release: v1.2.0

CHANGELOG.md

@@ -1,3 +1,6 @@
+* [v1.2.0](https://github.com/newt-sc/a4kSubtitles/releases/tag/service.subtitles.a4ksubtitles%2Fservice.subtitles.a4ksubtitles-1.2.0):
+  * Use internal ZipFile for extraction with a fallback to vfs.libarchive
+
 * [v1.1.0](https://github.com/newt-sc/a4kSubtitles/releases/tag/service.subtitles.a4ksubtitles%2Fservice.subtitles.a4ksubtitles-1.1.0):
   * Improve imdb id scraping
 

a4kSubtitles/core.py

@@ -11,9 +11,11 @@
 import time
 import hashlib
 import re
+import zipfile
 from datetime import datetime
 from base64 import b64encode
 from xml.etree import ElementTree
+from io import BytesIO
 
 from .lib import (
     cache,

a4kSubtitles/download.py

@@ -28,10 +28,20 @@ def __extract_zip(core, archivepath, filename, episodeid):
     sub_exts = ['.srt', '.sub']
     sub_exts_secondary = ['.smi', '.ssa', '.aqt', '.jss', '.ass', '.rt', '.txt']
 
-    archivepath_ = core.utils.quote_plus(archivepath)
-    subfile = core.utils.find_file_in_archive(core, archivepath_, sub_exts, episodeid)
+    try:
+        using_libvfs = False
+        with open(archivepath, 'rb') as f:
+            zipfile = core.zipfile.ZipFile(core.BytesIO(f.read()))
+        namelist = core.utils.get_zipfile_namelist(zipfile)
+    except:
+        using_libvfs = True
+        archivepath_ = core.utils.quote_plus(archivepath)
+        (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, archivepath_, sub_exts_secondary, episodeid)
+        subfile = core.utils.find_file_in_archive(core, namelist, sub_exts_secondary, episodeid)
 
     dest = core.os.path.join(core.utils.temp_dir, filename)
     if not subfile:
@@ -41,12 +51,17 @@ def __extract_zip(core, archivepath, filename, episodeid):
             core.os.rename(archivepath, dest)
             return dest
 
-    src = 'archive://' + archivepath_ + '/' + subfile
-    core.kodi.xbmcvfs.copy(src, dest)
+    if not using_libvfs:
+        src = core.utils.extract_zipfile_member(zipfile, subfile, core.utils.temp_dir)
+        core.os.rename(src, dest)
+    else:
+        src = 'archive://' + archivepath_ + '/' + subfile
+        core.kodi.xbmcvfs.copy(src, dest)
+
     return dest
 
 def __insert_lang_code_in_filename(core, filename, lang):
-    filename_chunks = filename.split('.')
+    filename_chunks = core.utils.strip_non_ascii_and_unprintable(filename).split('.')
     lang_code = core.kodi.xbmc.convertLanguage(lang, core.kodi.xbmc.ISO_639_2)
     filename_chunks.insert(-1, lang_code)
     return '.'.join(filename_chunks)

a4kSubtitles/lib/utils.py

@@ -18,7 +18,7 @@
     from urllib.parse import quote_plus, unquote, parse_qsl
     from io import StringIO
     import queue
-    unicode = lambda v, e: v
+    unicode = lambda v: v
 
 __url_regex = r'[a-z0-9][a-z0-9-]{0,5}[a-z0-9]\.[a-z0-9]{2,20}\.[a-z]{2,5}'
 __credit_part_regex = r'(sync|synced|fix|fixed|corrected|corrections)'
@@ -29,6 +29,9 @@
 cp1251_garbled = u'аеио'.encode('cp1251').decode('raw_unicode_escape')
 koi8r_garbled = u'аеио'.encode('koi8-r').decode('raw_unicode_escape')
 
+zip_utf8_flag = 0x800
+py3_zip_missing_utf8_flag_fallback_encoding = 'cp437'
+
 py2 = sys.version_info[0] == 2
 py3 = not py2
 
@@ -134,15 +137,10 @@ def get_json(path, filename):
     with open(json_path) as json_result:
         return json.load(json_result)
 
-def find_file_in_archive(core, archivepath, exts, part_of_filename=''):
-    (dirs, files) = core.kodi.xbmcvfs.listdir('archive://%s' % archivepath)
-
+def find_file_in_archive(core, namelist, exts, part_of_filename=''):
     first_ext_match = None
     exact_file = None
-    for file in files:
-        if core.utils.py2:
-            file = file.decode('utf8')
-
+    for file in namelist:
         file_lower = file.lower()
         if any(file_lower.endswith(ext) for ext in exts):
             if not first_ext_match:
@@ -152,3 +150,29 @@ def find_file_in_archive(core, archivepath, exts, part_of_filename=''):
                 break
 
     return exact_file if exact_file is not None else first_ext_match
+
+def get_zipfile_namelist(zipfile):
+    infolist = zipfile.infolist()
+    namelist = []
+
+    if py2:
+        for info in infolist:
+            namelist.append(info.filename.decode(default_encoding))
+    else:
+        for info in infolist:
+            filename = info.filename
+            if not info.flag_bits & zip_utf8_flag:
+                filename = info.filename.encode(py3_zip_missing_utf8_flag_fallback_encoding).decode(default_encoding)
+            namelist.append(filename)
+
+    return namelist
+
+def extract_zipfile_member(zipfile, filename, dest):
+    if py2:
+        return zipfile.extract(filename.encode(default_encoding), dest)
+    else:
+        try:
+            return zipfile.extract(filename, dest)
+        except:
+            filename = filename.encode(default_encoding).decode(py3_zip_missing_utf8_flag_fallback_encoding)
+            return zipfile.extract(filename, dest)

addon.xml

@@ -1,7 +1,7 @@
 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
 <addon id="service.subtitles.a4ksubtitles"
        name="a4kSubtitles"
-       version="1.1.0"
+       version="1.2.0"
        provider-name="Unknown">
     <requires>
       <import addon="xbmc.python" version="2.25.0"/>
@@ -27,6 +27,9 @@ Supports: OpenSubtitles, BSPlayer, Podnadpisi.NET, SubDB, Subscene, Addic7ed
             <screenshot>screenshot-03.png</screenshot>
         </assets>
         <news>
+[v1.2.0]:
+  * Use internal ZipFile for extraction with a fallback to vfs.libarchive
+
 [v1.1.0]:
   * Improve imdb id scraping
 

packages/addons.xml

@@ -4,7 +4,7 @@
 <addons>
 <addon id="service.subtitles.a4ksubtitles"
        name="a4kSubtitles"
-       version="1.1.0"
+       version="1.2.0"
        provider-name="Unknown">
     <requires>
       <import addon="xbmc.python" version="2.25.0"/>
@@ -30,6 +30,9 @@ Supports: OpenSubtitles, BSPlayer, Podnadpisi.NET, SubDB, Subscene, Addic7ed
             <screenshot>screenshot-03.png</screenshot>
         </assets>
         <news>
+[v1.2.0]:
+  * Use internal ZipFile for extraction with a fallback to vfs.libarchive
+
 [v1.1.0]:
   * Improve imdb id scraping
 

packages/addons.xml.crc

@@ -1 +1 @@
-cf0e071b1e659fd1525d23b9445319824f58c943
+d114c77b06cd3bc4c8e557687867387d148363d4

tests/test_suite.py

@@ -45,6 +45,19 @@
     "subdb_hash": "2aec1b70afe702e67ab39a0af776ba5a",
 }
 
+__tvshow_unicode_video_meta = {
+    "year": "2019",
+    "title": "In the Dark Night of the Soul It's Always 3:30 in the Morning",
+    "tvshow": "The Morning Show",
+    "imdb_id": "tt7203552",
+    "season": "1",
+    "episode": "1",
+    "filename": "The Morning Show (2019) S01E01 In the Dark Night of the Soul it's Always 330 in the Morning.mkv",
+    "filesize": 2135679767,
+    "filehash": "631fb55f3db5709c",
+    "subdb_hash": "b35c5f8e4afffd8ee09442e98f943410",
+}
+
 def __remove_meta_cache(a4ksubtitles_api):
     try:
         os.remove(a4ksubtitles_api.core.cache.__meta_cache_filepath)
@@ -79,7 +92,7 @@ def __search(a4ksubtitles_api, settings={}, video_meta={}):
     }
 
     search.settings = {
-        'general.timeout': '20',
+        'general.timeout': '30',
         'general.results_limit': '20',
         'opensubtitles.enabled': 'false',
         'opensubtitles.username': '',
@@ -109,6 +122,11 @@ def __search_tvshow(a4ksubtitles_api, settings={}, video_meta={}):
     tvshow_video_meta.update(video_meta)
     return __search(a4ksubtitles_api, settings, tvshow_video_meta)
 
+def __search_unicode_tvshow(a4ksubtitles_api, settings={}, video_meta={}):
+    tvshow_video_meta = __tvshow_unicode_video_meta.copy()
+    tvshow_video_meta.update(video_meta)
+    return __search(a4ksubtitles_api, settings, tvshow_video_meta)
+
 def test_api():
     def get_error_msg(e):
         return str(e.value).replace('\'', '')
@@ -221,6 +239,29 @@ def test_opensubtitles_tvshow():
 
     assert filepath != ''
 
+def test_opensubtitles_unicode_tvshow():
+    a4ksubtitles_api = api.A4kSubtitlesApi({'kodi': True})
+    __remove_all_cache(a4ksubtitles_api)
+
+    # search
+    settings = {
+        'opensubtitles.enabled': 'true',
+    }
+    search = __search_unicode_tvshow(a4ksubtitles_api, settings)
+
+    # download
+    item = search.results[0]
+
+    params = {
+        'action': 'download',
+        'service_name': 'opensubtitles',
+        'action_args': item['action_args']
+    }
+
+    filepath = a4ksubtitles_api.download(params, search.settings)
+
+    assert filepath != ''
+
 def test_opensubtitles_missing_imdb_id():
     a4ksubtitles_api = api.A4kSubtitlesApi({'kodi': True})
     __remove_all_cache(a4ksubtitles_api)