def is_valid_mkv(path):
with open(path, 'rb') as f:
try:
enzyme.MKV(f)
return True
except enzyme.exceptions.MalformedMKVError:
return False
def check_mkv(path):
mkv = None
if not os.path.isfile(path):
return None
try:
with open(path, "rb") as f:
mkv = enzyme.MKV(f)
except (IOError, OverflowError, enzyme.exceptions.MalformedMKVError, AttributeError) as e:
print os.path.basename(path), str(e)
if mkv is None:
return None
newname = fix_makemkv_name(os.path.basename(path))
audio_info = []
for track in mkv.audio_tracks:
audio_info.append('%s %s' % (track.language, track.name))
subtitle_info = []
for track in mkv.subtitle_tracks:
subtitle_info.append('%s %s%s%s' % (track.language,
"E" if track.enabled else "",
"D" if track.default else "",
"F" if track.forced else ""))
track = mkv.video_tracks[0]
print "%s - %s - %ux%u%s %s - %s - %s" % (
newname, mkv.info.duration,
track.width, track.height, 'i' if track.interlaced else '', track.codec_id,
', '.join(audio_info), ", ".join(subtitle_info))
return mkv
def list_languages(self, file):
subtitles_list = []
if self.ffprobe:
parser = VideoFileParser(ffprobe=self.ffprobe,
includeMissing=True,
rawMode=False)
data = parser.parseFfprobe(file)
for detected_language in data['subtitles']:
subtitles_list.append([
detected_language['language'], detected_language['forced'],
detected_language["codec"]
])
else:
if os.path.splitext(file)[1] == '.mkv':
with open(file, 'rb') as f:
try:
mkv = enzyme.MKV(f)
except MalformedMKVError:
logging.error(
'BAZARR cannot analyze this MKV with our built-in MKV parser, you should install ffmpeg: '
+ file)
else:
for subtitle_track in mkv.subtitle_tracks:
subtitles_list.append([
subtitle_track.language, subtitle_track.forced,
subtitle_track.codec_id
])
return subtitles_list
def list_languages(self, file):
subtitles_list = []
if os.path.splitext(file)[1] == '.mkv':
with open(file, 'rb') as f:
mkv = enzyme.MKV(f)
for subtitle_track in mkv.subtitle_tracks:
subtitles_list.append(
[subtitle_track.language, subtitle_track.forced])
else:
if self.ffprobe:
detected_languages = []
try:
detected_languages = subprocess.check_output(
[
self.ffprobe, "-loglevel", "error",
"-select_streams", "s", "-show_entries",
"stream_tags=language", "-of", "csv=p=0",
file.encode(locale.getpreferredencoding())
],
universal_newlines=True,
stderr=subprocess.STDOUT).strip().split("\n")
except subprocess.CalledProcessError as e:
raise FFprobeError(e.output)
else:
for detected_language in detected_languages:
subtitles_list.append([detected_language, False])
# I can't get the forced flag from ffprobe so I always assume it isn't forced
return subtitles_list
def describe(self, video_path, context):
"""Return video metadata."""
try:
with open(video_path, 'rb') as f:
data = defaultdict(dict)
ff = todict(enzyme.MKV(f))
data.update(ff)
if 'info' in data and data['info'] is None:
return {}
data['info']['complete_name'] = video_path
data['info']['file_size'] = os.path.getsize(video_path)
except enzyme.MalformedMKVError: # pragma: no cover
logger.warning('Invalid file %r', video_path)
if context.get('fail_on_error'):
raise MalformedFileError
return {}
if context.get('raw'):
return data
result = self._describe_tracks(data.get('info'),
data.get('video_tracks'),
data.get('audio_tracks'),
data.get('subtitle_tracks'), context)
result['provider'] = 'Enzyme {0}'.format(enzyme.__version__)
return result
def list_languages(self, file):
subtitles_list = []
if self.ffprobe:
api.initialize({'provider': 'ffmpeg', 'ffmpeg': self.ffprobe})
data = api.know(file)
if 'subtitle' in data:
for detected_language in data['subtitle']:
if 'language' in detected_language:
language = detected_language['language'].alpha3
forced = detected_language[
'forced'] if 'forced' in detected_language else None
codec = detected_language[
'format'] if 'format' in detected_language else None
subtitles_list.append([language, forced, codec])
else:
continue
else:
if os.path.splitext(file)[1] == '.mkv':
with open(file, 'rb') as f:
try:
mkv = enzyme.MKV(f)
except MalformedMKVError:
logging.error(
'BAZARR cannot analyze this MKV with our built-in MKV parser, you should install ffmpeg: '
+ file)
else:
for subtitle_track in mkv.subtitle_tracks:
subtitles_list.append([
subtitle_track.language, subtitle_track.forced,
subtitle_track.codec_id
])
return subtitles_list
def list_languages(self, file):
subtitles_list = []
if self.ffprobe:
parser = VideoFileParser(ffprobe=self.ffprobe,
includeMissing=True,
rawMode=False)
data = parser.parseFfprobe(file)
for detected_language in data['subtitles']:
subtitles_list.append([
detected_language['language'], detected_language['forced'],
detected_language["codec"]
])
else:
if os.path.splitext(file)[1] == '.mkv':
with open(file, 'rb') as f:
mkv = enzyme.MKV(f)
for subtitle_track in mkv.subtitle_tracks:
subtitles_list.append([
subtitle_track.language, subtitle_track.forced,
subtitle_track.codec_id
])
return subtitles_list
def __is_uhd_mkv(target):
with open(target, 'rb') as mkv_f:
import enzyme
mkv = enzyme.MKV(mkv_f)
if len(mkv.video_tracks) > 0:
uhd_track = next((v for v in mkv.video_tracks if v.display_width > 1920), None)
return uhd_track is not None
return False
def mkv_for_filenames(**kwargs):
for fname in cli_filenames(**kwargs):
with open(fname, 'rb') as f:
try:
mkv = enzyme.MKV(f)
except enzyme.exceptions.MalformedMKVError as ex:
print('Error parsing %s: %s' % (fname, ex))
continue
yield (fname, mkv)
def rename_files(movie_dir, kind="movie"):
print("Renaming files in: {}".format(movie_dir))
files = sorted(
[x for x in os.listdir(movie_dir) if os.path.isfile(os.path.join(movie_dir, x)) and x != "MANUALLY_NAMED"],
key=functools.cmp_to_key(logical_sort))
prefix, imdb_id = create_file_name_prefix(os.path.basename(movie_dir), kind, files)
runtime = int(get_imdb_runtime(imdb_id))
main_file_found = 0
filename_maps = []
for old_file in files:
old_name, extension = os.path.splitext(old_file)
new_name = prefix
is_main_file = False
movie = None
if extension == ".mkv":
if len(files) == 1:
new_name += " - Main file"
else:
with open(os.path.join(movie_dir, old_file), "rb") as fi:
try:
movie = enzyme.MKV(fi)
except:
movie = None
if movie and (movie.info.duration.seconds - (movie.info.duration.seconds % 60)) / 60 in [runtime - 1,
runtime,
runtime + 1]:
if not main_file_found:
new_name += " - Main file"
else:
new_name += " - Main file ({})".format(main_file_found)
print("\tFound main file")
main_file_found += 1
is_main_file = True
part = get_file_part_name(old_name)
if part and not is_main_file:
new_name += " - pt{}".format(part[-1])
new_name += extension
filename_maps.append((old_file, new_name, is_main_file, get_aspect_ratio(movie, new_name)))
aspect_ratios_found = list(set([x[3] for x in filename_maps if x[2]]))
for name_map in filename_maps:
old_file = name_map[0]
new_name = name_map[1]
is_main_file = name_map[2]
if is_main_file and main_file_found == 2 and len(aspect_ratios_found) == 2:
name, extension = os.path.splitext(new_name)
new_name = name.replace(" (1)", "") + " - " + name_map[3] + extension
if not is_main_file and main_file_found:
if not os.path.isdir(os.path.join(movie_dir, "Other")):
os.mkdir(os.path.join(movie_dir, "Other"))
os.rename(os.path.join(movie_dir, old_file), os.path.join(movie_dir, "Other", new_name))
elif old_file != new_name:
try:
os.rename(os.path.join(movie_dir, old_file), os.path.join(movie_dir, new_name))
except:
pdb.set_trace()
def getMatroska(vfile):
"""
Get metadata and track information from Matroska containers
"""
try:
with open(vfile) as f:
mkv = enzyme.MKV(f)
except enzyme.MalformedMKVError as e:
logexc(e, "Not a Matroska container or segment is corrupt")
return False
return mkv.to_dict()
def store_subtitles(file):
languages = []
actual_subtitles = []
if os.path.exists(file):
if os.path.splitext(file)[1] == '.mkv':
try:
with open(file, 'rb') as f:
mkv = enzyme.MKV(f)
for subtitle_track in mkv.subtitle_tracks:
try:
if alpha2_from_alpha3(subtitle_track.language) != None:
actual_subtitles.append([str(alpha2_from_alpha3(subtitle_track.language)),None])
except:
pass
except:
pass
brazilian_portuguese = [".pt-br", ".pob", "pb"]
try:
subtitles = core.search_external_subtitles(file)
except:
pass
else:
for subtitle, language in subtitles.iteritems():
if str(os.path.splitext(subtitle)[0]).lower().endswith(tuple(brazilian_portuguese)) is True:
actual_subtitles.append([str("pb"), path_replace_reverse(os.path.join(os.path.dirname(file), subtitle))])
elif str(language) != 'und':
actual_subtitles.append([str(language), path_replace_reverse(os.path.join(os.path.dirname(file), subtitle))])
else:
with open(path_replace(os.path.join(os.path.dirname(file), subtitle)), 'r') as f:
text = list(islice(f, 100))
text = ' '.join(text)
encoding = UnicodeDammit(text)
try:
text = text.decode(encoding.original_encoding)
except Exception as e:
logging.exception('Error trying to detect character encoding for this subtitles file: ' + path_replace(os.path.join(os.path.dirname(file), subtitle)) + ' You should try to delete this subtitles file manually and ask Bazarr to download it again.')
else:
detected_language = langdetect.detect(text)
if len(detected_language) > 0:
actual_subtitles.append([str(detected_language), path_replace_reverse(os.path.join(os.path.dirname(file), subtitle))])
conn_db = sqlite3.connect(os.path.join(os.path.dirname(__file__), 'data/db/bazarr.db'), timeout=30)
c_db = conn_db.cursor()
c_db.execute("UPDATE table_episodes SET subtitles = ? WHERE path = ?", (str(actual_subtitles), path_replace_reverse(file)))
conn_db.commit()
c_db.close()
return actual_subtitles
def main(args):
parser = argparse.ArgumentParser()
parser.add_argument("--select",
"-s",
action="store_true",
help="Use zenity to select witch tracks to export")
parser.add_argument(
"--notify",
action="store_true",
help="Use notify-send to notify when each jobe is done")
parser.add_argument("--dry-run",
"-n",
action="store_true",
help="Don't run the mkvextract command")
parser.add_argument("--progress",
action="store_true",
help="Use zenity to draw a progress bar")
parser.add_argument("filename", nargs="+")
args = parser.parse_args()
for filename in args.filename:
with open(filename, 'rb') as f:
mkv = enzyme.MKV(f)
subtitle_tracks = mkv.subtitle_tracks
if args.select:
cmd = construct_zenity_cmd(subtitle_tracks)
print(cmd)
selected_numbers = [
int(x) for x in subprocess.check_output(
cmd, shell=True).decode().strip().split("|")
]
print(selected_numbers)
subtitle_tracks = subtitle_track_by_number(subtitle_tracks,
selected_numbers)
if not subtitle_tracks:
print(f"No subtitle tracks found in {filename}", file=sys.stderr)
continue
print(
f"Will run mkvextract with the following tracks:\n{subtitle_tracks}\n"
)
cmd = construct_mkvextract_cmd(filename, subtitle_tracks)
if args.progress:
cmd += " | " + ZENITY_PROGRESS_CMD.format(filename=filename)
print("Command to be executed:\n" + cmd)
if not args.dry_run:
result = subprocess.call(cmd, shell=True)
result = "Done" if result == 0 else "Failed"
if args.notify:
subprocess.call(
f"notify-send 'Subtitle Extraction {result}' '{result} with {filename}'",
shell=True)
return 0
def read_mkv_file(source_file):
"""Retrieve MKV file information"""
import sys
import enzyme
from enzyme.exceptions import MalformedMKVError
with open(source_file, "rb") as mkv_source:
try:
mkv_file = enzyme.MKV(mkv_source)
except MalformedMKVError:
return SABResult(False,
error="[ERROR] {0}".format(sys.exc_info()[0]))
return SABResult(True, data=mkv_file)
def get_vid_duraiton(file_name):
"""
use enzyme library to extract video duration object
param: file_name
"""
try:
with open(file_name, "rb") as f:
meta = enzyme.MKV(f)
t_delta = meta.info.duration
except OSError:
f.close()
return None
f.close()
return t_delta
def read_mkv_file(source_file):
"""Retrieve MKV file information"""
import sys
import collections
import enzyme
from enzyme.exceptions import MalformedMKVError
result = collections.namedtuple('Result', 'mkv_file error')
with open(source_file, 'rb') as mkv_source:
try:
mkv_file = enzyme.MKV(mkv_source)
except MalformedMKVError:
return result(None, '[ERROR] %s' % sys.exc_info()[0])
return result(mkv_file, None)
def mkv_metadata(file):
"""retrieve info from MKV's metadata file"""
with open(file, 'rb') as f:
mkv = enzyme.MKV(f)
resolution = str(mkv.video_tracks[0].width) + 'x' + str(
mkv.video_tracks[0].height)
audio_channels = mkv.audio_tracks[0].channels
runtime = str(mkv.info.duration).split('.', 1)[0]
return {
'resolution': resolution,
'audio_channels': audio_channels,
'runtime': runtime
}
def store_subtitles_movie(file):
languages = []
actual_subtitles = []
if os.path.exists(file):
if os.path.splitext(file)[1] == '.mkv':
try:
with open(file, 'rb') as f:
mkv = enzyme.MKV(f)
for subtitle_track in mkv.subtitle_tracks:
try:
actual_subtitles.append([str(pycountry.languages.lookup(subtitle_track.language).alpha_2), None])
except:
pass
except:
pass
subtitles = core.search_external_subtitles(file)
for subtitle, language in subtitles.iteritems():
if str(language) != 'und':
actual_subtitles.append([str(language), path_replace_reverse_movie(os.path.join(os.path.dirname(file), subtitle))])
else:
if os.path.splitext(subtitle)[1] != ".sub":
with open(path_replace_movie(os.path.join(os.path.dirname(file), subtitle)), 'r') as f:
text = list(islice(f, 100))
text = ' '.join(text)
encoding = UnicodeDammit(text)
try:
text = text.decode(encoding.original_encoding)
except Exception as e:
logging.exception('Error trying to detect character encoding for this subtitles file: ' + path_replace_movie(os.path.join(os.path.dirname(file), subtitle)) + ' You should try to delete this subtitles file manually and ask Bazarr to download it again.')
else:
detected_language = langdetect.detect(text)
if len(detected_language) > 0:
actual_subtitles.append([str(detected_language), path_replace_reverse_movie(os.path.join(os.path.dirname(file), subtitle))])
conn_db = sqlite3.connect(os.path.join(os.path.dirname(__file__), 'data/db/bazarr.db'), timeout=30)
c_db = conn_db.cursor()
c_db.execute("UPDATE table_movies SET subtitles = ? WHERE path = ?", (str(actual_subtitles), path_replace_reverse_movie(file)))
conn_db.commit()
c_db.close()
return actual_subtitles
def prepare_subtitles(self, s, offset=None, keep_on_screen=False):
if s.is_internal and not keep_on_screen:
full_path = os.path.join(s.directory, s.file_name)
with open(full_path, 'rb') as fd:
mkv = enzyme.MKV(fd)
# TODO: check which track number instead of iterate
for sub_track in mkv.subtitle_tracks:
sub_track = mkv.subtitle_tracks[0]
if sub_track.codec_id == 'S_TEXT/UTF8':
new_file = 'subtitle_{id}.srt'.format(id=s.id)
cmd = [
'mkvextract', 'tracks', full_path,
'{n}:{e}'.format(n=sub_track.number - 1,
e=new_file)
]
subprocess.check_output(cmd)
subtitle_path = new_file
break
if not s.is_internal:
subtitle_path = os.path.join(s.directory, s.file_name)
if keep_on_screen:
subtitle_path = '{id}_keep.srt'.format(id=s.id)
call_command('export_subtitles', s.id, subtitle_path, True)
res = subtitle_path
output = subprocess.check_output(['file', subtitle_path])
cmd = None
extension = s.extension if s.extension != 'mkv' else 'srt'
new_name = '{id}.utf8.{ext}'.format(id=s.id, ext=extension)
if 'ISO-8859' in output or 'Non-ISO extended-ASCII' in output:
cmd = 'iconv --from-code=ISO-8859-1 --to-code=UTF-8 "{s}" > "{n}"'\
.format(s=subtitle_path, n=new_name)
elif 'ASCII' in output:
cmd = 'iconv --from-code=ASCII --to-code=UTF-8 "{s}" > "{n}"'\
.format(s=subtitle_path, n=new_name)
if cmd:
os.system(cmd)
res = new_name
if offset:
split = new_name.split('.')
n = '.'.join(split[:-1])
new_name = '{n}.ss.{ext}'.format(n=n, ext=split[-1])
cmd = 'ffmpeg -i "{res}" -ss {offset} -f {ext} -y "{nn}"'.format(
res=res, offset=offset, ext=split[-1], nn=new_name)
os.system(cmd)
res = new_name
return res
def list_languages(self, file):
from utils import get_binary
self.ffprobe = get_binary("ffprobe")
subtitles_list = []
if self.ffprobe:
api.initialize({'provider': 'ffmpeg', 'ffmpeg': self.ffprobe})
data = api.know(file)
traditional_chinese = ["cht", "tc", "traditional", "zht", "hant", "big5", u"繁", u"雙語"]
brazilian_portuguese = ["pt-br", "pob", "pb", "brazilian", "brasil", "brazil"]
if 'subtitle' in data:
for detected_language in data['subtitle']:
if 'language' in detected_language:
language = detected_language['language'].alpha3
if language == 'zho' and 'name' in detected_language:
if any (ext in (detected_language['name'].lower()) for ext in traditional_chinese):
language = 'zht'
if language == 'por' and 'name' in detected_language:
if any (ext in (detected_language['name'].lower()) for ext in brazilian_portuguese):
language = 'pob'
forced = detected_language['forced'] if 'forced' in detected_language else False
hearing_impaired = detected_language['hearing_impaired'] if 'hearing_impaired' in \
detected_language else False
codec = detected_language['format'] if 'format' in detected_language else None
subtitles_list.append([language, forced, hearing_impaired, codec])
else:
continue
else:
if os.path.splitext(file)[1] == '.mkv':
with open(file, 'rb') as f:
try:
mkv = enzyme.MKV(f)
except MalformedMKVError:
logging.error('BAZARR cannot analyze this MKV with our built-in MKV parser, you should install ffmpeg: ' + file)
else:
for subtitle_track in mkv.subtitle_tracks:
hearing_impaired = False
if subtitle_track.name:
if 'sdh' in subtitle_track.name.lower():
hearing_impaired = True
subtitles_list.append([subtitle_track.language, subtitle_track.forced, hearing_impaired,
subtitle_track.codec_id])
return subtitles_list
def parse_video(file, guid):
with open(file, 'rb') as f:
mkv = enzyme.MKV(f)
length = mkv.info.duration
bitrate = mkv.audio_tracks[0].sampling_frequency
channels = mkv.audio_tracks[0].channels
f_format = get_extension(file)
width = mkv.video_tracks[0].width
height = mkv.video_tracks[0].height
with connection.cursor() as cursor:
cursor.execute(
"""
INSERT INTO video_metadata (
file_guid, file_format, length, bit_rate, mono_or_stereo, width, height
) VALUES (
%s, %s, %s, %s, %s, %s, %s
)
""", [guid, f_format, length, bitrate, channels, width, height])
return True
def get_mkv_ttid_map(self):
mkv_ttid_map = dict()
for path in Path(self.download_dir).rglob('*.mkv'):
try:
with open(path, 'rb') as f:
mkv = enzyme.MKV(f)
if mkv.tags:
for x in mkv.tags:
for y in x.simpletags:
if y.name == 'TTID':
mkv_ttid_map[y.string] = str(path)
raise GetOutOfLoop
except GetOutOfLoop:
pass
except enzyme.MalformedMKVError as ex:
self.logger.warning("Exception while parsing file {}".format(str(path)))
self.logger.warning("You may want to delete and re-download this file.")
self.logger.warning("Exception: {}".format(ex))
pass
return mkv_ttid_map
def weebify_mkv(mkv_path, noop=False):
print('Opening {}'.format(mkv_path), file=sys.stderr)
with open(mkv_path, 'rb') as fp:
mkv = enzyme.MKV(fp)
jpn_audio_track = find_jpn_audio_track(mkv)
def_audio_track = find_default_audio_track(mkv)
set_default_sub = should_avoid_default_sub(mkv)
set_default_audio = jpn_audio_track and jpn_audio_track != def_audio_track
track_changes = {track.number: [] for track in mkv.audio_tracks + mkv.subtitle_tracks}
if set_default_audio:
if noop:
print('Would flag audio track {} as default'.format(jpn_audio_track.number), file=sys.stderr)
track_changes[jpn_audio_track.number].append('flag-default=1')
for track in mkv.audio_tracks:
if track != jpn_audio_track and track.default:
if noop:
print('Would unflag audio track {} as default'.format(track.number), file=sys.stderr)
track_changes[track.number].append('flag-default=0')
if set_default_sub:
primary_eng_sub = find_eng_sub(mkv)
if noop:
print('Would flag subtitle track {} as default'.format(primary_eng_sub.number), file=sys.stderr)
track_changes[primary_eng_sub.number].append('flag-default=1')
for track in mkv.subtitle_tracks:
if track != primary_eng_sub and track.default:
if noop:
print('Would unflag subtitle track {} as default'.format(track.number), file=sys.stderr)
track_changes[track.number].append('flag-default=0')
args = build_args(track_changes)
if args:
if noop:
return 1
else:
subprocess.check_call(['mkvpropedit', '-v', mkv_path, '-v'] + args)
return 0
else:
print('Nothing to do', file=sys.stderr)
return 0
def has_subtitle(self, language):
""" Returns true if the video has already
a subtitle for a specific language. """
has_subtitle = False
# Look for embedded subtitle in mkv video
if Video.is_mkv(self.signature):
with open(self.filename, 'rb') as file_handler:
mkv_video = enzyme.MKV(file_handler)
for sub in mkv_video.subtitle_tracks:
try:
if sub.language and \
Language.fromalpha3b(sub.language) == language:
has_subtitle = True
break
elif sub.name and \
Language.fromname(sub.name) == language:
has_subtitle = True
break
except BabelfishError:
LOG.error("Embedded subtitle track"
"language {} is not a valid language".format(
sub.language))
# Look for external subtitle
dir_name = os.path.dirname(self.filename)
base_name, _ = os.path.splitext(os.path.basename(self.filename))
search_subtitle = os.path.join(
dir_name, "{}.{}.*".format(base_name, language.alpha2))
existing_subtitles = [
sub_file for sub_file in glob.glob(search_subtitle)
if os.path.splitext(sub_file)[1] in Subtitle.EXTENSIONS
]
if existing_subtitles:
has_subtitle = True
return has_subtitle
def list_languages(self, file):
if self.ffprobe:
try:
if not settings.general.getboolean('ignore_pgs_subs'):
subtitle_languages = subprocess.check_output([self.ffprobe, "-loglevel", "error", "-select_streams", "s", "-show_entries", "stream_tags=language", "-of", "csv=p=0", file.encode(locale.getpreferredencoding())], universal_newlines=True, stderr=subprocess.STDOUT).strip()
if not subtitle_languages:
return []
return subtitle_languages.split('\n')
subtitle_tracks = subprocess.check_output([self.ffprobe, "-loglevel", "error", "-select_streams", "s", "-show_entries", "stream=codec_name:stream_tags=language", "-of", "csv=p=0", file.encode(locale.getpreferredencoding())], universal_newlines=True, stderr=subprocess.STDOUT).strip()
if not subtitle_tracks:
return []
subtitle_tracks = subtitle_tracks.split('\n')
return [lang for (sub_type, lang) in map(lambda subtitle_track: subtitle_track.split(','), subtitle_tracks) if sub_type != 'hdmv_pgs_subtitle']
except subprocess.CalledProcessError as e:
raise FFprobeError(e.output)
if os.path.splitext(file)[1] != '.mkv':
raise NotMKVAndNoFFprobe()
with open(file, 'rb') as f:
mkv = enzyme.MKV(f)
if not settings.general.getboolean('ignore_pgs_subs'):
return [subtitle_track.language for subtitle_track in mkv.subtitle_tracks]
return [subtitle_track.language for subtitle_track in mkv.subtitle_tracks if subtitle_track.codec_id != "S_HDMV/PGS"]
def collect_mkv(self, directory, file_name):
full_path = os.path.join(directory, file_name)
with open(full_path, 'rb') as fd:
try:
mkv = enzyme.MKV(fd)
except Exception as e:
print(full_path)
print(e)
return
if mkv.subtitle_tracks:
# TODO: collect all subtitles
s = mkv.subtitle_tracks[0]
# TODO: collect S_TEXT/ASS too
if s.codec_id == 'S_TEXT/UTF8':
temp_file = 'temp_subtitle.srt'
cmd = [
'mkvextract', 'tracks', full_path,
'{n}:{e}'.format(n=s.number - 1, e=temp_file)
]
subprocess.check_output(cmd)
# TODO: guess the language
lang = 'simple'
rel_mediafile = MediaFile.objects.filter(
directory=directory, file_name=file_name).first()
subtitle_file = SubtitlesFile.objects.create(
file_name=file_name,
directory=directory,
extension='mkv',
mediafile=rel_mediafile,
language=lang)
self.collect_subtitles_lines(subtitle_file, temp_file)
os.remove(temp_file)
return subtitle_file
else:
print 'Found other codec id {id} in {f}'.format(
id=s.codec_id, f=full_path)
def scan_video(path, subtitles=True, embedded_subtitles=True, original=None):
"""Scan a video and its subtitle languages from a video `path`
:param string path: absolute path to the video
:param bool subtitles: scan for subtitles with the same name
:param bool embedded_subtitles: scan for embedded subtitles
:return: the scanned video
:rtype: :class:`Video`
:raise: ValueError if cannot guess enough information from the path
"""
if not original:
original = path
dirpath, filename = os.path.split(path)
logger.info('Scanning video %r in %r', filename, dirpath)
video = Video.fromguess(path,
guessit.guess_file_info(original, 'autodetect'))
video.size = os.path.getsize(path)
if video.size > 10485760:
logger.debug('Size is %d', video.size)
video.hashes['opensubtitles'] = hash_opensubtitles(path)
video.hashes['thesubdb'] = hash_thesubdb(path)
logger.debug('Computed hashes %r', video.hashes)
else:
logger.warning('Size is lower than 10MB: hashes not computed')
if subtitles:
video.subtitle_languages |= scan_subtitle_languages(path)
# enzyme
try:
if filename.endswith('.mkv'):
with open(path, 'rb') as f:
mkv = enzyme.MKV(f)
if mkv.video_tracks:
video_track = mkv.video_tracks[0]
# resolution
if video_track.height in (480, 720, 1080):
if video_track.interlaced:
video.resolution = '%di' % video_track.height
logger.debug('Found resolution %s with enzyme',
video.resolution)
else:
video.resolution = '%dp' % video_track.height
logger.debug('Found resolution %s with enzyme',
video.resolution)
# video codec
if video_track.codec_id == 'V_MPEG4/ISO/AVC':
video.video_codec = 'h264'
logger.debug('Found video_codec %s with enzyme',
video.video_codec)
elif video_track.codec_id == 'V_MPEG4/ISO/SP':
video.video_codec = 'DivX'
logger.debug('Found video_codec %s with enzyme',
video.video_codec)
elif video_track.codec_id == 'V_MPEG4/ISO/ASP':
video.video_codec = 'XviD'
logger.debug('Found video_codec %s with enzyme',
video.video_codec)
else:
logger.warning('MKV has no video track')
if mkv.audio_tracks:
audio_track = mkv.audio_tracks[0]
# audio codec
if audio_track.codec_id == 'A_AC3':
video.audio_codec = 'AC3'
logger.debug('Found audio_codec %s with enzyme',
video.audio_codec)
elif audio_track.codec_id == 'A_DTS':
video.audio_codec = 'DTS'
logger.debug('Found audio_codec %s with enzyme',
video.audio_codec)
elif audio_track.codec_id == 'A_AAC':
video.audio_codec = 'AAC'
logger.debug('Found audio_codec %s with enzyme',
video.audio_codec)
else:
logger.warning('MKV has no audio track')
if mkv.subtitle_tracks:
# embedded subtitles
if embedded_subtitles:
embedded_subtitle_languages = set()
for st in mkv.subtitle_tracks:
if st.language:
try:
embedded_subtitle_languages.add(
babelfish.Language.fromalpha3b(
st.language))
except babelfish.Error:
logger.error(
'Embedded subtitle track language %r is not a valid language',
st.language)
embedded_subtitle_languages.add(
babelfish.Language('und'))
elif st.name:
try:
embedded_subtitle_languages.add(
babelfish.Language.fromname(st.name))
except babelfish.Error:
logger.debug(
'Embedded subtitle track name %r is not a valid language',
st.name)
embedded_subtitle_languages.add(
babelfish.Language('und'))
else:
embedded_subtitle_languages.add(
babelfish.Language('und'))
logger.debug('Found embedded subtitle %r with enzyme',
embedded_subtitle_languages)
video.subtitle_languages |= embedded_subtitle_languages
else:
logger.debug('MKV has no subtitle track')
except enzyme.Error:
logger.exception('Parsing video metadata with enzyme failed')
return video
def guess_video_metadata(filename):
"""Gets the video metadata properties out of a given file. The file needs to
exist on the filesystem to be able to be analyzed. An empty guess is
returned otherwise.
You need to have the Enzyme python package installed for this to work."""
result = Guess()
def found(prop, value):
result[prop] = value
log.debug('Found with enzyme %s: %s' % (prop, value))
# first get the size of the file, in bytes
try:
size = os.stat(filename).st_size
found('fileSize', size)
except Exception as e:
log.error('Cannot get video file size: %s' % e)
# file probably does not exist, we might as well return now
return result
# then get additional metadata from the file using enzyme, if available
try:
import enzyme
with open(filename) as f:
mkv = enzyme.MKV(f)
found('duration', mkv.info.duration.total_seconds())
if mkv.video_tracks:
video_track = mkv.video_tracks[0]
# resolution
if video_track.height in (480, 720, 1080):
if video_track.interlaced:
found('screenSize', '%di' % video_track.height)
else:
found('screenSize', '%dp' % video_track.height)
else:
# TODO: do we want this?
#found('screenSize', '%dx%d' % (video_track.width, video_track.height))
pass
# video codec
if video_track.codec_id == 'V_MPEG4/ISO/AVC':
found('videoCodec', 'h264')
elif video_track.codec_id == 'V_MPEG4/ISO/SP':
found('videoCodec', 'DivX')
elif video_track.codec_id == 'V_MPEG4/ISO/ASP':
found('videoCodec', 'XviD')
else:
log.warning('MKV has no video track')
if mkv.audio_tracks:
audio_track = mkv.audio_tracks[0]
# audio codec
if audio_track.codec_id == 'A_AC3':
found('audioCodec', 'AC3')
elif audio_track.codec_id == 'A_DTS':
found('audioCodec', 'DTS')
elif audio_track.codec_id == 'A_AAC':
found('audioCodec', 'AAC')
else:
log.warning('MKV has no audio track')
if mkv.subtitle_tracks:
embedded_subtitle_languages = set()
for st in mkv.subtitle_tracks:
try:
if st.language:
lang = babelfish.Language.fromalpha3b(st.language)
elif st.name:
lang = babelfish.Language.fromname(st.name)
else:
lang = babelfish.Language('und')
except babelfish.Error:
lang = babelfish.Language('und')
embedded_subtitle_languages.add(lang)
found('subtitleLanguage', embedded_subtitle_languages)
else:
log.debug('MKV has no subtitle track')
return result
except ImportError:
log.error('Cannot get video file metadata, missing dependency: enzyme')
log.error(
'Please install it from PyPI, by doing eg: pip install enzyme')
return result
except IOError as e:
log.error('Could not open file: %s' % filename)
log.error(
'Make sure it exists and is available for reading on the filesystem'
)
log.error('Error: %s' % e)
return result
except enzyme.Error as e:
log.error('Cannot guess video file metadata')
log.error('enzyme.Error while reading file: %s' % filename)
log.error('Error: %s' % e)
return result
def parse_video_metadata(file,
file_size,
episode_file_id=None,
movie_file_id=None):
# Define default data keys value
data = {
'ffprobe': {},
'enzyme': {},
'file_id': episode_file_id if episode_file_id else movie_file_id,
'file_size': file_size
}
# Get the actual cache value form database
if episode_file_id:
cache_key = database.execute(
'SELECT ffprobe_cache FROM table_episodes WHERE episode_file_id=? AND file_size=?',
(episode_file_id, file_size),
only_one=True)
elif movie_file_id:
cache_key = database.execute(
'SELECT ffprobe_cache FROM table_movies WHERE movie_file_id=? AND file_size=?',
(movie_file_id, file_size),
only_one=True)
else:
cache_key = None
# check if we have a value for that cache key
if not isinstance(cache_key, dict):
return data
else:
try:
# Unpickle ffprobe cache
cached_value = pickle.loads(cache_key['ffprobe_cache'])
except:
pass
else:
# Check if file size and file id matches and if so, we return the cached value
if cached_value['file_size'] == file_size and cached_value[
'file_id'] in [episode_file_id, movie_file_id]:
return cached_value
# if not, we retrieve the metadata from the file
from utils import get_binary
ffprobe_path = get_binary("ffprobe")
# if we have ffprobe available
if ffprobe_path:
api.initialize({'provider': 'ffmpeg', 'ffmpeg': ffprobe_path})
data['ffprobe'] = api.know(file)
# if nto, we use enzyme for mkv files
else:
if os.path.splitext(file)[1] == '.mkv':
with open(file, 'rb') as f:
try:
mkv = enzyme.MKV(f)
except MalformedMKVError:
logging.error(
'BAZARR cannot analyze this MKV with our built-in MKV parser, you should install '
'ffmpeg/ffprobe: ' + file)
else:
data['enzyme'] = mkv
# we write to db the result and return the newly cached ffprobe dict
if episode_file_id:
database.execute(
'UPDATE table_episodes SET ffprobe_cache=? WHERE episode_file_id=?',
(pickle.dumps(data, pickle.HIGHEST_PROTOCOL), episode_file_id))
elif movie_file_id:
database.execute(
'UPDATE table_movies SET ffprobe_cache=? WHERE movie_file_id=?',
(pickle.dumps(data, pickle.HIGHEST_PROTOCOL), movie_file_id))
return data
def extract_info(cls, video_path):
"""Extract info from the video."""
with open(video_path, 'rb') as f:
return to_dict(enzyme.MKV(f))
