support repositories to manage backends (closes #747)

modules/phpbb/__init__.py

@@ -0,0 +1,24 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2010-2011 Romain Bignon
+#
+# 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 .browser import PhpBB
+from .backend import PhpBBBackend
+
+__all__ = ['PhpBB', 'PhpBBBackend']

modules/phpbb/backend.py

@@ -0,0 +1,199 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2010-2011  Romain Bignon
+#
+# 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 __future__ import with_statement
+
+from weboob.tools.backend import BaseBackend, BackendConfig
+from weboob.tools.newsfeed import Newsfeed
+from weboob.tools.value import Value, ValueInt, ValueBackendPassword
+from weboob.tools.misc import limit
+from weboob.capabilities.messages import ICapMessages, ICapMessagesPost, Message, Thread, CantSendMessage
+
+from .browser import PhpBB
+from .tools import rssid, url2id, id2url, id2topic
+
+
+__all__ = ['PhpBBBackend']
+
+
+class PhpBBBackend(BaseBackend, ICapMessages, ICapMessagesPost):
+    NAME = 'phpbb'
+    MAINTAINER = 'Romain Bignon'
+    EMAIL = 'romain@weboob.org'
+    VERSION = '0.a'
+    LICENSE = 'AGPLv3+'
+    DESCRIPTION = "phpBB forum"
+    CONFIG = BackendConfig(Value('url',                     label='URL of forum', regexp='https?://.*'),
+                           Value('username',                label='Username', default=''),
+                           ValueBackendPassword('password', label='Password', default=''),
+                           ValueInt('thread_unread_messages', label='Limit number of unread messages to retrieve for a thread', default=500)
+                          )
+    STORAGE = {'seen': {}}
+    BROWSER = PhpBB
+
+    def create_default_browser(self):
+        username = self.config['username'].get()
+        if len(username) > 0:
+            password = self.config['password'].get()
+        else:
+            password = None
+        return self.create_browser(self.config['url'].get(),
+                                   username, password)
+
+    #### ICapMessages ##############################################
+
+    def _iter_threads(self, root_link=None):
+        with self.browser:
+            links = list(self.browser.iter_links(root_link.url if root_link else None))
+
+        for link in links:
+            if link.type == link.FORUM:
+                link.title = '%s[%s]' % (root_link.title if root_link else '', link.title)
+                for thread in self._iter_threads(link):
+                    yield thread
+            if link.type == link.TOPIC:
+                thread = Thread(url2id(link.url))
+                thread.title = ('%s ' % root_link.title if root_link else '') + link.title
+                thread.date = link.date
+                thread.nb_messages = link.nb_messages
+                thread.flags = thread.IS_DISCUSSION
+                yield thread
+
+    def iter_threads(self):
+        return self._iter_threads()
+
+    def get_thread(self, id):
+        thread = None
+        parent = None
+
+        if isinstance(id, Thread):
+            thread = id
+            id = thread.id
+
+        thread_id = url2id(id, nopost=True) or id
+        try:
+            last_seen_id = self.storage.get('seen', default={})[id2topic(thread_id)]
+        except KeyError:
+            last_seen_id = 0
+
+        with self.browser:
+            for post in self.browser.iter_posts(id):
+                if not thread:
+                    thread = Thread(thread_id)
+                    thread.title = post.title
+
+                m = self._post2message(thread, post)
+                m.parent = parent
+                if last_seen_id < post.id:
+                    m.flags |= Message.IS_UNREAD
+
+                if parent:
+                    parent.children = [m]
+                else:
+                    thread.root = m
+
+                parent = m
+
+        return thread
+
+    def _post2message(self, thread, post):
+        signature = post.signature
+        if signature:
+            signature += '<br />'
+        signature += 'URL: %s' % self.browser.absurl(id2url('%s.%s' % (thread.id, post.id)))
+        return Message(thread=thread,
+                       id=post.id,
+                       title=post.title,
+                       sender=post.author,
+                       receivers=None,
+                       date=post.date,
+                       parent=None,
+                       content=post.content,
+                       signature=signature,
+                       children=[],
+                       flags=Message.IS_HTML)
+
+    def iter_unread_messages(self, thread=None):
+        with self.browser:
+            url = self.browser.get_root_feed_url()
+            for article in Newsfeed(url, rssid).iter_entries():
+                id = url2id(article.link)
+                thread = None
+
+                try:
+                    last_seen_id = self.storage.get('seen', default={})[id2topic(id)]
+                except KeyError:
+                    last_seen_id = 0
+
+                child = None
+                iterator = self.browser.riter_posts(id, last_seen_id)
+                if self.config['thread_unread_messages'].get() > 0:
+                    iterator = limit(iterator, self.config['thread_unread_messages'].get())
+                for post in iterator:
+                    if not thread:
+                        thread = Thread('%s.%s' % (post.forum_id, post.topic_id))
+                    message = self._post2message(thread, post)
+
+                    if child:
+                        message.children.append(child)
+                        child.parent = message
+
+                    if post.parent:
+                        message.parent = Message(thread=thread,
+                                                 id=post.parent)
+                    else:
+                        thread.root = message
+                    yield message
+
+    def set_message_read(self, message):
+        try:
+            last_seen_id = self.storage.get('seen', default={})[id2topic(message.thread.id)]
+        except KeyError:
+            last_seen_id = 0
+
+        if message.id > last_seen_id:
+            self.storage.set('seen', id2topic(message.thread.id), message.id)
+            self.storage.save()
+
+    def fill_thread(self, thread, fields):
+        return self.get_thread(thread)
+
+    #### ICapMessagesReply #########################################
+    def post_message(self, message):
+        assert message.thread
+
+        forum = 0
+        topic = 0
+        if message.thread:
+            try:
+                if '.' in message.thread.id:
+                    forum, topic = [int(i) for i in message.thread.id.split('.', 1)]
+                else:
+                    forum = int(message.thread.id)
+            except ValueError:
+                raise CantSendMessage('Thread ID must be in form "FORUM_ID[.TOPIC_ID]".')
+
+        with self.browser:
+            return self.browser.post_answer(forum,
+                                            topic,
+                                            message.title,
+                                            message.content)
+
+    OBJECTS = {Thread: fill_thread}

modules/phpbb/browser.py

@@ -0,0 +1,201 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2010-2011  Romain Bignon
+#
+# 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/>.
+
+
+import re
+import urllib
+from urlparse import urlsplit
+
+from weboob.tools.browser import BaseBrowser, BrowserIncorrectPassword
+from weboob.capabilities.messages import CantSendMessage
+
+from .pages.index import LoginPage
+from .pages.forum import ForumPage, TopicPage, PostingPage
+from .tools import id2url, url2id
+
+
+__all__ = ['PhpBB']
+
+
+# Browser
+class PhpBB(BaseBrowser):
+    PAGES = {'https?://.*/index.php':                 ForumPage,
+             'https?://.*/':                          ForumPage,
+             'https?://.*/viewforum.php\?f=(\d+)':    ForumPage,
+             'https?://.*/search.php\?.*':            ForumPage,
+             'https?://.*/viewtopic.php\?.*':         TopicPage,
+             'https?://.*/posting.php\?.*':           PostingPage,
+             'https?://.*/ucp.php\?mode=login.*':     LoginPage,
+            }
+
+    last_board_msg_id = None
+
+    def __init__(self, url, *args, **kwargs):
+        self.url = url
+        v = urlsplit(url)
+        self.PROTOCOL = v.scheme
+        self.DOMAIN = v.netloc
+        self.BASEPATH = v.path[:v.path.rfind('/')]
+        BaseBrowser.__init__(self, *args, **kwargs)
+
+    def absurl(self, rel):
+        return BaseBrowser.absurl(self, '%s/%s' % (self.BASEPATH, rel))
+
+    def home(self):
+        self.location(self.url)
+
+    def is_logged(self):
+        return not self.page or self.page.is_logged()
+
+    def login(self):
+        data = {'login': 'Connexion',
+                'username': self.username,
+                'password': self.password,
+               }
+        self.location('%s/ucp.php?mode=login' % self.BASEPATH, urllib.urlencode(data), no_login=True)
+
+        assert self.is_on_page(LoginPage)
+
+        if not self.page.is_logged():
+            raise BrowserIncorrectPassword(self.page.get_error_message())
+
+    def get_root_feed_url(self):
+        self.home()
+        return self.page.get_feed_url()
+
+    def iter_links(self, url):
+        if url:
+            self.location(url)
+        else:
+            self.home()
+
+        assert self.is_on_page(ForumPage)
+        return self.page.iter_links()
+
+    def iter_posts(self, id, stop_id=None):
+        if id.startswith('http'):
+            self.location(id)
+        else:
+            self.location('%s/%s' % (self.BASEPATH, id2url(id)))
+        assert self.is_on_page(TopicPage)
+
+        parent = 0
+        while 1:
+            for post in self.page.iter_posts():
+                if stop_id and post.id >= stop_id:
+                    return
+
+                post.parent = parent
+                yield post
+                parent = post.id
+
+            if self.page.cur_page == self.page.tot_pages:
+                return
+            self.location(self.page.next_page_url())
+
+    def riter_posts(self, id, stop_id=None):
+        if id.startswith('http'):
+            self.location(id)
+        else:
+            self.location('%s/%s' % (self.BASEPATH, id2url(id)))
+        assert self.is_on_page(TopicPage)
+
+        child = None
+        while 1:
+            for post in self.page.riter_posts():
+                if child:
+                    child.parent = post.id
+                    yield child
+                if post.id <= stop_id:
+                    return
+                child = post
+
+            if self.page.cur_page == 1:
+                if child:
+                    yield child
+                return
+            self.location(self.page.prev_page_url())
+
+    def get_post(self, id):
+        if id.startswith('http'):
+            self.location(id)
+            id = url2id(id)
+        else:
+            self.location('%s/%s' % (self.BASEPATH, id2url(id)))
+        assert self.is_on_page(TopicPage)
+
+        post = self.page.get_post(int(id.split('.')[-1]))
+        if not post:
+            return None
+
+        if post.parent == 0 and self.page.cur_page > 1:
+            self.location(self.page.prev_page_url())
+            post.parent = self.page.get_last_post_id()
+
+        return post
+
+    def get_forums(self):
+        self.home()
+        return dict(self.page.iter_all_forums())
+
+    def post_answer(self, forum_id, topic_id, title, content):
+        if topic_id == 0:
+            if not forum_id:
+                forums = self.get_forums()
+                forums_prompt = 'Forums list:\n%s' % ('\n'.join(['\t- %s' % f for f in forums.itervalues()]))
+                m = re.match('\[(.*)\] (.*)', title or '')
+                if not m:
+                    raise CantSendMessage('Please enter a title formatted like that:\n\t"[FORUM] SUBJECT"\n\n%s' % forums_prompt)
+
+                forum_id = None
+                for k,v in forums.iteritems():
+                    if v.lower() == m.group(1).lower():
+                        forum_id = k
+                        break
+
+            if not forum_id:
+                raise CantSendMessage('Forum "%s" not found.\n\n%s' % (m.group(1), forums_prompt))
+
+            self.location('%s/posting.php?mode=post&f=%d' % (self.BASEPATH, forum_id))
+
+            assert self.is_on_page(PostingPage)
+            self.page.post(title, content)
+
+            assert self.is_on_page(PostingPage)
+            error = self.page.get_error_message()
+            if error:
+                raise CantSendMessage(u'Unable to send message: %s' % error)
+        else:
+            self.location('%s/%s' % (self.BASEPATH, id2url(topic_id)))
+            assert self.is_on_page(TopicPage)
+
+            self.page.go_reply()
+            assert self.is_on_page(PostingPage)
+
+            # Don't send title because it isn't needed in real use case
+            # and with monboob title is something like:
+            #   Re: [Forum Name] Re: Topic Name
+            if title is not None and title.startswith('Re:'):
+                title = None
+            self.page.post(title, content)
+
+            assert self.is_on_page(PostingPage)
+            error = self.page.get_error_message()
+            if error:
+                raise CantSendMessage(u'Unable to send message: %s' % error)

modules/phpbb/pages/forum.py

@@ -0,0 +1,235 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2010-2011  Romain Bignon
+#
+# 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 time import sleep
+from urlparse import urlsplit, parse_qs
+
+from weboob.tools.browser import BrokenPageError
+
+from .index import PhpBBPage
+from ..tools import parse_date
+
+
+__all__ = ['Link', 'ForumPage', 'TopicPage', 'PostingPage']
+
+
+class Link(object):
+    (FORUM,
+     TOPIC) = xrange(2)
+
+    def __init__(self, type, url):
+        self.type = type
+        self.url = url
+        self.title = u''
+        self.date = None
+        self.nb_messages = 0
+
+class ForumPage(PhpBBPage):
+    def iter_links(self):
+        for li in self.parser.select(self.document.getroot(), 'ul.forums li.row'):
+            title = li.cssselect('a.forumtitle')[0]
+            link = Link(Link.FORUM, title.attrib['href'])
+            link.title = title.text.strip()
+            yield link
+
+        for li in self.parser.select(self.document.getroot(), 'ul.topics li.row'):
+            title = li.cssselect('a.topictitle')[0]
+            link = Link(Link.TOPIC, title.attrib['href'])
+            link.title = title.text.strip()
+            for a in li.find('dl').find('dt').findall('a'):
+                for text in (a.text, a.tail):
+                    if text is None:
+                        continue
+                    try:
+                        link.date = parse_date(text.strip(u'» \r\n'))
+                    except ValueError:
+                        continue
+                    else:
+                        break
+            # it only lists number of answers, so we add 1.
+            link.nb_messages = int(li.cssselect('dd.posts')[0].text.strip()) + 1
+            yield link
+
+    def iter_all_forums(self):
+        for option in self.parser.select(self.document.getroot(), 'select#f', 1).findall('option'):
+            value = int(option.attrib['value'])
+            if value < 0 or not option.text:
+                continue
+
+            yield value, option.text.strip(u'» \xa0\n\r')
+
+class Post(object):
+    def __init__(self, forum_id, topic_id, id):
+        self.id = int(id)
+        self.forum_id = forum_id
+        self.topic_id = topic_id
+        self.title = u''
+        self.author = u''
+        self.date = None
+        self.content = u''
+        self.signature = u''
+        self.parent = 0
+
+class TopicPage(PhpBBPage):
+    def on_loaded(self):
+        div = self.document.getroot().cssselect('div.pagination')[0]
+        strongs = div.cssselect('strong')
+        self.cur_page = int(strongs[0].text.strip())
+        self.tot_pages = int(strongs[1].text.strip())
+
+        try:
+            url = self.document.xpath('//h2/a')[-1].attrib['href']
+        except BrokenPageError:
+            url = self.url
+        v = urlsplit(url)
+        args = parse_qs(v.query)
+        self.topic_id = int(args['t'][0])
+        self.forum_id = int(args['f'][0]) if 'f' in args else 0
+
+        self.forum_title = u''
+        nav = self.parser.select(self.document.getroot(), 'li.icon-home')
+        if len(nav) > 0:
+            text = nav[0].findall('a')[-1].text.strip()
+            if len(text) >= 20:
+                text = text[:20] + u'…'
+            self.forum_title = '[%s] ' % text
+
+    def go_reply(self):
+        self.browser.follow_link(url_regex='posting\.php')
+
+    def next_page_url(self):
+        try:
+            return self.parser.select(self.document.getroot(), 'a.right-box', 1).attrib['href']
+        except BrokenPageError:
+            a_list = self.parser.select(self.document.getroot(), 'div.pagination', 1).findall('a')
+            if self.cur_page == self.tot_pages:
+                return '#'
+            return a_list[-1].attrib['href']
+
+    def prev_page_url(self):
+        try:
+            return self.parser.select(self.document.getroot(), 'a.left-box', 1).attrib['href']
+        except BrokenPageError:
+            a_list = self.parser.select(self.document.getroot(), 'div.pagination', 1).findall('a')
+            if self.cur_page == self.tot_pages:
+                a = a_list[-1]
+            else:
+                a = a_list[-2]
+            return a.attrib['href']
+
+    def iter_posts(self):
+        for div in self.parser.select(self.document.getroot(), 'div.post'):
+            yield self._get_post(div)
+
+    def riter_posts(self):
+        for div in reversed(self.parser.select(self.document.getroot(), 'div.post')):
+            yield self._get_post(div)
+
+    def get_post(self, id):
+        parent = 0
+        for div in self.parser.select(self.document.getroot(), 'div.post'):
+            if div.attrib['id'] == 'p%d' % id:
+                post = self._get_post(div)
+                post.parent = parent
+                return post
+            else:
+                parent = int(div.attrib['id'][1:])
+
+    def _get_post(self, div):
+        body = div.cssselect('div.postbody')[0]
+        profile = div.cssselect('dl.postprofile')[0]
+
+        id = div.attrib['id'][1:]
+        post = Post(self.forum_id, self.topic_id, id)
+
+        title_tags = body.xpath('//h3/a')
+        if len(title_tags) == 0:
+            title_tags = self.document.xpath('//h2/a')
+        if len(title_tags) == 0:
+            title = u''
+            self.logger.warning('Unable to parse title')
+        else:
+            title = title_tags[-1].text.strip()
+
+        post.title = self.forum_title + title
+        for a in profile.cssselect('dt a'):
+            if a.text:
+                post.author = a.text.strip()
+
+        p_tags = body.cssselect('p.author')
+        if len(p_tags) == 0:
+            p_tags = body.find('p')
+        if len(p_tags) == 0:
+            post.date = None
+            self.logger.warning('Unable to parse datetime')
+        else:
+            p = p_tags[0]
+            text = p.find('strong') is not None and p.find('strong').tail
+            if not text:
+                text = p.text[4:]
+
+            text = text.strip(u'» \n\r')
+            try:
+                post.date = parse_date(text)
+            except ValueError:
+                self.logger.warning(u'Unable to parse datetime "%s"' % text)
+
+        post.content = self.parser.tostring(body.cssselect('div.content')[0])
+
+        signature = body.cssselect('div.signature')
+        if len(signature) > 0:
+            post.signature = self.parser.tostring(signature[0])
+        return post
+
+    def get_last_post_id(self):
+        id = 0
+        for div in self.parser.select(self.document.getroot(), 'div.post'):
+            id = int(div.attrib['id'][1:])
+        return id
+
+
+class PostingPage(PhpBBPage):
+    def post(self, title, content):
+        self.browser.select_form(predicate=lambda form: form.attrs.get('id', '') == 'postform')
+        self.browser.set_all_readonly(False)
+
+        if title:
+            self.browser['subject'] = title.encode('utf-8')
+        self.browser['message'] = content.encode('utf-8')
+
+        # This code on phpbb:
+        #   if ($cancel || ($current_time - $lastclick < 2 && $submit))
+        #   {
+        #       /* ... */
+        #       redirect($redirect);
+        #   }
+        # To prevent that shit because weboob is too fast, we simulate
+        # a value of lastclick 10 seconds before.
+        self.browser['lastclick'] = str(int(self.browser['lastclick']) - 10)
+
+        # Likewise for create_time, with this check:
+        #   $diff = time() - $creation_time;
+        #   // If creation_time and the time() now is zero we can assume it was not a human doing this (the check for if ($diff)...
+        #   if ($diff && ($diff <= $timespan || $timespan === -1))
+        # But as the form_token depends on the create_time value, I can't
+        # change it. But I can wait a second before posting...
+        sleep(1)
+
+        self.browser.submit(name='post')

modules/phpbb/pages/index.py

@@ -0,0 +1,39 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2010-2011  Romain Bignon
+#
+# 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.tools.browser import BasePage
+
+class PhpBBPage(BasePage):
+    def is_logged(self):
+        return len(self.document.getroot().cssselect('li.icon-register')) == 0
+
+    def get_feed_url(self):
+        links = self.document.getroot().cssselect('link[type="application/atom+xml"]')
+        return links[-1].attrib['href']
+
+    def get_error_message(self):
+        errors = []
+        for div in self.parser.select(self.document.getroot(), 'div.error,p.error'):
+            if div.text:
+                errors.append(div.text.strip())
+        return ', '.join(errors)
+
+class LoginPage(PhpBBPage):
+    pass

modules/phpbb/test.py

@@ -0,0 +1,36 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2010-2011  Romain Bignon
+#
+# 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.tools.test import BackendTest
+
+
+__all__ = ['PhpBBTest']
+
+
+class PhpBBTest(BackendTest):
+    BACKEND = 'phpbb'
+
+    def testthreads(self):
+        for thread in self.backend.iter_threads():
+            pass
+
+    def test_unread_messages(self):
+        for message in self.backend.iter_unread_messages():
+            pass

modules/phpbb/tools.py

@@ -0,0 +1,75 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2010-2011  Romain Bignon
+#
+# 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 dateutil.parser import parse as _parse_dt
+from urlparse import urlsplit, parse_qs
+
+from weboob.tools.misc import local2utc
+
+
+def url2id(url, nopost=False):
+    v = urlsplit(url)
+    pagename = v.path.split('/')[-1]
+    args = parse_qs(v.query)
+    if pagename == 'viewforum.php':
+        return '%d' % int(args['f'][0])
+    if pagename == 'viewtopic.php':
+        if 'f' in args:
+            s = '%d' % int(args['f'][0])
+        else:
+            s = '0'
+        s += '.%d' % int(args['t'][0])
+        if 'p' in args and not nopost:
+            s += '.%d' % int(args['p'][0])
+        return s
+
+    return None
+
+def id2url(id):
+    v = id.split('.')
+    if len(v) == 1:
+        return 'viewforum.php?f=%d' % int(v[0])
+    if len(v) == 2:
+        return 'viewtopic.php?f=%d&t=%d' % (int(v[0]), int(v[1]))
+    if len(v) == 3:
+        return 'viewtopic.php?f=%d&t=%d&p=%d#p%d' % (int(v[0]),
+                                                     int(v[1]),
+                                                     int(v[2]),
+                                                     int(v[2]))
+
+def id2topic(id):
+    try:
+        return int(id.split('.')[1])
+    except IndexError:
+        return None
+
+def rssid(id):
+    return id
+
+def parse_date(s):
+    s = s.replace(u'Fév', 'Feb') \
+         .replace(u'Avr', 'Apr') \
+         .replace(u'Mai', 'May') \
+         .replace(u'Juin', 'Jun') \
+         .replace(u'Juil', 'Jul') \
+         .replace(u'Aoû', 'Aug') \
+         .replace(u'Ao\xfbt', 'Aug') \
+         .replace(u'Déc', 'Dec')
+    return local2utc(_parse_dt(s))