chooses most complete result from api (#116)

src/auto_archiver/core/media.py

@@ -65,7 +65,9 @@ class Media:
 
     @property  # getter .mimetype
     def mimetype(self) -> str:
-        assert self.filename is not None and len(self.filename) > 0, "cannot get mimetype from media without filename"
+        if not self.filename or len(self.filename) == 0:
+            logger.warning(f"cannot get mimetype from media without filename: {self}")
+            return ""
         if not self._mimetype:
             self._mimetype = mimetypes.guess_type(self.filename)[0]
         return self._mimetype or ""

src/auto_archiver/core/metadata.py

@@ -165,3 +165,16 @@ class Metadata:
 
     def __str__(self) -> str:
         return self.__repr__()
+
+
+    @staticmethod
+    def choose_most_complete(results: List[Metadata]) -> Metadata:
+        # returns the most complete result from a list of results
+        # prioritizes results with more media, then more metadata
+        if len(results) == 0: return None
+        if len(results) == 1: return results[0]
+        most_complete = results[0]
+        for r in results[1:]:
+            if len(r.media) > len(most_complete.media): most_complete = r
+            elif len(r.media) == len(most_complete.media) and len(r.metadata) > len(most_complete.metadata): most_complete = r
+        return most_complete

src/auto_archiver/databases/api_db.py

@@ -35,16 +35,17 @@ class AAApiDb(Database):
         """ query the database for the existence of this item"""
         if not self.allow_rearchive: return
         
-        params = {"url": item.get_url(), "limit": 1}
+        params = {"url": item.get_url(), "limit": 15}
         headers = {"Authorization": f"Bearer {self.api_token}", "accept": "application/json"}
         response = requests.get(os.path.join(self.api_endpoint, "tasks/search-url"), params=params, headers=headers)
 
         if response.status_code == 200:
-            logger.success(f"API returned a previously archived instance: {response.json()}")
-            # TODO: can we do better than just returning the first result?
-            return Metadata.from_dict(response.json()[0]["result"])
-        
-        logger.error(f"AA API FAIL ({response.status_code}): {response.json()}")
+            if len(response.json()):
+                logger.success(f"API returned {len(response.json())} previously archived instance(s)")
+                fetched_metadata = [Metadata.from_dict(r["result"]) for r in response.json()]
+                return Metadata.choose_most_complete(fetched_metadata)
+        else:
+            logger.error(f"AA API FAIL ({response.status_code}): {response.json()}")
         return False
 
 

src/auto_archiver/version.py

@@ -3,7 +3,7 @@ _MAJOR = "0"
 _MINOR = "7"
 # On main and in a nightly release the patch should be one ahead of the last
 # released build.
-_PATCH = "2"
+_PATCH = "5"
 # This is mainly for nightly builds which have the suffix ".dev$DATE". See
 # https://semver.org/#is-v123-a-semantic-version for the semantics.
 _SUFFIX = ""