nojhan/weboob-devel
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
weboob-devel/modules/radiofrance/backend.py
Laurent Bachelier 682e14c86a Make CapCollection understandable and useable by humans 
* Make the declaration of fct and it in the constructor Collection,
 instead of adding them from the outside
* Add a function to flatten a list containing collection (solves the
 radioob search crash)
* Better display of collections in the "ls" command (and display both id
 and title)
* The "cd" command goes to the root of the path (like the UNIX cd)
* Move the Video object of canalplus in a correct path
* Make Collection iterable
* Add comments to CapCollection
* Cache the result of fct in a Collection; it is only called once
* CollectionNotFound errors can be more explicit by providing a path
* Require utf-8 in collection paths
* Code cleanups
2012-02-02 23:44:13 +01:00

227 lines
9.2 KiB
Python
Raw Blame History

# * -*- coding: utf-8 -*-
# Copyright(C) 2011-2012 Johann Broudin, 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.capabilities.base import NotLoaded
from weboob.capabilities.radio import ICapRadio, Radio, Stream, Emission
from weboob.capabilities.collection import ICapCollection, CollectionNotFound, Collection
from weboob.tools.backend import BaseBackend
from weboob.tools.browser import BaseBrowser, BasePage
from StringIO import StringIO
from time import time
try:
import json
except ImportError:
import simplejson as json
__all__ = ['RadioFranceBackend']
class DataPage(BasePage):
def get_title(self):
for metas in self.parser.select(self.document.getroot(), 'div.metas'):
title = unicode(metas.text_content()).strip()
if len(title):
return title
class RssPage(BasePage):
def get_title(self):
titles = []
for heading in self.parser.select(self.document.getroot(), 'h1, h2, h3, h4'):
# Remove newlines/multiple spaces
words = heading.text_content()
if words:
for word in unicode(words).split():
titles.append(word)
if len(titles):
return ' '.join(titles)
class RadioFranceBrowser(BaseBrowser):
DOMAIN = None
ENCODING = 'UTF-8'
PAGES = {r'/playerjs/direct/donneesassociees/html\?guid=$': DataPage,
r'http://players.tv-radio.com/radiofrance/metadatas/([a-z]+)RSS.html': RssPage}
def get_current_playerjs(self, id):
self.location('http://www.%s.fr/playerjs/direct/donneesassociees/html?guid=' % id)
assert self.is_on_page(DataPage)
return self.page.get_title()
def get_current_rss(self, id):
self.location('http://players.tv-radio.com/radiofrance/metadatas/%sRSS.html' % id)
assert self.is_on_page(RssPage)
return self.page.get_title()
def get_current_direct(self, id):
json_data = self.openurl('http://www.%s.fr/sites/default/files/direct.json?_=%s' % (id, int(time())))
data = json.load(json_data)
document = self.parser.parse(StringIO(data.get('html')))
artist = document.findtext('//span[@class="artiste"]')
title = document.findtext('//span[@class="titre"]')
artist = unicode(artist) if artist else None
title = unicode(title) if title else None
return (artist, title)
class RadioFranceBackend(BaseBackend, ICapRadio, ICapCollection):
NAME = 'radiofrance'
MAINTAINER = 'Laurent Bachelier'
EMAIL = 'laurent@bachelier.name'
VERSION = '0.a'
DESCRIPTION = u'The radios of Radio France (Inter, Culture, Le Mouv\', etc.)'
LICENSE = 'AGPLv3+'
BROWSER = RadioFranceBrowser
_MP3_URL = u'http://mp3.live.tv-radio.com/%s/all/%s.mp3'
_MP3_HD_URL = u'http://mp3.live.tv-radio.com/%s/all/%shautdebit.mp3'
_RADIOS = {'franceinter': (u'France Inter', True),
'franceculture': (u'France Culture', True),
'franceinfo': (u'France Info', False),
'fbidf': (u'France Bleu Île-de-France (Paris)', True),
'fip': (u'FIP', True),
'francemusique': (u'France Musique', True),
'lemouv': (u'Le Mouv\'', True),
'fbalsace': (u'France Bleu Alsace (Strasbourg)', False),
'fbarmorique': (u'France Bleu Armorique (Rennes)', False),
'fbauxerre': (u'France Bleu Auxerre', False),
'fbazur': (u'France Bleu Azur (Nice)', False),
'fbbassenormandie': (u'France Bleu Basse Normandie (Caen)', False),
'fbbearn': (u'France Bleu Bearn (Pau)', False),
'fbbelfort': (u'France Bleu Belfort', False),
'fbberry': (u'France Bleu Berry (Châteauroux)', False),
'fbbesancon': (u'France Bleu Besancon', False),
'fbbourgogne': (u'France Bleu Bourgogne (Dijon)', False),
'fbbreizizel': (u'France Bleu Breiz Izel (Quimper)', False),
'fbchampagne': (u'France Bleu Champagne (Reims)', False),
'fbcotentin': (u'France Bleu Cotentin (Cherbourg)', False),
'fbcreuse': (u'France Bleu Creuse (Gueret)', False),
'fbdromeardeche': (u'France Bleu Drome Ardeche (Valence)', False),
'fbfrequenzamora': (u'France Bleu Frequenza Mora (Bastia - Corse)', False),
'fbgardlozere': (u'France Bleu Gard Lozère (Nîmes)', False),
'fbgascogne': (u'France Bleu Gascogne (Mont-de-Marsan)', False),
'fbgironde': (u'France Bleu Gironde (Bordeaux)', False),
'fbhautenormandie': (u'France Bleu Haute Normandie (Rouen)', False),
'fbherault': (u'France Bleu Hérault (Montpellier)', False),
'fbisere': (u'France Bleu Isère (Grenoble)', False),
'fblarochelle': (u'France Bleu La Rochelle', False),
'fblimousin': (u'France Bleu Limousin (Limoges)', False),
'fbloireocean': (u'France Bleu Loire Océan (Nantes)', False),
'fblorrainenord': (u'France Bleu Lorraine Nord (Metz)', False),
'fbmayenne': (u'France Bleu Mayenne (Laval)', False),
'fbnord': (u'France Bleu Nord (Lille)', False),
'fborleans': (u'France Bleu Orléans', False),
'fbpaysbasque': (u'France Bleu Pays Basque (Bayonne)', False),
'fbpaysdauvergne': (u'France Bleu Pays d\'Auvergne (Clermont-Ferrand)', False),
'fbpaysdesavoie': (u'France Bleu Pays de Savoie (Chambery)', False),
'fbperigord': (u'France Bleu Périgord (Périgueux)', False),
'fbpicardie': (u'France Bleu Picardie (Amiens)', False),
'fbpoitou': (u'France Bleu Poitou (Poitiers)', False),
'fbprovence': (u'France Bleu Provence (Aix-en-Provence)', False),
'fbroussillon': (u'France Bleu Roussillon (Perpigan)', False),
'fbsudlorraine': (u'France Bleu Sud Lorraine (Nancy)', False),
'fbtoulouse': (u'France Bleu Toulouse', False),
'fbtouraine': (u'France Bleu Touraine (Tours)', False),
'fbvaucluse': (u'France Bleu Vaucluse (Avignon)', False),
}
_PLAYERJS_RADIOS = ('franceinter',
'franceculture',
'franceinfo',
'lemouv',
)
_DIRECTJSON_RADIOS = ('lemouv', 'franceinter', )
_RSS_RADIOS = ('francemusique', )
def iter_resources(self, split_path):
if len(split_path) == 1 and split_path[0] == 'francebleu':
for _id in sorted(self._RADIOS.iterkeys()):
if _id.startswith('fb'):
yield self.get_radio(_id)
elif len(split_path) == 0:
for _id in sorted(self._RADIOS.iterkeys()):
if not _id.startswith('fb'):
yield self.get_radio(_id)
yield Collection('francebleu', 'France Bleu',
children=self.iter_resources(['francebleu']))
else:
raise CollectionNotFound(split_path)
def iter_radios_search(self, pattern):
for radio in self._flatten_resources(self.iter_resources([])):
if pattern.lower() in radio.title.lower() or pattern.lower() in radio.description.lower():
yield radio
def get_radio(self, radio):
if not isinstance(radio, Radio):
radio = Radio(radio)
if not radio.id in self._RADIOS:
return None
title, hd = self._RADIOS[radio.id]
radio.title = title
radio.description = title
if hd:
url = self._MP3_HD_URL % (radio.id, radio.id)
else:
url = self._MP3_URL % (radio.id, radio.id)
# This should be asked demand, but is required for now as Radioob
# does not require it.
self.fillobj(radio, ('current', ))
stream = Stream(0)
stream.title = u'128kbits/s' if hd else u'32kbits/s'
stream.url = url
radio.streams = [stream]
return radio
def fill_radio(self, radio, fields):
if 'current' in fields:
artist = None
title = None
if radio.id in self._PLAYERJS_RADIOS:
title = self.browser.get_current_playerjs(radio.id)
if radio.id in self._DIRECTJSON_RADIOS:
artist, dtitle = self.browser.get_current_direct(radio.id)
if dtitle:
if title:
title = "%s [%s]" % (dtitle, title)
else:
title = dtitle
if radio.id in self._RSS_RADIOS:
title = self.browser.get_current_rss(radio.id)
if title:
if not radio.current or radio.current is NotLoaded:
radio.current = Emission(0)
radio.current.title = title
radio.current.arist = artist
return radio
OBJECTS = {Radio: fill_radio}
Reference in a new issue View git blame Copy permalink