First implementation of taz.de module

modules/taz/__init__.py

@@ -0,0 +1,24 @@
+"NewspaperTazBackend init"
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012  Florent Fourcot
+#
+# 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 .backend import NewspaperTazBackend
+
+
+__all__ = ['NewspaperTazBackend']

modules/taz/backend.py

@@ -0,0 +1,39 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Florent Fourcot
+#
+# 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/>.
+"backend for http://www.taz.de"
+
+from weboob.capabilities.messages import ICapMessages
+from weboob.tools.capabilities.messages.GenericBackend import GenericNewspaperBackend
+from .browser import NewspaperTazBrowser
+from .tools import rssid, url2id
+
+
+class NewspaperTazBackend(GenericNewspaperBackend, ICapMessages):
+    MAINTAINER = 'Florent Fourcot'
+    EMAIL = 'weboob@flo.fourcot.fr'
+    VERSION = '0.c'
+    LICENSE = 'AGPLv3+'
+    STORAGE = {'seen': {}}
+    NAME = 'taz'
+    DESCRIPTION = u'Taz newspaper website'
+    BROWSER = NewspaperTazBrowser
+    RSSID = staticmethod(rssid)
+    URL2ID = staticmethod(url2id)
+    RSSSIZE = 30
+    RSS_FEED = "http://www.taz.de/!p3270;rss/"

modules/taz/browser.py

@@ -0,0 +1,43 @@
+"browser for taz website"
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Florent Fourcot
+#
+# 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 .pages.article import ArticlePage
+from weboob.tools.browser import BaseBrowser
+
+
+class NewspaperTazBrowser(BaseBrowser):
+    "NewspaperTazBrowser class"
+    PAGES = {
+             "http://www.taz.de/.*": ArticlePage,
+            }
+
+    def is_logged(self):
+        return False
+
+    def login(self):
+        pass
+
+    def fillobj(self, obj, fields):
+        pass
+
+    def get_content(self, _id):
+        "return page article content"
+        self.location(_id)
+        return self.page.get_article(_id)

modules/taz/pages/article.py

@@ -0,0 +1,51 @@
+"ArticlePage object for Taz newspaper"
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Florent Fourcot
+#
+# 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.capabilities.messages.genericArticle import GenericNewsPage
+
+
+class ArticlePage(GenericNewsPage):
+    "ArticlePage object for taz"
+
+    def on_loaded(self):
+        self.main_div = self.document.getroot()
+        self.element_title_selector = "title"
+        self.element_author_selector = ".content-author>a"
+
+    def get_body(self):
+        part = self.document.getroot().xpath('//p[@class="article"]')
+        total = ""
+        for p in part:
+            if p.text:
+                total += "<p>"
+                total += self.browser.parser.tostring(p)
+                total += "</p>"
+        return total
+
+    def get_title(self):
+        title = GenericNewsPage.get_title(self)
+        return title
+
+    def get_author(self):
+        author = self.document.getroot().xpath('//span[@class="author"]')
+        if author:
+            return author[0].text.replace('von ', '')
+        else:
+            return ""

modules/taz/test.py

@@ -0,0 +1,32 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Florent Fourcot
+#
+# 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__ = ['TazTest']
+
+
+class TazTest(BackendTest):
+    BACKEND = 'taz'
+
+    def test_new_messages(self):
+        for message in self.backend.iter_unread_messages():
+            pass

modules/taz/tools.py

@@ -0,0 +1,32 @@
+"tools for taz module"
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012  Florent Fourcot
+# 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
+
+
+def url2id(url):
+    "return an id from an url"
+    regexp = re.compile(".*/!([0-9]+)/")
+    id = regexp.match(url).group(1)
+    return id
+
+
+def rssid(entry):
+    return url2id(entry.link)