rename parser/parsers module, add get_parsers() with preference_order

weboob/tools/parsers/__init__.py

@@ -0,0 +1,63 @@
+# -*- coding: utf-8 -*-
+
+"""
+Copyright(C) 2010  Christophe Benz, 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.
+
+"""
+
+import logging
+
+
+__all__ = ['get_parser', 'NoParserFound']
+
+
+class NoParserFound(Exception): pass
+
+
+def get_parser(preference_order=['lxml', 'html5lib', 'elementtidy', 'builtin'], *args, **kwargs):
+    """
+    Get a parser from a preference order list.
+    This allows Weboob to run on systems without lxml, which is the default parser.
+    Return a parser implementing IParser.
+    """
+    if not isinstance(preference_order, (tuple, list)):
+        preference_order = [preference_order]
+    for kind in preference_order:
+        if kind == 'lxml':
+            try:
+                from .lxmlparser import LxmlHtmlParser
+                return LxmlHtmlParser()
+            except ImportError:
+                logging.debug('%s is not installed.' % kind)
+        elif kind == 'html5lib':
+            try:
+                from .html5libparser import Html5libParser
+                return Html5libParser(*args, **kwargs)
+            except ImportError:
+                logging.debug('%s is not installed.' % kind)
+        elif kind == 'elementtidy':
+            try:
+                from .elementtidyparser import ElementTidyParser
+                return ElementTidyParser()
+            except ImportError:
+                logging.debug('%s is not installed.' % kind)
+        elif kind == 'builtin':
+            try:
+                from .htmlparser import HTMLParser
+                return HTMLParser()
+            except ImportError:
+                logging.debug('%s is not installed.' % kind)
+    raise NoParserFound()

weboob/tools/parsers/elementtidyparser.py

@@ -0,0 +1,67 @@
+# -*- 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.
+
+"""
+
+# XXX Currently, elementtidy segfaults when there are no error, because of
+# the behavior of libtidy.
+# A patch has been sent to Debian:
+# http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=576343
+#
+# As it is not integrated in Debian yet, and as this problem persists on other
+# systems, using elementtidy is for now to avoid.
+
+from elementtidy import TidyHTMLTreeBuilder
+try:
+    from xml.etree import cElementTree as ElementTree
+except ImportError:
+    from xml.etree import ElementTree
+
+from .iparser import IParser
+
+
+__all__ = ['ElementTidyParser']
+
+
+class ElementTidyParser(IParser):
+    def parse(self, data, encoding=None):
+        TidyHTMLTreeBuilder.ElementTree = ElementTree
+        HTMLTreeBuilder = TidyHTMLTreeBuilder.TidyHTMLTreeBuilder
+        parser = HTMLTreeBuilder(encoding)
+        tree = ElementTree.parse(data, parser)
+        for elem in tree.getiterator():
+            if elem.tag.startswith('{'):
+                elem.tag = elem.tag[elem.tag.find('}')+1:]
+        return tree
+
+    def tostring(self, element):
+        e = ElementTree.Element('body')
+        e.text = element.text
+        e.tail = element.tail
+        for sub in element.getchildren():
+            e.append(sub)
+        s = ''
+        # XXX OK if it doesn't work with utf-8, the result will be fucking ugly.
+        for encoding in ('utf-8', 'ISO-8859-1'):
+            try:
+                s = ElementTree.tostring(e, encoding)
+            except UnicodeError:
+                continue
+            else:
+                break
+        return unicode(s)

weboob/tools/parsers/html5libparser.py

@@ -0,0 +1,54 @@
+# -*- coding: utf-8 -*-
+
+"""
+Copyright(C) 2010  Romain Bignon, Christophe Benz
+
+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.
+
+"""
+
+from html5lib import treebuilders, HTMLParser
+try:
+    from xml.etree import cElementTree as ElementTree
+except ImportError:
+    from xml.etree import ElementTree
+
+from .iparser import IParser
+
+
+__all__ = ['Html5libParser']
+
+
+class Html5libParser(HTMLParser, IParser):
+    """
+    Parser using html5lib.
+
+    Note that it is not available on every systems.
+    """
+
+    # Default implementation for each type of API.
+    defaults = {'etree': ElementTree,
+               }
+    def __init__(self, api='etree'):
+        # if no default implementation is defined for this api, set it to None
+        # to let getTreeBuilder() using the corresponding implementation.
+        implementation = self.defaults.get(api, None)
+        HTMLParser.__init__(self, tree=treebuilders.getTreeBuilder(api, implementation))
+
+    def parse(self, data, encoding):
+        return HTMLParser.parse(self, data, encoding=encoding)
+
+    def tostring(self, elem):
+        # TODO
+        raise NotImplementedError()

weboob/tools/parsers/htmlparser.py

@@ -0,0 +1,95 @@
+# -*- 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.
+
+"""
+
+from HTMLParser import HTMLParser as _HTMLParser
+import htmlentitydefs
+try:
+    from xml.etree import cElementTree as ElementTree
+except ImportError:
+    from xml.etree import ElementTree
+
+from .iparser import IParser
+
+
+__all__ = ['HTMLParser']
+
+
+class HTMLTreeBuilder(_HTMLParser):
+    def __init__(self, encoding=None):
+        _HTMLParser.__init__(self)
+        self._target = ElementTree.TreeBuilder()
+
+    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_charref(self, name):
+        self._target.data(unichr(int(name)))
+
+    def handle_entityref(self, name):
+        try:
+            self._target.data(unichr(htmlentitydefs.name2codepoint[name]))
+        except KeyError:
+            self._target.data('&' + name)
+
+    def handle_data(self, data):
+        self._target.data(data)
+
+    def handle_endtag(self, tag):
+        try:
+            self._target.end(tag)
+        except:
+            pass
+
+class HTMLParser(IParser):
+    def parse(self, data, encoding=None):
+        parser = HTMLTreeBuilder(encoding)
+        tree = ElementTree.parse(data, parser)
+        for elem in tree.getiterator():
+            if elem.tag.startswith('{'):
+                elem.tag = elem.tag[elem.tag.find('}')+1:]
+        return tree
+
+    def tostring(self, element):
+        e = ElementTree.Element('body')
+        e.text = element.text
+        e.tail = element.tail
+        for sub in element.getchildren():
+            e.append(sub)
+        s = ''
+        # XXX OK if it doesn't work with utf-8, the result will be fucking ugly.
+        for encoding in ('utf-8', 'ISO-8859-1'):
+            try:
+                s = ElementTree.tostring(e, encoding)
+            except UnicodeError:
+                continue
+            else:
+                break
+        return unicode(s)

weboob/tools/parsers/iparser.py

@@ -0,0 +1,36 @@
+# -*- 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.
+
+"""
+
+class IParser(object):
+    def parse(self, data, encoding=None):
+        """
+        Parse a HTML document with a specific encoding to get a tree.
+
+        @param data  [str] HTML document
+        @param encoding  [str] encoding to use
+        @return  an object with the structured document
+        """
+        raise NotImplementedError()
+
+    def tostring(self, elem):
+        """
+        Get HTML string from an element.
+        """
+        raise NotImplementedError()

weboob/tools/parsers/lxmlparser.py

@@ -0,0 +1,41 @@
+# -*- coding: utf-8 -*-
+
+"""
+Copyright(C) 2010  Christophe Benz
+
+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.
+
+"""
+
+import lxml.html
+
+from .iparser import IParser
+
+
+__all__ = ['LxmlHtmlParser']
+
+
+class LxmlHtmlParser(IParser):
+    """
+    Parser using lxml.
+
+    Note that it is not available on every systems.
+    """
+
+    def parse(self, data, encoding=None):
+        parser = lxml.html.HTMLParser(encoding=encoding)
+        return lxml.html.parse(data, parser)
+
+    def tostring(self, element):
+        return lxml.html.tostring(element, encoding=unicode)