PICARD-1589: Support language for ID3 comments.

Adds syntax "comment:{language}:{description}" in addition to existing "comment:{description}" for comment tag names.

picard/formats/id3.py

@@ -46,6 +46,7 @@ from picard.util import (
     encode_filename,
     sanitize_date,
 )
+from picard.util.tags import parse_comment_tag
 
 
 id3.GRP1 = compatid3.GRP1
@@ -254,7 +255,11 @@ class ID3File(File):
                 elif frameid == 'COMM':
                     for text in frame.text:
                         if text:
-                            metadata.add('%s:%s' % (name, frame.desc), text)
+                            if frame.lang == 'eng':
+                                name = '%s:%s' % (name, frame.desc)
+                            else:
+                                name = '%s:%s:%s' % (name, frame.lang, frame.desc)
+                            metadata.add(name, text)
                 else:
                     metadata.add(name, frame)
             elif frameid == 'TIT1':
@@ -411,12 +416,12 @@ class ID3File(File):
                     else:
                         tmcl.people.append([role, value])
             elif name.startswith('comment:'):
-                desc = name.split(':', 1)[1]
+                (lang, desc) = parse_comment_tag(name)
                 if desc.lower()[:4] == 'itun':
                     tags.delall('COMM:' + desc)
                     tags.add(id3.COMM(encoding=0, desc=desc, lang='eng', text=[v + '\x00' for v in values]))
                 else:
-                    tags.add(id3.COMM(encoding=encoding, desc=desc, lang='eng', text=values))
+                    tags.add(id3.COMM(encoding=encoding, desc=desc, lang=lang, text=values))
             elif name.startswith('lyrics:') or name == 'lyrics':
                 if ':' in name:
                     desc = name.split(':', 1)[1]
@@ -528,10 +533,11 @@ class ID3File(File):
                                 if people[0] == role:
                                     frame.people.remove(people)
                 elif name.startswith('comment:'):
-                    desc = name.split(':', 1)[1]
+                    (lang, desc) = parse_comment_tag(name)
                     if desc.lower()[:4] != 'itun':
                         for key, frame in list(tags.items()):
-                            if frame.FrameID == 'COMM' and frame.desc == desc:
+                            if (frame.FrameID == 'COMM' and frame.desc == desc
+                                and frame.lang == lang):
                                 del tags[key]
                 elif name.startswith('lyrics:') or name == 'lyrics':
                     if ':' in name:

picard/util/tags.py

@@ -17,6 +17,9 @@
 # along with this program; if not, write to the Free Software
 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
 
+import re
+
+
 TAG_NAMES = {
     'acoustid_fingerprint': N_('AcoustID Fingerprint'),
     'acoustid_id': N_('AcoustID'),
@@ -32,7 +35,7 @@ TAG_NAMES = {
     'barcode': N_('Barcode'),
     'bpm': N_('BPM'),
     'catalognumber': N_('Catalog Number'),
-    'comment:': N_('Comment'),
+    'comment': N_('Comment'),
     'compilation': N_('Compilation (iTunes)'),
     'composer': N_('Composer'),
     'composersort': N_('Composer Sort Order'),
@@ -127,3 +130,19 @@ def display_tag_name(name):
         if desc:
             return '%s [%s]' % (_(TAG_NAMES.get(name, name)), desc)
     return _(TAG_NAMES.get(name, name))
+
+
+RE_COMMENT_LANG = re.compile('^([a-zA-Z]{3}):')
+def parse_comment_tag(name):  # noqa: E302
+    """
+    Parses a tag name like "comment:XXX:desc", where XXX is the language.
+    If language is not set ("comment:desc") "eng" is assumed as default.
+    Returns a (lang, desc) tuple.
+    """
+    desc = name.split(':', 1)[1]
+    lang = 'eng'
+    match = RE_COMMENT_LANG.match(desc)
+    if match:
+        lang = match.group(1)
+        desc = desc[4:]
+    return (lang, desc)

test/formats/common.py

@@ -87,6 +87,7 @@ TAGS = {
     'catalognumber': 'Foo',
     'comment:': 'Foo',
     'comment:foo': 'Foo',
+    'comment:deu:foo': 'Foo',
     'compilation': '1',
     'composer': 'Foo',
     'composersort': 'Foo',

test/formats/test_id3.py

@@ -101,14 +101,18 @@ class CommonId3Tests:
         def test_comment_delete(self):
             metadata = Metadata(self.tags)
             metadata['comment:bar'] = 'Foo'
+            metadata['comment:XXX:withlang'] = 'Foo'
             original_metadata = save_and_load_metadata(self.filename, metadata)
             del metadata['comment:bar']
+            del metadata['comment:XXX:withlang']
             new_metadata = save_and_load_metadata(self.filename, metadata)
 
             self.assertIn('comment:foo', original_metadata)
             self.assertIn('comment:bar', original_metadata)
+            self.assertIn('comment:XXX:withlang', original_metadata)
             self.assertIn('comment:foo', new_metadata)
             self.assertNotIn('comment:bar', new_metadata)
+            self.assertNotIn('comment:XXX:withlang', new_metadata)
 
         @skipUnlessTestfile
         def test_id3v23_simple_tags(self):

test/test_util_tags.py

@@ -0,0 +1,18 @@
+# -*- coding: utf-8 -*-
+from test.picardtestcase import PicardTestCase
+
+from picard.util.tags import (
+    display_tag_name,
+    parse_comment_tag,
+)
+
+
+class UtilTagsTest(PicardTestCase):
+    def test_display_tag_name(self):
+        self.assertEqual('Artist', display_tag_name('artist'))
+        self.assertEqual('Lyrics', display_tag_name('lyrics:'))
+        self.assertEqual('Comment [Foo]', display_tag_name('comment:Foo'))
+
+    def test_parse_comment_tag(self):
+        self.assertEqual(('XXX', 'foo'), parse_comment_tag('comment:XXX:foo'))
+        self.assertEqual(('eng', 'foo'), parse_comment_tag('comment:foo'))