[patreon] improve 'campaign_id' handling (#4699, #4715)

- add ways to directly specify a 'campaign_id'
  - 'campaign-id' config option
  - 'c' or 'campaign_id' URL query parameter
- more descriptive error messages
- show 'campaign_id' value in debug log

docs/configuration.rst

@@ -2440,6 +2440,20 @@ Description
     Note: This requires 1 additional HTTP request per post.
 
 
+extractor.patreon.campaign-id
+-----------------------------
+Type
+    ``string``
+Default
+    ``"auto"``
+Description
+    Alternative way of specifying the ``campaign_id`` value of a creator
+    in case the automatic extraction method no longer functions.
+
+    Another way of specifying this value is using a ``c`` or ``campaign_id``
+    URL query parameter, e,g, ``https://www.patreon.com/NAME?c=12345``.
+
+
 extractor.patreon.files
 -----------------------
 Type

gallery_dl/extractor/patreon.py

@@ -267,34 +267,52 @@ def __init__(self, match):
 
     def posts(self):
         query = text.parse_query(self.query)
+        campaign_id = self._get_campaign_id(query)
+        filters = self._get_filters(query)
 
-        creator_id = query.get("u")
-        if creator_id:
-            url = "{}/user/posts?u={}".format(self.root, creator_id)
+        self.log.debug("campaign_id: %s", campaign_id)
+
+        url = self._build_url("posts", (
+            "&filter[campaign_id]=" + campaign_id +
+            "&filter[contains_exclusive_posts]=true"
+            "&filter[is_draft]=false" + filters +
+            "&sort=" + query.get("sort", "-published_at")
+        ))
+        return self._pagination(url)
+
+    def _get_campaign_id(self, query):
+        campaign_id = self.config("campaign-id")
+        if campaign_id and campaign_id != "auto":
+            return str(campaign_id)
+
+        campaign_id = query.get("c") or query.get("campaign_id")
+        if campaign_id:
+            return campaign_id
+
+        user_id = query.get("u")
+        if user_id:
+            url = "{}/user/posts?u={}".format(self.root, user_id)
         else:
             url = "{}/{}/posts".format(self.root, self.creator)
         page = self.request(url, notfound="creator").text
 
         try:
+            data = None
             data = self._extract_bootstrap(page)
-            campaign_id = data["campaign"]["data"]["id"]
-        except (KeyError, ValueError):
-            raise exception.NotFoundError("creator")
-
-        filters = "".join(
+            return data["campaign"]["data"]["id"]
+        except (KeyError, ValueError) as exc:
+            self.log.debug(data)
+            raise exception.StopExtraction(
+                "Unable to extract campaign ID (%s: %s)",
+                exc.__class__.__name__, exc)
+
+    def _get_filters(self, query):
+        return "".join(
             "&filter[{}={}".format(key[8:], text.escape(value))
             for key, value in query.items()
             if key.startswith("filters[")
         )
 
-        url = self._build_url("posts", (
-            "&filter[campaign_id]=" + campaign_id +
-            "&filter[contains_exclusive_posts]=true"
-            "&filter[is_draft]=false" + filters +
-            "&sort=" + query.get("sort", "-published_at")
-        ))
-        return self._pagination(url)
-
 
 class PatreonUserExtractor(PatreonExtractor):
     """Extractor for media from creators supported by you"""