Akkariin/youtube-dl
1
0
Fork 0
mirror of https://github.com/l1ving/youtube-dl synced 2025-02-11 01:13:00 +08:00
Code Issues Releases Wiki Activity

[wdr] refactor so that JSONP download is reusable

Browse Source 
We want to parse multiple JSONP URLs on a page later, so we have to
change _html_search_regex() into a re.findall().
This commit is contained in:
Roland Hieber 2017-05-21 16:27:19 +02:00
parent bca4965919
commit 9436165909
1 changed files with 16 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
youtube_dl/extractor/wdr.py
View File

@ -16,21 +16,29 @@ from ..utils import (
class WDRBaseIE(InfoExtractor): class WDRBaseIE(InfoExtractor):
def _extract_wdr_video(self, webpage, display_id): def _extract_wdr_jsonp_urls(self, webpage, display_id):
""" returns list of jsonp urls """
# for wdr.de the data-extension is in a tag with the class "mediaLink" # for wdr.de the data-extension is in a tag with the class "mediaLink"
# for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn" # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
# for wdrmaus, in a tag with the class "videoButton" (previously a link # for wdrmaus, in a tag with the class "videoButton" (previously a link
# to the page in a multiline "videoLink"-tag) # to the page in a multiline "videoLink"-tag)
json_metadata = self._html_search_regex( json_metadata = re.findall(
r'class=(?:"(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b[^"]*"[^>]+|"videoLink\b[^"]*"[\s]*>\n[^\n]*)data-extension="([^"]+)"', r'class=(?:"(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b[^"]*"[^>]+|"videoLink\b[^"]*"[\s]*>\n[^\n]*)data-extension="([^"]+)"',
webpage, 'media link', default=None, flags=re.MULTILINE) webpage, flags=re.MULTILINE)
if not json_metadata: if not json_metadata:
return return
media_link_obj = self._parse_json(json_metadata, display_id, urls = []
transform_source=js_to_json) for json in json_metadata:
jsonp_url = media_link_obj['mediaObj']['url'] media_link_obj = self._parse_json(json, display_id,
transform_source=js_to_json)
urls.append(media_link_obj['mediaObj']['url'])
return urls
def _extract_wdr_video_from_jsonp_url(self, jsonp_url, display_id):
""" returns info dict """
metadata = self._download_json( metadata = self._download_json(
jsonp_url, display_id, transform_source=strip_jsonp) jsonp_url, display_id, transform_source=strip_jsonp)
@ -199,9 +207,9 @@ class WDRIE(WDRBaseIE):
display_id = mobj.group('display_id') display_id = mobj.group('display_id')
webpage = self._download_webpage(url, display_id) webpage = self._download_webpage(url, display_id)
info_dict = self._extract_wdr_video(webpage, display_id) jsonp_urls = self._extract_wdr_jsonp_urls(webpage, display_id)
if not info_dict: if not jsonp_urls:
entries = [ entries = [
self.url_result(page_url + href[0], 'WDR') self.url_result(page_url + href[0], 'WDR')
for href in re.findall( for href in re.findall(