Truncate tags_string until it fits in 255 bytes (#286)

* truncate tags_string until it fits in 255 bytes

IA's subject field has a 255 bytes length limit

* test: add test for tags_string truncation

* style: autopep8

* fix: test: tags_string truncate to 255 bytes (not chars)

Author: yzqzss

tests/test_tubeup.py

@@ -298,7 +298,7 @@ def test_generate_ydl_options_with_verbose_mode(self):
     def test_create_archive_org_metadata_from_youtubedl_meta(self):
         with open(get_testfile_path(
                 'Mountain_3_-_Video_Background_HD_1080p-6iRV8liah8A.info.json')
-                ) as f:
+        ) as f:
             vid_meta = json.load(f)
 
         result = TubeUp.create_archive_org_metadata_from_youtubedl_meta(
@@ -332,7 +332,7 @@ def test_create_archive_org_metadata_from_youtubedl_meta(self):
     def test_create_archive_org_metadata_from_youtubedl_meta_description_text_null(self):
         with open(get_testfile_path(
                 'description_text_null.json')
-                ) as f:
+        ) as f:
             vid_meta = json.load(f)
 
         result = TubeUp.create_archive_org_metadata_from_youtubedl_meta(
@@ -419,7 +419,7 @@ def test_create_archive_org_metadata_from_youtubedl_meta_no_date(self):
     def test_create_archive_org_metadata_from_youtubedl_meta_twitch_clips(self):
         with open(get_testfile_path(
                 'EA_Play_2016_Live_from_the_Novo_Theatre-42850523.info.json')
-                ) as f:
+        ) as f:
             vid_meta = json.load(f)
 
         result = TubeUp.create_archive_org_metadata_from_youtubedl_meta(
@@ -441,6 +441,21 @@ def test_create_archive_org_metadata_from_youtubedl_meta_twitch_clips(self):
 
         self.assertEqual(expected_result, result)
 
+    def test_create_archive_org_metadata_from_youtubedl_meta_mass_of_tags(self):
+        with open(get_testfile_path(
+                'Mountain_3_-_Video_Background_HD_1080p-6iRV8liah8A.info.json')
+        ) as f:
+            vid_meta = json.load(f)
+
+        vid_meta['tags'] = [f't{i}' for i in range(0, 300)]
+
+        result = TubeUp.create_archive_org_metadata_from_youtubedl_meta(
+            vid_meta
+        )
+
+        self.assertLessEqual(len(result['subject'].encode(encoding='utf-8')),
+                             255, msg='tags_string not truncated to <= 255 bytes')
+
     def test_get_resource_basenames(self):
         tu = TubeUp(dir_path=os.path.join(current_path,
                                           'test_tubeup_rootdir'))

tubeup/TubeUp.py

@@ -517,6 +517,12 @@ def create_archive_org_metadata_from_youtubedl_meta(vid_meta):
             except Exception:
                 print("Unable to process tags successfully.")
 
+        # IA's subject field has a 255 bytes length limit, so we need to truncate tags_string
+        while len(tags_string.encode('utf-8')) > 255:
+            tags_list = tags_string.split(';')
+            tags_list.pop()
+            tags_string = ';'.join(tags_list)
+
         # license
         licenseurl = TubeUp.determine_licenseurl(vid_meta)