upgrade to browser2

modules/youjizz/backend.py

@@ -41,14 +41,12 @@ class YoujizzBackend(BaseBackend, ICapVideo, ICapCollection):
     BROWSER = YoujizzBrowser
 
     def get_video(self, _id):
-        with self.browser:
         video = self.browser.get_video(_id)
         return video
 
     def search_videos(self, pattern, sortby=ICapVideo.SEARCH_RELEVANCE, nsfw=False):
         if not nsfw:
             return set()
-        with self.browser:
         return self.browser.search_videos(pattern)
 
     def fill_video(self, video, fields):

modules/youjizz/browser.py

@@ -18,9 +18,7 @@
 # along with weboob. If not, see <http://www.gnu.org/licenses/>.
 
 
-import urllib
+from weboob.tools.browser2 import PagesBrowser, URL
-
-from weboob.tools.browser import BaseBrowser
 from weboob.tools.browser.decorators import id2url
 
 from .pages.index import IndexPage
@@ -31,27 +29,28 @@ from .video import YoujizzVideo
 __all__ = ['YoujizzBrowser']
 
 
-class YoujizzBrowser(BaseBrowser):
+class YoujizzBrowser(PagesBrowser):
-    DOMAIN = 'youjizz.com'
+    BASEURL = 'http://www.youjizz.com'
-    ENCODING = None
+
-    PAGES = {r'http://.*youjizz\.com/?': IndexPage,
+    index = URL(r'/?(index.php)?$', IndexPage)
-             r'http://.*youjizz\.com/index.php': IndexPage,
+    search = URL(r'/search/(?P<pattern>.+)-(?P<pagenum>\d+).html', IndexPage)
-             r'http://.*youjizz\.com/search/(?P<pattern>.+)\.html': IndexPage,
+    video = URL(r'/videos/(?P<id>.*).html', VideoPage)
-             r'http://.*youjizz\.com/videos/(?P<id>.+)\.html': VideoPage,
-            }
 
     @id2url(YoujizzVideo.id2url)
     def get_video(self, url, video=None):
         self.location(url)
-        assert self.is_on_page(VideoPage), 'Should be on video page.'
+        assert self.video.is_here()
+
         return self.page.get_video(video)
 
     def search_videos(self, pattern):
-        self.location('/search/%s-1.html' % (urllib.quote_plus(pattern.encode('utf-8'))))
+        self.search.go(pattern=pattern, pagenum=1)
-        assert self.is_on_page(IndexPage)
+        assert self.search.is_here()
-        return self.page.iter_videos()
+
+        return self.pagination(lambda: self.page.iter_videos())
 
     def latest_videos(self):
-        self.home()
+        self.index.go()
-        assert self.is_on_page(IndexPage)
+        assert self.index.is_here()
-        return self.page.iter_videos()
+
+        return self.pagination(lambda: self.page.iter_videos())

modules/youjizz/pages/index.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright(C) 2010-2012 Roger Philibert
+# Copyright(C) 2010-2014 Roger Philibert
 #
 # This file is part of weboob.
 #
@@ -21,9 +21,10 @@
 import datetime
 import re
 
-from weboob.tools.browser import BasePage, BrokenPageError
+from weboob.tools.browser2 import HTMLPage
+from weboob.tools.browser2.page import ListElement, method, ItemElement
+from weboob.tools.browser2.filters import Filter, Link, CleanText
 from weboob.capabilities.image import BaseImage
-from weboob.tools.misc import to_unicode
 
 from ..video import YoujizzVideo
 
@@ -31,24 +32,23 @@ from ..video import YoujizzVideo
 __all__ = ['IndexPage']
 
 
-class IndexPage(BasePage):
+class IndexPage(HTMLPage):
-    def iter_videos(self):
+    @method
-        span_list = self.parser.select(self.document.getroot(), 'span#miniatura')
+    class iter_videos(ListElement):
-        for span in span_list:
+        item_xpath = '//span[@id="miniatura"]'
-            a = self.parser.select(span, 'a', 1)
-            url = a.attrib['href']
-            _id = re.sub(r'/videos/(.+)\.html', r'\1', url)
 
-            video = YoujizzVideo(_id)
+        next_page = Link(u'//a[text()="Next »"]')
 
-            video.thumbnail = BaseImage(span.find('.//img').attrib['data-original'])
+        class item(ItemElement):
-            video.thumbnail.url = video.thumbnail.id
+            klass = YoujizzVideo
 
-            title_el = self.parser.select(span, 'span#title1', 1)
+            class Id(Filter):
-            video.title = to_unicode(title_el.text.strip())
+                def filter(self, link):
+                    return re.sub(r'/videos/(.+)\.html', r'\1', link)
 
-            time_span = self.parser.select(span, 'span.thumbtime span', 1)
+            class Duration(Filter):
-            time_txt = time_span.text.strip().replace(';', ':')
+                def filter(self, txt):
+                    time_txt = txt.replace(';', ':')
                     hours, minutes, seconds = 0, 0, 0
                     if ':' in time_txt:
                         t = time_txt.split(':')
@@ -58,8 +58,16 @@ class IndexPage(BasePage):
                         if len(t) == 3:
                             hours = int(t[2])
                     elif time_txt != 'N/A':
-                raise BrokenPageError('Unable to parse the video duration: %s' % time_txt)
+                        raise ValueError('Unable to parse the video duration: %s' % time_txt)
 
-            video.duration = datetime.timedelta(hours=hours, minutes=minutes, seconds=seconds)
+                    return datetime.timedelta(hours=hours, minutes=minutes, seconds=seconds)
 
-            yield video
+
+            obj_id = Id(Link('.//a'))
+            obj_title = CleanText('.//span[@id="title1"]')
+            obj_duration = Duration(CleanText('.//span[@class="thumbtime"]//span'))
+
+            def obj_thumbnail(self):
+                thumbnail = BaseImage(self.xpath('.//img')[0].attrib['data-original'])
+                thumbnail.url = thumbnail.id
+                return thumbnail

modules/youjizz/pages/video.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright(C) 2010-2011 Roger Philibert
+# Copyright(C) 2010-2014 Roger Philibert
 #
 # This file is part of weboob.
 #
@@ -19,11 +19,12 @@
 
 
 import datetime
-import lxml.html
 import re
 
+from weboob.tools.browser2 import HTMLPage
+from weboob.tools.browser2.page import method, ItemElement
+from weboob.tools.browser2.filters import CleanText, Env
 from weboob.capabilities.base import NotAvailable
-from weboob.tools.browser import BasePage, BrokenPageError
 from weboob.tools.misc import to_unicode
 
 from ..video import YoujizzVideo
@@ -32,36 +33,36 @@ from ..video import YoujizzVideo
 __all__ = ['VideoPage']
 
 
-class VideoPage(BasePage):
+class VideoPage(HTMLPage):
-    def get_video(self, video=None):
+    @method
-        _id = to_unicode(self.group_dict['id'])
+    class get_video(ItemElement):
-        if video is None:
+        klass = YoujizzVideo
-            video = YoujizzVideo(_id)
-        title_el = self.parser.select(self.document.getroot(), 'title', 1)
-        video.title = to_unicode(title_el.text.strip())
 
+        obj_id = Env('id')
+        obj_title = CleanText('//title')
+
+        def obj_duration(self):
             # youjizz HTML is crap, we must parse it with regexps
-        data = lxml.html.tostring(self.document.getroot())
+            m = re.search(r'<strong>.*?Runtime.*?</strong> (.+?)</div>', self.page.response.text)
-        m = re.search(r'<strong>.*?Runtime.*?</strong> (.+?)</div>', data)
             if m:
                 txt = m.group(1).strip()
                 if txt == 'Unknown':
-                video.duration = NotAvailable
+                    return NotAvailable
                 else:
                     minutes, seconds = (int(v) for v in to_unicode(txt).split(':'))
-                video.duration = datetime.timedelta(minutes=minutes, seconds=seconds)
+                    return datetime.timedelta(minutes=minutes, seconds=seconds)
             else:
-            raise BrokenPageError('Unable to retrieve video duration')
+                raise ValueError('Unable to retrieve video duration')
 
-        real_id = int(_id.split('-')[-1])
+        def obj_url(self):
-        data = self.browser.readurl('http://www.youjizz.com/videos/embed/%s' % real_id)
+            real_id = int(self.env['id'].split('-')[-1])
+            response = self.page.browser.open('http://www.youjizz.com/videos/embed/%s' % real_id)
+            data = response.text
 
             video_file_urls = re.findall(r'"(http://[^",]+\.youjizz\.com[^",]+\.flv(?:\?[^"]*)?)"', data)
             if len(video_file_urls) == 0:
-            raise BrokenPageError('Video URL not found')
+                raise ValueError('Video URL not found')
             elif len(video_file_urls) > 1:
-            raise BrokenPageError('Many video file URL found')
+                raise ValueError('Many video file URL found')
             else:
-            video.url = to_unicode(video_file_urls[0])
+                return to_unicode(video_file_urls[0])
-
-        return video