[turner] Extract all formats

This commit is contained in:
Sergey M․ 2016-08-29 22:13:49 +07:00
parent 547993dcd0
commit cd10b3ea63
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D

View file

@ -4,6 +4,7 @@
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_str
from ..utils import ( from ..utils import (
xpath_text, xpath_text,
int_or_none, int_or_none,
@ -30,11 +31,11 @@ def _extract_cvp_info(self, data_src, video_id, path_data={}):
tokens = {} tokens = {}
urls = [] urls = []
formats = [] formats = []
rex = re.compile(r'''(?x) rex = re.compile(
(?P<width>[0-9]+)x(?P<height>[0-9]+) r'(?P<width>[0-9]+)x(?P<height>[0-9]+)(?:_(?P<bitrate>[0-9]+))?')
(?:_(?P<bitrate>[0-9]+))? # Possible formats locations: files/file, files/groupFiles/files
''') # and maybe others
for video_file in video_data.findall('files/file'): for video_file in video_data.findall('.//file'):
video_url = video_file.text.strip() video_url = video_file.text.strip()
if not video_url: if not video_url:
continue continue
@ -84,12 +85,14 @@ def _extract_cvp_info(self, data_src, video_id, path_data={}):
if video_url in urls: if video_url in urls:
continue continue
urls.append(video_url) urls.append(video_url)
format_id = video_file.attrib['bitrate'] format_id = video_file.get('bitrate')
if ext == 'smil': if ext == 'smil':
formats.extend(self._extract_smil_formats(video_url, video_id, fatal=False)) formats.extend(self._extract_smil_formats(
video_url, video_id, fatal=False))
elif ext == 'm3u8': elif ext == 'm3u8':
m3u8_formats = self._extract_m3u8_formats( m3u8_formats = self._extract_m3u8_formats(
video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False) video_url, video_id, 'mp4', m3u8_id=format_id or 'hls',
fatal=False)
if m3u8_formats: if m3u8_formats:
# Sometimes final URLs inside m3u8 are unsigned, let's fix this # Sometimes final URLs inside m3u8 are unsigned, let's fix this
# ourselves # ourselves
@ -103,7 +106,7 @@ def _extract_cvp_info(self, data_src, video_id, path_data={}):
elif ext == 'f4m': elif ext == 'f4m':
formats.extend(self._extract_f4m_formats( formats.extend(self._extract_f4m_formats(
update_url_query(video_url, {'hdcore': '3.7.0'}), update_url_query(video_url, {'hdcore': '3.7.0'}),
video_id, f4m_id=format_id, fatal=False)) video_id, f4m_id=format_id or 'hds', fatal=False))
else: else:
f = { f = {
'format_id': format_id, 'format_id': format_id,
@ -117,18 +120,19 @@ def _extract_cvp_info(self, data_src, video_id, path_data={}):
'height': int(mobj.group('height')), 'height': int(mobj.group('height')),
'tbr': int_or_none(mobj.group('bitrate')), 'tbr': int_or_none(mobj.group('bitrate')),
}) })
elif format_id.isdigit(): elif isinstance(format_id, compat_str):
f['tbr'] = int(format_id) if format_id.isdigit():
else: f['tbr'] = int(format_id)
mobj = re.match(r'ios_(audio|[0-9]+)$', format_id) else:
if mobj: mobj = re.match(r'ios_(audio|[0-9]+)$', format_id)
if mobj.group(1) == 'audio': if mobj:
f.update({ if mobj.group(1) == 'audio':
'vcodec': 'none', f.update({
'ext': 'm4a', 'vcodec': 'none',
}) 'ext': 'm4a',
else: })
f['tbr'] = int(mobj.group(1)) else:
f['tbr'] = int(mobj.group(1))
formats.append(f) formats.append(f)
self._sort_formats(formats) self._sort_formats(formats)