[npo] Improve quality metadata extraction

This commit is contained in:
Sergey M․ 2018-02-17 20:31:52 +07:00
parent c4e7496421
commit ee706f1009
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 25 additions and 4 deletions

View File

@ -11,6 +11,7 @@ from ..utils import (
determine_ext, determine_ext,
ExtractorError, ExtractorError,
fix_xml_ampersands, fix_xml_ampersands,
int_or_none,
orderedSet, orderedSet,
parse_duration, parse_duration,
qualities, qualities,
@ -194,7 +195,11 @@ class NPOIE(NPOBaseIE):
formats = [] formats = []
urls = set() urls = set()
quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std']) QUALITY_LABELS = ('Laag', 'Normaal', 'Hoog')
QUALITY_FORMATS = ('adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std')
quality_from_label = qualities(QUALITY_LABELS)
quality_from_format_id = qualities(QUALITY_FORMATS)
items = self._download_json( items = self._download_json(
'http://ida.omroep.nl/app.php/%s' % video_id, video_id, 'http://ida.omroep.nl/app.php/%s' % video_id, video_id,
'Downloading formats JSON', query={ 'Downloading formats JSON', query={
@ -210,11 +215,27 @@ class NPOIE(NPOBaseIE):
r'video/ida/([^/]+)', item_url, 'format id', r'video/ida/([^/]+)', item_url, 'format id',
default=None) default=None)
item_label = item.get('label')
def add_format_url(format_url): def add_format_url(format_url):
width = int_or_none(self._search_regex(
r'(\d+)[xX]\d+', format_url, 'width', default=None))
height = int_or_none(self._search_regex(
r'\d+[xX](\d+)', format_url, 'height', default=None))
if item_label in QUALITY_LABELS:
quality = quality_from_label(item_label)
f_id = item_label
elif item_label in QUALITY_FORMATS:
quality = quality_from_format_id(format_id)
f_id = format_id
else:
quality, f_id = None
formats.append({ formats.append({
'url': format_url, 'url': format_url,
'format_id': format_id, 'format_id': f_id,
'quality': quality(format_id), 'width': width,
'height': height,
'quality': quality,
}) })
# Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706 # Example: http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706
@ -226,7 +247,7 @@ class NPOIE(NPOBaseIE):
stream_info = self._download_json( stream_info = self._download_json(
item_url + '&type=json', video_id, item_url + '&type=json', video_id,
'Downloading %s stream JSON' 'Downloading %s stream JSON'
% item.get('label') or item.get('format') or format_id or num) % item_label or item.get('format') or format_id or num)
except ExtractorError as ee: except ExtractorError as ee:
if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404: if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
error = (self._parse_json( error = (self._parse_json(