aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorPhilipp Hagemeister <phihag@phihag.de>2014-09-28 10:34:55 +0200
committerPhilipp Hagemeister <phihag@phihag.de>2014-09-28 10:35:19 +0200
commitb14f3a4c1da00cbee8775904c24c4d0547018ae0 (patch)
tree6777376907e13636db4f1536c1ef8512b0453025
parent92f7963f6e5dd9d6f2c8805f8ca51c70df15a776 (diff)
downloadyoutube-dl-b14f3a4c1da00cbee8775904c24c4d0547018ae0.tar.xz
[golem] Simplify (#3828)
-rw-r--r--youtube_dl/extractor/common.py23
-rw-r--r--youtube_dl/extractor/golem.py124
2 files changed, 56 insertions, 91 deletions
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 8d6a6f601..f43a0a569 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -22,6 +22,7 @@ from ..utils import (
clean_html,
compiled_regex_type,
ExtractorError,
+ float_or_none,
int_or_none,
RegexNotFoundError,
sanitize_filename,
@@ -720,6 +721,28 @@ class InfoExtractor(object):
now_str = now.strftime("%Y-%m-%d %H:%M")
return name + ' ' + now_str
+ def _int(self, v, name, fatal=False, **kwargs):
+ res = int_or_none(v, **kwargs)
+ if 'get_attr' in kwargs:
+ print(getattr(v, kwargs['get_attr']))
+ if res is None:
+ msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
+ if fatal:
+ raise ExtractorError(msg)
+ else:
+ self._downloader.report_warning(msg)
+ return res
+
+ def _float(self, v, name, fatal=False, **kwargs):
+ res = float_or_none(v, **kwargs)
+ if res is None:
+ msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
+ if fatal:
+ raise ExtractorError(msg)
+ else:
+ self._downloader.report_warning(msg)
+ return res
+
class SearchInfoExtractor(InfoExtractor):
"""
diff --git a/youtube_dl/extractor/golem.py b/youtube_dl/extractor/golem.py
index 6a64b5d95..a237f19ee 100644
--- a/youtube_dl/extractor/golem.py
+++ b/youtube_dl/extractor/golem.py
@@ -4,7 +4,10 @@ from __future__ import unicode_literals
import re
from .common import InfoExtractor
-from ..utils import compat_urlparse
+from ..utils import (
+ compat_urlparse,
+ determine_ext,
+)
class GolemIE(InfoExtractor):
@@ -17,115 +20,54 @@ class GolemIE(InfoExtractor):
'format_id': 'high',
'ext': 'mp4',
'title': 'iPhone 6 und 6 Plus - Test',
- 'duration': 300,
+ 'duration': 300.44,
'filesize': 65309548,
}
}
- _CONFIG = 'https://video.golem.de/xml/{0}.xml'
_PREFIX = 'http://video.golem.de'
- def _warn(self, fmt, *args):
- self.report_warning(fmt.format(*args), self._id)
-
- def _extract_format(self, elem):
- format_id = elem.tag
-
- url = elem.findtext('./url')
- if url == '':
- self._warn("{0}: url: empty, skipping", format_id)
- return None
-
- fmt = {
- 'format_id': format_id,
- 'url': compat_urlparse.urljoin(self._PREFIX, url)
- }
-
- try:
- _, ext = elem.findtext('./filename', '').rsplit('.', 1)
- except ValueError:
- self._warn('{0}: ext: missing extension', format_id)
- else:
- fmt['ext'] = ext
-
- filesize = elem.findtext('./filesize')
- if filesize is not None:
- try:
- fmt['filesize'] = int(filesize)
- except ValueError as e:
- self._warn('{0}: filesize: {1}', format_id, e)
-
- width = elem.get('width')
- if width is not None:
- try:
- fmt['width'] = int(width)
- except ValueError as e:
- self._warn('{0}: width: {1}', format_id, e)
-
- height = elem.get('height')
- if height is not None:
- try:
- fmt['height'] = int(height)
- except ValueError as e:
- self._warn('{0}: height: {1}', format_id, e)
-
- return fmt
-
- def _extract_thumbnail(self, elem):
- url = elem.findtext('./url')
- if url == '':
- return None
- thumb = {
- 'url': compat_urlparse.urljoin(self._PREFIX, url)
- }
-
- width = elem.get('width')
- if width is not None:
- try:
- thumb['width'] = int(width)
- except ValueError as e:
- self._warn('thumbnail: width: {0}', e)
-
- height = elem.get('height')
- if height is not None:
- try:
- thumb['height'] = int(height)
- except ValueError as e:
- self._warn('thumbnail: height: {0}', e)
-
- return thumb
-
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- self._id = mobj.group('id')
+ video_id = self._match_id(url)
- config = self._download_xml(self._CONFIG.format(self._id), self._id)
+ config = self._download_xml(
+ 'https://video.golem.de/xml/{0}.xml'.format(video_id), video_id)
info = {
- 'id': self._id,
- 'title': config.findtext('./title', 'golem')
+ 'id': video_id,
+ 'title': config.findtext('./title', 'golem'),
+ 'duration': self._float(config.findtext('./playtime'), 'duration'),
}
formats = []
for e in config.findall('./*[url]'):
- fmt = self._extract_format(e)
- if fmt is not None:
- formats.append(fmt)
+ url = e.findtext('./url')
+ if not url:
+ self._downloader.report_warning(
+ "{0}: url: empty, skipping".format(e.tag))
+ continue
+
+ formats.append({
+ 'format_id': e.tag,
+ 'url': compat_urlparse.urljoin(self._PREFIX, url),
+ 'height': self._int(e.get('height'), 'height'),
+ 'width': self._int(e.get('width'), 'width'),
+ 'filesize': self._int(e.findtext('filesize'), 'filesize'),
+ 'ext': determine_ext(e.findtext('./filename')),
+ })
self._sort_formats(formats)
info['formats'] = formats
thumbnails = []
for e in config.findall('.//teaser[url]'):
- thumb = self._extract_thumbnail(e)
- if thumb is not None:
- thumbnails.append(thumb)
+ url = e.findtext('./url')
+ if not url:
+ continue
+ thumbnails.append({
+ 'url': compat_urlparse.urljoin(self._PREFIX, url),
+ 'width': self._int(e.get('width'), 'thumbnail width'),
+ 'height': self._int(e.get('height'), 'thumbnail height'),
+ })
info['thumbnails'] = thumbnails
- playtime = config.findtext('./playtime')
- if playtime is not None:
- try:
- info['duration'] = round(float(playtime))
- except ValueError as e:
- self._warn('duration: {0}', e)
-
return info