that match a key in this table will be converted to the corresponding value, so only the modes above will actually be
+# present in the spots.
+MODE_ALIASES = {
+    "USB": "SSB",
+    "LSB": "SSB",
+    "DIGITALVOICE": "DV",
+    "DIGITALVOI": "DV",
+    "DMR": "DV",
+    "DSTAR": "DV",
+    "C4FM": "DV",
+    "FUSION": "DV",
+    "M17": "DV",
+    "RTT": "RTTY",
+    "BPSK": "PSK",
+    "PSK31": "PSK",
+    "BPSK31": "PSK",
+    "MFSK": "FSK",
+    "MFSK32": "FSK",
+    "DIGI": "DATA",
+    "DIGITAL": "DATA",
+}

core/geo_utils.py

@@ -7,6 +7,8 @@ from shapely.geometry import Point, Polygon
 
 from core.data_store import DATA_STORE
 
+logger = logging.getLogger(__name__)
+
 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")
@@ -71,7 +73,7 @@ def lat_lon_for_grid_sw_corner(grid):
     """Convert a Maidenhead grid reference of arbitrary precision to the lat/long of the southwest corner of the square.
     Returns None if the grid format is invalid."""
 
-    lat, lon, lat_cell_size, lon_cell_size = lat_lon_for_grid_sw_corner_plus_size(grid)
+    lat, lon, _lat_cell_size, _lon_cell_size = lat_lon_for_grid_sw_corner_plus_size(grid)
     if lat is not None and lon is not None:
         return [lat, lon]
     else:
@@ -113,8 +115,8 @@ def lat_lon_for_grid_sw_corner_plus_size(grid):
     while block * 2 < length:
         if block % 2 == 0:
             # Letters in this block
-            lon_cell_no = ord(grid[block * 2]) - ord('A')
-            lat_cell_no = ord(grid[block * 2 + 1]) - ord('A')
+            lon_cell_no = ord(grid[block * 2]) - ord("A")
+            lat_cell_no = ord(grid[block * 2 + 1]) - ord("A")
             # Bail if the values aren't in range. Allowed values are A-R (0-17) for the first letter block, or
             # A-X (0-23) thereafter.
             max_cell_no = 17 if block == 0 else 23
@@ -154,7 +156,7 @@ def lat_lon_for_grid_sw_corner_plus_size(grid):
     lat -= 90.0
 
     # Return None values on maths errors
-    if any(x != x for x in [lat, lon, lat_cell_size, lon_cell_size]):  # NaN check
+    if any(x != x for x in [lat, lon, lat_cell_size, lon_cell_size]):
         return None, None, None, None
 
     return lat, lon, lat_cell_size, lon_cell_size
@@ -172,7 +174,7 @@ def wab_wai_square_to_lat_lon(ref):
     elif re.match(r"^W[AV][0-9]{2}$", ref):
         return utm_grid_square_to_lat_lon(ref)
     else:
-        logging.warning("Invalid WAB/WAI square: " + ref)
+        logger.warning(f"Invalid WAB/WAI square: {ref}")
         return None
 
 

core/live_data_cache.py

@@ -5,6 +5,8 @@ import time
 import diskcache
 from cachetools import TTLCache
 
+logger = logging.getLogger(__name__)
+
 
 class LiveDataCache:
     """Cache for spots and alerts. Uses the fast in-memory TTLCache for normal data I/O, including the TTL to enforce
@@ -20,6 +22,8 @@ class LiveDataCache:
         self._listeners_lock = threading.Lock()
         self._snapshot_dir = snapshot_dir
         self._disk_cache = diskcache.Cache(str(snapshot_dir))
+        self._stop_event = threading.Event()
+        self._snapshot_thread = None
         self._load_snapshot()
         self._start_periodic_snapshot(snapshot_interval_sec)
 
@@ -34,8 +38,7 @@ class LiveDataCache:
             try:
                 callback(value)
             except Exception:
-                logging.error("Listener raised an exception for key %s", key, exc_info=True)
-
+                logger.exception(f"Listener raised an exception for key {key}")
 
     def get(self, key, default=None):
         with self._lock:
@@ -70,11 +73,16 @@ class LiveDataCache:
             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)
+        except Exception:
+            logger.exception(f"Failed to write snapshot to {self._snapshot_dir}")
 
     def _load_snapshot(self):
-        data = self._disk_cache.get("snapshot")
+        try:
+            data = self._disk_cache.get("snapshot")
+        except Exception:
+            logger.warning(f"Failed to load snapshot from {self._snapshot_dir}, clearing it.")
+            self._disk_cache.clear()
+            return
         if not data:
             return
 
@@ -84,17 +92,23 @@ class LiveDataCache:
                 # 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)
+        logger.info(f"Loaded snapshot from {self._snapshot_dir}")
 
     def _start_periodic_snapshot(self, interval):
         def loop():
-            while True:
-                time.sleep(interval)
+            while not self._stop_event.wait(timeout=interval):
                 self.save_snapshot()
 
-        t = threading.Thread(target=loop, name=f"LiveDataCache-Snapshot-{self._snapshot_dir}")
-        t.start()
+        self._snapshot_thread = threading.Thread(
+            target=loop, name=f"LiveDataCache-Snapshot-{self._snapshot_dir}", daemon=True
+        )
+        self._snapshot_thread.start()
 
     def close(self):
+        self._stop_event.set()
+        if self._snapshot_thread:
+            self._snapshot_thread.join(timeout=15)
+            if self._snapshot_thread.is_alive():
+                logger.warning(f"LiveDataCache snapshot thread for {self._snapshot_dir} did not exit on time.")
         self.save_snapshot()
         self._disk_cache.close()

core/prometheus_metrics_handler.py

@@ -1,4 +1,10 @@
-from prometheus_client import CollectorRegistry, generate_latest, Counter, disable_created_metrics, Gauge
+from prometheus_client import (
+    CollectorRegistry,
+    Counter,
+    Gauge,
+    disable_created_metrics,
+    generate_latest,
+)
 
 disable_created_metrics()
 # Prometheus metrics registry
@@ -9,25 +15,13 @@ page_requests_counter = Counter(
     "Total number of page requests received",
     registry=registry,
 )
-api_requests_counter = Counter(
-    "spothole_api_requests",
-    "Total number of API requests received",
-    registry=registry
-)
-spots_gauge = Gauge(
-    "spothole_spots",
-    "Number of spots currently in the software",
-    registry=registry
-)
-alerts_gauge = Gauge(
-    "spothole_alerts",
-    "Number of alerts currently in the software",
-    registry=registry
-)
+api_requests_counter = Counter("spothole_api_requests", "Total number of API requests received", registry=registry)
+spots_gauge = Gauge("spothole_spots", "Number of spots currently in the software", registry=registry)
+alerts_gauge = Gauge("spothole_alerts", "Number of alerts currently in the software", registry=registry)
 memory_use_gauge = Gauge(
     "spothole_memory_usage_bytes",
     "Current memory usage of the software in bytes",
-    registry=registry
+    registry=registry,
 )
 
 

core/sig_lookup_helper.py

@@ -1,53 +1,81 @@
 import logging
+import re
 
-from pyhamtools.locator import locator_to_latlong, latlong_to_locator
+from pyhamtools.locator import latlong_to_locator, locator_to_latlong
 
+from core.constants import SIGS
 from core.data_store import DATA_STORE
+from core.enums import SIGRefType
 from core.geo_utils import wab_wai_square_to_lat_lon
 from data.sig_ref import SIGRef
 
+logger = logging.getLogger(__name__)
 
-def get_sig_ref_info(sig, ref_id):
+
+def get_sig_ref_info(sig_name, 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.")
+    if sig_name is None or sig_name == "" or ref_id is None or ref_id == "":
+        logger.debug("Failed to look up sig_ref info, sig or ref were not set.")
         return None
 
-    sig_ref = SIGRef(sig=sig, id=ref_id)
+    # Sometimes we allow spaces instead of dashes in references due to common usage that way, but official reference
+    # lists never do, so convert them here.
+    ref_id = ref_id.replace(" ", "-")
+
+    # Prepare the object to be returned
+    sig_ref = SIGRef(sig=sig_name, id=ref_id)
+
+    # We can always get the reference type and the icon from the SIG itself
+    for sig in SIGS:
+        if sig.name.upper() == sig_name.upper():
+            sig_ref.ref_type = sig.ref_type
+            sig_ref.icon = sig.icon
 
     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)
+        # activators add leading zeros. We also need to normalise "DME 01234" to "DME-01234" to match what's in our
+        # database.
+        if sig_name.upper() == "DME":
+            match = re.match(r"DME[\- ](\d{3,5})", ref_id, re.IGNORECASE)
+            if match:
+                number = match.group(1)
+                ref_id = f"DME-{number.zfill(5)}"
 
-        # If the SIG is HEMA, we have no current lookup for this so just skip the lookup here.
-        if sig.upper() == "HEMA":
+        # DTMBA spotters sometimes include spaces and dashes, our regex allows them but they must be removed here so we
+        # can look up against the official list which doesn't have them
+        if sig_name.upper() == "DTMBA":
+            ref_id = ref_id.replace("-", "").replace(" ", "")
+
+        ### NO DATA SIGS ###
+        #
+        # If the SIG is HEMA or BIWOTA, we have no way to either generate useful data or look it up on a reference list,
+        # so just skip the lookup here.
+        if sig_name.upper() == "HEMA" or sig_name.upper() == "BIWOTA":
             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":
+        if sig_name.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:
+            if sig_ref.grid and (not sig_ref.latitude or not sig_ref.longitude):
                 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":
+        elif sig_name.upper() == "WAB" or sig_name.upper() == "WAI":
             ll = wab_wai_square_to_lat_lon(ref_id)
             if ll:
                 sig_ref.name = ref_id
@@ -55,33 +83,57 @@ def get_sig_ref_info(sig, ref_id):
                     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")
+                except Exception:
+                    logger.warning("Invalid lat/lon received for WAB/WAI reference")
             return sig_ref
 
-        elif sig.upper() == "BOTA":
+        elif sig_name.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(" ", "-")
+            if sig_ref.name:
+                sig_ref.url = f"https://www.beachesontheair.com/beaches/{sig_ref.name.lower().replace(' ', '-')}"
+            return sig_ref
+
+        elif sig_name.upper() == "GMA Islands":
+            # GMA Islands is a bit of a mess of GMA and IOTA references. Try looking them both up and see what returns
+            # the best result.
+            iota_lookup = get_sig_ref_info("IOTA", ref_id)
+            gma_lookup = get_sig_ref_info("GMA", ref_id)
+            for key, value in iota_lookup.__dict__.items():
+                if value is not None and sig_ref.__dict__.get(key) is None:
+                    sig_ref.__dict__[key] = value
+            for key, value in gma_lookup.__dict__.items():
+                if value is not None and sig_ref.__dict__.get(key) is None:
+                    sig_ref.__dict__[key] = value
+            sig_ref.ref_type = SIGRefType.ISLAND
             return sig_ref
 
         ### ACTUAL LOOKUP ###
         #
-        # OK, this is something we have to look up. Now check to see if our data store contains 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
+        # OK, this is something we have to look up. Now check to see if our data store contains reference data and if
+        # so, copy the data into the sig_ref object
+        key = f"{sig_name}:{ref_id}"
+        try:
+            lookup_data = DATA_STORE.sigrefs.get(key) if key in DATA_STORE.sigrefs else None
+            if lookup_data:
+                for attr, value in lookup_data.__dict__.items():
+                    if value is not None and sig_ref.__dict__.get(attr) is None:
+                        sig_ref.__dict__[attr] = value
 
-        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)
+            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.
+                logger.debug(f"{sig_name} database did not contain data for ref {ref_id}")
+
+        except (ValueError, KeyError):
+            # Catch exceptions due to e.g. old versions of objects in the cache that are no longer compatible,
+            # and remove them from the cache.
+            del DATA_STORE.sigrefs[key]
+            return None
 
     except Exception:
-        logging.error("Exception when looking up sig_ref info for " + sig + " ref " + ref_id, exc_info=True)
+        logger.exception(f"Exception when looking up sig_ref info for {sig_name} ref {ref_id}")
     return sig_ref
 
 

core/sig_utils.py

@@ -21,4 +21,4 @@ def get_sig_name_from_comment_name(sig):
 
 
 # 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")"
+ANY_SIG_REGEX = rf"({'|'.join(n for s in SIGS for n in s.comment_names)})"

core/single_object_data_cache.py

@@ -0,0 +1,45 @@
+import logging
+import threading
+
+import diskcache
+
+logger = logging.getLogger(__name__)
+
+
+class SingleObjectDataCache:
+    """Cache for status and solar conditions. This uses DiskCache, but unlike the standard DiskCache users like SIG and
+    callsign lookup handlers, status and solar conditions are persisted as a single object. If we just load the object
+    from DiskCache and modify it, DiskCache doesn't know that it's been updated and needs re-caching, so we provide a
+    store() method that any functions updating the object can call afterwards."""
+
+    def __init__(self, cache_dir, object_if_empty):
+        """Initialize a SingleObjectDataCache. Provide the directory to load the cache from and save it to. If the cache
+        is empty, the provided object_if_empty parameter will be used to initialise it."""
+
+        self._lock = threading.Lock()
+        self._cache = diskcache.Cache(cache_dir)
+        # This cache stores a single object, doesn't matter what it's called so "object" will do
+        if "object" not in self._cache:
+            self._cache.add("object", object_if_empty)
+        try:
+            self._obj = self._cache.get("object")
+        except Exception:
+            logger.warning(f"Failed to load cache from {cache_dir}, clearing it.")
+            self._cache.clear()
+            self._cache.add("object", object_if_empty)
+            self._obj = object_if_empty
+
+    def get(self):
+        """Get the data object. This can then be manipulated as necessary across multiple threads. Any function
+        modifying the object must remember to call store() afterwards."""
+        return self._obj
+
+    def store(self):
+        """Store the updated object in the cache. Any function modifying the object must remember to call this
+        afterwards."""
+        with self._lock:
+            self._cache.set("object", self._obj)
+
+    def close(self):
+        self.store()
+        self._cache.close()

core/status_reporter.py

@@ -1,16 +1,21 @@
+import logging
 import os
 from datetime import datetime
-from threading import Thread, Event
+from threading import Event, Thread
 
 import psutil
 import pytz
 
+from core.cleanup import CLEANUP_TIMER
 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
+from core.prometheus_metrics_handler import alerts_gauge, memory_use_gauge, spots_gauge
+from telnetserver.telnetserver import TELNET_SERVER
+from webserver.webserver import WEB_SERVER
+
+logger = logging.getLogger(__name__)
 
 
 class StatusReporter:
@@ -24,19 +29,24 @@ class StatusReporter:
         self._stop_event = Event()
         self._startup_time = datetime.now(pytz.UTC)
 
-        DATA_STORE.status_data["software_version"] = SOFTWARE_VERSION
-        DATA_STORE.status_data["server_owner_callsign"] = SERVER_OWNER_CALLSIGN
+        DATA_STORE.status.get()["software_version"] = SOFTWARE_VERSION
+        DATA_STORE.status.get()["server_owner_callsign"] = SERVER_OWNER_CALLSIGN
+        DATA_STORE.status.store()
 
     def start(self):
         """Start the reporter thread"""
 
-        self._thread = Thread(target=self._run, name="StatusReporter")
+        self._thread = Thread(target=self._run, name="StatusReporter", daemon=True)
         self._thread.start()
 
     def stop(self):
         """Stop any threads and prepare for application shutdown"""
 
         self._stop_event.set()
+        if self._thread:
+            self._thread.join(timeout=15)
+            if self._thread.is_alive():
+                logger.warning("Status reporter worker thread did not exit on time and will be killed.")
 
     def _run(self):
         """Thread entry point: report immediately on startup, then on each interval until stopped"""
@@ -49,57 +59,98 @@ class StatusReporter:
     def _report(self):
         """Write status information"""
 
-        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,
-                           "enabled_by_default_in_web_ui": p.enabled_by_default_in_web_ui, "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},
-                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},
-                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},
-                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 WEB_SERVER.web_server_metrics[
-                                                   "last_api_access_time"] else 0,
-                                               "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 WEB_SERVER.web_server_metrics[
-                                                   "last_page_access_time"] else 0,
-                                               "page_access_count": WEB_SERVER.web_server_metrics[
-                                                   "page_access_counter"]}
+        DATA_STORE.status.get()["uptime"] = (datetime.now(pytz.UTC) - self._startup_time).total_seconds()
+        DATA_STORE.status.get()["mem_use_mb"] = round(psutil.Process(os.getpid()).memory_info().rss / (1024 * 1024), 3)
+        DATA_STORE.status.get()["num_spots"] = len(DATA_STORE.spots.values())
+        DATA_STORE.status.get()["num_alerts"] = len(DATA_STORE.alerts.values())
+        DATA_STORE.status.get()["spot_providers"] = [
+            {
+                "name": p.name,
+                "enabled": p.enabled,
+                "enabled_by_default_in_web_ui": p.enabled_by_default_in_web_ui,
+                "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,
+            }
+            for p in DATA_PROVIDERS.spot_providers
+        ]
+        DATA_STORE.status.get()["alert_providers"] = [
+            {
+                "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,
+            }
+            for p in DATA_PROVIDERS.alert_providers
+        ]
+        DATA_STORE.status.get()["solar_condition_providers"] = [
+            {
+                "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,
+            }
+            for p in DATA_PROVIDERS.solar_condition_providers
+        ]
+        DATA_STORE.status.get()["static_data_providers"] = [
+            {
+                "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,
+            }
+            for p in DATA_PROVIDERS.static_data_providers
+        ]
+        DATA_STORE.status.get()["sig_ref_data_providers"] = [
+            {
+                "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,
+            }
+            for p in DATA_PROVIDERS.sig_ref_data_providers
+        ]
+        DATA_STORE.status.get()["callsign_data_providers"] = [
+            {
+                "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,
+            }
+            for p in DATA_PROVIDERS.callsign_data_providers
+        ]
+        DATA_STORE.status.get()["cleanup"] = {
+            "status": CLEANUP_TIMER.status,
+            "last_ran": CLEANUP_TIMER.last_cleanup_time.replace(tzinfo=pytz.UTC).timestamp()
+            if CLEANUP_TIMER.last_cleanup_time
+            else 0,
+        }
+        DATA_STORE.status.get()["webserver"] = {
+            "status": WEB_SERVER.web_server_metrics.status,
+            "api_requests_per_hour": WEB_SERVER.web_server_metrics.api_requests_per_hour(),
+            "page_requests_per_hour": WEB_SERVER.web_server_metrics.page_requests_per_hour(),
+            "sse_client_count": WEB_SERVER.sse_client_count,
+        }
+        DATA_STORE.status.get()["telnet"] = {
+            "client_count": TELNET_SERVER.client_count,
+        }
+        DATA_STORE.status.store()
 
         # Update Prometheus metrics
         memory_use_gauge.set(psutil.Process(os.getpid()).memory_info().rss)

core/url_data_cache.py

@@ -14,12 +14,14 @@ class URLDataCache(CachedSession):
     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))
+        super().__init__(
+            f"{CACHE_DIR}urls/{name}",
+            expire_after=timedelta(days=1),
+            allowable_codes=(200, 400, 401, 403, 404),
+        )
+        self._lock = threading.Lock()
 
     def get(self, *args, **kwargs):
         with self._lock:
-            return super().get(*args, **kwargs)
+            return super().get(*args, **kwargs)

core/utils.py

@@ -1,12 +1,16 @@
 import logging
+import re
 
 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.constants import BANDS, UNKNOWN_BAND
 from core.data_store import DATA_STORE
-from data.callsign import Callsign
+from core.enums import MODE_ALIASES, Continent, Mode, ModeType
+from data.callsign import Callsign, LocationSourceForCallsign
+
+logger = logging.getLogger(__name__)
 
 
 def safe_json_dumps(obj):
@@ -16,30 +20,42 @@ def safe_json_dumps(obj):
     return simplejson.dumps(obj, ensure_ascii=False, ignore_nan=True, default=lambda o: o.__dict__)
 
 
-def infer_mode_from_comment(comment):
+def infer_mode_from_comment(comment: str) -> Mode | None:
     """Infer a mode from the comment"""
 
-    for mode in ALL_MODES:
-        if mode in comment.upper():
+    if not comment:
+        return None
+
+    for mode in Mode:
+        if re.search(r"(^|\W)" + mode + r"($|\W)", comment, re.IGNORECASE):
             return mode
-    for mode in MODE_ALIASES.keys():
-        if mode in comment.upper():
-            return MODE_ALIASES[mode]
+    for alias in MODE_ALIASES:
+        if re.search(r"(^|\W)" + alias + r"($|\W)", comment, re.IGNORECASE):
+            return Mode(MODE_ALIASES[alias])
+
     return None
 
 
-def infer_mode_type_from_mode(mode):
-    """Infer a "mode family" from a mode."""
+def infer_mode_type_from_mode(mode: str) -> ModeType | None:
+    """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.")
+    if not mode:
+        return None
+
+    if mode in MODE_ALIASES:
+        mode = MODE_ALIASES[mode]
+
+    try:
+        mode = Mode(mode.upper())
+        if mode.is_cw:
+            return ModeType.CW
+        if mode.is_phone:
+            return ModeType.PHONE
+        return ModeType.DATA
+
+    except ValueError:
+        if mode.upper() != "OTHER" and mode != "?":
+            logger.warning(f"Found an unrecognised mode: {mode}. Developer should categorise this.")
         return None
 
 
@@ -60,16 +76,39 @@ def infer_mode_from_frequency(freq):
         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
+        # a spot at 7074.5 kHz will be indicated as SSB, 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):
+        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):
+        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
+
+        if not mode:
+            return None
+
+        mode = mode.upper()
+        if mode in MODE_ALIASES:
+            mode = MODE_ALIASES[mode]
+
+        return Mode(mode)
+
     except KeyError:
         return None
 
@@ -77,7 +116,7 @@ def infer_mode_from_frequency(freq):
 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
+    dxcc_data = DATA_STORE.dxcc_data.get(dxcc, None)
     return dxcc_data["flag"] if dxcc_data else None
 
 
@@ -89,28 +128,30 @@ def get_callsign_object_from_pyhamtools_callinfo(callsign, callinfo):
         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
+        country = data.get("country", None)
+        dxcc_id = data.get("adif", None)
+        continent = Continent(data["continent"]) if "continent" in data else None
+        cq_zone = data.get("cqz", None)
+        itu_zone = data.get("ituz", 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")
+        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=LocationSourceForCallsign.DXCC,
+        )
 
     except (KeyError, ValueError):
         # Unknown callsign, can't look anything up, return a Callsign object with basic data so that gets cached

data/alert.py

@@ -1,16 +1,18 @@
-import copy
 import hashlib
 import json
 import logging
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from datetime import datetime, timedelta
 
 import pytz
 
 from core.call_lookup_helper import get_call_info
+from core.enums import AlertType, Continent
 from core.sig_lookup_helper import populate_missing_sig_ref_info
 from core.utils import get_flag_for_dxcc
 
+logger = logging.getLogger(__name__)
+
 
 @dataclass
 class Alert:
@@ -18,6 +20,9 @@ class Alert:
 
     # Unique identifier for the alert
     id: str | None = None
+
+    # DX (alerting) operator info
+
     # Callsigns of the operators that has been alerted
     dx_calls: list | None = None
     # Names of the operators that has been alerted
@@ -27,13 +32,16 @@ class Alert:
     # Country flag of the DX operator
     dx_flag: str | None = None
     # Continent of the DX operator
-    dx_continent: str | None = None
+    dx_continent: Continent | None = None
     # DXCC ID of the DX operator
     dx_dxcc_id: int | None = None
     # CQ zone of the DX operator
     dx_cq_zone: int | None = None
     # ITU zone of the DX operator
     dx_itu_zone: int | None = None
+
+    # General alert info
+
     # Intended frequencies & modes of operation. Essentially just a different kind of comment field.
     freqs_modes: str | None = None
     # Start time of the activation, UTC seconds since UNIX epoch
@@ -44,25 +52,41 @@ class Alert:
     end_time: float | None = None
     # End time of the activation of the alert, ISO 8601
     end_time_iso: str | None = None
+    # Comment made by the alerter, if any
+    comment: str | None = None
+    # The type of alert this is: xOTA, DXpedition, or Contest.
+    alert_type: AlertType | None = None
+    # A URL link to more information, if any
+    url: str | None = None
+
+    # Special Interest Group info
+
+    # Special Interest Group (SIG), e.g. outdoor activity programme such as POTA
+    sig: str | None = None
+    # SIG references. We allow multiple here for e.g. n-fer activations, unlike ADIF SIG_INFO
+    sig_refs: list = field(default_factory=list)
+
+    # Timing info
+
     # 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 = None
     # Time that this software received the alert, ISO 8601
     received_time_iso: str | None = None
-    # Comment made by the alerter, if any
-    comment: str | None = None
-    # Special Interest Group (SIG), e.g. outdoor activity programme such as POTA
-    sig: str | None = None
-    # SIG references. We allow multiple here for e.g. n-fer activations, unlike ADIF SIG_INFO
-    sig_refs: list | None = None
-    # Whether this alert is for a DXpedition, as opposed to e.g. an xOTA programme.
-    is_dxpedition: bool = False
+
+    # Source info
+
     # Where we got the alert from, e.g. "POTA", "SOTA"...
     source: str | None = None
     # The ID the source gave it, if any.
     source_id: str | None = None
 
+    # Display info
+
+    # Icon to use when displaying this alert in the web UI. Chosen from the Font Awesome set.
+    icon: str | None = None
+
     def infer_missing(self, credentials=None):
         """Infer missing parameters where possible"""
 
@@ -90,8 +114,8 @@ class Alert:
                 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 = call_info.country
-                if self.dx_calls and self.dx_calls[0] and not self.dx_continent:
-                    self.dx_continent = call_info.continent
+                if self.dx_calls and self.dx_calls[0] and call_info.continent and not self.dx_continent:
+                    self.dx_continent = Continent(call_info.continent)
                 if self.dx_calls and self.dx_calls[0] and not self.dx_cq_zone:
                     self.dx_cq_zone = call_info.cq_zone
                 if self.dx_calls and self.dx_calls[0] and not self.dx_itu_zone:
@@ -104,35 +128,43 @@ class Alert:
             # 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:
+            if self.sig_refs:
                 for sig_ref in self.sig_refs:
                     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.
-            if self.sig_refs and len(self.sig_refs) > 0 and self.sig_refs[0] and not self.sig:
+            if self.sig_refs and self.sig_refs[0] and not self.sig:
                 self.sig = self.sig_refs[0].sig
 
-            # Always create an ID based on a hash of every parameter *except* received_time. This is used as the index
-            # to a map, which as a byproduct avoids us having multiple duplicate copies of the object that are identical
-            # apart from that they were retrieved from the API at different times. Note that the simple Python hash()
-            # function includes a seed randomly generated at runtime; this is therefore not consistent between runs. But we
-            # use diskcache to store our data between runs, so we use SHA256 which does not include this random element.
-            # The ID is computed before the online lookups below so that it is stable regardless of whether credentials
-            # are provided, allowing the enriched API response to be matched to the stored alert by ID.
+            # Create an ID based on the source and source ID if possible, as these guaranee uniqueness. If there is no
+            # source ID, use a combination of callsign and start time. Excluding things like the comment here allows for
+            # user updates of their alert comments without duplicating in the system.
             if not self.id:
-                self_copy = copy.deepcopy(self)
-                self_copy.received_time = 0
-                self_copy.received_time_iso = ""
-                self.id = hashlib.sha256(str(self_copy).encode("utf-8")).hexdigest()
+                if self.source and self.source_id:
+                    self.id = hashlib.sha256(str({"s": self.source, "sid": self.source_id}).encode("utf-8")).hexdigest()
+                else:
+                    self.id = hashlib.sha256(
+                        str({"s": self.source, "c": self.dx_calls, "t": self.start_time}).encode("utf-8")
+                    ).hexdigest()
 
             # DX operator name lookup, using QRZ.com/HamQTH.
             if self.dx_calls and not self.dx_names:
-                self.dx_names = list(
-                    map(lambda c: get_call_info(c, credentials).name, self.dx_calls))
+                self.dx_names = [get_call_info(c, credentials).name for c in self.dx_calls]
 
-        except Exception as e:
-            logging.error("Exception while inferring missing data from spot", e, exc_info=True)
+            # Icon for the spot should be the icon of the first SIG ref if present, otherwise a radio tower
+            self.icon = "fa-tower-cell"
+            if self.alert_type == AlertType.DXPEDITION:
+                self.icon = "fa-globe-africa"
+            elif self.alert_type == AlertType.CONTEST:
+                self.icon = "fa-trophy"
+            elif self.alert_type == AlertType.SATELLITE:
+                self.icon = "fa-satellite"
+            elif self.sig_refs and self.sig_refs[0].icon:
+                self.icon = self.sig_refs[0].icon
+
+        except Exception:
+            logger.exception("Exception while inferring missing data from spot")
 
     def to_json(self):
         """JSON serialise"""
@@ -145,5 +177,8 @@ class Alert:
         either having an end_time in the past, or if it only has a start_time, then that start time was more than 3 hours
         ago. If it somehow doesn't have a start_time either, it is considered to be expired."""
 
-        return not self.start_time or (self.end_time and self.end_time < datetime.now(pytz.UTC).timestamp()) or (
-                not self.end_time and self.start_time < (datetime.now(pytz.UTC) - timedelta(hours=3)).timestamp())
+        return (
+            not self.start_time
+            or (self.end_time and self.end_time < datetime.now(pytz.UTC).timestamp())
+            or (not self.end_time and self.start_time < (datetime.now(pytz.UTC) - timedelta(hours=3)).timestamp())
+        )

data/callsign.py

@@ -1,5 +1,7 @@
 from dataclasses import dataclass
 
+from core.enums import Continent, LocationSourceForCallsign
+
 
 @dataclass
 class Callsign:
@@ -13,35 +15,45 @@ class Callsign:
     # "Home" call, i.e. with any prefixes and suffixes stripped off
     home_call: str | None = None
     # Operator name
-    name : str | None = None
+    name: str | None = None
     # QTH (location), free text
-    qth : str | None = None
+    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
+    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
+    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
+    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
+    continent: Continent | 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
+    # Location source
+    location_source: LocationSourceForCallsign | 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
+        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/lookup_credentials.py

@@ -4,6 +4,7 @@ from dataclasses import dataclass
 @dataclass
 class LookupCredentials:
     """Per-request credentials for QRZ.com and HamQTH online callsign lookups."""
+
     qrz_username: str = ""
     qrz_password: str = ""
     qrz_session_key: str = ""  # alternative to username/password

data/sig.py

@@ -1,21 +1,39 @@
 from dataclasses import dataclass, field
 
+from core.enums import SIGRefType, SIGType
+
 
 @dataclass
 class SIG:
     """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."""
+    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 as used in the UI and API, e.g. "Towers"
     name: str
     # Description, e.g. "Towers on the Air"
     description: str
+    # Type, either Worldwide, Regional or Event. Used for sorting in the web UI.
+    sig_type: SIGType
+    # Identifies that the SIG's reference ID structure defined by its regex is unique across all programmes and
+    # anything else we expect a user to put in a spot comment, and therefore we can pull references out of
+    # spot comments without also needing to see the SIG name first. For example, "OHFF-1234" or "B/G-1234" are
+    # obviously WWFF and WWBOTA, nothing else looks like those. But "SZ09" could be WAB or Tiles, "GB1234" could
+    # conceivably be POTA or ILLW, etc.
+    refs_globally_unique: bool
     # SIG names as they might appear in cluster spot comments, e.g. ["TOTA"]
     comment_names: list[str] = field(default_factory=list)
+    # Reference type, what gets activated e.g. Park, Summit. May be None if the SIG is for multiple types of things, in
+    # which case the spot data will have to provide this instead.
+    ref_type: SIGRefType | None = None
     # Regex matcher for references, e.g. for POTA r"[A-Z]{2}\-\d+".
     ref_regex: str | None = None
+    # Icon to use in the UI when referencing this SIG. Chosen from the Font Awesome set.
+    icon: str | None = None
+    # Emoji flag for the country or region where this SIG is relevant, if any. If None, this implies the SIG is in
+    # worldwide usage.
+    region_flag: str | None = None

data/sig_ref.py

@@ -1,24 +1,32 @@
 from dataclasses import dataclass
 
+from core.enums import SIGRefType
+
 
 @dataclass
 class SIGRef:
     """Data class that defines a Special Interest Group "info" or reference. As well as the basic reference ID we include a
     name and a lookup URL."""
 
-    # Reference ID, e.g. "GB-0001".
-    id: str
     # SIG that this reference is in, e.g. "POTA".
     sig: str
+    # Reference ID, e.g. "GB-0001".
+    id: str | None = None
     # Name of the reference, e.g. "Null Country Park", if known.
     name: str | None = None
+    # Type of the reference, e.g. "Park", if known.
+    ref_type: SIGRefType | None = None
     # URL to look up more information about the reference, if known.
     url: str | None = None
-    # Latitude of the reference, if known.
+    # Icon to use for the reference, derived from the SIG. Chosen from the Font Awesome set.
+    icon: str | None = None
+    # Latitude of the reference, in degrees, if known.
     latitude: float | None = None
-    # Longitude of the reference, if known.
+    # Longitude of the reference, in degrees, if known.
     longitude: float | None = None
-    # Maidenhead grid reference of the reference, if known.
+    # Altitude of the reference, in metres, if known.
+    altitude: float | None = None
+    # Maidenhead grid ref of the reference, if known.
     grid: str | None = None
     # Activation score. SOTA only
     activation_score: int | None = None

data/solar_conditions.py

@@ -82,9 +82,9 @@ def _xray_blackout_scale(xray):
         number = float(xray[1:])
     except ValueError:
         return 0
-    if letter == 'M':
+    if letter == "M":
         return 1 if number < 5 else 2
-    if letter == 'X':
+    if letter == "X":
         if number < 10:
             return 3
         if number < 20:

data/spot.py

@@ -1,24 +1,37 @@
-import copy
 import hashlib
 import json
 import logging
 import re
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from datetime import datetime, timedelta
+from math import isnan
 
 import pytz
-from pyhamtools.locator import locator_to_latlong, latlong_to_locator
+from pyhamtools.locator import latlong_to_locator, locator_to_latlong
 
-from core.config import MAX_SPOT_AGE
-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.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.config import MAX_SPOT_AGE
+from core.constants import PROPAGATION_MODES, SIGS
+from core.data_store import DATA_STORE
+from core.enums import Continent, LocationSourceForSpot, Mode, ModeSource, ModeType
+from core.geo_utils import lat_lon_to_cq_zone, lat_lon_to_itu_zone
 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 core.sig_utils import (
+    ANY_SIG_REGEX,
+    get_ref_regex_for_sig,
+    get_sig_name_from_comment_name,
+)
+from core.utils import (
+    get_flag_for_dxcc,
+    infer_band_from_freq,
+    infer_mode_from_comment,
+    infer_mode_from_frequency,
+    infer_mode_type_from_mode,
+)
 from data.sig_ref import SIGRef
 
+logger = logging.getLogger(__name__)
+
 
 @dataclass
 class Spot:
@@ -41,7 +54,7 @@ class Spot:
     # Country flag of the DX operator
     dx_flag: str | None = None
     # Continent of the DX operator
-    dx_continent: str | None = None
+    dx_continent: Continent | None = None
     # DXCC ID of the DX operator
     dx_dxcc_id: int | None = None
     # CQ zone of the DX operator
@@ -57,9 +70,8 @@ class Spot:
     # lookup
     dx_latitude: float | None = None
     dx_longitude: float | None = None
-    # DX Location source. Indicates how accurate the location might be. Values: "SPOT", "GRID", "SIG REF LOOKUP",
-    # "HOME QTH", "DXCC", "NONE"
-    dx_location_source: str = "NONE"
+    # DX Location source. Indicates how accurate the location might be.
+    dx_location_source: LocationSourceForSpot | None = None
     # DX Location good. Indicates that the software thinks the location data is good enough to plot on a map. This is
     # true if the location source is "SPOT", "SIG REF LOOKUP" or "GRID", or if the location source is "HOME QTH" and the
     # DX callsign doesn't have a suffix like /P.
@@ -74,7 +86,7 @@ class Spot:
     # Country flag of the spotter
     de_flag: str | None = None
     # Continent of the spotter
-    de_continent: str | None = None
+    de_continent: Continent | None = None
     # DXCC ID of the spotter
     de_dxcc_id: int | None = None
     # If this is an APRS/Packet/etc spot, what SSID was the spotter/receiver using?
@@ -92,11 +104,11 @@ class Spot:
     # General QSO info
 
     # Reported mode, such as SSB, PHONE, CW, FT8...
-    mode: str | None = None
-    # Inferred mode "family". One of "CW", "PHONE" or "DIGI".
-    mode_type: str | None = None
-    # Source of the mode information. "SPOT", "COMMENT", "BANDPLAN" or "NONE"
-    mode_source: str = "NONE"
+    mode: Mode | None = None
+    # Inferred mode "family".
+    mode_type: ModeType | None = None
+    # Source of the mode information.
+    mode_source: ModeSource | None = None
     # Frequency, in Hz
     freq: float | None = None
     # Band, defined by the frequency, e.g. "40m" or "70cm"
@@ -113,7 +125,7 @@ class Spot:
     # Special Interest Group (SIG), e.g. outdoor activity programme such as POTA
     sig: str | None = None
     # SIG references. We allow multiple here for e.g. n-fer activations, unlike ADIF SIG_INFO
-    sig_refs: list | None = None
+    sig_refs: list = field(default_factory=list)
 
     # Timing info
 
@@ -135,16 +147,18 @@ class Spot:
     # The ID the source gave it, if any.
     source_id: str | None = None
 
+    # Display info
+
+    # Icon to use when displaying this spot in the web UI. Chosen from the Font Awesome set.
+    icon: 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
-            ]
+            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"""
@@ -176,8 +190,8 @@ class Spot:
             dx_call_info = get_call_info(self.dx_call, credentials)
             if self.dx_call and not self.dx_country:
                 self.dx_country = dx_call_info.country
-            if self.dx_call and not self.dx_continent:
-                self.dx_continent = dx_call_info.continent
+            if self.dx_call and dx_call_info.continent and not self.dx_continent:
+                self.dx_continent = Continent(dx_call_info.continent)
             if self.dx_call and not self.dx_dxcc_id:
                 self.dx_dxcc_id = dx_call_info.dxcc_id
             if self.dx_dxcc_id and not self.dx_flag:
@@ -207,19 +221,22 @@ 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 (
+                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 = de_call_info.country
-                if not self.de_continent:
-                    self.de_continent = de_call_info.continent
+                if de_call_info.continent and not self.de_continent:
+                    self.de_continent = Continent(de_call_info.continent)
                 if not self.de_dxcc_id:
                     self.de_dxcc_id = de_call_info.dxcc_id
                 if self.de_dxcc_id and not self.de_flag:
                     self.de_flag = get_flag_for_dxcc(self.de_dxcc_id)
 
             # Remove NaNs in frequency
-            if self.freq and self.freq == float("nan"):
+            if self.freq and isnan(self.freq):
                 self.freq = None
 
             # Band from frequency
@@ -229,17 +246,13 @@ class Spot:
 
             # Mode from comments or bandplan
             if self.mode:
-                self.mode_source = "SPOT"
+                self.mode_source = ModeSource.SPOT
             if self.comment and not self.mode:
                 self.mode = infer_mode_from_comment(self.comment)
-                self.mode_source = "COMMENT"
+                self.mode_source = ModeSource.COMMENT
             if self.freq and not self.mode:
                 self.mode = infer_mode_from_frequency(self.freq)
-                self.mode_source = "BANDPLAN"
-
-            # Normalise mode if necessary.
-            if self.mode in MODE_ALIASES:
-                self.mode = MODE_ALIASES[self.mode]
+                self.mode_source = ModeSource.BANDPLAN
 
             # Mode type from mode
             if self.mode and not self.mode_type:
@@ -247,19 +260,19 @@ class Spot:
 
             # If we have a latitude or grid at this point, it can only have been provided by the spot itself
             if self.dx_latitude or self.dx_grid:
-                self.dx_location_source = "SPOT"
+                self.dx_location_source = LocationSourceForSpot.SPOT
 
             # Set the top-level "SIG" if it is missing but we have at least one SIG ref.
-            if not self.sig and self.sig_refs and len(self.sig_refs) > 0:
+            if not self.sig and self.sig_refs:
                 self.sig = self.sig_refs[0].sig.upper()
 
             # See if we already have a SIG reference, but the comment looks like it contains more for the same SIG. This
             # should catch e.g. POTA comments like "2-fer: GB-0001 GB-0002".
-            if self.comment and self.sig_refs and len(self.sig_refs) > 0 and self.sig_refs[0].sig:
+            if self.comment and self.sig_refs and self.sig_refs[0].sig:
                 sig = self.sig_refs[0].sig.upper()
                 regex = get_ref_regex_for_sig(sig)
                 if regex:
-                    all_comment_ref_matches = re.finditer(r"(^|\W)(" + regex + r")(^|\W)", self.comment, re.IGNORECASE)
+                    all_comment_ref_matches = re.finditer(r"(^|\W)(" + regex + r")($|\W)", self.comment, re.IGNORECASE)
                     for ref_match in all_comment_ref_matches:
                         self._append_sig_ref_if_missing(SIGRef(id=ref_match.group(2).upper(), sig=sig))
 
@@ -279,31 +292,47 @@ 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"([ -])(" + ref_regex + r")($|\W)", self.comment,
-                                                  re.IGNORECASE)
+                        ref_matches = re.finditer(
+                            r"(^|\W)" + found_sig + r"([ -])(" + ref_regex + r")($|\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))
 
+            # See if the comment looks like it contains any SIG references *without* the corresponding SIG name, but
+            # where the SIG reference is unique-looking enough that we can't confuse it with any other SIG.
+            if self.comment:
+                for sig in SIGS:
+                    if sig.refs_globally_unique and sig.ref_regex:
+                        ref_matches = re.finditer(r"(^|\W)(" + sig.ref_regex + r")($|\W)", self.comment, re.IGNORECASE)
+                        for ref_match in ref_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 "OHFF-1234", now we know it's WWFF.
+                            if not self.sig:
+                                self.sig = sig.name
+                            self._append_sig_ref_if_missing(SIGRef(id=ref_match.group(2).upper(), sig=sig.name))
+
             # 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:
+            if self.sig_refs:
                 for sig_ref in self.sig_refs:
                     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
-                    if sig_ref.latitude and not self.dx_latitude:
+                    if sig_ref.latitude and not self.dx_latitude and sig_ref.longitude and not self.dx_longitude:
                         self.dx_latitude = sig_ref.latitude
                         self.dx_longitude = sig_ref.longitude
                         if self.sig == "WAB" or self.sig == "WAI" or self.sig == "Tiles":
-                            self.dx_location_source = "GRID"
+                            self.dx_location_source = LocationSourceForSpot.GRID
                         else:
-                            self.dx_location_source = "SIG REF LOOKUP"
+                            self.dx_location_source = LocationSourceForSpot.SIG_REF_LOOKUP
 
             # 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.
-            if self.sig_refs and len(self.sig_refs) > 0 and not self.sig:
+            if self.sig_refs 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".
@@ -311,15 +340,16 @@ class Spot:
             # 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)
+                    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 = "GRID"
+                        self.dx_location_source = LocationSourceForSpot.GRID
 
                     # 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()
@@ -328,58 +358,69 @@ class Spot:
                             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)
+                            logger.info(f"Seen a new propagation mode tag not yet in the system: {mode_tag}")
+
+            # Set SIGs based on propagation mode
+            if self.propagation_mode == "Satellite":
+                if not self.sig:
+                    self.sig = "AMSAT"
+                if not any(sig_ref.sig == "AMSAT" for sig_ref in self.sig_refs):
+                    self.sig_refs.append(SIGRef(sig="AMSAT"))
+            if self.propagation_mode == "Earth-Moon-Earth":
+                if not self.sig:
+                    self.sig = "EME"
+                if not any(sig_ref.sig == "EME" for sig_ref in self.sig_refs):
+                    self.sig_refs.append(SIGRef(sig="EME"))
 
             # Parse "de_grid -> dx_grid" structures from the comment
             if self.comment:
                 grid_mode_grid_match = re.search(
-                    r'\b([A-Ra-r]{2}\d{2}(?:[A-Xa-x]{2}(?:\d{2})?)?)\s*->\s*([A-Ra-r]{2}\d{2}(?:[A-Xa-x]{2}(?:\d{2})?)?)\b',
-                    self.comment)
+                    r"\b([A-Ra-r]{2}\d{2}(?:[A-Xa-x]{2}(?:\d{2})?)?)\s*->\s*([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.dx_grid:
                         self.dx_grid = grid_mode_grid_match.group(1).upper()
-                        self.dx_location_source = "GRID"
+                        self.dx_location_source = LocationSourceForSpot.GRID
                     if not self.de_grid:
                         self.de_grid = grid_mode_grid_match.group(2).upper()
 
             # DX Grid to lat/lon and vice versa in case one is missing
-            if self.dx_grid and not self.dx_latitude:
+            if self.dx_grid and (not self.dx_latitude or not self.dx_longitude):
                 try:
                     ll = locator_to_latlong(self.dx_grid)
                     self.dx_latitude = ll[0]
                     self.dx_longitude = ll[1]
-                except:
-                    logging.debug("Invalid grid received for spot")
+                except Exception:
+                    logger.debug("Invalid grid received for spot")
             if self.dx_latitude and self.dx_longitude and not self.dx_grid:
                 try:
                     self.dx_grid = latlong_to_locator(self.dx_latitude, self.dx_longitude, 8)
-                except:
-                    logging.debug("Invalid lat/lon received for spot")
+                except Exception:
+                    logger.debug("Invalid lat/lon received for spot")
 
             # QRT comment detection
             if self.comment and not self.qrt:
                 self.qrt = "QRT" in self.comment.upper()
 
-            # Always create an ID based on a hash of every parameter *except* received_time. This is used as the index
-            # to a map, which as a byproduct avoids us having multiple duplicate copies of the object that are identical
-            # apart from that they were retrieved from the API at different times. Note that the simple Python hash()
-            # function includes a seed randomly generated at runtime; this is therefore not consistent between runs. But we
-            # use diskcache to store our data between runs, so we use SHA256 which does not include this random element.
-            # The ID is computed before the online lookups below so that it is stable regardless of whether credentials
-            # are provided, allowing the enriched API response to be matched to the stored spot by ID.
+            # Create an ID based on the source and source ID if possible, as these guaranee uniqueness. If there is no
+            # source ID, use a combination of callsign and spot time. Spot time is down to the second or even the
+            # millisecond, so we can be reasonably sure two spots that match are the same spot.
             if not self.id:
-                self_copy = copy.deepcopy(self)
-                self_copy.received_time = 0
-                self_copy.received_time_iso = ""
-                self.id = hashlib.sha256(str(self_copy).encode("utf-8")).hexdigest()
+                if self.source and self.source_id:
+                    self.id = hashlib.sha256(str({"s": self.source, "sid": self.source_id}).encode("utf-8")).hexdigest()
+                else:
+                    self.id = hashlib.sha256(
+                        str({"s": self.source, "c": self.dx_call, "t": self.time}).encode("utf-8")
+                    ).hexdigest()
 
             # 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 = dx_call_info.name
-            if self.dx_call and not self.dx_latitude:
+            if self.dx_call and (not self.dx_latitude or not self.dx_longitude):
                 self.dx_latitude = dx_call_info.latitude
                 self.dx_longitude = dx_call_info.longitude
                 self.dx_grid = dx_call_info.grid
@@ -387,44 +428,67 @@ class Spot:
 
             # 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:
+            if self.sig_refs:
                 qth = self.sig_refs[0].id
                 if self.sig_refs[0].name:
-                    qth += " " + self.sig_refs[0].name
+                    qth += f" {self.sig_refs[0].name}"
                 self.dx_qth = qth
             else:
                 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:
+                if self.dx_latitude and self.dx_longitude:
                     self.dx_cq_zone = lat_lon_to_cq_zone(self.dx_latitude, self.dx_longitude)
                 elif self.dx_call:
                     self.dx_cq_zone = dx_call_info.cq_zone
             if not self.dx_itu_zone:
-                if self.dx_latitude:
+                if self.dx_latitude and self.dx_longitude:
                     self.dx_itu_zone = lat_lon_to_itu_zone(self.dx_latitude, self.dx_longitude)
                 elif self.dx_call:
                     self.dx_itu_zone = dx_call_info.itu_zone
 
+            # DXCC lookup from callsign if nothing else has provided it
+            if self.dx_call and not self.dx_dxcc_id:
+                for regex, entity_code in DATA_STORE.dxcc_lookup_by_call_regex:
+                    if regex.pattern and regex.match(self.dx_call):
+                        self.dx_dxcc_id = entity_code
+                        break
+            if self.dx_dxcc_id and not self.dx_flag:
+                self.dx_flag = get_flag_for_dxcc(self.dx_dxcc_id)
+
             # 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 = 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 == "GRID"
-                    or (self.dx_location_source == "HOME QTH" and "/" not in (self.dx_call or ""))))
+            self.dx_location_good = bool(
+                self.dx_latitude
+                and self.dx_longitude
+                and (
+                    self.dx_location_source == LocationSourceForSpot.SPOT
+                    or self.dx_location_source == LocationSourceForSpot.SIG_REF_LOOKUP
+                    or self.dx_location_source == LocationSourceForSpot.GRID
+                    or (self.dx_location_source == LocationSourceForSpot.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"):
+            if (
+                self.de_call
+                and any(char.isdigit() for char in str(self.de_call))
+                and not (self.de_call.startswith("T2") and self.source == "APRS-IS")
+                and (not self.de_latitude or not self.de_longitude)
+            ):
                 # DE operator location lookup
-                if not self.de_latitude:
-                    self.de_latitude = de_call_info.latitude
-                    self.de_longitude = de_call_info.longitude
-                    self.de_grid = de_call_info.grid
+                self.de_latitude = de_call_info.latitude
+                self.de_longitude = de_call_info.longitude
+                self.de_grid = de_call_info.grid
 
-        except Exception as e:
-            logging.error("Exception while inferring missing data from spot", e, exc_info=True)
+            # Icon for the spot should be the icon of the first SIG ref if present, otherwise a radio tower
+            self.icon = "fa-tower-cell"
+            if self.sig_refs and self.sig_refs[0].icon:
+                self.icon = self.sig_refs[0].icon
+
+        except Exception:
+            logger.exception("Exception while inferring missing data from spot")
 
     def to_json(self):
         """JSON serialise"""
@@ -434,16 +498,14 @@ 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."""
 
-        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 sig_refs:
+        for sig_ref in self.sig_refs:
             if sig_ref.id == new_sig_ref.id and sig_ref.sig == new_sig_ref.sig:
                 return
-        sig_refs.append(new_sig_ref)
+        self.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/didbase-stations.csv

@@ -1,12 +1,9 @@
 AA343,"Almaty, Kazakhstan"
 AL945,"Alpena, United States"
 AT138,"Athens, Greece"
-AU930,"Austin, United States"
-BR52P,"Brisbane, Australia"
 BVJ03,"Boa Vista, Brazil"
 CAJ2M,"Cachoeira Paulista, Brazil"
 CB53N,"Canberra, Australia"
-CGK21,"Campo Grande, Brazil"
 DB049,"Dourbes, Belgium"
 DW41K,"Darwin, Australia"
 EA036,"El Arenosillo, Spain"
@@ -18,24 +15,18 @@ GM037,"Gibilmanna, Italy"
 GR13L,"Grahamstown, South Africa"
 HE13N,"Hermanus, South Africa"
 HO54K,"Hobart, Australia"
-IC437,"I-Cheon, South Korea"
 IF843,"Idaho National Laboratory, United States"
 JI91J,"Jicamarca, Peru"
 JR055,"Juliusruh, Germany"
 LAA38,"Lajes Terceira Island, Portugal"
 LL721,"Lualualei, United States"
-LM42B,"Learmonth, Australia"
-LV12P,"Louisvale, South Africa"
 MHJ45,"Millstone Hill, United States"
-ML10L,"Malindi, Kenya"
 NI135,"Nicosia, Cyprus"
 NI63_,"Norfolk, Australia"
-PA836,"Portt Arguello, United States"
+PA836,"Port Arguello, United States"
 PE43K,"Perth, Australia"
-PF765,"Poker Flat, United States"
 PQ052,"Pruhonice, Czechia"
 RO041,"Rome, Italy"
-SAA0K,"Saoluis, Brazil"
 SO148,"Sopron, Hungary"
 TR169,"Tromso, Norway"
 TV51R,"Townsville, Australia"

datafiles/toilets.csv

@@ -27,4 +27,9 @@ 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
+EH T-51,50.3636495,7.5584857
+HOPE T-01,40.7527998,-73.9935015
+HOPE T-02,40.7526614,-73.9935752
+HOPE T-03,40.7528663,-73.9936566
+HOPE T-04,40.7527845,-73.9937284
+HOPE T-05,40.7518292,-73.9932781

docs/docker.md

@@ -13,6 +13,7 @@ services:
     restart: unless-stopped
     ports:
       - "8080:8080"
+      - "7373:7373" # For telnet if required
     volumes:
       - ./config.yml:/app/config.yml
       - ./cache:/app/cache
@@ -23,15 +24,16 @@ You can replace `#main` with any other branch or tag reference, for example `#1.
 
 Save the file. You will still need to create a copy of `config-example.yml` and name it `config.yml`, though with the
 Docker setup nothing has actually been downloaded yet, so you will have to copy the example from the repository some
-other way, e.g. [from the repo in a web browser](https://git.ianrenton.com/ian/spothole/src/branch/main/config-example.yml).
+other way,
+e.g. [from the repo in a web browser](https://git.ianrenton.com/ian/spothole/src/branch/main/config-example.yml).
 
 With that in place, run `docker compose up` and you should be good to go. To detach, press `d` or run the command with
 the `-d` flag.
 
 ### nginx Reverse Proxy with Docker
 
-In a containerised setup, it's typical to run an nginx reverse proxy in one container, alongside certbot for renewal
-of HTTPS certificates, and then applications like Spothole in a separate container. In this case, there are a couple of
+In a containerised setup, it's typical to run an nginx reverse proxy in one container, alongside certbot for renewal of
+HTTPS certificates, and then applications like Spothole in a separate container. In this case, there are a couple of
 variations of the docker compose file above, and the nginx reverse proxy configuration covered [here](./nginx.md), that
 you will want to make.
 
@@ -60,8 +62,8 @@ networks:
 ```
 
 In your nginx site configuration, you'll want to refer to the Spothole container directly, and drop the block that
-allows nginx to access static files directly, as these will be inaccessible in another container. So you may end up
-with something like:
+allows nginx to access static files directly, as these will be inaccessible in another container. So you may end up with
+something like:
 
 ```nginx
 server {
@@ -146,13 +148,13 @@ server {
 ```
 
 If desired, you could even change the port on which Spothole runs from 8080 to a plain 80, in which case your
-`proxy_pass` statements could drop the `:8080` suffix. Since Spothole is in a container, it can serve HTTP on port 80
-if desired, because it doesn't conflict with the host system.
+`proxy_pass` statements could drop the `:8080` suffix. Since Spothole is in a container, it can serve HTTP on port 80 if
+desired, because it doesn't conflict with the host system.
 
 ### Restoring the static files bypass
 
-If you would still like to bypass Spothole's web server for the static files, and serve them with nginx, you can
-do. The easiest way is to run another nginx container to serve the files, so your Spothole `compose.yaml` becomes:
+If you would still like to bypass Spothole's web server for the static files, and serve them with nginx, you can do. The
+easiest way is to run another nginx container to serve the files, so your Spothole `compose.yaml` becomes:
 
 ```yaml
 services:
@@ -183,8 +185,8 @@ networks:
     external: true
 ```
 
-Then you can re-add the block that handles the `/static` path in your nginx reverse proxy config, but this time
-point it at the new container rather than at a filesystem path:
+Then you can re-add the block that handles the `/static` path in your nginx reverse proxy config, but this time point it
+at the new container rather than at a filesystem path:
 
 ```nginx
     # Load static assets from the spothole-static-nginx container

docs/modifying.md

@@ -16,9 +16,12 @@ To navigate your way around the source code, this list may help.
 * `/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/callsign` - Classes providing callsign 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
+* `/webserver` - Classes for running Spothole's own web server
+* `/telnetserver` - Classes for running Spothole's telnet server
 * `spothole.py` - Main application script
 
 *Templates*
@@ -48,8 +51,8 @@ To navigate your way around the source code, this list may help.
 ### 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 `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.)
+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.

docs/nginx.md

@@ -2,8 +2,8 @@
 
 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
+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.
 
@@ -89,19 +89,14 @@ server {
 ```
 
 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.)
+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:
 

docs/running.md

@@ -3,7 +3,7 @@
 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
+You will require Python version 3.10 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

providers/alert/alert_provider.py

@@ -29,7 +29,7 @@ class AlertProvider:
 
         # Sort the batch so that earliest ones go in first. This helps keep the ordering correct when alerts are fired
         # off to SSE listeners.
-        alerts = sorted(alerts, key=lambda a: (a.start_time if a and a.start_time else 0))
+        alerts = sorted(alerts, key=lambda a: a.start_time if a and a.start_time else 0)
         for alert in alerts:
             # Fill in any blanks and add to the list
             alert.infer_missing()

providers/alert/bota.py

@@ -3,9 +3,10 @@ from datetime import datetime, timedelta
 import pytz
 from bs4 import BeautifulSoup
 
-from providers.alert.http_alert_provider import HTTPAlertProvider
+from core.enums import AlertType
 from data.alert import Alert
 from data.sig_ref import SIGRef
+from providers.alert.http_alert_provider import HTTPAlertProvider
 
 
 class BOTA(HTTPAlertProvider):
@@ -23,17 +24,17 @@ class BOTA(HTTPAlertProvider):
         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'})
+        div = bs.body.find("div", attrs={"class": "view-activations-public"})
         if div:
-            table = div.find('table', attrs={'class': 'views-table'})
+            table = div.find("table", attrs={"class": "views-table"})
             if table:
-                tbody = table.find('tbody')
+                tbody = table.find("tbody")
                 if not tbody:
                     return new_alerts
-                for row in tbody.find_all('tr'):
-                    cells = row.find_all('td')
-                    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
+                for row in tbody.find_all("tr"):
+                    cells = row.find_all("td")
+                    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()
@@ -41,7 +42,7 @@ class BOTA(HTTPAlertProvider):
                     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_span = cells[2].find('span') if len(cells) > 2 else None
+                    date_span = cells[2].find("span") if len(cells) > 2 else None
                     if not date_span:
                         continue
                     date_text = date_span.get_text().strip()
@@ -52,11 +53,13 @@ class BOTA(HTTPAlertProvider):
                         date_time = date_time.replace(year=datetime.now(pytz.UTC).year + 1)
 
                     # Convert to our alert format
-                    alert = Alert(source=self.name,
-                                  dx_calls=[dx_call],
-                                  sig_refs=[SIGRef(id=ref_name, sig="BOTA")],
-                                  start_time=date_time.timestamp(),
-                                  is_dxpedition=False)
+                    alert = Alert(
+                        source=self.name,
+                        dx_calls=[dx_call],
+                        sig_refs=[SIGRef(id=ref_name, sig="BOTA")],
+                        start_time=date_time.timestamp(),
+                        alert_type=AlertType.XOTA,
+                    )
 
                     new_alerts.append(alert)
         return new_alerts

providers/alert/hamsat.py

@@ -0,0 +1,56 @@
+from datetime import datetime
+
+import pytz
+
+from core.enums import AlertType
+from data.alert import Alert
+from data.sig_ref import SIGRef
+from providers.alert.http_alert_provider import HTTPAlertProvider
+
+
+class Hamsat(HTTPAlertProvider):
+    """Alert provider for Hamsat (hams.at)"""
+
+    POLL_INTERVAL_SEC = 1800
+    ALERTS_URL = "https://hams.at/api/alerts"
+
+    def __init__(self, provider_config):
+        super().__init__("Hamsat", 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()["data"]:
+            # Convert to our alert format
+            freqs_modes = source_alert.get("mode", "")
+            if "mhz" in source_alert:
+                if "mhz_direction" in source_alert:
+                    freqs_modes = f"{source_alert['mhz']!s} {source_alert['mhz_direction']}, {freqs_modes}"
+                else:
+                    freqs_modes = f"{source_alert['mhz']!s}, {freqs_modes}"
+
+            alert = Alert(
+                source=self.name,
+                source_id=source_alert["id"],
+                dx_calls=[source_alert["callsign"].upper()],
+                freqs_modes=freqs_modes,
+                comment=source_alert["comment"],
+                # Fudge a SIG ref to provide the remaining bits of data we need: the satellite and the operator's grid
+                sig_refs=[
+                    SIGRef(
+                        sig="AMSAT",
+                        id=f"{source_alert['satellite']['name']} from {source_alert['grids'][0]}",
+                    )
+                ],
+                start_time=datetime.strptime(source_alert["aos_at"], "%Y-%m-%dT%H:%M:%SZ")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp(),
+                end_time=datetime.strptime(source_alert["los_at"], "%Y-%m-%dT%H:%M:%SZ")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp(),
+                alert_type=AlertType.SATELLITE,
+            )
+
+            # Add to our list
+            new_alerts.append(alert)
+        return new_alerts

providers/alert/http_alert_provider.py

@@ -1,13 +1,15 @@
 import logging
 from datetime import datetime
-from threading import Thread, Event
+from threading import Event, Thread
 
 import pytz
 import requests
-from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout, JSONDecodeError, ReadTimeout
 
-from providers.alert.alert_provider import AlertProvider
 from core.constants import HTTP_HEADERS
+from providers.alert.alert_provider import AlertProvider
+
+logger = logging.getLogger(__name__)
 
 
 class HTTPAlertProvider(AlertProvider):
@@ -24,12 +26,16 @@ class HTTPAlertProvider(AlertProvider):
     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 + " alert API every " + str(self._poll_interval) + " seconds.")
-        self._thread = Thread(target=self._run, name=f"HTTPAlertProvider-{self.name}")
+        logger.info(f"Set up query of {self.name} alert API every {self._poll_interval!s} seconds.")
+        self._thread = Thread(target=self._run, name=f"HTTPAlertProvider-{self.name}", daemon=True)
         self._thread.start()
 
     def stop(self):
         self._stop_event.set()
+        if self._thread:
+            self._thread.join(timeout=35)
+            if self._thread.is_alive():
+                logger.warning(f"{self.name} alert worker thread did not exit on time and will be killed.")
 
     def _run(self):
         while True:
@@ -40,7 +46,7 @@ class HTTPAlertProvider(AlertProvider):
     def _poll(self):
         try:
             # Request data from API
-            logging.debug("Polling " + self.name + " alert API...")
+            logger.debug(f"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:
@@ -52,18 +58,23 @@ class HTTPAlertProvider(AlertProvider):
 
                 self.status = "OK"
                 self.last_update_time = datetime.now(pytz.UTC)
-                logging.debug("Received data from " + self.name + " alert API.")
+                logger.debug(f"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.")
+                logger.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.")
+            self.status = "Error"
+            logger.warning(f"Connection error when accessing {self.name} alerts API.")
         except (ConnectTimeout, ReadTimeout):
-            logging.warning(f"Timeout when accessing {self.name} alerts API.")
+            self.status = "Error"
+            logger.warning(f"Timeout when accessing {self.name} alerts API.")
+        except JSONDecodeError:
+            self.status = "Error"
+            logger.warning(f"Invalid or empty JSON response from {self.name} alert API.")
         except Exception:
             self.status = "Error"
-            logging.exception("Exception in HTTP JSON Alert Provider (" + self.name + ")")
+            logger.exception(f"Exception in HTTP JSON Alert Provider ({self.name})")
             # Brief pause on error before the next poll, but still respond promptly to stop()
             self._stop_event.wait(timeout=1)
 

providers/alert/ical_alert_provider.py

@@ -0,0 +1,47 @@
+from datetime import datetime, time
+from typing import cast
+
+import pytz
+from icalendar import Calendar, Event
+
+from data.alert import Alert
+from providers.alert.http_alert_provider import HTTPAlertProvider
+
+
+class ICALAlertProvider(HTTPAlertProvider):
+    """Generic alert provider for iCal calendars. Defines an abstract method event_to_alert(event) that subclasses must
+    implement, and use it to convert an iCal event to an Alert object based on whatever format their iCal events use."""
+
+    def __init__(self, name, provider_config, url, poll_interval):
+        super().__init__(name, provider_config, url, poll_interval)
+
+    def _http_response_to_alerts(self, http_response):
+        new_alerts = []
+        cal = Calendar.from_ical(http_response.content)
+
+        # Iterate through events, passing each one in turn to the subclass' event_to_alert method to turn it into
+        # a Spothole alert object
+        for component in cal.walk():
+            if component.name != "VEVENT":
+                continue
+
+            event = cast(Event, component)
+            alert = self.event_to_alert(event)
+            new_alerts.append(alert)
+        return new_alerts
+
+    def event_to_alert(self, event: Event) -> Alert:
+        """Convert an ICal event to an Alert object. Subclasses must implement this method."""
+
+    @staticmethod
+    def _to_utc_timestamp(value):
+        """Convert a date or datetime value from an iCal field into a UTC UNIX timestamp."""
+
+        # Datetime object so we can treat it as-is, check if it has a non-UTC tz and convert it if necessary
+        if isinstance(value, datetime):
+            if value.tzinfo is None:
+                value = pytz.UTC.localize(value)
+            return value.astimezone(pytz.UTC).timestamp()
+
+        # Date object so this is an all day event
+        return pytz.UTC.localize(datetime.combine(value, time.min)).timestamp()

providers/alert/ng3k.py

@@ -6,19 +6,20 @@ import pytz
 from rss_parser import Parser
 from rss_parser.models.rss import RSS
 
-from providers.alert.http_alert_provider import HTTPAlertProvider
+from core.enums import AlertType
 from data.alert import Alert
+from providers.alert.http_alert_provider import HTTPAlertProvider
 
 
 class NG3K(HTTPAlertProvider):
-    """Alert provider NG3K DXpedition list"""
+    """Alert provider for NG3K DXpedition list"""
 
-    POLL_INTERVAL_SEC = 1800
+    POLL_INTERVAL_DAYS = 1
     ALERTS_URL = "https://www.ng3k.com/adxo.xml"
     AS_CALL_PATTERN = re.compile("as ([a-z0-9/]+)", re.IGNORECASE)
 
     def __init__(self, provider_config):
-        super().__init__("NG3K", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_SEC)
+        super().__init__("NG3K", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_DAYS * 24 * 60 * 60)
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []
@@ -49,11 +50,16 @@ class NG3K(HTTPAlertProvider):
                 end_day = end_string.split(", ")[0].strip()
                 end_mon = start_mon
 
-            start_timestamp = datetime.strptime(start_year + " " + start_mon + " " + start_day, "%Y %b %d").replace(
-                tzinfo=pytz.UTC).timestamp()
-            end_timestamp = datetime.strptime(end_year + " " + end_mon + " " + end_day + " 23:59",
-                                              "%Y %b %d %H:%M").replace(
-                tzinfo=pytz.UTC).timestamp()
+            start_timestamp = (
+                datetime.strptime(f"{start_year} {start_mon} {start_day}", "%Y %b %d")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp()
+            )
+            end_timestamp = (
+                datetime.strptime(f"{end_year} {end_mon} {end_day} 23:59", "%Y %b %d %H:%M")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp()
+            )
 
             # Sometimes the DX callsign is "real", sometimes you just get a prefix with the real working callsigns being
             # provided in the "by" field. e.g. call="JW", by="By LA7XK as JW7XK, LA6VM as JW6VM, LA9DL as JW9DL". So
@@ -75,14 +81,16 @@ class NG3K(HTTPAlertProvider):
             comment = extra_parts[3] if len(extra_parts) > 3 else ""
 
             # Convert to our alert format
-            alert = Alert(source=self.name,
-                          dx_calls=dx_calls,
-                          dx_country=dx_country,
-                          freqs_modes=bands + (("; " + modes) if modes != "" else ""),
-                          comment=by + "; " + comment + "; " + qsl_info,
-                          start_time=start_timestamp,
-                          end_time=end_timestamp,
-                          is_dxpedition=True)
+            alert = Alert(
+                source=self.name,
+                dx_calls=dx_calls,
+                dx_country=dx_country,
+                freqs_modes=bands + (f"; {modes}" if modes != "" else ""),
+                comment=f"{by}; {comment}; {qsl_info}",
+                start_time=start_timestamp,
+                end_time=end_timestamp,
+                alert_type=AlertType.DXPEDITION,
+            )
 
             # Add to our list.
             new_alerts.append(alert)

providers/alert/parksnpeaks.py

@@ -3,9 +3,12 @@ from datetime import datetime
 
 import pytz
 
-from providers.alert.http_alert_provider import HTTPAlertProvider
+from core.enums import AlertType
 from data.alert import Alert
 from data.sig_ref import SIGRef
+from providers.alert.http_alert_provider import HTTPAlertProvider
+
+logger = logging.getLogger(__name__)
 
 
 class ParksNPeaks(HTTPAlertProvider):
@@ -30,8 +33,9 @@ class ParksNPeaks(HTTPAlertProvider):
             else:
                 sig_ref = source_alert["WWFFID"]
                 sig_ref_name = source_alert["Location"]
-            start_time = datetime.strptime(source_alert["alTime"], "%Y-%m-%d %H:%M:%S").replace(
-                tzinfo=pytz.UTC).timestamp()
+            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,
@@ -40,18 +44,31 @@ class ParksNPeaks(HTTPAlertProvider):
                 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=sigrefs,
-                          start_time=start_time,
-                          is_dxpedition=False)
+            alert = Alert(
+                source=self.name,
+                source_id=source_alert["alID"],
+                dx_calls=[source_alert["CallSign"].upper()],
+                freqs_modes=f"{source_alert['Freq']} {source_alert['MODE']}",
+                comment=source_alert["Comments"],
+                sig_refs=sigrefs,
+                start_time=start_time,
+                alert_type=AlertType.XOTA,
+            )
 
             # 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", "SANPCPA", "LLOTA", "QRP"]:
-                logging.warning("PNP alert found with sig " + sig + ", developer needs to add support for this!")
+            if sig and sig not in [
+                "POTA",
+                "SOTA",
+                "WWFF",
+                "HEMA",
+                "SIOTA",
+                "ZLOTA",
+                "KRMNPA",
+                "SANPCPA",
+                "LLOTA",
+                "QRP",
+            ]:
+                logger.warning(f"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
             # the alert list. Note that while ZLOTA has its own spots API, it doesn't have its own alerts API. So that

providers/alert/pota.py

@@ -2,9 +2,10 @@ from datetime import datetime
 
 import pytz
 
-from providers.alert.http_alert_provider import HTTPAlertProvider
+from core.enums import AlertType
 from data.alert import Alert
 from data.sig_ref import SIGRef
+from providers.alert.http_alert_provider import HTTPAlertProvider
 
 
 class POTA(HTTPAlertProvider):
@@ -21,18 +22,31 @@ class POTA(HTTPAlertProvider):
         # Iterate through source data
         for source_alert in http_response.json():
             # Convert to our alert format
-            alert = Alert(source=self.name,
-                          source_id=source_alert["scheduledActivitiesId"],
-                          dx_calls=[source_alert["activator"].upper()],
-                          freqs_modes=source_alert["frequencies"],
-                          comment=source_alert["comments"],
-                          sig_refs=[SIGRef(id=source_alert["reference"], sig="POTA", name=source_alert["name"],
-                                           url="https://pota.app/#/park/" + source_alert["reference"])],
-                          start_time=datetime.strptime(source_alert["startDate"] + source_alert["startTime"],
-                                                       "%Y-%m-%d%H:%M").replace(tzinfo=pytz.UTC).timestamp(),
-                          end_time=datetime.strptime(source_alert["endDate"] + source_alert["endTime"],
-                                                     "%Y-%m-%d%H:%M").replace(tzinfo=pytz.UTC).timestamp(),
-                          is_dxpedition=False)
+            alert = Alert(
+                source=self.name,
+                source_id=source_alert["scheduledActivitiesId"],
+                dx_calls=[source_alert["activator"].upper()],
+                freqs_modes=source_alert["frequencies"],
+                comment=source_alert["comments"],
+                sig_refs=[
+                    SIGRef(
+                        id=source_alert["reference"],
+                        sig="POTA",
+                        name=source_alert["name"],
+                        url=f"https://pota.app/#/park/{source_alert['reference']}",
+                    )
+                ],
+                start_time=datetime.strptime(
+                    source_alert["startDate"] + source_alert["startTime"],
+                    "%Y-%m-%d%H:%M",
+                )
+                .replace(tzinfo=pytz.UTC)
+                .timestamp(),
+                end_time=datetime.strptime(source_alert["endDate"] + source_alert["endTime"], "%Y-%m-%d%H:%M")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp(),
+                alert_type=AlertType.XOTA,
+            )
 
             # 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

providers/alert/rsgb_ical_alert_provider.py

@@ -0,0 +1,75 @@
+import re
+
+from icalendar import Event
+
+from core.enums import AlertType, Continent
+from data.alert import Alert
+from providers.alert.ical_alert_provider import ICALAlertProvider
+
+
+class RSGBICALAlertProvider(ICALAlertProvider):
+    """Generic alert provider for RSGB contest iCal calendars. Builds on the generic iCal alert provider by adding
+    handling specific to how RSGB's iCal events are formatted. This is still effectively an abstract class itself;
+    RSGB has two contest calendars (HF & VHF) that each subclass this."""
+
+    def __init__(self, name, provider_config, url, poll_interval):
+        super().__init__(name, provider_config, url, poll_interval)
+
+    FREQ_PATTERN = re.compile(r"([\d.]+(?:MHz|GHz))|SHF")
+
+    def event_to_alert(self, event: Event) -> Alert:
+        """Convert an iCal event in RSGB's format to an Alert object."""
+
+        summary = str(event.get("summary", "")).strip()
+
+        # Ensure summaries start with "RSGB" to avoid any confusion
+        if not summary.startswith("RSGB "):
+            summary = "RSGB " + summary
+
+        # Extract freq from summary. HF contests don't give frequencies, all VHF ones do
+        match = self.FREQ_PATTERN.search(summary)
+        if match:
+            freqs_modes = match.group()
+        else:
+            freqs_modes = "HF bands"
+        freqs_modes = freqs_modes + ", "
+
+        # Extract mode from summary
+        if "FMAC" in summary:
+            freqs_modes = freqs_modes + "FM"
+        elif "CW" in summary:
+            freqs_modes = freqs_modes + "CW"
+        elif "SSB" in summary:
+            freqs_modes = freqs_modes + "SSB"
+        elif "FT8" in summary:
+            freqs_modes = freqs_modes + "FT8"
+        elif "FT4" in summary:
+            freqs_modes = freqs_modes + "FT4"
+        elif "DATA" in summary:
+            freqs_modes = freqs_modes + "Data modes"
+        else:
+            freqs_modes = freqs_modes + "All modes"
+
+        dtstart = event.get("dtstart")
+        dtend = event.get("dtend")
+        start_timestamp = self._to_utc_timestamp(dtstart.dt)
+        end_timestamp = self._to_utc_timestamp(dtend.dt) - 1 if dtend is not None else start_timestamp
+
+        # Convert to our alert format
+        alert = Alert(
+            source=self.name,
+            dx_calls=[],
+            dx_country="United Kingdom",
+            dx_dxcc_id=235,
+            dx_continent=Continent.EU,
+            dx_cq_zone=14,
+            dx_itu_zone=27,
+            dx_flag="🇬🇧",
+            freqs_modes=freqs_modes,
+            comment=summary,
+            start_time=start_timestamp,
+            end_time=end_timestamp,
+            alert_type=AlertType.CONTEST,
+        )
+
+        return alert

providers/alert/rsgbhfcontests.py

@@ -0,0 +1,11 @@
+from providers.alert.rsgb_ical_alert_provider import RSGBICALAlertProvider
+
+
+class RSGBHFContests(RSGBICALAlertProvider):
+    """Alert provider for RSGB HF Contest calendar"""
+
+    POLL_INTERVAL_DAYS = 30
+    ALERTS_URL = "https://calendar.google.com/calendar/ical/a5ff31ebb1b4834dc7fff4c5415ae8251c6a9aa11f98c6af6e472b6c552b1915%40group.calendar.google.com/public/basic.ics"
+
+    def __init__(self, provider_config):
+        super().__init__("RSGB HF Contests", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_DAYS * 24 * 60 * 60)

providers/alert/rsgbvhfcontests.py

@@ -0,0 +1,11 @@
+from providers.alert.rsgb_ical_alert_provider import RSGBICALAlertProvider
+
+
+class RSGBVHFContests(RSGBICALAlertProvider):
+    """Alert provider for RSGB VHF Contest calendar"""
+
+    POLL_INTERVAL_DAYS = 30
+    ALERTS_URL = "https://calendar.google.com/calendar/ical/40f3552bff39a016f1cdca205864177070dcad68d55be17eb061cb021f39f96c%40group.calendar.google.com/public/basic.ics"
+
+    def __init__(self, provider_config):
+        super().__init__("RSGB VHF Contests", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_DAYS * 24 * 60 * 60)

providers/alert/sota.py

@@ -2,9 +2,10 @@ from datetime import datetime
 
 import pytz
 
-from providers.alert.http_alert_provider import HTTPAlertProvider
+from core.enums import AlertType
 from data.alert import Alert
 from data.sig_ref import SIGRef
+from providers.alert.http_alert_provider import HTTPAlertProvider
 
 
 class SOTA(HTTPAlertProvider):
@@ -26,18 +27,26 @@ class SOTA(HTTPAlertProvider):
             summit_points = None
             if len(details) > 2:
                 summit_points = int(details[-1].split(" ")[0])
-            alert = Alert(source=self.name,
-                          source_id=source_alert["id"],
-                          dx_calls=[source_alert["activatingCallsign"].upper()],
-                          dx_names=[source_alert["activatorName"].upper()],
-                          freqs_modes=source_alert["frequency"],
-                          comment=source_alert["comments"],
-                          sig_refs=[
-                              SIGRef(id=source_alert["associationCode"] + "/" + source_alert["summitCode"], sig="SOTA",
-                                     name=summit_name, activation_score=summit_points)],
-                          start_time=datetime.strptime(source_alert["dateActivated"],
-                                                       "%Y-%m-%dT%H:%M:%SZ").replace(tzinfo=pytz.UTC).timestamp(),
-                          is_dxpedition=False)
+            alert = Alert(
+                source=self.name,
+                source_id=source_alert["id"],
+                dx_calls=[source_alert["activatingCallsign"].upper()],
+                dx_names=[source_alert["activatorName"].upper()],
+                freqs_modes=source_alert["frequency"],
+                comment=source_alert["comments"],
+                sig_refs=[
+                    SIGRef(
+                        id=f"{source_alert['associationCode']}/{source_alert['summitCode']}",
+                        sig="SOTA",
+                        name=summit_name,
+                        activation_score=summit_points,
+                    )
+                ],
+                start_time=datetime.strptime(source_alert["dateActivated"], "%Y-%m-%dT%H:%M:%SZ")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp(),
+                alert_type=AlertType.XOTA,
+            )
 
             # Add to our list
             new_alerts.append(alert)

providers/alert/wa7bnm.py

@@ -0,0 +1,41 @@
+from icalendar import Event
+
+from core.enums import AlertType
+from data.alert import Alert
+from providers.alert.ical_alert_provider import ICALAlertProvider
+
+
+class WA7BNM(ICALAlertProvider):
+    """Alert provider for the WA7BNM contest calendar."""
+
+    POLL_INTERVAL_DAYS = 1
+    ALERTS_URL = "https://contestcalendar.com/weeklycontcustom.php"
+
+    def __init__(self, provider_config):
+        super().__init__(
+            "WA7BNM Contest Calendar", provider_config, self.ALERTS_URL, self.POLL_INTERVAL_DAYS * 24 * 60 * 60
+        )
+
+    def event_to_alert(self, event: Event) -> Alert:
+        """Convert an iCal event in WA7BNM's format to an Alert object."""
+
+        summary = str(event.get("summary", ""))
+        url = str(event.get("url", ""))
+
+        dtstart = event.get("dtstart")
+        dtend = event.get("dtend")
+        start_timestamp = self._to_utc_timestamp(dtstart.dt)
+        end_timestamp = self._to_utc_timestamp(dtend.dt) - 1 if dtend is not None else start_timestamp
+
+        # Convert to our alert format
+        alert = Alert(
+            source=self.name,
+            dx_calls=[],
+            comment=summary,
+            url=url,
+            start_time=start_timestamp,
+            end_time=end_timestamp,
+            alert_type=AlertType.CONTEST,
+        )
+
+        return alert

providers/alert/wota.py

@@ -1,13 +1,17 @@
+import logging
 from datetime import datetime
 from typing import cast
+from xml.parsers.expat import ExpatError
 
 import pytz
 from rss_parser import Parser as RSSParser
 from rss_parser.models.rss import RSS
 
-from providers.alert.http_alert_provider import HTTPAlertProvider
 from data.alert import Alert
 from data.sig_ref import SIGRef
+from providers.alert.http_alert_provider import HTTPAlertProvider
+
+logger = logging.getLogger(__name__)
 
 
 class WOTA(HTTPAlertProvider):
@@ -22,43 +26,62 @@ class WOTA(HTTPAlertProvider):
 
     def _http_response_to_alerts(self, http_response):
         new_alerts = []
-        rss = cast(RSS, RSSParser.parse(http_response.content.decode("utf-8-sig")))
-        # Iterate through source data
-        for source_alert in rss.channel.items:
 
-            # Reject GUID missing or zero
-            if not source_alert.guid or not source_alert.guid.content or source_alert.guid.content == "http://www.wota.org.uk/alerts/0":
-                continue
+        try:
+            rss = cast(RSS, RSSParser.parse(http_response.content.decode("utf-8-sig")))
+            # Iterate through source data
+            for source_alert in rss.channel.items:
+                # Reject GUID missing or zero
+                if (
+                    not source_alert.guid
+                    or not source_alert.guid.content
+                    or source_alert.guid.content == "http://www.wota.org.uk/alerts/0"
+                ):
+                    continue
 
-            # Pick apart the title
-            title_split = source_alert.title.split(" on ")
-            dx_call = title_split[0]
-            ref = None
-            ref_name = None
-            if len(title_split) > 1:
-                ref_split = title_split[1].split(" - ")
-                ref = str(ref_split[0])
-                if len(ref_split) > 1:
-                    ref_name = str(ref_split[1])
+                # Pick apart the title
+                dx_call = None
+                ref = None
+                ref_name = None
+                try:
+                    title_split = source_alert.title.split(" on ")
+                    dx_call = title_split[0]
+                    if len(title_split) > 1:
+                        ref_split = title_split[1].split(" - ")
+                        ref = str(ref_split[0])
+                        if len(ref_split) > 1:
+                            ref_name = str(ref_split[1])
+                except Exception:
+                    logger.warning(f"Could not parse WOTA alert title: {source_alert.description}")
 
-            # Pick apart the description
-            desc_split = source_alert.description.split(". ")
-            freqs_modes = desc_split[0].replace("Frequencies/modes:", "").strip()
-            comment = None
-            if len(desc_split) > 1:
-                comment = desc_split[1].strip()
+                # Pick apart the description
+                comment = None
+                freqs_modes = None
+                try:
+                    desc_split = source_alert.description.split(". ")
+                    freqs_modes = desc_split[0].replace("Frequencies/modes:", "").strip()
+                    if len(desc_split) > 1:
+                        comment = desc_split[1].strip()
+                except Exception:
+                    logger.warning(f"Could not parse WOTA alert description: {source_alert.description}")
 
-            time = datetime.strptime(source_alert.pub_date.content, self.RSS_DATE_TIME_FORMAT).astimezone(pytz.UTC)
+                time = datetime.strptime(source_alert.pub_date.content, self.RSS_DATE_TIME_FORMAT).astimezone(pytz.UTC)
 
-            # Convert to our alert format
-            alert = Alert(source=self.name,
-                          source_id=source_alert.guid.content,
-                          dx_calls=[dx_call],
-                          freqs_modes=freqs_modes,
-                          comment=comment,
-                          sig_refs=[SIGRef(id=ref, sig="WOTA", name=ref_name)] if ref else [],
-                          start_time=time.timestamp())
+                # Convert to our alert format
+                alert = Alert(
+                    source=self.name,
+                    source_id=source_alert.guid.content,
+                    dx_calls=[dx_call],
+                    freqs_modes=freqs_modes,
+                    comment=comment,
+                    sig_refs=[SIGRef(id=ref, sig="WOTA", name=ref_name)] if ref else [],
+                    start_time=time.timestamp(),
+                )
+
+                # Add to our list.
+                new_alerts.append(alert)
+
+        except ExpatError:
+            logger.warning("WOTA alert RSS feed was fetched but was invalid")
 
-            # Add to our list.
-            new_alerts.append(alert)
         return new_alerts

providers/alert/wwff.py

@@ -2,9 +2,10 @@ from datetime import datetime
 
 import pytz
 
-from providers.alert.http_alert_provider import HTTPAlertProvider
+from core.enums import AlertType
 from data.alert import Alert
 from data.sig_ref import SIGRef
+from providers.alert.http_alert_provider import HTTPAlertProvider
 
 
 class WWFF(HTTPAlertProvider):
@@ -21,17 +22,21 @@ class WWFF(HTTPAlertProvider):
         # Iterate through source data
         for source_alert in http_response.json():
             # Convert to our alert format
-            alert = Alert(source=self.name,
-                          source_id=source_alert["id"],
-                          dx_calls=[source_alert["activator_call"].upper()],
-                          freqs_modes=source_alert["band"] + " " + source_alert["mode"],
-                          comment=source_alert["remarks"],
-                          sig_refs=[SIGRef(id=source_alert["reference"], sig="WWFF")],
-                          start_time=datetime.strptime(source_alert["utc_start"],
-                                                       "%Y-%m-%d %H:%M:%S").replace(tzinfo=pytz.UTC).timestamp(),
-                          end_time=datetime.strptime(source_alert["utc_end"],
-                                                     "%Y-%m-%d %H:%M:%S").replace(tzinfo=pytz.UTC).timestamp(),
-                          is_dxpedition=False)
+            alert = Alert(
+                source=self.name,
+                source_id=source_alert["id"],
+                dx_calls=[source_alert["activator_call"].upper()],
+                freqs_modes=f"{source_alert['band']} {source_alert['mode']}",
+                comment=source_alert["remarks"],
+                sig_refs=[SIGRef(id=source_alert["reference"], sig="WWFF")],
+                start_time=datetime.strptime(source_alert["utc_start"], "%Y-%m-%d %H:%M:%S")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp(),
+                end_time=datetime.strptime(source_alert["utc_end"], "%Y-%m-%d %H:%M:%S")
+                .replace(tzinfo=pytz.UTC)
+                .timestamp(),
+                alert_type=AlertType.XOTA,
+            )
 
             # Add to our list
             new_alerts.append(alert)

providers/callsigndata/api_query_callsign_data_provider.py

@@ -1,13 +1,3 @@
-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
 
 
@@ -15,7 +5,7 @@ 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."""
+        """Set up the provider."""
         super().__init__(name, provider_config, storage)
 
         if self.enabled:

providers/callsigndata/callsign_data_provider.py

@@ -1,4 +1,3 @@
-import logging
 from datetime import datetime
 
 import pytz
@@ -43,7 +42,13 @@ class CallsignDataProvider:
 
         if self.enabled:
             if callsign in self._storage:
-                return self._storage[callsign]
+                try:
+                    return self._storage[callsign]
+                except (ValueError, KeyError):
+                    # Catch exceptions due to e.g. old versions of objects in the cache that are no longer compatible,
+                    # and remove them from the cache.
+                    del self._storage[callsign]
+                    return None
             else:
                 c = self._perform_new_lookup(callsign, lookup_credentials)
                 if c:
@@ -52,7 +57,6 @@ class CallsignDataProvider:
         else:
             return None
 
-
     def _perform_new_lookup(self, callsign, lookup_credentials):
         """Makes a new request to the data source for callsign data."""
 

providers/callsigndata/clublogapi.py

@@ -2,12 +2,16 @@ import logging
 from datetime import datetime
 
 import pytz
-from pyhamtools import LookupLib, Callinfo
+from pyhamtools import Callinfo, LookupLib
 
 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
+from providers.callsigndata.api_query_callsign_data_provider import (
+    APIQueryCallsignDataProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 
 class ClublogAPI(APIQueryCallsignDataProvider):
@@ -23,12 +27,12 @@ class ClublogAPI(APIQueryCallsignDataProvider):
             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.")
+            logger.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)
 
@@ -41,8 +45,8 @@ class ClublogAPI(APIQueryCallsignDataProvider):
             else:
                 return None
 
-        except Exception as e:
+        except Exception:
             self.status = "Error"
-            logging.error("Exception when looking up data from Clublog API", e, exc_info=True)
+            logger.exception("Exception when looking up data from Clublog API")
 
         return callsign_data

providers/callsigndata/clublogxml.py

@@ -1,14 +1,16 @@
 import gzip
 import logging
-from datetime import datetime
 
-import pytz
-from pyhamtools import LookupLib, Callinfo
+from pyhamtools import Callinfo, LookupLib
 
 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
+from providers.callsigndata.file_download_callsign_data_provider import (
+    FileDownloadCallsignDataProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 
 class ClublogXML(FileDownloadCallsignDataProvider):
@@ -25,11 +27,18 @@ class ClublogXML(FileDownloadCallsignDataProvider):
         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.")
+            logger.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)
+        super().__init__(
+            "Clublog XML",
+            provider_config,
+            f"{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:
@@ -47,8 +56,8 @@ class ClublogXML(FileDownloadCallsignDataProvider):
             self._callinfo = Callinfo(lookuplib)
             return True
 
-        except Exception as e:
-            logging.error("Exception when loading Clublog XML.", e, exc_info=True)
+        except Exception:
+            logger.exception("Exception when loading Clublog XML.")
             return False
 
     def _perform_new_lookup(self, callsign, lookup_credentials):
@@ -62,8 +71,8 @@ class ClublogXML(FileDownloadCallsignDataProvider):
             else:
                 return None
 
-        except Exception as e:
+        except Exception:
             self.status = "Error"
-            logging.error("Exception when looking up data from Clublog XML data", e, exc_info=True)
+            logger.exception("Exception when looking up data from Clublog XML data")
 
         return callsign_data

providers/callsigndata/countryfiles.py

@@ -1,11 +1,15 @@
 import logging
 
-from pyhamtools import LookupLib, Callinfo
+from pyhamtools import Callinfo, LookupLib
 
 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
+from providers.callsigndata.file_download_callsign_data_provider import (
+    FileDownloadCallsignDataProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 
 class CountryFiles(FileDownloadCallsignDataProvider):
@@ -17,8 +21,14 @@ class CountryFiles(FileDownloadCallsignDataProvider):
     _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)
+        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:
@@ -26,8 +36,8 @@ class CountryFiles(FileDownloadCallsignDataProvider):
             self._callinfo = Callinfo(lookuplib)
             return True
 
-        except Exception as e:
-            logging.error("Exception when loading Country Files cty.plist.", e, exc_info=True)
+        except Exception:
+            logger.exception("Exception when loading Country Files cty.plist.")
             return False
 
     def _perform_new_lookup(self, callsign, lookup_credentials):
@@ -41,8 +51,8 @@ class CountryFiles(FileDownloadCallsignDataProvider):
             else:
                 return None
 
-        except Exception as e:
+        except Exception:
             self.status = "Error"
-            logging.error("Exception when looking up data from Country file", e, exc_info=True)
+            logger.exception("Exception when looking up data from Country file")
 
         return callsign_data

providers/callsigndata/file_download_callsign_data_provider.py

@@ -1,28 +1,29 @@
 import logging
 from datetime import datetime
-from threading import Thread, Event
+from threading import Event, Thread
 
 import pytz
-from requests import ReadTimeout
-from requests.exceptions import ConnectionError, ConnectTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout, ReadTimeout
 
 from core.constants import HTTP_HEADERS
 from core.url_data_cache import URLDataCache
 from providers.callsigndata.callsign_data_provider import CallsignDataProvider
 
+logger = logging.getLogger(__name__)
+
 
 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*."""
+        """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)
+        self._url_data_cache = URLDataCache(f"callsigndata_{name}")
 
         if self.enabled:
             self.status = "Ready"
@@ -30,13 +31,16 @@ class FileDownloadCallsignDataProvider(CallsignDataProvider):
     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}")
+        logger.info(f"Set up query of {self.name} callsign reference data every {self._poll_interval!s} days.")
+        self._thread = Thread(target=self._run, name=f"FileDownloadCallsignDataProvider-{self.name}", daemon=True)
         self._thread.start()
 
     def stop(self):
         self._stop_event.set()
+        if self._thread:
+            self._thread.join(timeout=35)
+            if self._thread.is_alive():
+                logger.warning(f"{self.name} callsign data worker thread did not exit on time and will be killed.")
 
     def _run(self):
         while True:
@@ -48,7 +52,7 @@ class FileDownloadCallsignDataProvider(CallsignDataProvider):
         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...")
+            logger.debug(f"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:
@@ -61,24 +65,26 @@ class FileDownloadCallsignDataProvider(CallsignDataProvider):
                     if ok:
                         self.status = "OK"
                         self.last_update_time = datetime.now(pytz.UTC)
-                        logging.info("Updated callsign reference data from " + self.name)
+                        logger.info(f"Updated callsign reference data from {self.name}")
                     else:
                         self.status = "Error"
-                        logging.warning(f"Error updating callsign reference data from {self.name}.")
+                        logger.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}.")
+                logger.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}.")
+            logger.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}.")
+            logger.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 + ")")
+            logger.exception(f"Exception in callsign reference data provider ({self.name})")
             self._stop_event.wait(timeout=1)
 
     def _handle_file(self, path):

providers/callsigndata/hamqth.py

@@ -1,19 +1,24 @@
 import logging
 import urllib.parse
-from datetime import timedelta, datetime
+from datetime import datetime, timedelta
 
 import pytz
 import xmltodict
 from pyhamtools import callinfo
-from requests import ConnectTimeout, ReadTimeout
+from requests.exceptions import ConnectionError, 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.data_store import CACHE_DIR, DATA_STORE
+from core.enums import Continent
 from core.url_data_cache import URLDataCache
-from data.callsign import Callsign
-from providers.callsigndata.api_query_callsign_data_provider import APIQueryCallsignDataProvider
+from data.callsign import Callsign, LocationSourceForCallsign
+from providers.callsigndata.api_query_callsign_data_provider import (
+    APIQueryCallsignDataProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 
 class HamQTH(APIQueryCallsignDataProvider):
@@ -22,18 +27,19 @@ class HamQTH(APIQueryCallsignDataProvider):
     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._PRG = f"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))
+        self._CREDENTIALS_CACHE = CachedSession(f"{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):
+        if not lookup_credentials or not (
+            (lookup_credentials.hamqth_username and lookup_credentials.hamqth_password)
+            or lookup_credentials.hamqth_session_id
+        ):
             return None
 
         try:
@@ -44,18 +50,18 @@ class HamQTH(APIQueryCallsignDataProvider):
             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
+                        f"{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.")
+                        logger.debug("HamQTH login details incorrect, failed to look up with HamQTH.")
                         return None
                 except Exception:
-                    logging.error("Exception when getting HamQTH session key")
+                    logger.error("Exception when getting HamQTH session key")
                     return None
 
             if not session_id:
@@ -68,14 +74,16 @@ class HamQTH(APIQueryCallsignDataProvider):
                 if home_call != callsign:
                     calls_to_try.append(home_call)
             except ValueError:
-                logging.debug("Could not look up home call for callsign %s", callsign)
+                logger.debug(f"Could not look up home call for callsign {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)
+                        f"{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"]
@@ -85,27 +93,26 @@ class HamQTH(APIQueryCallsignDataProvider):
                         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)
+                        logger.warning(f"HTTP {response.status_code} looking up callsign {lookup_call} using HamQTH")
 
                 except (KeyError, ValueError):
                     continue
                 except ConnectionError:
-                    logging.warning(f"Connection error when looking up callsign %s using HamQTH", lookup_call)
+                    logger.warning(f"Connection error when looking up callsign {lookup_call} using HamQTH")
                     continue
                 except (ConnectTimeout, ReadTimeout):
-                    logging.warning(f"Timeout when looking up callsign %s using HamQTH", lookup_call)
+                    logger.warning(f"Timeout when looking up callsign {lookup_call} using HamQTH")
                     continue
                 except Exception:
-                    logging.error("Exception when looking up callsign %s using HamQTH", lookup_call, exc_info=True)
+                    logger.exception(f"Exception when looking up callsign {lookup_call} using HamQTH")
                     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:
+        except Exception:
             self.status = "Error"
-            logging.error("Exception when looking up data from HamQTH", e, exc_info=True)
+            logger.exception("Exception when looking up data from HamQTH")
             # Return None, this won't be cached so we will be asked to query data again for this call next time.
             return None
 
@@ -116,27 +123,32 @@ class HamQTH(APIQueryCallsignDataProvider):
         # 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:
+        if (
+            data.get("latitude") is not None
+            and data.get("longitude") is not None
+            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"):
+        if data.get("grid") 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")
+        return Callsign(
+            call=callsign,
+            home_call=callinfo.Callinfo.get_homecall(callsign),
+            name=data.get("nick", None),
+            qth=data.get("qth", None),
+            country=data.get("country", None),
+            continent=Continent(data["continent"]) if "continent" in data else None,
+            latitude=lat,
+            longitude=lon,
+            grid=grid,
+            dxcc_id=int(data["adif"]) if data.get("adif") is not None else None,
+            cq_zone=int(data["cq"]) if data.get("cq") is not None else None,
+            itu_zone=int(data["itu"]) if data.get("itu") is not None else None,
+            location_source=LocationSourceForCallsign.HOME_QTH,
+        )

providers/callsigndata/qrz.py

@@ -1,19 +1,22 @@
 import logging
 import urllib.parse
-from datetime import timedelta, datetime
+from datetime import datetime, timedelta
 
 import pytz
 import xmltodict
 from pyhamtools import callinfo
-from requests import ConnectTimeout, ReadTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout, ReadTimeout
 from requests_cache import CachedSession
 
 from core.constants import HTTP_HEADERS
-from core.data_store import DATA_STORE, CACHE_DIR
+from core.data_store import CACHE_DIR, DATA_STORE
+from core.enums import Continent, LocationSourceForCallsign
 from core.url_data_cache import URLDataCache
 from data.callsign import Callsign
 from providers.callsigndata.api_query_callsign_data_provider import APIQueryCallsignDataProvider
 
+logger = logging.getLogger(__name__)
+
 
 class QRZ(APIQueryCallsignDataProvider):
     """Callsign data provider for QRZ.com."""
@@ -24,14 +27,14 @@ class QRZ(APIQueryCallsignDataProvider):
         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))
+        self._CREDENTIALS_CACHE = CachedSession(f"{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):
+        if not lookup_credentials or not (
+            (lookup_credentials.qrz_username and lookup_credentials.qrz_password) or lookup_credentials.qrz_session_key
+        ):
             return None
 
         try:
@@ -42,19 +45,19 @@ class QRZ(APIQueryCallsignDataProvider):
             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
+                        f"{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.")
+                        logger.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")
+                    logger.error("Exception when getting QRZ.com session key")
                     return None
 
             if not session_key:
@@ -67,14 +70,16 @@ class QRZ(APIQueryCallsignDataProvider):
                 if home_call != callsign:
                     calls_to_try.append(home_call)
             except ValueError:
-                logging.debug("Could not look up home call for callsign %s", callsign)
+                logger.debug(f"Could not look up home call for callsign {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)
+                        f"{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:
@@ -89,32 +94,33 @@ class QRZ(APIQueryCallsignDataProvider):
                             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"))
+                                logger.debug(
+                                    f"QRZ returned an error looking up callsign {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)
+                            logger.warning(f"QRZ returned a malformed response looking up callsign {lookup_call}")
                     elif not response.from_cache:
-                        logging.warning("HTTP %d looking up callsign %s using QRZ", lookup_call)
+                        logger.warning(f"HTTP {response.status_code} looking up callsign {lookup_call} using QRZ")
 
                 except (KeyError, ValueError):
                     continue
                 except ConnectionError:
-                    logging.warning(f"Connection error when looking up callsign %s using QRZ", lookup_call)
+                    logger.warning(f"Connection error when looking up callsign {lookup_call} using QRZ")
                     continue
                 except (ConnectTimeout, ReadTimeout):
-                    logging.warning(f"Timeout when looking up callsign %s using QRZ.", lookup_call)
+                    logger.warning(f"Timeout when looking up callsign {lookup_call} using QRZ.")
                     continue
                 except Exception:
-                    logging.error("Exception when looking up callsign %s using QRZ", lookup_call, exc_info=True)
+                    logger.exception(f"Exception when looking up callsign {lookup_call} using QRZ")
                     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:
+        except Exception:
             self.status = "Error"
-            logging.error("Exception when looking up data from QRZ.com", e, exc_info=True)
+            logger.exception("Exception when looking up data from QRZ.com")
             # Return None, this won't be cached so we will be asked to query data again for this call next time.
             return None
 
@@ -122,6 +128,13 @@ class QRZ(APIQueryCallsignDataProvider):
     def qrz_response_to_callsign(callsign, data):
         """Convert the "Callsign" block in QRZ's API response to our own Callsign object."""
 
+        # I have encountered a user passing multiple callsigns to the QRZ lookup function in a way that QRZ actually
+        # does handle, and returns a list of callsign data. If we encounter that, just take the first one, as our own
+        # functions can't deal with multiple calls this way.
+        if isinstance(data, list):
+            data = data[0]
+            callsign = data["call"]
+
         # Get a name
         name = None
         if "name_fmt" in data:
@@ -129,34 +142,39 @@ class QRZ(APIQueryCallsignDataProvider):
         if "fname" in data:
             name = data["fname"]
             if "nick" in data:
-                name = name + " \"" + data["nick"] + "\""
+                name = f'{name} "{data["nick"]}"'
             if "name" in data:
-                name = name + " " + data["name"]
+                name = f"{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:
+        if (
+            data.get("latitude") is not None
+            and data.get("longitude") is not None
+            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"):
+        if data.get("grid") 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")
+        return Callsign(
+            call=callsign,
+            home_call=callinfo.Callinfo.get_homecall(callsign),
+            name=name,
+            qth=data.get("addr2", None),
+            country=data.get("country", None),
+            continent=Continent(data["continent"]) if "continent" in data else None,
+            latitude=lat,
+            longitude=lon,
+            grid=grid,
+            dxcc_id=int(data["adif"]) if data.get("adif") is not None else None,
+            cq_zone=int(data["cqzone"]) if data.get("cqzone") is not None else None,
+            itu_zone=int(data["ituzone"]) if data.get("ituzone") is not None else None,
+            location_source=LocationSourceForCallsign.HOME_QTH,
+        )

providers/sigrefdata/arlhs.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class ARLHS(FileDownloadSIGRefDataProvider):
@@ -20,13 +23,18 @@ class ARLHS(FileDownloadSIGRefDataProvider):
         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"]))
+                new_data.append(
+                    SIGRef(
+                        sig=self.SIG,
+                        id=ref_id,
+                        name=row.get("Name", None),
+                        ref_type=SIGRefType.LIGHTHOUSE,
+                        url=f"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

providers/sigrefdata/cota.py

@@ -0,0 +1,52 @@
+from time import sleep
+
+from pyhamtools.locator import latlong_to_locator
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class COTA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Castles on the Air"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "COTA"
+    DATA_URL = "https://www.cotagroup.org/cotagroup/map/data/castles-all-7d90ee2a5e1175e5dece1bbf9dc87504.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[2]:
+                ref_id = ref[3]
+                name = ref[4]
+                lat = float(ref[0])
+                lon = float(ref[1])
+                grid = latlong_to_locator(lat, lon)
+
+                new_data.append(
+                    SIGRef(
+                        sig=self.SIG,
+                        id=ref_id,
+                        name=name,
+                        ref_type=SIGRefType.CASTLE,
+                        grid=grid,
+                        latitude=lat,
+                        longitude=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_event.is_set():
+                    break
+
+                # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+                # is available for other threads e.g. the web server.
+                sleep(0.001)
+
+        return new_data

providers/sigrefdata/dce.py

@@ -0,0 +1,42 @@
+import io
+from time import sleep
+
+import pandas as pd
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class DCE(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Diploma Castillos de España"""
+
+    POLL_INTERVAL_DAYS = 365
+    SIG = "DCE"
+    DATA_URL = "https://www.acracb.org/dce/descargas/General/directorio_referencias_dce.xls"
+
+    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 = []
+
+        file_stream = io.BytesIO(http_response.content)
+        df = pd.read_excel(file_stream, engine="xlrd", header=None)
+
+        for index, row in df.iterrows():
+            if row.iloc[0] and row.iloc[2]:
+                new_data.append(
+                    SIGRef(sig=self.SIG, id=row.iloc[0].strip(), name=row.iloc[2].strip(), ref_type=SIGRefType.CASTLE)
+                )
+
+            # 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
+
+            # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+            # is available for other threads e.g. the web server.
+            sleep(0.001)
+
+        return new_data

providers/sigrefdata/defe.py

@@ -0,0 +1,46 @@
+import io
+from time import sleep
+
+import pandas as pd
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class DEFE(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Diploma Estationes de Ferrocarril de España"""
+
+    POLL_INTERVAL_DAYS = 365
+    SIG = "DEFE"
+    DATA_URL = "https://www.acracb.org/defe/descargas/General/directorio_referencias_defe.xls"
+
+    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 = []
+
+        file_stream = io.BytesIO(http_response.content)
+        df = pd.read_excel(file_stream, engine="xlrd", header=None)
+
+        for index, row in df.iterrows():
+            # Skip the header row
+            if str(row.iloc[0]) == "NºDEFE":
+                continue
+
+            if row.iloc[0] and row.iloc[1]:
+                new_data.append(
+                    SIGRef(sig=self.SIG, id=row.iloc[0].strip(), name=row.iloc[1].strip(), ref_type=SIGRefType.BUILDING)
+                )
+
+            # 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
+
+            # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+            # is available for other threads e.g. the web server.
+            sleep(0.001)
+
+        return new_data

providers/sigrefdata/dme.py

@@ -3,8 +3,11 @@ from time import sleep
 
 from pyhamtools.locator import latlong_to_locator
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.local_file_sig_ref_data_provider import LocalFileSIGRefDataProvider
+from providers.sigrefdata.local_file_sig_ref_data_provider import (
+    LocalFileSIGRefDataProvider,
+)
 
 
 class DME(LocalFileSIGRefDataProvider):
@@ -20,23 +23,37 @@ class DME(LocalFileSIGRefDataProvider):
         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
+                # Store reference IDs with the "DME-" prefix rather than just the number. This will prevent Spothole
+                # from agressively thinking every number in a spot comment is DME after it's seen "DME" once. The only
+                # numbers that count are straight after "DME " or "DME-". The dash versus space is normalised in
+                # sig_lookup_helper.py.
+                ref_id = "DME-" + 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)
+                ref = SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    ref_type=SIGRefType.TOWN,
+                    name=f"{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:
+                if self._stop_event.is_set():
                     break
 
                 # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time

providers/sigrefdata/dmue.py

@@ -0,0 +1,37 @@
+import csv
+from time import sleep
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class DMUE(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Diploma Museos de España"""
+
+    POLL_INTERVAL_DAYS = 365
+    SIG = "DMUE"
+    DATA_URL = "https://dmue.radiogalena.es/nom_dmue.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.reader(http_response.content.decode("utf-8-sig").splitlines(), delimiter=";"):
+            if len(row) > 1 and row[0] and row[1]:
+                new_data.append(
+                    SIGRef(sig=self.SIG, id=row[0].strip(), name=row[1].strip(), ref_type=SIGRefType.BUILDING)
+                )
+
+            # 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
+
+            # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+            # is available for other threads e.g. the web server.
+            sleep(0.001)
+
+        return new_data

providers/sigrefdata/dmve.py

@@ -0,0 +1,50 @@
+import io
+from time import sleep
+
+import pandas as pd
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class DMVE(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Diploma Monumentos y Vestigios de España"""
+
+    POLL_INTERVAL_DAYS = 365
+    SIG = "DMVE"
+    DATA_URL = "https://www.acracb.org/dmve/descargas/General/directorio_referencias_dmve.xls"
+
+    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 = []
+
+        file_stream = io.BytesIO(http_response.content)
+        # Despide the .xls extension this is actually an xlsx file, so we need openpyxl not xlrd
+        df = pd.read_excel(file_stream, engine="openpyxl", header=None)
+
+        for index, row in df.iterrows():
+            ref = row.iloc[0]
+            name = row.iloc[1]
+
+            # Skip the header row and blank rows
+            if str(ref) == "REF.":
+                continue
+            if pd.isna(ref) or pd.isna(name):
+                continue
+
+            if ref and name:
+                new_data.append(SIGRef(sig=self.SIG, id=ref.strip(), name=name.strip(), ref_type=SIGRefType.BUILDING))
+
+            # 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
+
+            # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+            # is available for other threads e.g. the web server.
+            sleep(0.001)
+
+        return new_data

providers/sigrefdata/dtmba.py

@@ -0,0 +1,35 @@
+from time import sleep
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class DTMBA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Diploma Teatri Musei Belle Arti"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "DTMBA"
+    DATA_URL = "https://www.iu1fig.com/share/iz0eik/dtmba/export.php"
+
+    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 http_response.content.decode("utf-8-sig").splitlines():
+            split = row.split(";")
+            ref_id = split[0]
+            ref_name = split[1]
+            new_data.append(SIGRef(sig=self.SIG, id=ref_id, name=ref_name, ref_type=SIGRefType.BUILDING))
+
+            # 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
+
+            # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+            # is available for other threads e.g. the web server.
+            sleep(0.001)
+
+        return new_data

providers/sigrefdata/fea.py

@@ -0,0 +1,56 @@
+from io import BytesIO
+from time import sleep
+
+import pdfplumber
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class FEA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Diploma Faros de España"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "FEA"
+    DATA_URL = "http://ea5ol.net/Lista%20Faros.pdf"
+
+    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 = []
+
+        # Use PDFPlumber to extract the tables in the PDF
+        with pdfplumber.open(BytesIO(http_response.content)) as pdf:
+            all_rows = []
+
+            for page_number, page in enumerate(pdf.pages, start=1):
+                tables = page.extract_tables()
+
+                for table_number, table in enumerate(tables, start=1):
+                    if not table:
+                        continue
+
+                    rows = [row for row in table if any(cell and cell.strip() for cell in row)]
+                    all_rows.extend(rows)
+
+        for row in all_rows:
+            if not "REF" in row[0] and not "\n" in row[0]:
+                # FEA references are technically [DE]\-\d{4}(\.\d)? but spotters always seem to miss out the D- or E-
+                # prefix and just use FEA-1234 or FEA 1234, so we add both copies to the database.
+                ref_id_1 = row[0].strip()
+                ref_id_2 = ref_id_1.replace("D-", "FEA-").replace("E-", "FEA-")
+                new_data.append(SIGRef(sig=self.SIG, id=ref_id_1, name=row[1].strip(), ref_type=SIGRefType.LIGHTHOUSE))
+                new_data.append(SIGRef(sig=self.SIG, id=ref_id_2, name=row[1].strip(), ref_type=SIGRefType.LIGHTHOUSE))
+
+            # 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
+
+            # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+            # is available for other threads e.g. the web server.
+            sleep(0.001)
+
+        return new_data

providers/sigrefdata/file_download_sig_ref_data_provider.py

@@ -1,39 +1,41 @@
 import logging
 from datetime import datetime
-from threading import Thread, Event
+from threading import Thread
 
 import pytz
-from requests import ReadTimeout
-from requests.exceptions import ConnectionError, ConnectTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout, ReadTimeout
 
 from core.constants import HTTP_HEADERS
 from core.url_data_cache import URLDataCache
 from providers.sigrefdata.sig_ref_data_provider import SIGRefDataProvider
 
+logger = logging.getLogger(__name__)
+
 
 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*."""
+        """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)
+        self._url_data_cache = URLDataCache(f"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}")
+        logger.info(f"Set up query of {self.sig_name} SIG ref data every {self._poll_interval!s} days.")
+        self._thread = Thread(target=self._run, name=f"FileDownloadSIGRefDataProvider-{self.sig_name}", daemon=True)
         self._thread.start()
 
     def stop(self):
         super().stop()
-        self._stop_event.set()
+        if self._thread:
+            self._thread.join(timeout=35)
+            if self._thread.is_alive():
+                logger.warning(f"{self.sig_name} SIG ref data worker thread did not exit on time and will be killed.")
 
     def _run(self):
         while True:
@@ -45,7 +47,7 @@ class FileDownloadSIGRefDataProvider(SIGRefDataProvider):
         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...")
+            logger.debug(f"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:
@@ -57,20 +59,20 @@ class FileDownloadSIGRefDataProvider(SIGRefDataProvider):
 
                 self.status = "OK"
                 self.last_update_time = datetime.now(pytz.UTC)
-                logging.debug("Received SIG ref data for " + self.sig_name)
+                logger.debug(f"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}.")
+                logger.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}.")
+            logger.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}.")
+            logger.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 + ")")
+            logger.exception(f"Exception in HTTP SIG Ref Data Provider ({self.sig_name})")
             self._stop_event.wait(timeout=1)
 
     def _http_response_to_data(self, http_response):

providers/sigrefdata/gma.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class GMA(FileDownloadSIGRefDataProvider):
@@ -19,13 +22,21 @@ class GMA(FileDownloadSIGRefDataProvider):
         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"]))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("Name", None),
+                    ref_type=SIGRefType.SUMMIT,
+                    url=f"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,
+                    altitude=float(row["Height (m)"].replace("m", ""))
+                    if "Height (m)" in row and row["Height (m)"] != ""
+                    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

providers/sigrefdata/illw.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class ILLW(FileDownloadSIGRefDataProvider):
@@ -20,13 +23,18 @@ class ILLW(FileDownloadSIGRefDataProvider):
         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"]))
+                new_data.append(
+                    SIGRef(
+                        sig=self.SIG,
+                        id=ref_id,
+                        name=row.get("Name", None),
+                        ref_type=SIGRefType.LIGHTHOUSE,
+                        url=f"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

providers/sigrefdata/iota.py

@@ -3,8 +3,13 @@ from time import sleep
 
 from pyhamtools.locator import latlong_to_locator
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 
 class IOTA(FileDownloadSIGRefDataProvider):
@@ -23,16 +28,29 @@ class IOTA(FileDownloadSIGRefDataProvider):
         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
+                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)
+                    logger.debug(
+                        "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))
+                new_data.append(
+                    SIGRef(
+                        sig=self.SIG,
+                        id=ref_id,
+                        name=ref["name"],
+                        ref_type=SIGRefType.ISLAND,
+                        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

providers/sigrefdata/krmnpa.py

@@ -1,4 +1,6 @@
-from providers.sigrefdata.pnp_kml_sig_ref_data_provider import ParksNPeaksKMLSIGRefDataProvider
+from providers.sigrefdata.pnp_kml_sig_ref_data_provider import (
+    ParksNPeaksKMLSIGRefDataProvider,
+)
 
 
 class KRMNPA(ParksNPeaksKMLSIGRefDataProvider):
@@ -9,4 +11,4 @@ class KRMNPA(ParksNPeaksKMLSIGRefDataProvider):
     DATA_URL = "https://parksnpeaks.org/getPOI.php?poiClass=KRMNPA&poiFormat=4"
 
     def __init__(self, provider_config):
-        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)
+        super().__init__(self.SIG, provider_config, self.DATA_URL, self.POLL_INTERVAL_DAYS)

providers/sigrefdata/llota.py

@@ -2,8 +2,11 @@ from time import sleep
 
 from pyhamtools.locator import locator_to_latlong
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class LLOTA(FileDownloadSIGRefDataProvider):
@@ -25,11 +28,18 @@ class LLOTA(FileDownloadSIGRefDataProvider):
                 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]))
+                new_data.append(
+                    SIGRef(
+                        sig=self.SIG,
+                        id=ref_id,
+                        name=str(ref["name"]),
+                        ref_type=SIGRefType.LAKE,
+                        url=f"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

providers/sigrefdata/local_file_sig_ref_data_provider.py

@@ -5,6 +5,8 @@ import pytz
 
 from providers.sigrefdata.sig_ref_data_provider import SIGRefDataProvider
 
+logger = logging.getLogger(__name__)
+
 
 class LocalFileSIGRefDataProvider(SIGRefDataProvider):
     """Generic SIG ref data provider class for providers that fetch their data from a local file on startup."""
@@ -14,7 +16,7 @@ class LocalFileSIGRefDataProvider(SIGRefDataProvider):
         self._path = path
 
     def start(self):
-        logging.debug("Loading " + self.sig_name + " SIG ref data from file.")
+        logger.debug(f"Loading {self.sig_name} SIG ref data from file.")
         try:
             new_data = self._file_to_data(self._path)
             if new_data:
@@ -23,10 +25,10 @@ class LocalFileSIGRefDataProvider(SIGRefDataProvider):
                 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:
+                logger.info(f"Failed to load SIG ref data for {self.sig_name}")
+        except Exception:
             self.status = "Error"
-            logging.error("Exception in local file SIG Ref Data Provider (" + self.sig_name + ")", e, exc_info=True)
+            logger.exception(f"Exception in local file SIG Ref Data Provider ({self.sig_name})")
 
     def _file_to_data(self, path):
         """Load a file on the given path and turn it into SIG Ref data."""

providers/sigrefdata/mota.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class MOTA(FileDownloadSIGRefDataProvider):
@@ -19,13 +22,18 @@ class MOTA(FileDownloadSIGRefDataProvider):
         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"]))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("Name", None),
+                    ref_type=SIGRefType.MILL,
+                    url=f"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

providers/sigrefdata/pga.py

@@ -0,0 +1,58 @@
+from time import sleep
+
+from bs4 import BeautifulSoup
+
+from core.enums import SIGRefType
+from data.sig_ref import SIGRef
+from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+
+
+class PGA(FileDownloadSIGRefDataProvider):
+    """SIG ref data provider for Polish Gmina Award"""
+
+    POLL_INTERVAL_DAYS = 30
+    SIG = "PGA"
+    DATA_URL = "http://www.spga.pl/lista_pga2.php"
+
+    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 = []
+        soup = BeautifulSoup(http_response.text, "html.parser")
+
+        # Iterate through tables in the page
+        for table in soup.find_all("table"):
+            header_cells = table.find_all(["th", "td"], limit=10)
+            header_texts = [c.get_text(strip=True) for c in header_cells]
+
+            # If it has "PGA" and "Nazwa" in the header, it's the main data table
+            if any("PGA" in t for t in header_texts) and any("Nazwa" in t for t in header_texts):
+                # Iterate through all rows except the first
+                rows = table.find_all("tr")
+                for row in rows[1:]:
+                    cells = row.find_all("td")
+                    ref_id = cells[0].get_text(strip=True)
+                    name = cells[1].get_text(strip=True)
+                    if not ref_id:
+                        continue
+
+                    new_data.append(
+                        SIGRef(
+                            sig=self.SIG,
+                            id=ref_id,
+                            name=name,
+                            ref_type=SIGRefType.REGION,
+                        )
+                    )
+
+                    # 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
+
+                    # Very short pause. This will extend the time to handle sig refs by a few seconds but will ensure some time
+                    # is available for other threads e.g. the web server.
+                    sleep(0.001)
+
+        return new_data

providers/sigrefdata/pnp_kml_sig_ref_data_provider.py

@@ -4,8 +4,11 @@ from time import sleep
 from fastkml import kml
 from pyhamtools.locator import latlong_to_locator
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class ParksNPeaksKMLSIGRefDataProvider(FileDownloadSIGRefDataProvider):
@@ -15,7 +18,7 @@ class ParksNPeaksKMLSIGRefDataProvider(FileDownloadSIGRefDataProvider):
     REF_PATTERN = re.compile(r"VKFF-\d+")
 
     def __init__(self, sig_name, provider_config, url, poll_interval):
-        """ Set up the provider, note poll_interval is in *days*."""
+        """Set up the provider, note poll_interval is in *days*."""
         super().__init__(sig_name, provider_config, url, poll_interval)
 
     def _http_response_to_data(self, http_response):
@@ -24,7 +27,9 @@ class ParksNPeaksKMLSIGRefDataProvider(FileDownloadSIGRefDataProvider):
         k = kml.KML.from_string(http_response.content)
 
         for document in k.features:
+            # noinspection unresolved-references
             for folder in document.features:
+                # noinspection unresolved-references
                 for placemark in folder.features:
                     description = placemark.description or ""
                     match = self.REF_PATTERN.search(description)
@@ -35,10 +40,15 @@ class ParksNPeaksKMLSIGRefDataProvider(FileDownloadSIGRefDataProvider):
 
                     longitude, latitude = placemark.geometry.x, placemark.geometry.y
 
-                    ref = SIGRef(sig=self.sig_name, id=ref_id, name=placemark.name,
-                                              url="https://parksnpeaks.org/getPark.php?actPark=" + ref_id,
-                                              latitude=latitude,
-                                              longitude=longitude)
+                    ref = SIGRef(
+                        sig=self.sig_name,
+                        id=ref_id,
+                        name=placemark.name,
+                        ref_type=SIGRefType.PARK,
+                        url=f"https://parksnpeaks.org/getPark.php?actPark={ref_id}",
+                        latitude=latitude,
+                        longitude=longitude,
+                    )
                     if latitude and longitude:
                         ref.grid = latlong_to_locator(latitude, longitude, 6)
                     new_data.append(ref)

providers/sigrefdata/pota.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class POTA(FileDownloadSIGRefDataProvider):
@@ -19,13 +22,18 @@ class POTA(FileDownloadSIGRefDataProvider):
         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))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("name", None),
+                    ref_type=SIGRefType.PARK,
+                    url=f"https://pota.app/#/park/{ref_id}",
+                    grid=row.get("grid", 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

providers/sigrefdata/sanpcpa.py

@@ -1,4 +1,6 @@
-from providers.sigrefdata.pnp_kml_sig_ref_data_provider import ParksNPeaksKMLSIGRefDataProvider
+from providers.sigrefdata.pnp_kml_sig_ref_data_provider import (
+    ParksNPeaksKMLSIGRefDataProvider,
+)
 
 
 class SANPCPA(ParksNPeaksKMLSIGRefDataProvider):

providers/sigrefdata/sig_ref_data_provider.py

@@ -1,10 +1,13 @@
 import logging
 from datetime import datetime
+from threading import Event
 
 import pytz
 
 from core.data_store import DATA_STORE
 
+logger = logging.getLogger(__name__)
+
 
 class SIGRefDataProvider:
     """Generic SIG reference data provider class. Subclasses of this query the individual URLs or files for data."""
@@ -17,34 +20,34 @@ class SIGRefDataProvider:
         self.last_update_time = datetime.min.replace(tzinfo=pytz.UTC)
         self.status = "Not Started" if self.enabled else "Disabled"
         self.reference_count = 0
-        self._stop = False
-
+        self._stop_event = Event()
 
     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. Subclasses should implement this method and call
         super()."""
 
-        self._stop = True
+        self._stop_event.set()
 
     def _add_data(self, new_data):
         """Add all the provided reference data objects to the data store."""
 
-        # with transact() batches all writes together to save making thousands of individual sqlite writes
-        with DATA_STORE.sigrefs.transact():
+        # with transact() batches all writes together to save making thousands of individual sqlite writes. However,
+        # that means that each provider holds the lock while it writes, and the default behaviour for other attempted
+        # transact()s is to fail if they can't get the lock (?!). This behaviour is fixed by retry=True.
+        with DATA_STORE.sigrefs.transact(retry=True):
             for d in new_data:
-                DATA_STORE.sigrefs.set(self.sig_name + ":" + d.id, d)
+                DATA_STORE.sigrefs.set(f"{self.sig_name}:{d.id}", d)
 
                 # For the big data sources, loading will take a few minutes. If we want to shut down the software neatly
                 # within the first few minutes of startup, we need a way to abort this expensive process of filling up the
                 # disk cache.
-                if self._stop:
+                if self._stop_event.is_set():
                     break
 
         self.reference_count = len(new_data)
-        logging.info(f"Loaded %d references for %s into the data store.", self.reference_count, self.sig_name)
+        logger.info(f"Loaded {self.reference_count} references for {self.sig_name} into the data store.")

providers/sigrefdata/siota.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class SIOTA(FileDownloadSIGRefDataProvider):
@@ -19,10 +22,17 @@ class SIOTA(FileDownloadSIGRefDataProvider):
         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))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("NAME", None),
+                    ref_type=SIGRefType.SILO,
+                    grid=row.get("LOCATOR", 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

providers/sigrefdata/sota.py

@@ -3,8 +3,11 @@ from time import sleep
 
 from pyhamtools.locator import latlong_to_locator
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class SOTA(FileDownloadSIGRefDataProvider):
@@ -23,11 +26,18 @@ class SOTA(FileDownloadSIGRefDataProvider):
             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)
+            altitude = float(row["AltM"]) if "AltM" in row and row["AltM"] != "" else None
+            ref = SIGRef(
+                sig=self.SIG,
+                id=ref_id,
+                name=row.get("SummitName", None),
+                ref_type=SIGRefType.SUMMIT,
+                url=f"https://www.sotadata.org.uk/en/summit/{ref_id}",
+                latitude=latitude,
+                longitude=longitude,
+                altitude=altitude,
+                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)

providers/sigrefdata/toilets.py

@@ -1,7 +1,10 @@
 import csv
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.local_file_sig_ref_data_provider import LocalFileSIGRefDataProvider
+from providers.sigrefdata.local_file_sig_ref_data_provider import (
+    LocalFileSIGRefDataProvider,
+)
 
 
 class Toilets(LocalFileSIGRefDataProvider):
@@ -19,12 +22,20 @@ class Toilets(LocalFileSIGRefDataProvider):
             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"])))
+                new_data.append(
+                    SIGRef(
+                        sig=self.SIG,
+                        id=row["ref"],
+                        name=row["ref"],
+                        ref_type=SIGRefType.TOILET,
+                        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:
+                if self._stop_event.is_set():
                     break
 
         return new_data

providers/sigrefdata/towers.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class Towers(FileDownloadSIGRefDataProvider):
@@ -19,11 +22,18 @@ class Towers(FileDownloadSIGRefDataProvider):
         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))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("Nazev", None),
+                    ref_type=SIGRefType.TOWER,
+                    url=f"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

providers/sigrefdata/wca.py

@@ -4,8 +4,13 @@ from time import sleep
 
 from pyhamtools.locator import latlong_to_locator
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 
 class WCA(FileDownloadSIGRefDataProvider):
@@ -34,13 +39,20 @@ class WCA(FileDownloadSIGRefDataProvider):
                     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)
+                logger.debug(f"Encountered dodgy formatting in WCA CSV, skipping location data for {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))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("CLEAN NAME", None),
+                    ref_type=SIGRefType.CASTLE,
+                    url=f"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

providers/sigrefdata/wota.py

@@ -1,7 +1,10 @@
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class WOTA(FileDownloadSIGRefDataProvider):
@@ -20,15 +23,24 @@ class WOTA(FileDownloadSIGRefDataProvider):
             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
+            url = f"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)
+                url = f"https://www.wota.org.uk/MM_LDO-{number + 214!s}"
 
-            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]))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=feature["properties"]["title"],
+                    url=url,
+                    ref_type=SIGRefType.SUMMIT,
+                    grid=feature["properties"]["qthLocator"],
+                    latitude=feature["geometry"]["coordinates"][1],
+                    longitude=feature["geometry"]["coordinates"][0],
+                    altitude=feature["properties"]["height"],
+                )
+            )
 
             # 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

providers/sigrefdata/wwbota.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class WWBOTA(FileDownloadSIGRefDataProvider):
@@ -19,11 +22,18 @@ class WWBOTA(FileDownloadSIGRefDataProvider):
         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))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("Name", None),
+                    ref_type=SIGRefType.BUNKER,
+                    url=f"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

providers/sigrefdata/wwff.py

@@ -1,8 +1,11 @@
 import csv
 from time import sleep
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class WWFF(FileDownloadSIGRefDataProvider):
@@ -19,14 +22,22 @@ class WWFF(FileDownloadSIGRefDataProvider):
         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))
+            new_data.append(
+                SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=row.get("name", None),
+                    ref_type=SIGRefType.PARK,
+                    url=f"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

providers/sigrefdata/zlota.py

@@ -2,8 +2,11 @@ from time import sleep
 
 from pyhamtools.locator import latlong_to_locator
 
+from core.enums import SIGRefType
 from data.sig_ref import SIGRef
-from providers.sigrefdata.file_download_sig_ref_data_provider import FileDownloadSIGRefDataProvider
+from providers.sigrefdata.file_download_sig_ref_data_provider import (
+    FileDownloadSIGRefDataProvider,
+)
 
 
 class ZLOTA(FileDownloadSIGRefDataProvider):
@@ -22,16 +25,33 @@ class ZLOTA(FileDownloadSIGRefDataProvider):
         if isinstance(data, list):
             for ref in data:
                 ref_id = ref["code"]
-                latitude = ref["y"]
-                longitude = ref["x"]
+                latitude = ref["latitude"]
+                longitude = ref["longitude"]
+                try:
+                    ref_type = SIGRefType(ref["asset_type"].title().upper())
+                except ValueError:
+                    ref_type = None
 
-                ref = SIGRef(sig=self.SIG, id=ref_id, name=ref["name"],
-                                          url="https://ontheair.nz/assets/" + ref_id.replace("/", "_"),
-                                          latitude=latitude,
-                                          longitude=longitude)
+                new_ref = SIGRef(
+                    sig=self.SIG,
+                    id=ref_id,
+                    name=ref["name"],
+                    ref_type=ref_type,
+                    url=f"https://ontheair.nz/assets/{ref_id.replace('/', '_')}",
+                    latitude=latitude,
+                    longitude=longitude,
+                )
+
+                # Check lat/lon validity and update grid accordingly
                 if latitude and longitude:
-                    ref.grid = latlong_to_locator(latitude, longitude, 6)
-                new_data.append(ref)
+                    try:
+                        new_ref.grid = latlong_to_locator(latitude, longitude, 6)
+                    except ValueError:
+                        # Junk lat/lon, remove from the spot
+                        new_ref.latitude = None
+                        new_ref.longitude = None
+
+                new_data.append(new_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

providers/solarconditions/giroionosonde.py

@@ -1,24 +1,26 @@
 import csv
 import logging
-from datetime import datetime, timezone, timedelta
-from threading import Thread, Event
+from datetime import datetime, timedelta, timezone
+from threading import Event, Thread
 
 import pytz
 import requests
-from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout, ReadTimeout
 
 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
+logger = logging.getLogger(__name__)
+
+# Each station gets polled roughly once every two hours (7200 seconds). Note that to avoid a burst of requests to the
+# server every two hours, the requests for data from each station are spaced out throughout the period, leading to one
+# request being sent every 3-4 minutes.
+POLL_INTERVAL = 7200
 # 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"
+LGDC_URL = "https://lgdc.uml.edu/fastchar/getbest"
 HISTORY_HOURS = 24
 
 
@@ -40,9 +42,16 @@ class GIROIonosonde(SolarConditionsProvider):
         # 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
+            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}})
@@ -50,19 +59,23 @@ class GIROIonosonde(SolarConditionsProvider):
     @staticmethod
     def _load_stations():
         stations = []
-        with open(STATIONS_INDEX, newline='') as f:
+        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")
+        logger.info(f"Set up query of GIRO ionosonde data API every {POLL_INTERVAL} seconds.")
+        self._thread = Thread(target=self._run, name="GIROIonosondeDataProvider", daemon=True)
         self._thread.start()
 
     def stop(self):
         self._stop_event.set()
+        if self._thread:
+            self._thread.join(timeout=35)
+            if self._thread.is_alive():
+                logger.warning("GIRO ionosonde worker thread did not exit on time and will be killed.")
 
     def _run(self):
         # Real interval at which we poll is the "once per hour" divided by the number of stations, so each one gets
@@ -79,7 +92,7 @@ class GIROIonosonde(SolarConditionsProvider):
         ursi = station["ursi"]
         name = station["name"]
         try:
-            logging.debug(f"Polling GIRO ionosonde data for {ursi} ({name})...")
+            logger.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()
@@ -94,8 +107,14 @@ class GIROIonosonde(SolarConditionsProvider):
 
             # 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_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}
@@ -104,7 +123,8 @@ class GIROIonosonde(SolarConditionsProvider):
 
             band_states = compute_band_states(merged_fof2, merged_muf, merged_luf)
             ionosonde_data[ursi] = {
-                "ursi": ursi, "name": name,
+                "ursi": ursi,
+                "name": name,
                 "fof2": merged_fof2 or None,
                 "muf": merged_muf or None,
                 "luf": merged_luf or None,
@@ -113,29 +133,29 @@ class GIROIonosonde(SolarConditionsProvider):
             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}).")
+            logger.debug(f"Updated ionosonde data for {ursi} ({name}).")
 
         except Exception:
             self.status = "Error"
-            logging.exception(f"Exception fetching GIRO ionosonde data for {ursi} ({name})")
+            logger.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}"
+        url = f"{LGDC_URL}?ursiCode={ursi}&charName=foF2,MUF%28D%29,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.")
+                logger.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.")
+            logger.warning("Timeout when accessing Giro ionosonde API.")
             return None, None, None
         except ConnectionError:
-            logging.warning("Connection error when accessing Giro ionosonde API.")
+            logger.warning("Connection error when accessing Giro ionosonde API.")
             return None, None, None
 
     @staticmethod
@@ -147,14 +167,14 @@ class GIROIonosonde(SolarConditionsProvider):
         luf_data = {}
         for line in text.splitlines():
             line = line.strip()
-            if not line or line.startswith('#'):
+            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()
+                    ts = datetime.fromisoformat(parts[0].replace("Z", "+00:00")).timestamp()
                 except ValueError:
                     continue
                 try:

providers/solarconditions/hamqsl.py

@@ -2,9 +2,14 @@ import logging
 from xml.etree import ElementTree
 
 import pytz
-from dateutil import parser as dateutil_parser, tz as dateutil_tz
+from dateutil import parser as dateutil_parser
+from dateutil import tz as dateutil_tz
 
-from providers.solarconditions.http_solar_conditions_provider import HTTPSolarConditionsProvider
+from providers.solarconditions.http_solar_conditions_provider import (
+    HTTPSolarConditionsProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 POLL_INTERVAL = 3600  # 1 hour
 URL = "https://www.hamqsl.com/solarxml.php"
@@ -21,14 +26,14 @@ class HamQSL(HTTPSolarConditionsProvider):
         root = ElementTree.fromstring(http_response.text)
         sd = root.find("solardata")
         if sd is None:
-            logging.warning("HamQSL solar conditions API returned unexpected XML structure")
+            logger.warning("HamQSL solar conditions API returned unexpected XML structure")
             return None
 
         # 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")
+                logger.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
@@ -72,11 +77,11 @@ class HamQSL(HTTPSolarConditionsProvider):
                 tz_abbr = updated_str.split()[-1]
                 timezone = dateutil_tz.gettz(tz_abbr)
                 if timezone is None:
-                    raise ValueError("Unknown timezone abbreviation: " + tz_abbr)
+                    raise ValueError(f"Unknown timezone abbreviation: {tz_abbr}")
                 dt = dateutil_parser.parse(updated_str, tzinfos={tz_abbr: timezone})
                 updated = dt.astimezone(pytz.UTC).timestamp()
             except (ValueError, IndexError):
-                logging.warning("HamQSL solar conditions API returned unrecognised timestamp format: " + updated_str)
+                logger.warning(f"HamQSL solar conditions API returned unrecognised timestamp format: {updated_str}")
 
         # Return the data ready to be put into the solar conditions object.
         return {
@@ -92,7 +97,8 @@ class HamQSL(HTTPSolarConditionsProvider):
             "aurora_latitude": float_val("latdegree"),
             "solar_wind": float_val("solarwind"),
             "magnetic_field": float_val("magneticfield"),
-            "geomag_field": text("geomagfield").title()
+            "geomag_field": text("geomagfield")
+            .title()
             .replace("Vr Quiet", "Very Quiet")
             .replace("Unsettld", "Unsettled")
             .replace("Min Strm", "Minor Storm")
@@ -102,8 +108,10 @@ class HamQSL(HTTPSolarConditionsProvider):
             "geomag_noise": text("signalnoise"),
             "hf_conditions": hf_conditions,
             "vhf_conditions": {
-                "vhf_aurora_northern_hemi": (vhf_map.get(("vhf-aurora", "northern_hemi")) or "").title().replace(
-                    "Lat Aur", "Latitude") or None,
+                "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

@@ -1,14 +1,16 @@
 import logging
 from datetime import datetime
-from threading import Thread, Event
+from threading import Event, Thread
 
 import pytz
 import requests
-from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout, ReadTimeout
 
 from core.constants import HTTP_HEADERS
 from providers.solarconditions.solar_conditions_provider import SolarConditionsProvider
 
+logger = logging.getLogger(__name__)
+
 
 class HTTPSolarConditionsProvider(SolarConditionsProvider):
     """Generic solar conditions provider for providers that request data via HTTP(S). Subclasses implement
@@ -22,13 +24,16 @@ class HTTPSolarConditionsProvider(SolarConditionsProvider):
         self._stop_event = Event()
 
     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, name=f"HTTPSolarConditionsProvider-{self.name}")
+        logger.info(f"Set up query of {self.name} solar conditions API every {self._poll_interval!s} seconds.")
+        self._thread = Thread(target=self._run, name=f"HTTPSolarConditionsProvider-{self.name}", daemon=True)
         self._thread.start()
 
     def stop(self):
         self._stop_event.set()
+        if self._thread:
+            self._thread.join(timeout=35)
+            if self._thread.is_alive():
+                logger.warning(f"{self.name} solar conditions worker thread did not exit on time and will be killed.")
 
     def _run(self):
         while True:
@@ -38,7 +43,7 @@ class HTTPSolarConditionsProvider(SolarConditionsProvider):
 
     def _poll(self):
         try:
-            logging.debug("Polling " + self.name + " solar conditions API...")
+            logger.debug(f"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:
@@ -47,18 +52,18 @@ class HTTPSolarConditionsProvider(SolarConditionsProvider):
 
                 self.status = "OK"
                 self.last_update_time = datetime.now(pytz.UTC)
-                logging.debug("Received data from " + self.name + " solar conditions API.")
+                logger.debug(f"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.")
+                logger.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.")
+            logger.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.")
+            logger.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 + ")")
+            logger.exception(f"Exception in HTTP Solar Conditions Provider ({self.name})")
             self._stop_event.wait(timeout=1)
 
     def _http_response_to_solar_conditions(self, http_response):

providers/solarconditions/kc2gprop.py

@@ -1,15 +1,17 @@
 import logging
-from datetime import datetime, timezone, timedelta
-from threading import Thread, Event
+from datetime import datetime, timedelta, timezone
+from threading import Event, Thread
 
 import pytz
 import requests
-from requests.exceptions import ConnectionError, ReadTimeout, ConnectTimeout
+from requests.exceptions import ConnectionError, ConnectTimeout, ReadTimeout
 
 from core.constants import HTTP_HEADERS
 from providers.solarconditions.ionosonde_utils import compute_band_states
 from providers.solarconditions.solar_conditions_provider import SolarConditionsProvider
 
+logger = logging.getLogger(__name__)
+
 POLL_INTERVAL = 900  # 15 minutes
 KC2G_URL = "https://prop.kc2g.com/api/stations.json"
 HISTORY_HOURS = 24
@@ -29,12 +31,16 @@ class KC2GProp(SolarConditionsProvider):
         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")
+        logger.info(f"Set up query of KC2G ionosonde data API every {POLL_INTERVAL} seconds.")
+        self._thread = Thread(target=self._run, name="KC2GPropProvider", daemon=True)
         self._thread.start()
 
     def stop(self):
         self._stop_event.set()
+        if self._thread:
+            self._thread.join(timeout=35)
+            if self._thread.is_alive():
+                logger.warning("KC2G ionosonde worker thread did not exit on time and will be killed.")
 
     def _run(self):
         while True:
@@ -44,10 +50,10 @@ class KC2GProp(SolarConditionsProvider):
 
     def _poll(self):
         try:
-            logging.debug("Polling KC2G ionosonde data...")
+            logger.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.")
+                logger.warning(f"HTTP {http_response.status_code} when calling KG2G ionosonde API.")
                 return
 
             now = datetime.now(timezone.utc)
@@ -114,13 +120,13 @@ class KC2GProp(SolarConditionsProvider):
             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.")
+            logger.debug(f"Updated KC2G ionosonde data for {updated_count} stations.")
 
         except ConnectionError:
-            logging.warning("Connection error when accessing KC2G ionosonde API.")
+            logger.warning("Connection error when accessing KC2G ionosonde API.")
         except (ConnectTimeout, ReadTimeout):
-            logging.warning(f"Timeout when accessing KC2G ionosonde API.")
+            logger.warning("Timeout when accessing KC2G ionosonde API.")
         except Exception:
             self.status = "Error"
-            logging.exception("Exception in KC2G ionosonde data provider")
+            logger.exception("Exception in KC2G ionosonde data provider")
             self._stop_event.wait(timeout=1)

providers/solarconditions/noaa3dayforecast.py

@@ -2,7 +2,11 @@ import logging
 import re
 from datetime import datetime, timezone
 
-from providers.solarconditions.http_solar_conditions_provider import HTTPSolarConditionsProvider
+from providers.solarconditions.http_solar_conditions_provider import (
+    HTTPSolarConditionsProvider,
+)
+
+logger = logging.getLogger(__name__)
 
 POLL_INTERVAL = 10800  # Every 3 hours
 URL = "https://services.swpc.noaa.gov/text/3-day-forecast.txt"
@@ -26,21 +30,21 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
                 start_idx = i
                 break
         if start_idx is None:
-            logging.warning(f"NOAA 3-day forecast: could not find '{section_header}' section")
+            logger.warning(f"NOAA 3-day forecast: could not find '{section_header}' section")
             return None
 
         # 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]):
+            if re.search(r"[A-Za-z]{3}\s+\d{2}", lines[j]):
                 date_header_idx = j
                 break
         if date_header_idx is None:
-            logging.warning(f"NOAA 3-day forecast: could not find date header after '{section_header}'")
+            logger.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])
+        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]}")
+            logger.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
@@ -50,25 +54,25 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
                 dt = datetime.strptime(f"{day_str} {month_str} {year}", "%d %b %Y").replace(tzinfo=timezone.utc)
                 column_timestamps.append(dt.timestamp())
             except ValueError:
-                logging.warning(f"NOAA 3-day forecast: could not parse date: {month_str} {day_str} {year}")
+                logger.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 followed by percentage values
         result = {}
-        for line in lines[date_header_idx + 1:]:
+        for line in lines[date_header_idx + 1 :]:
             line_stripped = line.strip()
             if not line_stripped:
                 if result:
                     break
                 continue
-            pct_matches = list(re.finditer(r'\b(\d+)%', line_stripped))
+            pct_matches = list(re.finditer(r"\b(\d+)%", line_stripped))
             if not pct_matches:
                 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_label = line_stripped[: line_stripped.index(pct_matches[0].group())].strip()
             row_data = {}
             for j, match in enumerate(pct_matches):
                 if j >= len(column_timestamps):
@@ -89,26 +93,26 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
                 start_idx = i
                 break
         if start_idx is None:
-            logging.warning("NOAA K-index forecast: could not find 'NOAA Kp index breakdown' section")
+            logger.warning("NOAA K-index forecast: could not find 'NOAA Kp index breakdown' section")
             return None
 
         # Extract the year from the header line, e.g. "NOAA Kp index breakdown Apr 2-Apr 4, 2026"
         header_line = lines[start_idx]
-        year_match = re.search(r'\b(\d{4})\b', header_line)
+        year_match = re.search(r"\b(\d{4})\b", header_line)
         if not year_match:
-            logging.warning("NOAA K-index forecast: could not extract year from: " + header_line)
+            logger.warning(f"NOAA K-index forecast: could not extract year from: {header_line}")
             return None
         year = int(year_match.group(1))
 
         # Parse the column date headers on the next line, e.g. "              Apr 02     Apr 03     Apr 04"
         if start_idx + 1 >= len(lines):
-            logging.warning("NOAA K-index forecast: missing date header line")
+            logger.warning("NOAA K-index forecast: missing date header line")
             return None
 
         date_header_line = lines[start_idx + 2]
-        date_matches = re.findall(r'([A-Za-z]{3})\s+(\d{2})', date_header_line)
+        date_matches = re.findall(r"([A-Za-z]{3})\s+(\d{2})", date_header_line)
         if not date_matches:
-            logging.warning("NOAA K-index forecast: could not parse date headers from: " + date_header_line)
+            logger.warning(f"NOAA K-index forecast: could not parse date headers from: {date_header_line}")
             return None
 
         column_dates = []
@@ -116,13 +120,13 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
             try:
                 column_dates.append(datetime.strptime(f"{day_str} {month_str} {year}", "%d %b %Y").date())
             except ValueError:
-                logging.warning(f"NOAA K-index forecast: could not parse date: {month_str} {day_str} {year}")
+                logger.warning(f"NOAA K-index forecast: could not parse date: {month_str} {day_str} {year}")
                 return None
 
         # Parse each data row, e.g. "00-03UT          2.00          3.00          2.00"
         k_index_forecast = {}
-        for line in lines[start_idx + 3:]:
-            time_match = re.match(r'^(\d{2})-(\d{2})UT\s+(.*)', line.strip())
+        for line in lines[start_idx + 3 :]:
+            time_match = re.match(r"^(\d{2})-(\d{2})UT\s+(.*)", line.strip())
             if not time_match:
                 if k_index_forecast:
                     break
@@ -130,7 +134,7 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
 
             start_hour = int(time_match.group(1))
             # Split on 2 or more spaces so that e.g. "5.67 (G2)" stays as one token per column
-            raw_values = re.split(r' {2,}', time_match.group(3).strip())
+            raw_values = re.split(r" {2,}", time_match.group(3).strip())
 
             for i, val in enumerate(raw_values):
                 if i >= len(column_dates):
@@ -142,14 +146,22 @@ class NOAA3dayForecast(HTTPSolarConditionsProvider):
                     continue
 
                 date = column_dates[i]
-                start_dt = datetime(date.year, date.month, date.day, start_hour, 0, 0, tzinfo=timezone.utc)
+                start_dt = datetime(
+                    date.year,
+                    date.month,
+                    date.day,
+                    start_hour,
+                    0,
+                    0,
+                    tzinfo=timezone.utc,
+                )
 
                 # Key the data dict by start time
                 key = start_dt.timestamp()
                 k_index_forecast[key] = kp
 
         if not k_index_forecast:
-            logging.warning("NOAA K-index forecast: no data rows parsed")
+            logger.warning("NOAA K-index forecast: no data rows parsed")
             return None
 
         # Parse Solar Radiation Storm Forecast (single row: "S1 or greater")