implement Gazelle.get_torrent() method

weboob/backends/gazelle/pages/torrents.py

@@ -19,7 +19,9 @@ Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 """
 
 import re
+from logging import warning
 
+from weboob.tools.misc import html2text
 from weboob.tools.browser import BasePage
 from weboob.capabilities.torrent import Torrent
 
@@ -37,11 +39,18 @@ class TorrentsPage(BasePage):
             }
         return float(n.replace(',', '')) * m.get(u, 1)
 
+    def format_url(self, url):
+        return '%s://%s/%s' % (self.browser.PROTOCOL,
+                               self.browser.DOMAIN,
+                               url)
+
     def iter_torrents(self):
-        table = self.document.getroot().cssselect('table#torrent_table')
+        table = self.document.getroot().cssselect('table.torrent_table')
         if not table:
             table = self.document.getroot().cssselect('table#browse_torrent_table')
-        if table:
+        if not table:
+            warning('No table found')
+        else:
             table = table[0]
             current_group = None
             for tr in table.findall('tr'):
@@ -57,7 +66,7 @@ class TorrentsPage(BasePage):
                         if current_group:
                             current_group += ' - '
                         current_group += a.text
-                elif tr.attrib.get('class', '').startswith('group_torrent ') or \
+                elif tr.attrib.get('class', '').startswith('group_torrent') or \
                      tr.attrib.get('class', '').startswith('torrent'):
                     tds = tr.findall('td')
 
@@ -87,8 +96,77 @@ class TorrentsPage(BasePage):
 
                     torrent = Torrent(id,
                                       title,
-                                      url=url,
+                                      url=self.format_url(url),
                                       size=size,
                                       seeders=seeders,
                                       leechers=leechers)
                     yield torrent
+                else:
+                    print tr.attrib
+
+    def get_torrent(self, id):
+        table = self.document.getroot().cssselect('div.thin')
+        if not table:
+            warning('No div.thin found')
+            return None
+
+        h2 = table[0].find('h2')
+        title = h2.text or ''
+        if h2.find('a') != None:
+            title += h2.find('a').text + h2.find('a').tail
+
+        torrent = Torrent(id, title)
+        table = self.document.getroot().cssselect('table.torrent_table')
+        if not table:
+            warning('No table found')
+            return None
+
+        for tr in table[0].findall('tr'):
+            if tr.attrib.get('class', '').startswith('group_torrent'):
+                tds = tr.findall('td')
+
+                if not len(tds) == 5:
+                    continue
+
+                url = tds[0].find('span').find('a').attrib['href']
+                id = self.TORRENTID_REGEXP.match(url)
+
+                if not id:
+                    warning('ID not found')
+                    continue
+
+                id = id.group(1)
+
+                if id != torrent.id:
+                    continue
+
+                torrent.url = self.format_url(url)
+                torrent.size = self.unit(*tds[1].text.split())
+                torrent.seeders = int(tds[3].text)
+                torrent.leechers = int(tds[4].text)
+                break
+
+        if not torrent.url:
+            warning('Torrent %d not found in list' % torrent.id)
+            return None
+
+        div = self.document.getroot().cssselect('div.main_column')
+        if not div:
+            warning('WTF')
+            return None
+
+        for box in div[0].cssselect('div.box'):
+            title = None
+            body = None
+
+            title_t = box.cssselect('div.head')
+            if title_t:
+                title = title_t[0].find('strong').text
+            body_t = box.cssselect('div.body')
+            if body_t:
+                body = html2text(self.browser.parser.tostring(body_t[0]))
+
+            if title and body:
+                torrent.description += '%s\n\n%s\n' % (title, body)
+
+        return torrent