def test_pprint(self):
frame = CTOC(element_id=u"foo", flags=3,
child_element_ids=[u"ch0"],
sub_frames=[TPE2(encoding=3, text=[u"foo"])])
self.assertEqual(
frame.pprint(),
"CTOC=foo flags=3 child_element_ids=ch0\n TPE2=foo")
def test_write(self):
frame = CTOC(element_id=u"foo", flags=3,
child_element_ids=[u"ch0"],
sub_frames=[TPE2(encoding=3, text=[u"f", u"b"])])
config = ID3SaveConfig(3, "/")
data = (b"foo\x00\x03\x01ch0\x00TPE2\x00\x00\x00\x0b\x00\x00\x01"
b"\xff\xfef\x00/\x00b\x00\x00\x00")
self.assertEqual(frame._writeData(config), data)
def generate_ctoc(num_chaps):
"""Generates a CTOC frame for `num_chaps` chapters."""
return CTOC(element_id=u"toc",
flags=CTOCFlags.TOP_LEVEL | CTOCFlags.ORDERED,
child_element_ids=id_list_gen(num_chaps),
sub_frames=[
TIT2(text=u"Primary Chapter List"),
])
def test_ctoc_subframes(self):
id3 = ID3()
id3.version = (2, 3)
id3.add(CTOC(sub_frames=[TYER(encoding=0, text="2006")]))
id3.update_to_v24()
ctoc = id3.getall("CTOC")[0]
self.assertEqual(ctoc.sub_frames.getall("TDRC")[0], u"2006")
self.assertFalse(ctoc.sub_frames.getall("TYER"))
id3.update_to_v23()
self.assertEqual(ctoc.sub_frames.getall("TYER")[0], u"2006")
def test_save_v23_recurse_restore(self):
self.realid3.add(CHAP(sub_frames=[TDRC(text="2006")]))
self.realid3.add(CTOC(sub_frames=[TDRC(text="2006")]))
self.id3.save(self.filename, v2_version=3)
for frame_id in ["CHAP", "CTOC"]:
chap = self.realid3.getall(frame_id)[0]
assert chap.sub_frames.getall("TDRC")[0] == "2006"
new = ID3(self.filename, translate=False)
assert new.version == (2, 3, 0)
chap = new.getall(frame_id)[0]
assert not chap.sub_frames.getall("TDRC")
assert chap.sub_frames.getall("TYER")[0] == "2006"
def add_chapters(filepath, chapters):
"""Adds chapters to the given audio file"""
audio = ID3(filepath)
# Adding table of contents
audio.add(
CTOC(element_id=u"toc",
flags=CTOCFlags.TOP_LEVEL | CTOCFlags.ORDERED,
child_element_ids=[
u"chp" + str(i + 1) for i in range(len(chapters))
],
sub_frames=[TIT2(text=[u"Table of Contents"])]))
# Adding chapters
for i in range(len(chapters) - 1):
add_chapter(audio, chapters[i][0], chapters[i + 1][0], chapters[i][1],
i + 1)
length = MutagenFile(filepath).info.length * 1000
add_chapter(audio, chapters[-1][0], length, chapters[-1][1], len(chapters))
audio.save()
def test_unknown_chap(self):
# add ctoc
id3 = ID3(self.filename)
id3.add(CTOC(element_id="foo", flags=3, child_element_ids=["ch0"],
sub_frames=[TIT2(encoding=3, text=["bla"])]))
id3.save()
# pretend we don't know ctoc and save
id3 = ID3(self.filename, known_frames={"CTOC": CTOC})
ctoc = id3.getall("CTOC")[0]
self.assertFalse(ctoc.sub_frames)
self.assertTrue(ctoc.sub_frames.unknown_frames)
id3.save()
# make sure we wrote all sub frames back
id3 = ID3(self.filename)
self.assertEqual(
id3.getall("CTOC")[0].sub_frames.getall("TIT2")[0].text, ["bla"])
def run():
parser = argparse.ArgumentParser(
prog='odmpy',
description='Download/return an Overdrive loan audiobook',
epilog=
'Version {version}. [Python {py_major}.{py_minor}.{py_micro}-{platform}] '
'Source at https://github.com/ping/odmpy/'.format(
version=__version__,
py_major=sys.version_info.major,
py_minor=sys.version_info.minor,
py_micro=sys.version_info.micro,
platform=sys.platform,
))
parser.add_argument('-v',
'--verbose',
dest='verbose',
action='store_true',
help='Enable more verbose messages for debugging')
parser.add_argument(
'-t',
'--timeout',
dest='timeout',
type=int,
default=10,
help='Timeout (seconds) for network requests. Default 10.')
subparsers = parser.add_subparsers(
title='Available commands',
dest='subparser_name',
help='To get more help, use the -h option with the command.')
parser_info = subparsers.add_parser(
'info',
description='Get information about a loan file.',
help='Get information about a loan file')
parser_info.add_argument('odm_file', type=str, help='ODM file path')
parser_dl = subparsers.add_parser('dl',
description='Download from a loan file.',
help='Download from a loan file')
parser_dl.add_argument('-d',
'--downloaddir',
dest='download_dir',
default='.',
help='Download folder path')
parser_dl.add_argument('-c',
'--chapters',
dest='add_chapters',
action='store_true',
help='Add chapter marks (experimental)')
parser_dl.add_argument(
'-m',
'--merge',
dest='merge_output',
action='store_true',
help='Merge into 1 file (experimental, requires ffmpeg)')
parser_dl.add_argument(
'--mergeformat',
dest='merge_format',
choices=['mp3', 'm4b'],
default='mp3',
help='Merged file format (m4b is slow, experimental, requires ffmpeg)')
parser_dl.add_argument(
'-k',
'--keepcover',
dest='always_keep_cover',
action='store_true',
help='Always generate the cover image file (cover.jpg)')
parser_dl.add_argument('-f',
'--keepmp3',
dest='keep_mp3',
action='store_true',
help='Keep downloaded mp3 files (after merging)')
parser_dl.add_argument('-j',
'--writejson',
dest='write_json',
action='store_true',
help='Generate a meta json file (for debugging)')
parser_dl.add_argument('odm_file', type=str, help='ODM file path')
parser_ret = subparsers.add_parser('ret',
description='Return a loan file.',
help='Return a loan file.')
parser_ret.add_argument('odm_file', type=str, help='ODM file path')
args = parser.parse_args()
if args.verbose:
logger.setLevel(logging.DEBUG)
try:
# test for odm file
args.odm_file
except AttributeError:
parser.print_help()
exit(0)
xml_doc = xml.etree.ElementTree.parse(args.odm_file)
root = xml_doc.getroot()
# Return Book
if args.subparser_name == 'ret':
logger.info('Returning {} ...'.format(args.odm_file))
early_return_url = root.find('EarlyReturnURL').text
try:
early_return_res = requests.get(early_return_url,
headers={'User-Agent': UA_LONG},
timeout=10)
early_return_res.raise_for_status()
logger.info('Loan returned successfully: {}'.format(args.odm_file))
except HTTPError as he:
if he.response.status_code == 403:
logger.warning('Loan is probably already returned.')
sys.exit()
logger.error(
'Unexpected HTTPError while trying to return loan {}'.format(
args.odm_file))
logger.error('HTTPError: {}'.format(str(he)))
logger.debug(he.response.content)
sys.exit(1)
except ConnectionError as ce:
logger.error('ConnectionError: {}'.format(str(ce)))
sys.exit(1)
sys.exit()
metadata = None
for t in root.itertext():
if not t.startswith('<Metadata>'):
continue
metadata = xml.etree.ElementTree.fromstring(
# remove invalid & char
re.sub(r'\s&\s', ' & ', t))
break
debug_meta = {}
title = metadata.find('Title').text
cover_url = metadata.find(
'CoverUrl').text if metadata.find('CoverUrl') != None else ''
authors = [
unescape_html(c.text) for c in metadata.find('Creators')
if 'Author' in c.attrib.get('role', '')
]
if not authors:
authors = [
unescape_html(c.text) for c in metadata.find('Creators')
if 'Editor' in c.attrib.get('role', '')
]
if not authors:
authors = [unescape_html(c.text) for c in metadata.find('Creators')]
publisher = metadata.find('Publisher').text
description = metadata.find('Description').text if metadata.find(
'Description') is not None else ''
debug_meta['meta'] = {
'title': title,
'coverUrl': cover_url,
'authors': authors,
'publisher': publisher,
'description': description,
}
# View Book Info
if args.subparser_name == 'info':
logger.info(u'{:10} {}'.format('Title:', colored.blue(title)))
logger.info(u'{:10} {}'.format(
'Creators:',
colored.blue(u', '.join([
u'{} ({})'.format(c.text, c.attrib['role'])
for c in metadata.find('Creators')
]))))
logger.info(u'{:10} {}'.format('Publisher:',
metadata.find('Publisher').text))
logger.info(u'{:10} {}'.format(
'Subjects:',
u', '.join([c.text for c in metadata.find('Subjects')])))
logger.info(u'{:10} {}'.format(
'Languages:',
u', '.join([c.text for c in metadata.find('Languages')])))
logger.info(u'{:10} \n{}'.format('Description:',
metadata.find('Description').text))
for formats in root.findall('Formats'):
for f in formats:
logger.info(u'\n{:10} {}'.format('Format:', f.attrib['name']))
parts = f.find('Parts')
for p in parts:
logger.info('* {} - {} ({:,.0f}kB)'.format(
p.attrib['name'], p.attrib['duration'],
math.ceil(1.0 * int(p.attrib['filesize']) / 1024)))
sys.exit()
# Download Book
download_baseurl = ''
download_parts = []
for formats in root.findall('Formats'):
for f in formats:
protocols = f.find('Protocols')
for p in protocols:
if p.attrib.get('method', '') != 'download':
continue
download_baseurl = p.attrib['baseurl']
break
parts = f.find('Parts')
for p in parts:
download_parts.append(p.attrib)
debug_meta['download_parts'] = download_parts
logger.info('Downloading "{}" by "{}" in {} parts...'.format(
colored.blue(title, bold=True), colored.blue(', '.join(authors)),
len(download_parts)))
# declare book folder/file names here together so we can catch problems from too long names
book_folder = os.path.join(
args.download_dir,
u'{} - {}'.format(title.replace(os.sep, '-'),
u', '.join(authors).replace(os.sep, '-')))
# for merged mp3
book_filename = os.path.join(
book_folder,
u'{} - {}.mp3'.format(title.replace(os.sep, '-'),
u', '.join(authors).replace(os.sep, '-')))
# for merged m4b
book_m4b_filename = os.path.join(
book_folder,
u'{} - {}.m4b'.format(title.replace(os.sep, '-'),
u', '.join(authors).replace(os.sep, '-')))
if not os.path.exists(book_folder):
try:
os.makedirs(book_folder)
except OSError as exc:
if exc.errno not in (
36, 63): # ref http://www.ioplex.com/~miallen/errcmpp.html
raise
# Ref OSError: [Errno 36] File name too long https://github.com/ping/odmpy/issues/5
# create book folder, file with just the title
book_folder = os.path.join(
args.download_dir, u'{}'.format(title.replace(os.sep, '-')))
os.makedirs(book_folder)
book_filename = os.path.join(
book_folder, u'{}.mp3'.format(title.replace(os.sep, '-')))
book_m4b_filename = os.path.join(
book_folder, u'{}.m4b'.format(title.replace(os.sep, '-')))
cover_filename = os.path.join(book_folder, 'cover.jpg')
debug_filename = os.path.join(book_folder, 'debug.json')
if not os.path.isfile(cover_filename) and cover_url:
cover_res = requests.get(cover_url, headers={'User-Agent': UA})
cover_res.raise_for_status()
with open(cover_filename, 'wb') as outfile:
outfile.write(cover_res.content)
acquisition_url = root.find('License').find('AcquisitionUrl').text
media_id = root.attrib['id']
client_id = str(uuid.uuid1()).upper()
raw_hash = '{client_id}|{omc}|{os}|ELOSNOC*AIDEM*EVIRDREVO'.format(
client_id=client_id, omc=OMC, os=OS)
m = hashlib.sha1(raw_hash.encode('utf-16-le'))
license_hash = base64.b64encode(m.digest())
# Extract license
# License file is downloadable only once per odm
# so we keep it in case downloads fail
_, odm_filename = os.path.split(args.odm_file)
license_file = os.path.join(args.download_dir,
odm_filename.replace('.odm', '.license'))
if os.path.isfile(license_file):
logger.warning(
'Already downloaded license file: {}'.format(license_file))
else:
# download license file
params = OrderedDict([('MediaID', media_id), ('ClientID', client_id),
('OMC', OMC), ('OS', OS),
('Hash', license_hash)])
license_res = requests.get(acquisition_url,
params=params,
headers={'User-Agent': UA},
timeout=args.timeout,
stream=True)
try:
license_res.raise_for_status()
with open(license_file, 'wb') as outfile:
for chunk in license_res.iter_content(1024):
outfile.write(chunk)
logger.debug('Saved license file {}'.format(license_file))
except HTTPError as he:
if he.response.status_code == 404:
# odm file has expired
logger.error('The loan file "{}" has expired.'
'Please download again.'.format(args.odm_file))
else:
logger.error(he.response.content)
sys.exit(1)
except ConnectionError as ce:
logger.error('ConnectionError: {}'.format(str(ce)))
sys.exit(1)
license_xml_doc = xml.etree.ElementTree.parse(license_file)
license_root = license_xml_doc.getroot()
ns = '{http://license.overdrive.com/2008/03/License.xsd}'
license_client = license_root.find('{}SignedInfo'.format(ns)).find(
'{}ClientID'.format(ns))
license_client_id = license_client.text
lic_file_contents = ''
with open(license_file, 'r') as lic_file:
lic_file_contents = lic_file.read()
cover_bytes = None
if os.path.isfile(cover_filename):
with open(cover_filename, 'rb') as f:
cover_bytes = f.read()
track_count = 0
file_tracks = []
keep_cover = args.always_keep_cover
audio_lengths_ms = []
for p in download_parts:
part_number = int(p['number'])
part_filename = os.path.join(
book_folder, u'{}.mp3'.format(
slugify(u'{} - Part {:02d}'.format(title, part_number),
allow_unicode=True)))
part_tmp_filename = u'{}.part'.format(part_filename)
part_file_size = int(p['filesize'])
part_url_filename = p['filename']
part_download_url = '{}/{}'.format(download_baseurl, part_url_filename)
part_markers = []
if os.path.isfile(part_filename):
logger.warning('Already saved {}'.format(
colored.magenta(part_filename)))
else:
try:
part_download_res = requests.get(part_download_url,
headers={
'User-Agent': UA,
'ClientID':
license_client_id,
'License':
lic_file_contents
},
timeout=args.timeout,
stream=True)
part_download_res.raise_for_status()
chunk_size = 1024 * 1024
expected_chunk_count = math.ceil(1.0 * part_file_size /
chunk_size)
with open(part_tmp_filename, 'wb') as outfile:
for chunk in progress.bar(
part_download_res.iter_content(
chunk_size=chunk_size),
label='Part {}'.format(part_number),
expected_size=expected_chunk_count):
if chunk:
outfile.write(chunk)
# try to remux file to remove mp3 lame tag errors
cmd = [
'ffmpeg', '-y', '-nostdin', '-hide_banner', '-loglevel',
'info' if logger.level == logging.DEBUG else 'error', '-i',
part_tmp_filename, '-c:a', 'copy', '-c:v', 'copy',
part_filename
]
try:
exit_code = subprocess.call(cmd)
if exit_code:
logger.warning(
'ffmpeg exited with the code: {0!s}'.format(
exit_code))
logger.warning('Command: {0!s}'.format(' '.join(cmd)))
os.rename(part_tmp_filename, part_filename)
else:
os.remove(part_tmp_filename)
except Exception as ffmpeg_ex:
logger.warning('Error executing ffmpeg: {}'.format(
str(ffmpeg_ex)))
os.rename(part_tmp_filename, part_filename)
except HTTPError as he:
logger.error('HTTPError: {}'.format(str(he)))
logger.debug(he.response.content)
sys.exit(1)
except ConnectionError as ce:
logger.error('ConnectionError: {}'.format(str(ce)))
sys.exit(1)
try:
# Fill id3 info for mp3 part
mutagen_audio = MP3(part_filename, ID3=ID3)
if not mutagen_audio.tags:
mutagen_audio.tags = ID3()
if 'TIT2' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TIT2(encoding=3, text=u'{}'.format(title)))
if 'TIT3' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TIT3(encoding=3, text=u'{}'.format(description)))
if 'TALB' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TALB(encoding=3, text=u'{}'.format(title)))
if 'TPE1' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TPE1(encoding=3, text=u'{}'.format(authors[0])))
if 'TPE2' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TPE2(encoding=3, text=u'{}'.format(authors[0])))
if 'TRCK' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TRCK(encoding=3, text=u'{}'.format(part_number)))
if 'TPUB' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TPUB(encoding=3, text=u'{}'.format(publisher)))
if 'COMM' not in mutagen_audio.tags:
mutagen_audio.tags.add(
COMM(encoding=3,
desc=u'Description',
text=u'{}'.format(description)))
if cover_bytes:
mutagen_audio.tags.add(
APIC(encoding=3,
mime=u'image/jpeg',
type=3,
desc=u'Cover',
data=cover_bytes))
mutagen_audio.save()
audio_lengths_ms.append(
int(round(mutagen_audio.info.length * 1000)))
# Extract OD chapter info from mp3s for use in merged file
if 'TXXX:OverDrive MediaMarkers' in mutagen_audio.tags \
and mutagen_audio.tags['TXXX:OverDrive MediaMarkers'].text:
marker_text = mutagen_audio.tags[
'TXXX:OverDrive MediaMarkers'].text[0]
try:
tree = xml.etree.ElementTree.fromstring(marker_text)
except UnicodeEncodeError:
tree = xml.etree.ElementTree.fromstring(
marker_text.encode('ascii', 'ignore').decode('ascii'))
for m in tree.iter('Marker'):
marker_name = m.find('Name').text.strip()
marker_timestamp = m.find('Time').text
timestamp = None
ts_mark = 0
# 2 timestamp formats found
for r in ('%M:%S.%f', '%H:%M:%S.%f'):
try:
timestamp = time.strptime(marker_timestamp, r)
ts = datetime.timedelta(hours=timestamp.tm_hour,
minutes=timestamp.tm_min,
seconds=timestamp.tm_sec)
ts_mark = int(1000 * ts.total_seconds())
break
except ValueError:
pass
if not timestamp:
# some invalid timestamp string, e.g. 60:15.00
mobj = re.match(MARKER_TIMESTAMP_HHMMSS,
marker_timestamp)
if mobj:
ts_mark = int(mobj.group('hr')) * 60 * 60 * 1000 + \
int(mobj.group('min')) * 60 * 1000 + \
int(mobj.group('sec')) * 1000 + \
int(mobj.group('ms'))
else:
mobj = re.match(MARKER_TIMESTAMP_MMSS,
marker_timestamp)
if mobj:
ts_mark = int(mobj.group('min')) * 60 * 1000 + \
int(mobj.group('sec')) * 1000 + \
int(mobj.group('ms'))
else:
raise ValueError(
'Invalid marker timestamp: {}'.format(
marker_timestamp))
track_count += 1
part_markers.append((u'ch{:02d}'.format(track_count),
marker_name, ts_mark))
if args.add_chapters and not args.merge_output:
# set the chapter marks
generated_markers = []
for j, file_marker in enumerate(part_markers):
generated_markers.append({
'id':
file_marker[0],
'text':
file_marker[1],
'start_time':
int(file_marker[2]),
'end_time':
int(
round(mutagen_audio.info.length *
1000) if j == (len(part_markers) -
1) else part_markers[j +
1][2]),
})
mutagen_audio.tags.add(
CTOC(element_id=u'toc',
flags=CTOCFlags.TOP_LEVEL | CTOCFlags.ORDERED,
child_element_ids=[
m['id'].encode('ascii') for m in generated_markers
],
sub_frames=[TIT2(text=[u'Table of Contents'])]))
for i, m in enumerate(generated_markers):
mutagen_audio.tags.add(
CHAP(element_id=m['id'].encode('ascii'),
start_time=m['start_time'],
end_time=m['end_time'],
sub_frames=[TIT2(text=[u'{}'.format(m['text'])])
]))
start_time = datetime.timedelta(
milliseconds=m['start_time'])
end_time = datetime.timedelta(milliseconds=m['end_time'])
logger.debug(
u'Added chap tag => {}: {}-{} "{}" to "{}"'.format(
colored.cyan(m['id']), start_time, end_time,
colored.cyan(m['text']),
colored.blue(part_filename)))
if len(generated_markers) == 1:
# Weird player problem on voice where title is shown instead of chapter title
mutagen_audio.tags.add(
TIT2(encoding=3, text=u'{}'.format(title)))
mutagen_audio.save()
except Exception as e:
logger.warning('Error saving ID3: {}'.format(
colored.red(str(e), bold=True)))
keep_cover = True
logger.info('Saved "{}"'.format(colored.magenta(part_filename)))
file_tracks.append({
'file': part_filename,
'markers': part_markers,
})
# end loop: for p in download_parts:
debug_meta['audio_lengths_ms'] = audio_lengths_ms
debug_meta['file_tracks'] = file_tracks
if args.merge_output:
if os.path.isfile(book_filename if args.merge_format ==
'mp3' else book_m4b_filename):
logger.warning('Already saved "{}"'.format(
colored.magenta(book_filename if args.merge_format ==
'mp3' else book_m4b_filename)))
sys.exit(0)
logger.info('Generating "{}"...'.format(
colored.magenta(book_filename if args.merge_format ==
'mp3' else book_m4b_filename)))
# We can't directly generate a m4b here even if specified because eyed3 doesn't support m4b/mp4
temp_book_filename = '{}.part'.format(book_filename)
cmd = [
'ffmpeg',
'-y',
'-nostdin',
'-hide_banner',
'-loglevel',
'info' if logger.level == logging.DEBUG else 'error',
'-stats',
'-i',
'concat:{}'.format('|'.join([ft['file'] for ft in file_tracks])),
'-acodec',
'copy',
'-b:a',
'64k', # explicitly set audio bitrate
'-f',
'mp3',
temp_book_filename
]
exit_code = subprocess.call(cmd)
if exit_code:
logger.error(
'ffmpeg exited with the code: {0!s}'.format(exit_code))
logger.error('Command: {0!s}'.format(' '.join(cmd)))
exit(exit_code)
os.rename(temp_book_filename, book_filename)
mutagen_audio = MP3(book_filename, ID3=ID3)
if not mutagen_audio.tags:
mutagen_audio.tags = ID3()
# Overwrite title since it gets picked up from the first track by default
mutagen_audio.tags.add(TIT2(encoding=3, text=u'{}'.format(title)))
if 'TALB' not in mutagen_audio.tags:
mutagen_audio.tags.add(TALB(encoding=3, text=u'{}'.format(title)))
if 'TPE1' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TPE1(encoding=3, text=u'{}'.format(authors[0])))
if 'TPE2' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TPE2(encoding=3, text=u'{}'.format(authors[0])))
if 'TPUB' not in mutagen_audio.tags:
mutagen_audio.tags.add(
TPUB(encoding=3, text=u'{}'.format(publisher)))
if 'COMM' not in mutagen_audio.tags:
mutagen_audio.tags.add(
COMM(encoding=3,
desc=u'Description',
text=u'{}'.format(description)))
mutagen_audio.save()
if args.add_chapters:
merged_markers = []
for i, f in enumerate(file_tracks):
prev_tracks_len_ms = 0 if i == 0 else reduce(
lambda x, y: x + y, audio_lengths_ms[0:i])
this_track_endtime_ms = int(
reduce(lambda x, y: x + y, audio_lengths_ms[0:i + 1]))
file_markers = f['markers']
for j, file_marker in enumerate(file_markers):
merged_markers.append({
'id':
file_marker[0],
'text':
u'{}'.format(file_marker[1]),
'start_time':
int(file_marker[2]) + prev_tracks_len_ms,
'end_time':
int(this_track_endtime_ms if j == (
len(file_markers) -
1) else file_markers[j + 1][2] +
prev_tracks_len_ms),
})
debug_meta['merged_markers'] = merged_markers
mutagen_audio.tags.add(
CTOC(element_id=u'toc',
flags=CTOCFlags.TOP_LEVEL | CTOCFlags.ORDERED,
child_element_ids=[
m['id'].encode('ascii') for m in merged_markers
],
sub_frames=[TIT2(text=[u"Table of Contents"])]))
for i, m in enumerate(merged_markers):
mutagen_audio.tags.add(
CHAP(element_id=m['id'].encode('ascii'),
start_time=m['start_time'],
end_time=m['end_time'],
sub_frames=[TIT2(text=[u'{}'.format(m['text'])])]))
start_time = datetime.timedelta(milliseconds=m['start_time'])
end_time = datetime.timedelta(milliseconds=m['end_time'])
logger.debug(
u'Added chap tag => {}: {}-{} "{}" to "{}"'.format(
colored.cyan(m['id']), start_time, end_time,
colored.cyan(m['text']), colored.blue(book_filename)))
mutagen_audio.save()
if args.merge_format == 'mp3':
logger.info('Merged files into "{}"'.format(
colored.magenta(book_filename if args.merge_format ==
'mp3' else book_m4b_filename)))
if args.merge_format == 'm4b':
temp_book_m4b_filename = '{}.part'.format(book_m4b_filename)
cmd = [
'ffmpeg',
'-y',
'-nostdin',
'-hide_banner',
'-loglevel',
'info' if logger.level == logging.DEBUG else 'error',
'-stats',
'-i',
book_filename,
]
if os.path.isfile(cover_filename):
cmd.extend(['-i', cover_filename])
cmd.extend([
'-map',
'0:a',
'-c:a',
'aac',
'-b:a',
'64k', # explicitly set audio bitrate
])
if os.path.isfile(cover_filename):
cmd.extend([
'-map',
'1:v',
'-c:v',
'copy',
'-disposition:v:0',
'attached_pic',
])
cmd.extend(['-f', 'mp4', temp_book_m4b_filename])
exit_code = subprocess.call(cmd)
if exit_code:
logger.error(
'ffmpeg exited with the code: {0!s}'.format(exit_code))
logger.error('Command: {0!s}'.format(' '.join(cmd)))
exit(exit_code)
os.rename(temp_book_m4b_filename, book_m4b_filename)
logger.info('Merged files into "{}"'.format(
colored.magenta(book_m4b_filename)))
try:
os.remove(book_filename)
except Exception as e:
logger.warning('Error deleting "{}": {}'.format(
book_filename, str(e)))
if not args.keep_mp3:
for f in file_tracks:
try:
os.remove(f['file'])
except Exception as e:
logger.warning('Error deleting "{}": {}'.format(
f['file'], str(e)))
if not keep_cover:
try:
os.remove(cover_filename)
except Exception as e:
logger.warning('Error deleting "{}": {}'.format(
cover_filename, str(e)))
if args.write_json:
with open(debug_filename, 'w') as outfile:
json.dump(debug_meta, outfile, indent=2)
def download(broadcast, targetDir, reliveUrlTemplate):
broadcastStartDT = parse(broadcast['start'])
broadcastEndDT = parse(broadcast['end'])
# build filename from channel, show title and broadcast datetime, while escaping "bad" characters
filename = os.path.join(
targetDir,
re.sub(
'[^\w\s\-\.\[\]]', '_',
broadcast['trackingInfos']['pageVars']['broadcast_service'] + ' ' +
broadcastStartDT.astimezone(
pytz.timezone('Europe/Berlin')).strftime("%Y-%m-%d %H:%M") +
' ' + broadcast['trackingInfos']['pageVars']['topline']) + ".mp3")
# skip broadcast if file is already exists
if os.path.isfile(filename) and os.path.getsize(filename) > 0:
print("%s already exists, skipping." % filename, flush=True)
return
# get links to all audio segments of this broadcast
segmentUrls = getSegmentUrls(broadcastStartDT, broadcastEndDT,
reliveUrlTemplate)
if segmentUrls is None:
# skip broadcast if no segments available
print("Skipping %s, not yet in relive" % filename)
return
# dowload all ts segments, and convert them to mp3
print("Downloading %s ..." % filename, end=" ", flush=True)
try:
sound = AudioSegment.empty()
for i in segmentUrls:
sound += AudioSegment.from_file(BytesIO(urlopen(i).read()))
sound.export(filename, format="mp3")
except:
print("failed.", flush=True)
return
else:
print("done.", flush=True)
# ID3: remove all tags
try:
tags = ID3(filename)
tags.delete()
except ID3NoHeaderError:
tags = ID3()
# ID3: save as much information as possible in the ID3 tags
tags.add(
TRSN(
text=[broadcast['trackingInfos']['pageVars']['broadcast_service']
]))
tags.add(
TPE1(
text=[broadcast['trackingInfos']['pageVars']['broadcast_service']
]))
tags.add(
TALB(text=[
" - ".join(
list(
dict.fromkeys([
broadcast['trackingInfos']['pageVars']['topline'],
broadcast['trackingInfos']['pageVars']['title']
])))
]))
tags.add(TRCK(text=['1/1']))
#tags.add(TIT2(text=[broadcastStartDT.astimezone(pytz.timezone('Europe/Berlin')).strftime("%Y-%m-%d %H:%M")]))
tags.add(TIT2(text=[broadcast['publicationOf']['title']]))
tags.add(
COMM(lang="deu",
desc="desc",
text=[broadcast['publicationOf']['description']]))
tags.add(
TYER(text=[
broadcastStartDT.astimezone(pytz.timezone(
'Europe/Berlin')).strftime("%Y")
]))
tags.add(
TDAT(text=[
broadcastStartDT.astimezone(pytz.timezone(
'Europe/Berlin')).strftime("%d%m")
]))
tags.add(
TIME(text=[
broadcastStartDT.astimezone(pytz.timezone(
'Europe/Berlin')).strftime("%H%M")
]))
tags.add(
TLEN(text=[
int((broadcastEndDT - broadcastStartDT).total_seconds() * 1000)
]))
tags.add(WOAS(url=broadcast['publicationOf']['canonicalUrl']))
tags.add(WORS(url="https://www.br.de/radio/"))
# ID3: chapters
chapterNr = 0
for chapter in broadcast['items']:
chapterStartDT = parse(chapter['start'])
if 'duration' in chapter and chapter['duration'] is not None:
chapterEndDT = chapterStartDT + timedelta(
seconds=chapter['duration'])
else:
chapterEndDT = broadcastEndDT
artists = []
for i in ['performer', 'author']:
if i in chapter and chapter[i] is not None and len(chapter[i]) > 0:
artists.append(chapter[i])
titles = []
for i in ['title']:
if i in chapter and chapter[i] is not None and len(chapter[i]) > 0:
titles.append(chapter[i])
tags.add(
CHAP(element_id=chapterNr,
start_time=floor(
(chapterStartDT - broadcastStartDT).total_seconds() *
1000),
end_time=ceil(
(chapterEndDT - broadcastStartDT).total_seconds() * 1000),
sub_frames=[
TIT2(text=[
" - ".join([" ".join(artists), " ".join(titles)])
])
]))
chapterNr += 1
tocList = ",".join([str(i) for i in range(0, chapterNr)])
tags.add(
CTOC(element_id="toc",
flags=CTOCFlags.TOP_LEVEL | CTOCFlags.ORDERED,
child_element_ids=[tocList],
sub_frames=[TIT2(text=["Table Of Contents"])]))
# ID3: cover image
response = requests.get(
broadcast['publicationOf']['defaultTeaserImage']['url'], timeout=5)
if response.status_code == 200:
tags.add(
APIC(mime=response.headers['content-type'],
desc="Front Cover",
data=response.content))
# save ID3 tags
tags.save(filename, v2_version=3)
tags.add(WOAS(url=showInfo['website']))
tags.add(WORS(url=stationInfo['website']))
for chapter in showInfo['parts'][partNo]['chapters']:
tags.add(
CHAP(element_id=chapter["id"],
start_time=chapter["start_ms"],
end_time=chapter["end_ms"],
sub_frames=[TIT2(text=[chapter["title"]])]))
tocList = ",".join([
chapter["id"] for chapter in showInfo['parts'][partNo]['chapters']
])
tags.add(
CTOC(element_id="toc",
flags=CTOCFlags.TOP_LEVEL | CTOCFlags.ORDERED,
child_element_ids=[tocList],
sub_frames=[TIT2(text=["Table Of Contents"])]))
if showInfo['image_mime'] is not None and showInfo[
'image_data'] is not None:
tags.add(
APIC(mime=showInfo['image_mime'],
desc="Front Cover",
data=showInfo['image_data']))
# save ID3 tags
tags.save(showInfo['parts'][partNo]['filepath'] + ".part",
v2_version=3)
# done!
os.rename(showInfo['parts'][partNo]['filepath'] + ".part",