HEMA support

providers/hema.py

@@ -0,0 +1,69 @@
+import re
+from datetime import datetime, timedelta
+
+import pytz
+import requests
+from requests_cache import CachedSession
+
+from data.spot import Spot
+from providers.http_provider import HTTPProvider
+
+
+# Provider for HuMPs Excluding Marilyns Award
+class HEMA(HTTPProvider):
+    POLL_INTERVAL_SEC = 300
+    # HEMA wants us to check for a "spot seed" from the API and see if it's actually changed before querying the main
+    # data API. So it's actually the SPOT_SEED_URL that we pass into the constructor and get the superclass to call on a
+    # timer. The actual data lookup all happens after parsing and checking the seed.
+    SPOT_SEED_URL = "http://www.hema.org.uk/spotSeed.jsp"
+    SPOTS_URL = "http://www.hema.org.uk/spotsMobile.jsp"
+    FREQ_MODE_PATTERN = re.compile("^([\\d.]*) \\((.*)\\)$")
+    SPOTTER_COMMENT_PATTERN = re.compile("^\\((.*)\\) (.*)$")
+
+    def __init__(self):
+        super().__init__(self.SPOT_SEED_URL, self.POLL_INTERVAL_SEC)
+        self.spot_seed = ""
+
+    def name(self):
+        return "HEMA"
+
+    def http_response_to_spots(self, http_response):
+        # OK, source data is actually just the spot seed at this point. We'll then go on to fetch real data if we know
+        # this has changed.
+        spot_seed_changed = http_response.text != self.spot_seed
+        self.spot_seed = http_response.text
+
+        new_spots = []
+        # OK, if the spot seed actually changed, now we make the real request for data.
+        if spot_seed_changed:
+            source_data = requests.get(self.SPOTS_URL, headers=self.HTTP_HEADERS)
+            source_data_items = source_data.text.split("=")
+            # Iterate through source data items.
+            for source_spot in source_data_items:
+                spot_items = source_spot.split(";")
+                # Any line with less than 9 items is not a proper spot line
+                if len(spot_items) >= 9:
+                    # Fiddle with some data to extract bits we need. Freq/mode and spotter/comment come in combined fields.
+                    freq_mode_match = re.search(self.FREQ_MODE_PATTERN, spot_items[5])
+                    spotter_comment_match = re.search(self.SPOTTER_COMMENT_PATTERN, spot_items[6])
+
+                    # Convert to our spot format
+                    spot = Spot(source=self.name(),
+                                dx_call=spot_items[2].upper(),
+                                de_call=spotter_comment_match.group(1).upper(),
+                                freq=float(freq_mode_match.group(1)) * 1000,
+                                mode=freq_mode_match.group(2).upper(),
+                                comment=spotter_comment_match.group(2),
+                                sig="HEMA",
+                                sig_refs=[spot_items[3].upper()],
+                                sig_refs_names=[spot_items[4]],
+                                time=datetime.strptime(spot_items[0], "%d/%m/%Y %H:%M").replace(tzinfo=pytz.UTC),
+                                latitude=float(spot_items[7]),
+                                longitude=float(spot_items[8]))
+
+                    # Fill in any missing data
+                    spot.infer_missing()
+                    # Add to our list. Don't worry about de-duping, removing old spots etc. at this point; other code will do
+                    # that for us.
+                    new_spots.append(spot)
+        return new_spots