nojhan/weboob-devel
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Fix francetelevisions for website updates

Browse source 
And also add support for existing data it lacked.
This commit is contained in:
Laurent Bachelier 2012-02-01 00:22:12 +01:00 • committed by Romain Bignon
commit f3470a0128
2 changed files with 65 additions and 43 deletions
Download patch file Download diff file Expand all files Collapse all files

47
modules/francetelevisions/browser.py
View file

@ -1,6 +1,6 @@
# -*- coding: utf-8 -*-
# Copyright(C) 2011 Romain Bignon
# Copyright(C) 2011-2012 Romain Bignon, Laurent Bachelier
#
# This file is part of weboob.
#
@ -17,10 +17,14 @@
# You should have received a copy of the GNU Affero General Public License
# along with weboob. If not, see <http://www.gnu.org/licenses/>.
import datetime
from lxml import etree
from weboob.tools.browser import BaseBrowser
from weboob.tools.browser.decorators import id2url
from .pages import IndexPage, VideoPage, MetaVideoPage
from .pages import IndexPage, VideoPage
from .video import PluzzVideo
@ -29,12 +33,11 @@ __all__ = ['PluzzBrowser']
class PluzzBrowser(BaseBrowser):
DOMAIN = 'pluzz.fr'
ENCODING = None
ENCODING = 'ISO-8859-1'
PAGES = {r'http://[w\.]*pluzz.fr/?': IndexPage,
r'http://[w\.]*pluzz.fr/recherche.html.*': IndexPage,
r'http://[w\.]*pluzz.fr/[-\w]+/.*': IndexPage,
r'http://[w\.]*pluzz.fr/((?!recherche).+)\.html': VideoPage,
r'http://info\.francetelevisions\.fr/\?id-video=.*': MetaVideoPage,
}
@id2url(PluzzVideo.id2url)
@ -42,15 +45,15 @@ class PluzzBrowser(BaseBrowser):
self.location(url)
assert self.is_on_page(VideoPage)
id = self.page.get_id()
metaurl = self.page.get_meta_url()
if metaurl is None:
return None
_id = self.page.get_id()
if video is None:
video = PluzzVideo(_id)
self.location(metaurl)
assert self.is_on_page(MetaVideoPage)
infourl = self.page.get_info_url()
if infourl is not None:
self.parse_info(self.openurl(infourl).read(), video)
return self.page.get_video(id, video)
return video
def iter_search_results(self, pattern):
if not pattern:
@ -60,3 +63,25 @@ class PluzzBrowser(BaseBrowser):
assert self.is_on_page(IndexPage)
return self.page.iter_videos()
def parse_info(self, data, video):
parser = etree.XMLParser(encoding='utf-8')
root = etree.XML(data, parser)
assert root.tag == 'oeuvre'
video.title = root.findtext('titre')
hours, minutes, seconds = root.findtext('duree').split(':')
video.duration = datetime.timedelta(hours=int(hours), minutes=int(minutes), seconds=int(seconds))
for vid in root.find('videos'):
if vid.findtext('statut') == 'ONLINE' and vid.findtext('format') == 'wmv':
video.url = vid.findtext('url')
date = root.findtext('diffusions/diffusion')
if date:
video.date = datetime.datetime.strptime(date, '%d/%m/%Y %H:%M')
video.description = root.findtext('synopsis')
return video

59
modules/francetelevisions/pages.py
View file

@ -1,6 +1,6 @@
# -*- coding: utf-8 -*-
# Copyright(C) 2011 Romain Bignon
# Copyright(C) 2011-2012 Romain Bignon, Laurent Bachelier
#
# This file is part of weboob.
#
@ -22,7 +22,6 @@ import re
from weboob.tools.capabilities.thumbnail import Thumbnail
from weboob.tools.browser import BasePage, BrokenPageError
from weboob.capabilities.base import NotAvailable
from .video import PluzzVideo
@ -34,57 +33,55 @@ __all__ = ['IndexPage', 'VideoPage']
class IndexPage(BasePage):
def iter_videos(self):
for div in self.parser.select(self.document.getroot(), 'li.vignette'):
url = self.parser.select(div, 'h4 a', 1).attrib['href']
m = re.match('http://www.pluzz.fr/([^/]+).html', url)
title = self.parser.select(div, 'h4 a', 1)
url = title.attrib['href']
m = re.match('^http://www.pluzz.fr/([^/]+)\.html$', url)
if not m:
print ':('
print ':( %s' % url
continue
_id = m.group(1)
video = PluzzVideo(_id)
video.title = self.parser.select(div, 'h4 a', 1).text
m = re.match('^(.+) - ([0-2][0-9])h([0-5][0-9])$', title.text)
if m:
video.title = m.group(1)
hour = int(m.group(2))
minute = int(m.group(3))
else:
video.title = title.text
hour = 0
minute = 0
# Date is not available anymore on search results.
video.date = NotAvailable
#m = re.match('(\d+)/(\d+)/(\d+)', self.parser.select(div, 'p.date', 1).text)
#if m:
# video.date = datetime.datetime(int(m.group(3)),
# int(m.group(2)),
# int(m.group(1)))
m = re.match('(\d+)/(\d+)/(\d+)', self.parser.select(div, 'p.date', 1).text)
if m:
video.date = datetime.datetime(int(m.group(3)),
int(m.group(2)),
int(m.group(1)),
hour,
minute)
url = self.parser.select(div, 'img.illustration', 1).attrib['src']
video.thumbnail = Thumbnail('http://www.pluzz.fr/%s' % url)
yield video
class VideoPage(BasePage):
def on_loaded(self):
p = self.parser.select(self.document.getroot(), 'p.alert')
if len(p) > 0:
raise Exception(p[0].text)
def get_meta_url(self):
def get_info_url(self):
try:
div = self.parser.select(self.document.getroot(), 'a#current_video', 1)
except BrokenPageError:
return None
else:
return div.attrib['href']
m = re.match(
'^%s(\d+)$' % re.escape('http://info.francetelevisions.fr/?id-video='),
div.attrib['href'])
if m:
return r'http://www.pluzz.fr/appftv/webservices/video/getInfosOeuvre.php?mode=zeri&id-diffusion=%s' % m.group(1)
def get_id(self):
return self.groups[0]
class MetaVideoPage(BasePage):
def get_meta(self, name):
return self.parser.select(self.document.getroot(), 'meta[name=%s]' % name, 1).attrib['content']
def get_video(self, id, video=None):
if video is None:
video = PluzzVideo(id)
video.title = self.get_meta('vignette-titre-court')
video.url = 'mms://a988.v101995.c10199.e.vm.akamaistream.net/7/988/10199/3f97c7e6/ftvigrp.download.akamai.com/10199/cappuccino/production/publication/%s' % self.get_meta('urls-url-video')
video.description = self.get_meta('description')
hours, minutes, seconds = self.get_meta('vignette-duree').split(':')
video.duration = datetime.timedelta(hours=int(hours), minutes=int(minutes), seconds=int(seconds))
return video