use cElementTree and elementtidy as default encoders, and added wrappers to ElementParser and HTMLParser if they are missing

weboob/backends/aum/adopte.py

@@ -20,6 +20,7 @@ Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 
 import time
 from logging import warning
+from html5lib import treebuilders, HTMLParser
 
 from weboob.tools.browser import Browser
 from weboob.backends.aum.exceptions import AdopteWait
@@ -34,6 +35,13 @@ from weboob.backends.aum.pages.login import LoginPage, RedirectPage, BanPage, Er
 from weboob.backends.aum.pages.edit import EditPhotoPage, EditPhotoCbPage, EditAnnouncePage, EditDescriptionPage, EditSexPage, EditPersonalityPage
 from weboob.backends.aum.pages.wait import WaitPage
 
+class AdopteParser(HTMLParser):
+    def __init__(self):
+        HTMLParser.__init__(self, tree=treebuilders.getTreeBuilder("dom"))
+
+    def parse(self, data):
+        return HTMLParser.parse(self, data, encoding='iso-8859-1')
+
 class AdopteUnMec(Browser):
     DOMAIN = 'www.adopteunmec.com'
     PROTOCOL = 'http'
@@ -66,6 +74,7 @@ class AdopteUnMec(Browser):
             }
 
     def __init__(self, *args, **kwargs):
+        kwargs['parser'] = AdopteParser
         Browser.__init__(self, *args, **kwargs)
         self.my_id = 0
 

weboob/backends/dlfp/pages/index.py

@@ -19,12 +19,12 @@ Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 """
 
 from weboob.tools.browser import BrowserIncorrectPassword, BasePage
+from weboob.capabilities.messages import Message
 
 class DLFPPage(BasePage):
     def is_logged(self):
-        forms = self.document.getElementsByTagName('form')
-        for form in forms:
-            if form.getAttribute('id') == 'formulaire':
+        for form in self.document.getiterator('form'):
+            if form.attrib.get('id', None) == 'formulaire':
                 return False
 
         return True
@@ -38,9 +38,7 @@ class LoginPage(DLFPPage):
             raise BrowserIncorrectPassword()
 
     def has_error(self):
-        plist = self.document.getElementsByTagName('p')
-        for p in plist:
-            p = p.childNodes[0]
-            if hasattr(p, 'data') and p.data.startswith(u'Vous avez rentré un mauvais mot de passe'):
+        for p in self.document.getiterator('p'):
+            if p.text and p.text.startswith(u'Vous avez rentré un mauvais mot de passe'):
                 return True
         return False

weboob/tools/browser.py

@@ -21,32 +21,23 @@ Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 import mechanize
 import urllib2
 import ClientForm
-try:
-    from html5lib import treebuilders, HTMLParser
-except ImportError:
-    # XXX change this to use another lib than html5lib
-    class StandardParser:
-        def parse(self, data):
-            return None
-else:
-    class StandardParser(HTMLParser):
-        def __init__(self):
-            HTMLParser.__init__(self, tree=treebuilders.getTreeBuilder("dom"))
-
-        def parse(self, data):
-            return HTMLParser.parse(data, encoding='iso-8859-1')
 import re
 import time
 from logging import warning, error, debug
 from copy import copy
 
+from weboob.tools.parser import StandardParser
+
+# Try to load cookies
 try:
     from weboob.tools.firefox_cookies import FirefoxCookieJar
-    HAVE_COOKIES = True
 except ImportError, e:
     warning("Unable to store cookies: %s" % e)
     HAVE_COOKIES = False
+else:
+    HAVE_COOKIES = True
 
+# Exceptions
 class BrowserIncorrectPassword(Exception):
     pass
 
@@ -57,6 +48,9 @@ class BrowserRetry(Exception):
     pass
 
 class NoHistory:
+    """
+    We don't want to fill memory with history
+    """
     def __init__(self): pass
     def add(self, request, response): pass
     def back(self, n, _response): pass
@@ -64,15 +58,24 @@ class NoHistory:
     def close(self): pass
 
 class BasePage:
+    """
+    Base page
+    """
     def __init__(self, browser, document, url=''):
         self.browser = browser
         self.document = document
         self.url = url
 
     def loaded(self):
+        """
+        Called when the page is loaded.
+        """
         pass
 
 class Browser(mechanize.Browser):
+    """
+    Base browser class to navigate on a website.
+    """
 
     # ------ Class attributes --------------------------------------
 
@@ -83,21 +86,27 @@ class Browser(mechanize.Browser):
 
     # ------ Abstract methods --------------------------------------
 
-    # Go to home
     def home(self):
+        """
+        Go to the home page.
+        """
         raise NotImplementedError()
 
-    # Login to the website
     def login(self):
+        """
+        Login to the website.
+        """
         raise NotImplementedError()
 
-    # Return True if we are logged on website
     def is_logged(self):
+        """
+        Return True if we are loggen on website.
+        """
         raise NotImplementedError()
 
     # ------ Browser methods ---------------------------------------
 
-    def __init__(self, username, password=None, firefox_cookies=None, parser=StandardParser):
+    def __init__(self, username=None, password=None, firefox_cookies=None, parser=StandardParser):
         mechanize.Browser.__init__(self, history=NoHistory())
         self.addheaders = [
                 ['User-agent', self.USER_AGENT]
@@ -127,7 +136,7 @@ class Browser(mechanize.Browser):
 
     def pageaccess(func):
         def inner(self, *args, **kwargs):
-            if not self.page or not self.page.is_logged() and self.password:
+            if not self.page or self.password and not self.page.is_logged():
                 self.home()
 
             return func(self, *args, **kwargs)
@@ -227,7 +236,7 @@ class Browser(mechanize.Browser):
         self.page = pageCls(self, document, result.geturl())
         self.page.loaded()
 
-        if not self.is_logged() and self.password:
+        if self.password and not self.is_logged():
             print '!! Relogin !!'
             self.login()
             return

weboob/tools/parser.py

@@ -0,0 +1,71 @@
+# -*- coding: utf-8 -*-
+
+"""
+Copyright(C) 2010  Romain Bignon
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation, version 3 of the License.
+
+This program 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 General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; if not, write to the Free Software
+Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+
+"""
+
+try:
+    from xml.etree import cElementTree as ElementTree
+except ImportError:
+    from xml.etree import ElementTree
+
+try:
+    from elementtidy import TidyHTMLTreeBuilder
+    TidyHTMLTreeBuilder.ElementTree = ElementTree # force cElementTree if using it.
+    class HTMLTreeBuilder(TidyHTMLTreeBuilder.TidyHTMLTreeBuilder):
+        def __init__(self):
+            TidyHTMLTreeBuilder.TidyHTMLTreeBuilder.__init__(self, 'utf-8')
+
+except ImportError:
+    from HTMLParser import HTMLParser
+
+    class HTMLTreeBuilder(HTMLParser):
+        def __init__(self, html=0, target=None):
+            HTMLParser.__init__(self)
+            if target is None:
+                target = ElementTree.TreeBuilder()
+            self._target = target
+
+        def doctype(self, name, pubid, system):
+            pass
+
+        def close(self):
+            tree = self._target.close()
+            return tree
+
+        def handle_starttag(self, tag, attrs):
+            self._target.start(tag, dict(attrs))
+
+        def handle_startendtag(self, tag, attrs):
+            self._target.start(tag, dict(attrs))
+            self._target.end(tag)
+
+        def handle_data(self, data):
+            self._target.data(data)
+
+        def handle_endtag(self, tag):
+            self._target.end(tag)
+
+class StandardParser(object):
+    def parse(self, data):
+        parser = HTMLTreeBuilder()
+        tree = ElementTree.parse(data, parser)
+
+        for elem in tree.getiterator():
+            if elem.tag.startswith('{'):
+                elem.tag = elem.tag[elem.tag.find('}')+1:]
+        return tree