PICARD-2935: apply genre filters and threshold before selecting minimal usage

picard/track.py

@@ -112,10 +112,17 @@ class TagGenreFilter:
                 return True
         return False
 
-    def filter(self, counter):
-        for name, count in counter:
+    def filter(self, counter: Counter, minusage=0) -> Counter:
+        result = Counter()
+        for name, count in counter.items():
             if not self.skip(name):
-                yield (name, count)
+                result[name] = count
+        topcount = result.most_common(1)[0][1]
+        for name, count in counter.items():
+            percent = 100 * count // topcount
+            if percent < minusage:
+                del result[name]
+        return result
 
     def format_errors(self):
         fmt = _("Error line %(lineno)d: %(error)s")
@@ -321,18 +328,13 @@ class Track(DataObject, FileListItem):
         if not genres:
             return []
 
-        # Find most common genres
-        most_common_genres = genres.most_common(limit)
-        topcount = most_common_genres[0][1]
-
         # Filter by name and usage
         genres_filter = TagGenreFilter(filters)
-        genres_list = []
-        for name, count in genres_filter.filter(most_common_genres):
-            percent = 100 * count // topcount
-            if percent < minusage:
-                break
-            genres_list.append(name.title())
+        genres = genres_filter.filter(genres, minusage=minusage)
+
+        # Find most common genres
+        most_common_genres = genres.most_common(limit)
+        genres_list = [name.title() for name, _count in most_common_genres]
         genres_list.sort()
 
         # And generate the genre metadata tag

test/test_taggenrefilter.py

@@ -20,6 +20,7 @@
 # along with this program; if not, write to the Free Software
 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
 
+from collections import Counter
 
 from test.picardtestcase import PicardTestCase
 
@@ -168,5 +169,12 @@ class TagGenreFilterTest(PicardTestCase):
 
     def test_filter_method(self):
         tag_filter = TagGenreFilter("-a*")
-        result = list(tag_filter.filter([("ax", 1), ("bx", 2), ("ay", 3), ("by", 4)]))
-        self.assertEqual([('bx', 2), ('by', 4)], result)
+        genres = Counter(ax=1, bx=2, ay=3, by=4)
+        result = tag_filter.filter(genres)
+        self.assertEqual([('bx', 2), ('by', 4)], list(result.items()))
+
+    def test_filter_method_minusage(self):
+        tag_filter = TagGenreFilter("-a*")
+        genres = Counter(ax=4, bx=5, ay=20, by=10, bz=4)
+        result = tag_filter.filter(genres, minusage=50)
+        self.assertEqual([('bx', 5), ('by', 10)], list(result.items()))

test/test_track.py

@@ -62,6 +62,12 @@ class TrackGenres2MetadataTest(PicardTestCase):
         ret = Track._genres_to_metadata(genres, limit=0)
         self.assertEqual(ret, [])
 
+    def test_limit_after_filter(self):
+        genres = Counter(rock=5, blues=7, pop=1, psychedelic=3)
+        filters = '-rock'
+        ret = Track._genres_to_metadata(genres, limit=3, filters=filters)
+        self.assertEqual(ret, ['Blues', 'Pop', 'Psychedelic'])
+
     def test_minusage(self):
         genres = Counter(pop=6, rock=7, blues=2)
         ret = Track._genres_to_metadata(genres, minusage=10)