Don’t silence syntax errors and other exceptions that occur while rea…

…ding metadata
getnikola · Apr 14, 2018 · e03fc62 · e03fc62
1 parent 8b43762
commit e03fc62
Show file tree

Hide file tree

Showing 3 changed files with 26 additions and 23 deletions.
diff --git a/CHANGES.txt b/CHANGES.txt
@@ -40,6 +40,8 @@ Features
 Bugfixes
 --------
 
+* Don’t silence syntax errors and other exceptions that occur while
+  reading metadata
 * Use documented dateutil API for time zone list (Issue #3006)
 * Handle trailing slash redirects with query strings correctly in
   ``nikola serve`` (Issue #3000)

diff --git a/nikola/metadata_extractors.py b/nikola/metadata_extractors.py
@@ -126,7 +126,7 @@ def default_metadata_extractors_by() -> dict:
     return d
 
 
-def _register_default(extractor: MetadataExtractor) -> MetadataExtractor:
+def _register_default(extractor: type) -> type:
     """Register a default extractor."""
     _default_extractors.append(extractor())
     return extractor

diff --git a/nikola/post.py b/nikola/post.py
@@ -952,28 +952,6 @@ def get_metadata_from_file(source_path, post, config, lang, metadata_extractors_
             source_path += '.' + lang
         with io.open(source_path, "r", encoding="utf-8-sig") as meta_file:
             source_text = meta_file.read()
-
-        meta = {}
-        used_extractor = None
-        for priority in metadata_extractors.MetaPriority:
-            found_in_priority = False
-            for extractor in metadata_extractors_by['priority'].get(priority, []):
-                if not metadata_extractors.check_conditions(post, source_path, extractor.conditions, config, source_text):
-                    continue
-                extractor.check_requirements()
-                new_meta = extractor.extract_text(source_text)
-                if new_meta:
-                    found_in_priority = True
-                    used_extractor = extractor
-                    # Map metadata from other platforms to names Nikola expects (Issue #2817)
-                    map_metadata(new_meta, extractor.map_from, config)
-
-                    meta.update(new_meta)
-                    break
-
-            if found_in_priority:
-                break
-        return meta, used_extractor
     except (UnicodeDecodeError, UnicodeEncodeError):
         msg = 'Error reading {0}: Nikola only supports UTF-8 files'.format(source_path)
         LOGGER.error(msg)
@@ -982,6 +960,29 @@ def get_metadata_from_file(source_path, post, config, lang, metadata_extractors_
         return {}, None
 
 
+    meta = {}
+    used_extractor = None
+    for priority in metadata_extractors.MetaPriority:
+        found_in_priority = False
+        for extractor in metadata_extractors_by['priority'].get(priority, []):
+            if not metadata_extractors.check_conditions(post, source_path, extractor.conditions, config, source_text):
+                continue
+            extractor.check_requirements()
+            new_meta = extractor.extract_text(source_text)
+            if new_meta:
+                found_in_priority = True
+                used_extractor = extractor
+                # Map metadata from other platforms to names Nikola expects (Issue #2817)
+                map_metadata(new_meta, extractor.map_from, config)
+
+                meta.update(new_meta)
+                break
+
+        if found_in_priority:
+            break
+    return meta, used_extractor
+
+
 def get_metadata_from_meta_file(path, post, config, lang, metadata_extractors_by=None):
     """Take a post path, and gets data from a matching .meta file."""
     meta_path = os.path.splitext(path)[0] + '.meta'