Fixes and Improvement parsing

CHANGELOG.md

@@ -1,3 +1,16 @@
+* [v3.17.0](https://github.com/newt-sc/a4kSubtitles/releases/tag/service.subtitles.a4ksubtitles%2Fservice.subtitles.a4ksubtitles-3.17.0):
+  * Fix: TV year being pulled incorrectly
+  * Fix: File name issues in both temp and media directories
+  * Fix: Subtitle file missing extension
+  * Fix: Incorrect episode selection when downloaded archive contains multiple subtitles
+  * Fix: Auto-download not working after the first selected episode in the playlist (A4K only works for the first media)
+  * Fix: Subtitle import issue due to "illegal characters"
+  * Improve: determination of subtitle episodes
+  * Improve: results parsing
+  * Feature: Auto-download now copies subtitles next to the video or to a custom location, based on Kodi's subtitle storage mode path
+  * SubSource: Now supports series in "absolute order", as used by some anime websites
+  * SubSource: Fixed issue of duplicated subtitle IDs with different names
+
 * [v3.16.1](https://github.com/newt-sc/a4kSubtitles/releases/tag/service.subtitles.a4ksubtitles%2Fservice.subtitles.a4ksubtitles-3.16.1):
   * Fix addons.xml.crc  
 

a4kSubtitles/download.py

@@ -1,4 +1,7 @@
 # -*- coding: utf-8 -*-
+subtitles_exts = ['.srt', '.sub']
+subtitles_exts_secondary = ['.smi', '.ssa', '.aqt', '.jss', '.ass', '.rt', '.txt']
+subtitles_exts_all = subtitles_exts + subtitles_exts_secondary
 
 def __download(core, filepath, request):
     request['stream'] = True
@@ -7,6 +10,10 @@ def __download(core, filepath, request):
             core.shutil.copyfileobj(r.raw, f)
 
 def __extract_gzip(core, archivepath, filename):
+    if not any(filename.lower().endswith(ext) for ext in subtitles_exts_all):
+        # For now, we will use 'srt' to mark unknown file extensions as subtitles.
+        filename = filename + ".srt"
+
     filepath = core.os.path.join(core.utils.temp_dir, filename)
 
     if core.utils.py2:
@@ -25,8 +32,8 @@ def __extract_gzip(core, archivepath, filename):
     return filepath
 
 def __extract_zip(core, archivepath, filename, episodeid):
-    sub_exts = ['.srt', '.sub']
-    sub_exts_secondary = ['.smi', '.ssa', '.aqt', '.jss', '.ass', '.rt', '.txt']
+    sub_exts = subtitles_exts
+    sub_exts_secondary = subtitles_exts_secondary
 
     try:
         using_libvfs = False
@@ -39,9 +46,13 @@ def __extract_zip(core, archivepath, filename, episodeid):
         (dirs, files) = core.kodi.xbmcvfs.listdir('archive://%s' % archivepath_)
         namelist = [file.decode(core.utils.default_encoding) if core.utils.py2 else file for file in files]
 
-    subfile = core.utils.find_file_in_archive(core, namelist, sub_exts, episodeid)
-    if not subfile:
-        subfile = core.utils.find_file_in_archive(core, namelist, sub_exts_secondary, episodeid)
+    subfile = core.utils.find_file_in_archive(core, namelist, sub_exts + sub_exts_secondary, episodeid)
+
+    if subfile:
+        # Add the subtitle file extension.
+        subfilename_and_ext = subfile.rsplit(".", 1)
+        if len(subfilename_and_ext) > 1:
+            filename = filename + "." + subfilename_and_ext[-1]
 
     dest = core.os.path.join(core.utils.temp_dir, filename)
     if not subfile:
@@ -67,9 +78,15 @@ def __extract_zip(core, archivepath, filename, episodeid):
     return dest
 
 def __insert_lang_code_in_filename(core, filename, lang_code):
-    filename_chunks = core.utils.strip_non_ascii_and_unprintable(filename).split('.')
-    filename_chunks.insert(-1, lang_code)
-    return '.'.join(filename_chunks)
+    name = core.utils.strip_non_ascii_and_unprintable(filename)
+    nameparts = name.rsplit(".", 1)
+
+    # Because this can be called via "raw" subtitles where sub ext exists we will ensure it ends with the subtitle ext.
+    # Otherwise we will use "filename.lang_code" later the ext will be added on unzip process.
+    if len(nameparts) > 1 and ("." + nameparts[1] in subtitles_exts_all):
+        return ".".join([nameparts[0], lang_code, nameparts[1]])
+
+    return "{0}.{1}".format(name, lang_code)
 
 def __postprocess(core, filepath, lang_code):
     try:
@@ -111,6 +128,24 @@ def __postprocess(core, filepath, lang_code):
             f.write(text.encode(core.utils.default_encoding))
     except: pass
 
+def __copy_sub_local(core, subfile):
+    # Copy the subfile to local.
+    if core.os.getenv('A4KSUBTITLES_TESTRUN') == 'true':
+        return
+
+    media_name = core.os.path.splitext(core.os.path.basename(core.kodi.xbmc.getInfoLabel('Player.Filename')))[0]
+    sub_name, lang_code, extension = core.os.path.basename(subfile).rsplit(".", 2)
+    file_dest, folder_dest = None, None
+    if core.kodi.get_kodi_setting("subtitles.storagemode") == 0:
+        folder_dest = core.kodi.xbmc.getInfoLabel('Player.Folderpath')
+        file_dest = core.os.path.join(folder_dest, ".".join([media_name, lang_code, extension]))
+    elif core.kodi.get_kodi_setting("subtitles.storagemode") == 1:
+        folder_dest = core.kodi.get_kodi_setting("subtitles.custompath")
+        file_dest = core.os.path.join(folder_dest, ".".join([media_name, lang_code, extension]))
+
+    if file_dest and core.kodi.xbmcvfs.exists(folder_dest):
+        core.kodi.xbmcvfs.copy(subfile, file_dest)
+
 def download(core, params):
     core.logger.debug(lambda: core.json.dumps(params, indent=2))
 
@@ -120,6 +155,7 @@ def download(core, params):
     actions_args = params['action_args']
     lang_code = core.utils.get_lang_id(actions_args['lang'], core.kodi.xbmc.ISO_639_2)
     filename = __insert_lang_code_in_filename(core, actions_args['filename'], lang_code)
+    filename = core.utils.slugify_filename(filename)
     archivepath = core.os.path.join(core.utils.temp_dir, 'sub.zip')
 
     service_name = params['service_name']
@@ -140,6 +176,7 @@ def download(core, params):
     __postprocess(core, filepath, lang_code)
 
     if core.api_mode_enabled:
+        __copy_sub_local(core, filepath)
         return filepath
 
     listitem = core.kodi.xbmcgui.ListItem(label=filepath, offscreen=True)

a4kSubtitles/lib/kodi_mock.py

@@ -28,6 +28,7 @@
 xbmc.ENGLISH_NAME = 'name'
 
 __player = lambda: None
+__player.isPlayingVideo = lambda: None
 __player.getPlayingFile = lambda: ''
 __player.getAvailableSubtitleStreams = lambda: []
 __player.setSubtitles = lambda s: None

a4kSubtitles/lib/utils.py

@@ -61,6 +61,9 @@ def strip_non_ascii_and_unprintable(text):
     result = ''.join(char for char in text if char in string.printable)
     return result.encode('ascii', errors='ignore').decode('ascii', errors='ignore')
 
+def slugify_filename(text):
+    return re.sub(r'[\\/*?:"<>|]', '_', text)
+
 def get_lang_id(language, lang_format):
     try:
         return get_lang_ids([language], lang_format)[0]
@@ -173,15 +176,16 @@ def get_json(path, filename):
     with open_file_wrapper(json_path)() as json_result:
         return json.load(json_result)
 
-def find_file_in_archive(core, namelist, exts, part_of_filename=''):
+def find_file_in_archive(core, namelist, exts, episode_number=''):
     first_ext_match = None
     exact_file = None
     for file in namelist:
         file_lower = file.lower()
         if any(file_lower.endswith(ext) for ext in exts):
+            sub_meta = extract_season_episode(file_lower, True)
             if not first_ext_match:
                 first_ext_match = file
-            if (part_of_filename == '' or part_of_filename in file_lower):
+            if (episode_number == '' or sub_meta.episode == episode_number):
                 exact_file = file
                 break
 
@@ -212,3 +216,46 @@ def extract_zipfile_member(zipfile, filename, dest):
         except:
             filename = filename.encode(default_encoding).decode(py3_zip_missing_utf8_flag_fallback_encoding)
             return zipfile.extract(filename, dest)
+
+def extract_season_episode(filename, episode_fallback=False, zfill=3):
+    episode_pattern = r'(?:e|ep.?|episode.?)(\d{1,5})'
+    season_pattern = r'(?:s|season.?)(\d{1,5})'
+    combined_pattern = r'\b(?:s|season)(\d{1,5})\s?[x|\-|\_|\s]\s?[a-z]?(\d{1,5})\b'
+    range_episodes_pattern = r'\b(?:.{1,4}e|ep|eps|episodes|\s)?(\d{1,5}?)(?:v.?)?\s?[\-|\~]\s?(\d{1,5})(?:v.?)?\b'
+    date_pattern = r'\b(\d{2,4}-\d{1,2}-\d{2,4})\b'
+
+    filename = re.sub(date_pattern, "", filename)
+    season_match = re.search(season_pattern, filename, re.IGNORECASE)
+    episode_match = re.search(episode_pattern, filename, re.IGNORECASE)
+    combined_match = re.search(combined_pattern, filename, re.IGNORECASE)
+    range_episodes_match = re.findall(range_episodes_pattern, filename, re.IGNORECASE)
+
+    season = season_match.group(1) if season_match else None
+    episode = episode_match.group(1) if episode_match else None
+    episodes_range = range(0)
+
+    if combined_match:
+        season = season if season else combined_match.group(1)
+        episode = episode if episode else combined_match.group(2)
+
+    if range_episodes_match:
+        range_start, range_end = map(int, range_episodes_match[-1])
+        episodes_range = range(range_start, range_end)
+
+    if episode_fallback and not episode:
+        # If no matches found, attempt to capture episode-like sequences
+        fallback_pattern = re.compile(r'\bE?P?(\d{1,5})v?\d?\b', re.IGNORECASE)
+        filename = re.sub(r'[\s\.\:\;\(\)\[\]\{\}\\\/\&\€\'\`\#\@\=\$\?\!\%\+\-\_\*\^]', " ", filename)
+        fallback_matches = fallback_pattern.findall(filename)
+
+        if fallback_matches:
+            # Assuming the last number in the fallback matches is the episode number
+            episode = fallback_matches[-1].lstrip("0").zfill(zfill)
+
+    return DictAsObject(
+        {
+            "season": season.lstrip("0").zfill(zfill) if season else "",
+            "episode": episode.lstrip("0").zfill(zfill) if episode else "",
+            "episodes_range": episodes_range
+        }
+    )

a4kSubtitles/lib/video.py

@@ -274,7 +274,8 @@ def __update_info_from_imdb(core, meta, pagination_token=''):
                 meta.episode = str(result['series']['episodeNumber']['episodeNumber'])
         else:
             meta.tvshow = result['titleText']['text']
-            meta.tvshow_year = str(result['releaseDate']['year'])
+            if meta.tvshow_year == '':
+                meta.tvshow_year = str(result['releaseDate']['year'])
 
             episodes = result['episodes']['result']['edges']
             s_number = int(meta.season)
@@ -316,15 +317,11 @@ def __get_basic_info():
         if regex_result:
             meta.imdb_id = regex_result.group(1)
 
-    if meta.season == '':
-        regex_result = re.search(r'.*season=(\d{1,}).*', filename_and_path, re.IGNORECASE)
-        if regex_result:
-            meta.season = regex_result.group(1)
-
-    if meta.episode == '':
-        regex_result = re.search(r'.*episode=(\d{1,}).*', filename_and_path, re.IGNORECASE)
-        if regex_result:
-            meta.episode = regex_result.group(1)
+    if meta.season == '' or meta.episode == '':
+        filename_info = utils.extract_season_episode(meta.filename, zfill=0)
+        filename_path_info = utils.extract_season_episode(filename_and_path, zfill=0)
+        meta.season = meta.season or filename_path_info.season or filename_info.season
+        meta.episode = meta.episode or filename_path_info.episode or filename_info.episode
 
     return meta