do not use thumbnail_url anymore (closes #810), use set_empty_fields() method, do not give fields in constructor of BaseVideo anymore

modules/arte/pages.py

@@ -23,6 +23,8 @@ import re
 import urllib
 
 from weboob.tools.browser import BasePage, BrokenPageError
+from weboob.tools.capabilities.thumbnail import Thumbnail
+from weboob.capabilities import NotAvailable
 
 
 from .video import ArteVideo
@@ -53,7 +55,7 @@ class IndexPage(BasePage):
             video.rating_max = rating_max
 
             thumb = self.parser.select(div, 'img[class=thumbnail]', 1)
-            video.thumbnail_url = 'http://videos.arte.tv' + thumb.attrib['src']
+            video.thumbnail = Thumbnail('http://videos.arte.tv' + thumb.attrib['src'])
 
             try:
                 parts = self.parser.select(div, 'div.duration_thumbnail', 1).text.split(':')
@@ -69,6 +71,8 @@ class IndexPage(BasePage):
             else:
                 video.duration = datetime.timedelta(hours=int(hours), minutes=int(minutes), seconds=int(seconds))
 
+            video.set_empty_fields(NotAvailable, ('url',))
+
             yield video
 
 class VideoPage(BasePage):
@@ -77,6 +81,7 @@ class VideoPage(BasePage):
             video = ArteVideo(self.group_dict['id'])
         video.title = self.get_title()
         video.url = self.get_url(lang, quality)
+        video.set_empty_fields(NotAvailable)
         return video
 
     def get_title(self):

modules/dailymotion/pages.py

@@ -22,7 +22,7 @@ import urllib
 import re
 
 from weboob.tools.capabilities.thumbnail import Thumbnail
-from weboob.capabilities.base import NotAvailable
+from weboob.capabilities import NotAvailable
 from weboob.tools.misc import html2text
 from weboob.tools.browser import BasePage, BrokenPageError
 
@@ -73,8 +73,8 @@ class IndexPage(BasePage):
             rating_div = self.parser.select(div, 'div.small_stars', 1)
             video.rating_max = self.get_rate(rating_div)
             video.rating = self.get_rate(rating_div.find('div'))
-            # XXX missing date
+
-            video.date = NotAvailable
+            video.set_empty_fields(NotAvailable, ('url',))
             yield video
 
     def get_rate(self, div):
@@ -109,4 +109,6 @@ class VideoPage(BasePage):
                     mediaURL = urllib.unquote(mobj.group(1))
                 video.url = mediaURL
 
+        video.set_empty_fields(NotAvailable)
+
         return video

modules/ina/backend.py

@@ -47,7 +47,6 @@ class InaBackend(BaseBackend, ICapVideo):
             return self.browser.search_videos(pattern)
 
     def fill_video(self, video, fields):
-        return video
         if fields != ['thumbnail']:
             # if we don't want only the thumbnail, we probably want also every fields
             with self.browser:

modules/ina/pages/search.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright(C) 2010-2011 Romain Bignon
+# Copyright(C) 2010-2012 Romain Bignon
 #
 # This file is part of weboob.
 #
@@ -21,8 +21,8 @@
 import datetime
 import re
 
-from weboob.tools.browser import BasePage
+from weboob.tools.browser import BasePage, BrokenPageError
-from weboob.tools.browser import BrokenPageError
+from weboob.tools.capabilities.thumbnail import Thumbnail
 
 from ..video import InaVideo
 
@@ -42,24 +42,21 @@ class SearchPage(BasePage):
         for li in ul.findall('li'):
             id = re.sub(self.URL_REGEXP, r'\1', li.find('a').attrib['href'])
 
-            thumbnail = 'http://boutique.ina.fr%s' % li.find('a').find('img').attrib['src']
+            video = InaVideo('boutique.%s' % id)
 
-            title = self.parser.select(li, 'p.titre', 1).text
+            video.thumbnail = Thumbnail('http://boutique.ina.fr%s' % li.find('a').find('img').attrib['src'])
+
+            video.title = self.parser.select(li, 'p.titre', 1).text
 
             date = self.parser.select(li, 'p.date', 1).text
             day, month, year = [int(s) for s in date.split('/')]
-            date = datetime.datetime(year, month, day)
+            video.date = datetime.datetime(year, month, day)
 
             duration = self.parser.select(li, 'p.duree', 1).text
             m = re.match(r'((\d+)h)?((\d+)min)?(\d+)s', duration)
             if m:
-                duration = datetime.timedelta(hours=int(m.group(2) or 0), minutes=int(m.group(4) or 0), seconds=int(m.group(5)))
+                video.duration = datetime.timedelta(hours=int(m.group(2) or 0), minutes=int(m.group(4) or 0), seconds=int(m.group(5)))
             else:
                 raise BrokenPageError('Unable to match duration (%r)' % duration)
 
-            yield InaVideo('boutique.%s' % id,
+            yield video
-                           title=title,
-                           date=date,
-                           duration=duration,
-                           thumbnail_url=thumbnail,
-                          )

modules/ina/pages/video.py

@@ -25,8 +25,8 @@ try:
 except ImportError:
     from cgi import parse_qs
 
-from weboob.tools.browser import BasePage
+from weboob.capabilities import NotAvailable
-from weboob.tools.browser import BrokenPageError
+from weboob.tools.browser import BasePage, BrokenPageError
 
 from ..video import InaVideo
 
@@ -45,6 +45,8 @@ class BaseVideoPage(BasePage):
         video.date = date
         video.duration = duration
         video.description = self.get_description()
+
+        video.set_empty_fields(NotAvailable)
         return video
 
     def get_id(self):

modules/youjizz/pages/index.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright(C) 2010-2011 Roger Philibert
+# Copyright(C) 2010-2012 Roger Philibert
 #
 # This file is part of weboob.
 #
@@ -21,8 +21,8 @@
 import datetime
 import re
 
-from weboob.tools.browser import BasePage
+from weboob.tools.browser import BasePage, BrokenPageError
-from weboob.tools.browser import BrokenPageError
+from weboob.tools.capabilities.thumbnail import Thumbnail
 
 from ..video import YoujizzVideo
 
@@ -38,10 +38,12 @@ class IndexPage(BasePage):
             url = a.attrib['href']
             _id = re.sub(r'/videos/(.+)\.html', r'\1', url)
 
-            thumbnail_url = span.find('.//img').attrib['src']
+            video = YoujizzVideo(_id)
+
+            video.thumbnail = Thumbnail(span.find('.//img').attrib['src'])
 
             title_el = self.parser.select(span, 'span#title1', 1)
-            title = title_el.text.strip()
+            video.title = title_el.text.strip()
 
             time_span = self.parser.select(span, 'span.thumbtime span', 1)
             time_txt = time_span.text.strip().replace(';', ':')
@@ -52,9 +54,6 @@ class IndexPage(BasePage):
             else:
                 raise BrokenPageError('Unable to parse the video duration: %s' % time_txt)
 
+            video.duration = datetime.timedelta(minutes=minutes, seconds=seconds)
 
-            yield YoujizzVideo(_id,
+            yield video
-                               title=title,
-                               duration=datetime.timedelta(minutes=minutes, seconds=seconds),
-                               thumbnail_url=thumbnail_url,
-                               )

modules/youporn/pages/index.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright(C) 2010-2011 Romain Bignon
+# Copyright(C) 2010-2012 Romain Bignon
 #
 # This file is part of weboob.
 #
@@ -21,6 +21,9 @@
 import re
 import datetime
 
+from weboob.capabilities.base import NotAvailable
+from weboob.tools.capabilities.thumbnail import Thumbnail
+
 from .base import PornPage
 from ..video import YoupornVideo
 
@@ -45,7 +48,10 @@ class IndexPage(PornPage):
             url = a.attrib['href']
             _id = url[len('/watch/'):]
             _id = _id[:_id.find('/')]
-            title = a.text.strip()
+
+            video = YoupornVideo(int(_id))
+            video.title = a.text.strip()
+            video.thumbnail = Thumbnail(thumbnail_url)
 
             hours = minutes = seconds = 0
             div = li.cssselect('h2[class=duration]')
@@ -56,19 +62,15 @@ class IndexPage(PornPage):
                 elif len(pack) == 2:
                     minutes, seconds = pack
 
-            rating = 0
+            video.duration = datetime.timedelta(hours=hours, minutes=minutes, seconds=seconds)
-            rating_max = 0
+
             div = li.cssselect('div.stars')
             if div:
                 m = re.match('.*star-(\d).*', div[0].attrib.get('class', ''))
                 if m:
-                    rating = int(m.group(1))
+                    video.rating = int(m.group(1))
-                    rating_max = 5
+                    video.rating_max = 5
 
-            yield YoupornVideo(int(_id),
+            video.set_empty_fields(NotAvailable, ('url', 'author'))
-                               title=title,
+
-                               rating=rating,
+            yield video
-                               rating_max=rating_max,
-                               duration=datetime.timedelta(hours=hours, minutes=minutes, seconds=seconds),
-                               thumbnail_url=thumbnail_url,
-                               )

modules/youporn/pages/video.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright(C) 2010-2011 Romain Bignon
+# Copyright(C) 2010-2012 Romain Bignon
 #
 # This file is part of weboob.
 #
@@ -22,6 +22,7 @@ import re
 import datetime
 from dateutil.parser import parse as parse_dt
 
+from weboob.capabilities.base import NotAvailable
 from weboob.tools.browser import BrokenPageError
 
 from .base import PornPage
@@ -37,6 +38,8 @@ class VideoPage(PornPage):
         video.title = self.get_title()
         video.url, video.ext = self.get_url()
         self.set_details(video)
+
+        video.set_empty_fields(NotAvailable)
         return video
 
     def get_url(self):

modules/youtube/backend.py

@@ -25,7 +25,9 @@ import gdata.youtube.service
 import re
 import urllib
 
+from weboob.capabilities.base import NotAvailable
 from weboob.capabilities.video import ICapVideo
+from weboob.tools.capabilities.thumbnail import Thumbnail
 from weboob.tools.backend import BaseBackend, BackendConfig
 from weboob.tools.misc import to_unicode
 from weboob.tools.value import ValueBackendPassword, Value
@@ -61,11 +63,11 @@ class YoutubeBackend(BaseBackend, ICapVideo):
         """
         Parse an entry returned by gdata and return a Video object.
         """
-        video = YoutubeVideo(to_unicode(entry.id.text.split('/')[-1].strip()),
+        video = YoutubeVideo(to_unicode(entry.id.text.split('/')[-1].strip()))
-                             title=to_unicode(entry.media.title.text.strip()),
+        video.title = to_unicode(entry.media.title.text.strip())
-                             duration=to_unicode(datetime.timedelta(seconds=int(entry.media.duration.seconds.strip()))),
+        video.duration = datetime.timedelta(seconds=int(entry.media.duration.seconds.strip()))
-                             thumbnail_url=to_unicode(entry.media.thumbnail[0].url.strip()),
+        video.thumbnail = Thumbnail(to_unicode(entry.media.thumbnail[0].url.strip()))
-                             )
+
         if entry.author[0].name.text:
             video.author = entry.author[0].name.text.strip()
         if entry.media.name:
@@ -104,6 +106,8 @@ class YoutubeBackend(BaseBackend, ICapVideo):
 
         video = self._entry2video(entry)
         self._set_video_url(video)
+
+        video.set_empty_fields(NotAvailable)
         return video
 
     def search_videos(self, pattern=None, sortby=ICapVideo.SEARCH_RELEVANCE, nsfw=False, max_results=None):

weboob/capabilities/__init__.py

@@ -1 +1,5 @@
+# -*- coding: utf-8 -*-
 
+from .base import NotLoaded, NotAvailable, CapBaseObject, IBaseCap
+
+__all__ = ['NotLoaded', 'NotAvailable', 'CapBaseObject', 'IBaseCap']

weboob/capabilities/video.py

@@ -20,7 +20,7 @@
 
 from datetime import datetime, timedelta
 
-from .base import IBaseCap, CapBaseObject, NotLoaded
+from .base import IBaseCap, CapBaseObject, NotAvailable
 from weboob.tools.capabilities.thumbnail import Thumbnail
 
 
@@ -31,25 +31,21 @@ class BaseVideo(CapBaseObject):
     Represents a video.
     This object has to be inherited to specify how to calculate the URL of the video from its ID.
     """
-    def __init__(self, _id, title=NotLoaded, url=NotLoaded, author=NotLoaded, duration=NotLoaded, date=NotLoaded,
+
-                 rating=NotLoaded, rating_max=NotLoaded, thumbnail=NotLoaded, thumbnail_url=None, nsfw=False):
+    def __init__(self, _id):
         CapBaseObject.__init__(self, unicode(_id))
 
-        self.add_field('title', basestring, title)
+        self.add_field('title', basestring)
-        self.add_field('url', basestring, url)
+        self.add_field('url', basestring)
         self.add_field('ext', basestring)
-        self.add_field('author', basestring, author)
+        self.add_field('author', basestring)
         self.add_field('description', basestring)
-        self.add_field('duration', (int,long,timedelta), duration)
+        self.add_field('duration', (int,long,timedelta))
-        self.add_field('date', datetime, date)
+        self.add_field('date', datetime)
-        self.add_field('rating', (int,long,float), rating)
+        self.add_field('rating', (int,long,float), NotAvailable)
-        self.add_field('rating_max', (int,long,float), rating_max)
+        self.add_field('rating_max', (int,long,float), NotAvailable)
-        self.add_field('thumbnail', Thumbnail, thumbnail)
+        self.add_field('thumbnail', Thumbnail)
-        self.add_field('nsfw', bool, nsfw)
+        self.add_field('nsfw', bool, False)
-
-        # XXX remove this and fix all backends
-        if thumbnail_url is not None and self.thumbnail is NotLoaded:
-            self.thumbnail = Thumbnail(thumbnail_url)
 
     @classmethod
     def id2url(cls, _id):