[booru] improve error handling

e.g. for posts without a valid 'file_url' (#1176)

CHANGELOG.md

@@ -1,5 +1,7 @@
 # Changelog
 
+## Unreleased
+
 ## 1.16.0 - 2020-12-12
 ### Additions
 - [booru] implement generalized extractors for `*booru` and `moebooru` sites

gallery_dl/extractor/booru.py

@@ -30,7 +30,11 @@ class BooruExtractor(Extractor):
         for post in self.posts():
             try:
                 url = self._prepare_post(post, extended_tags)
-            except KeyError:
+                if url[0] == "/":
+                    url = self.root + url
+            except Exception:
+                self.log.debug("Unable to fetch download URL for post %s "
+                               "(md5: %s)", post.get("id"), post.get("md5"))
                 continue
             post.update(data)
             text.nameext_from_url(url, post)

gallery_dl/extractor/moebooru.py

@@ -25,8 +25,6 @@ class MoebooruExtractor(BooruExtractor):
 
     def _prepare_post(self, post, extended_tags=False):
         url = post["file_url"]
-        if url[0] == "/":
-            url = self.root + url
         if extended_tags:
             self._fetch_extended_tags(post)
         post["date"] = text.parse_timestamp(post["created_at"])

gallery_dl/extractor/sankaku.py

@@ -38,8 +38,6 @@ class SankakuExtractor(BooruExtractor):
 
     def _prepare_post(self, post, extended_tags=False):
         url = post["file_url"]
-        if url[0] == "/":
-            url = self.root + url
         if extended_tags:
             self._fetch_extended_tags(post)
         post["date"] = text.parse_timestamp(post["created_at"]["s"])

gallery_dl/version.py

@@ -6,4 +6,4 @@
 # it under the terms of the GNU General Public License version 2 as
 # published by the Free Software Foundation.
 
-__version__ = "1.16.0"
+__version__ = "1.16.1-dev"