Move all canalplus pages in one file

There isn't enough code to justify multiple files.
Also includes pep8 fixes.
This commit is contained in:
Laurent Bachelier 2012-03-09 03:39:53 +01:00
commit ec689b1e7f
4 changed files with 29 additions and 78 deletions

View file

@ -25,7 +25,7 @@ import lxml.etree
from weboob.tools.browser import BaseBrowser from weboob.tools.browser import BaseBrowser
from weboob.tools.browser.decorators import id2url from weboob.tools.browser.decorators import id2url
from .pages import InitPage, VideoPage from .pages import ChannelsPage, VideoPage
from .video import CanalplusVideo from .video import CanalplusVideo
from weboob.capabilities.collection import CollectionNotFound from weboob.capabilities.collection import CollectionNotFound
@ -46,7 +46,7 @@ class CanalplusBrowser(BaseBrowser):
DOMAIN = u'service.canal-plus.com' DOMAIN = u'service.canal-plus.com'
ENCODING = 'utf-8' ENCODING = 'utf-8'
PAGES = { PAGES = {
r'http://service.canal-plus.com/video/rest/initPlayer/cplus/': InitPage, r'http://service.canal-plus.com/video/rest/initPlayer/cplus/': ChannelsPage,
r'http://service.canal-plus.com/video/rest/search/cplus/.*': VideoPage, r'http://service.canal-plus.com/video/rest/search/cplus/.*': VideoPage,
r'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/(?P<id>.+)': VideoPage, r'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/(?P<id>.+)': VideoPage,
r'http://service.canal-plus.com/video/rest/getMEAs/cplus/.*': VideoPage, r'http://service.canal-plus.com/video/rest/getMEAs/cplus/.*': VideoPage,
@ -76,7 +76,7 @@ class CanalplusBrowser(BaseBrowser):
return self.page.get_video(video, self.quality) return self.page.get_video(video, self.quality)
def iter_resources(self, split_path): def iter_resources(self, split_path):
if not self.is_on_page(InitPage): if not self.is_on_page(ChannelsPage):
self.home() self.home()
channels = self.page.get_channels() channels = self.page.get_channels()

115
modules/canalplus/pages.py Normal file
View file

@ -0,0 +1,115 @@
# -*- coding: utf-8 -*-
# Copyright(C) 2010-2012 Nicolas Duhamel, Laurent Bachelier
#
# This file is part of weboob.
#
# weboob is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# weboob is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with weboob. If not, see <http://www.gnu.org/licenses/>.
from datetime import datetime
from weboob.tools.browser import BasePage
from weboob.capabilities.collection import Collection
from weboob.capabilities.base import NotAvailable
from weboob.tools.capabilities.thumbnail import Thumbnail
from .video import CanalplusVideo
__all__ = ['ChannelsPage', 'VideoPage']
class ChannelsPage(BasePage):
def get_channels(self):
"""
Extract all possible channels (paths) from the page
"""
channels = list()
for elem in self.document[2].getchildren():
for e in elem.getchildren():
if e.tag == "NOM":
name = unicode(e.text.strip())
channels.append(Collection([name]))
elif e.tag == "SELECTIONS":
for select in e:
subname = unicode(select[1].text.strip())
sub = Collection([name, subname])
sub._link_id = select[0].text
channels.append(sub)
return channels
class VideoPage(BasePage):
def parse_video(self, el, video=None, quality=None):
_id = el.find('ID').text
if _id == '-1':
# means the video is not found
return None
if not video:
video = CanalplusVideo(_id)
infos = el.find('INFOS')
video.title = u''
for part in infos.find('TITRAGE'):
if len(part.text.strip()) == 0:
continue
if len(video.title) > 0:
video.title += u' — '
video.title += part.text.strip()
video.description = infos.find('DESCRIPTION').text
media = el.find('MEDIA')
url = media.find('IMAGES').find('PETIT').text
if url:
video.thumbnail = Thumbnail(url)
else:
video.thumbnail = NotAvailable
lastest_format = None
for format in media.find('VIDEOS'):
if format.text is None:
continue
if format.tag == quality:
video.url = format.text
break
lastest_format = format
if not video.url and lastest_format is not None:
video.url = lastest_format.text
day, month, year = map(int, infos.find('PUBLICATION').find('DATE').text.split('/'))
hour, minute, second = map(int, infos.find('PUBLICATION').find('HEURE').text.split(':'))
video.date = datetime(year, month, day, hour, minute, second)
return video
def iter_results(self):
for vid in self.document.getchildren():
yield self.parse_video(vid)
def iter_channel(self):
for vid in self.document.getchildren():
yield self.parse_video_channel(vid)
def parse_video_channel(self, el):
_id = el[0].text
video = CanalplusVideo(_id)
video.title = el[2][3][0].text
video.date = datetime.now()
return video
def get_video(self, video, quality):
_id = self.group_dict['id']
for vid in self.document.getchildren():
if not _id in vid.find('ID').text:
continue
return self.parse_video(vid, video, quality)

View file

@ -1,23 +0,0 @@
# -*- coding: utf-8 -*-
# Copyright(C) 2010-2011 Nicolas Duhamel
#
# This file is part of weboob.
#
# weboob is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# weboob is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with weboob. If not, see <http://www.gnu.org/licenses/>.
from .initpage import InitPage
from .videopage import VideoPage
__all__ = ['InitPage', 'VideoPage']

View file

@ -1,45 +0,0 @@
# -*- coding: utf-8 -*-
# Copyright(C) 2010-2012 Nicolas Duhamel, Laurent Bachelier
#
# This file is part of weboob.
#
# weboob is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# weboob is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with weboob. If not, see <http://www.gnu.org/licenses/>.
from weboob.tools.browser import BasePage
from weboob.capabilities.collection import Collection
__all__ = ['InitPage']
class InitPage(BasePage):
def get_channels(self):
"""
Extract all possible channels (paths) from the page
"""
channels = list()
for elem in self.document[2].getchildren():
for e in elem.getchildren():
if e.tag == "NOM":
name = unicode(e.text.strip())
channels.append(Collection([name]))
elif e.tag == "SELECTIONS":
for select in e:
subname = unicode(select[1].text.strip())
sub = Collection([name, subname])
sub._link_id = select[0].text
channels.append(sub)
return channels