From scoder at codespeak.net  Tue Oct 13 09:15:03 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Tue, 13 Oct 2009 09:15:03 +0200 (CEST)
Subject: [Lxml-checkins] r68360 - in lxml/trunk: . src/lxml/tests
Message-ID: <20091013071503.F09A0168053@codespeak.net>

Author: scoder
Date: Tue Oct 13 09:15:02 2009
New Revision: 68360

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/tests/common_imports.py
   lxml/trunk/src/lxml/tests/test_elementtree.py
Log:
 r5272 at delle:  sbehnel | 2009-10-01 12:25:59 +0200
 enable compatible comparison tests for ET 1.2.x


Modified: lxml/trunk/src/lxml/tests/common_imports.py
==============================================================================
--- lxml/trunk/src/lxml/tests/common_imports.py	(original)
+++ lxml/trunk/src/lxml/tests/common_imports.py	Tue Oct 13 09:15:02 2009
@@ -22,9 +22,9 @@
         ElementTree = None
 
 if hasattr(ElementTree, 'VERSION'):
-    if make_version_tuple(ElementTree.VERSION)[:2] < (1,3):
-        # compatibility tests require ET 1.3+
-        ElementTree = None
+    ET_VERSION = make_version_tuple(ElementTree.VERSION)
+else:
+    ET_VERSION = (0,0,0)
 
 try:
     import cElementTree # standard ET
@@ -35,9 +35,18 @@
         cElementTree = None
 
 if hasattr(cElementTree, 'VERSION'):
-    if make_version_tuple(cElementTree.VERSION)[:2] <= (1,0):
-        # compatibility tests do not run with cET 1.0.7
-        cElementTree = None
+    CET_VERSION = make_version_tuple(cElementTree.VERSION)
+else:
+    CET_VERSION = (0,0,0)
+
+def filter_by_version(test_class, version_dict, current_version):
+    """Remove test methods that do not work with the current lib version.
+    """
+    find_required_version = version_dict.get
+    for name in dir(test_class):
+        expected_version = find_required_version(name, (0,0,0))
+        if expected_version > current_version:
+            setattr(test_class, name, None)
 
 try:
     import doctest

Modified: lxml/trunk/src/lxml/tests/test_elementtree.py
==============================================================================
--- lxml/trunk/src/lxml/tests/test_elementtree.py	(original)
+++ lxml/trunk/src/lxml/tests/test_elementtree.py	Tue Oct 13 09:15:02 2009
@@ -15,14 +15,19 @@
 if this_dir not in sys.path:
     sys.path.insert(0, this_dir) # needed for Py3
 
-from common_imports import StringIO, BytesIO, etree, ElementTree, cElementTree
-from common_imports import fileInTestDir, canonicalize, HelperTestCase
+from common_imports import StringIO, BytesIO, etree
+from common_imports import ElementTree, cElementTree, ET_VERSION, CET_VERSION
+from common_imports import filter_by_version, fileInTestDir, canonicalize, HelperTestCase
 from common_imports import _str, _bytes
 
+if cElementTree is not None and CET_VERSION <= (1,0,7):
+    cElementTree = None
+
+if ElementTree is not None:
+    print("Comparing with ElementTree %s" % getattr(ElementTree, "VERSION", "?"))
+
 if cElementTree is not None:
-    if tuple([int(n) for n in
-              getattr(cElementTree, "VERSION", "0.0").split(".")]) <= (1,0,7):
-        cElementTree = None
+    print("Comparing with cElementTree %s" % getattr(cElementTree, "VERSION", "?"))
 
 try:
     reversed
@@ -34,6 +39,8 @@
 
 class ETreeTestCaseBase(HelperTestCase):
     etree = None
+    required_versions_ET = {}
+    required_versions_cET = {}
 
     def test_element(self):
         for i in range(10):
@@ -658,6 +665,7 @@
         self.assertEquals(0, len(root))
         self.assertEquals('This is a text.', root.text)
 
+    required_versions_ET['test_fromstringlist'] = (1,3)
     def test_fromstringlist(self):
         fromstringlist = self.etree.fromstringlist
 
@@ -666,6 +674,7 @@
         self.assertEquals(0, len(root))
         self.assertEquals('This is a text.', root.text)
 
+    required_versions_ET['test_fromstringlist_characters'] = (1,3)
     def test_fromstringlist_characters(self):
         fromstringlist = self.etree.fromstringlist
 
@@ -673,6 +682,7 @@
         self.assertEquals(0, len(root))
         self.assertEquals('This is a text.', root.text)
 
+    required_versions_ET['test_fromstringlist_single'] = (1,3)
     def test_fromstringlist_single(self):
         fromstringlist = self.etree.fromstringlist
 
@@ -780,6 +790,7 @@
                 result.append(el1.tag)
         self.assertEquals(['one','one', 'two', 'two', 'one', 'two'], result)
 
+    required_versions_ET['test_itertext'] = (1,3)
     def test_itertext(self):
         # ET 1.3+
         XML = self.etree.XML
@@ -789,6 +800,7 @@
         self.assertEquals(["RTEXT", "ATAIL", "CTEXT", "CTAIL"],
                           text)
 
+    required_versions_ET['test_itertext_child'] = (1,3)
     def test_itertext_child(self):
         # ET 1.3+
         XML = self.etree.XML
@@ -868,6 +880,7 @@
                 _bytes('<doc%s>This is a test.</doc%s>' % (i, i)),
                 canonicalize(data))
 
+    required_versions_ET['test_write_method_html'] = (1,3)
     def test_write_method_html(self):
         ElementTree = self.etree.ElementTree
         Element = self.etree.Element
@@ -887,6 +900,7 @@
         self.assertEquals(_bytes('<html><body><p>html<br>test</p></body></html>'),
                           data)
 
+    required_versions_ET['test_write_method_text'] = (1,3)
     def test_write_method_text(self):
         ElementTree = self.etree.ElementTree
         Element = self.etree.Element
@@ -1062,6 +1076,7 @@
             a.tail)
         self.assertXML(_bytes('<a></a>'), a)
 
+    required_versions_ET['test_extend'] = (1,3)
     def test_extend(self):
         root = self.etree.Element('foo')
         for i in range(3):
@@ -1099,6 +1114,7 @@
         self.assertEquals(a[0].text, 'foo')
 
     # ElementTree < 1.3 adds whitespace around comments
+    required_versions_ET['test_comment_text'] = (1,3)
     def test_comment_text(self):
         Element = self.etree.Element
         SubElement = self.etree.SubElement
@@ -1119,8 +1135,9 @@
         self.assertEquals(
             _bytes('<a><!--TEST--></a>'),
             tostring(a))
-        
+
     # ElementTree < 1.3 adds whitespace around comments
+    required_versions_ET['test_comment_whitespace'] = (1,3)
     def test_comment_whitespace(self):
         Element = self.etree.Element
         SubElement = self.etree.SubElement
@@ -1654,6 +1671,7 @@
             _bytes('<c hoi="dag"></c>'),
             b)
 
+    required_versions_ET['test_iter'] = (1,3)
     def test_iter(self):
         Element = self.etree.Element
         SubElement = self.etree.SubElement
@@ -2607,6 +2625,7 @@
         self.assert_(tostring(b) == _bytes('<b/>Foo') or
                      tostring(b) == _bytes('<b />Foo'))
 
+    required_versions_ET['test_tostring_method_html'] = (1,3)
     def test_tostring_method_html(self):
         tostring = self.etree.tostring
         Element = self.etree.Element
@@ -2621,6 +2640,7 @@
         self.assertEquals(_bytes('<html><body><p>html<br>test</p></body></html>'),
                           tostring(html, method="html"))
 
+    required_versions_ET['test_tostring_method_text'] = (1,3)
     def test_tostring_method_text(self):
         tostring = self.etree.tostring
         Element = self.etree.Element
@@ -2784,6 +2804,7 @@
         parse = self.etree.parse
         self.assertRaises(TypeError, parse, None)
 
+    required_versions_ET['test_parse_error'] = (1,3)
     def test_parse_error(self):
         # ET < 1.3 raises ExpatError
         parse = self.etree.parse
@@ -2791,6 +2812,7 @@
         self.assertRaises(SyntaxError, parse, f)
         f.close()
 
+    required_versions_ET['test_parse_error_from_file'] = (1,3)
     def test_parse_error_from_file(self):
         parse = self.etree.parse
         # from file
@@ -2890,6 +2912,7 @@
         self.assertEquals(_str('<a>S?k p? nettet</a>').encode('iso-8859-1'),
                           result)
 
+    required_versions_ET['test_parse_encoding_8bit_explicit'] = (1,3)
     def test_parse_encoding_8bit_explicit(self):
         XMLParser = self.etree.XMLParser
 
@@ -2905,6 +2928,7 @@
         a = tree.getroot()
         self.assertEquals(a.text, text)
 
+    required_versions_ET['test_parse_encoding_8bit_override'] = (1,3)
     def test_parse_encoding_8bit_override(self):
         XMLParser = self.etree.XMLParser
 
@@ -3302,11 +3326,28 @@
         self.assertEquals(root[0].tag, "a")
         self.assertEquals(root[0].get("test"), "works")
 
+    def test_feed_parser_unicode(self):
+        parser = self.etree.XMLParser()
+
+        parser.feed(_str('<ro'))
+        parser.feed(_str('ot><'))
+        parser.feed(_str('a test="works"/'))
+        parser.feed(_str('></root'))
+        parser.feed(_str('>'))
+
+        root = parser.close()
+
+        self.assertEquals(root.tag, "root")
+        self.assertEquals(root[0].tag, "a")
+        self.assertEquals(root[0].get("test"), "works")
+
+    required_versions_ET['test_feed_parser_error_close_empty'] = (1,3)
     def test_feed_parser_error_close_empty(self):
         ParseError = self.etree.ParseError
         parser = self.etree.XMLParser()
         self.assertRaises(ParseError, parser.close)
 
+    required_versions_ET['test_feed_parser_error_close_incomplete'] = (1,3)
     def test_feed_parser_error_close_incomplete(self):
         ParseError = self.etree.ParseError
         parser = self.etree.XMLParser()
@@ -3316,6 +3357,7 @@
 
         self.assertRaises(ParseError, parser.close)
 
+    required_versions_ET['test_feed_parser_error_broken'] = (1,3)
     def test_feed_parser_error_broken(self):
         ParseError = self.etree.ParseError
         parser = self.etree.XMLParser()
@@ -3330,6 +3372,7 @@
 
         self.assertRaises(ParseError, parser.close)
 
+    required_versions_ET['test_feed_parser_error_position'] = (1,3)
     def test_feed_parser_error_position(self):
         ParseError = self.etree.ParseError
         parser = self.etree.XMLParser()
@@ -3344,6 +3387,7 @@
 
     # parser target interface
 
+    required_versions_ET['test_parser_target_property'] = (1,3)
     def test_parser_target_property(self):
         class Target(object):
             pass
@@ -3488,6 +3532,7 @@
                            "end-sub", "start-sub", "end-sub", "end-root"],
                           events)
 
+    required_versions_ET['test_parser_target_entity_unknown'] = (1,3)
     def test_parser_target_entity_unknown(self):
         events = []
         class Target(object):
@@ -3645,10 +3690,18 @@
     class ElementTreeTestCase(ETreeTestCaseBase):
         etree = ElementTree
 
+    filter_by_version(
+        ElementTreeTestCase,
+        ElementTreeTestCase.required_versions_ET, ET_VERSION)
+
 if cElementTree:
     class CElementTreeTestCase(ETreeTestCaseBase):
         etree = cElementTree
 
+    filter_by_version(
+        CElementTreeTestCase,
+        CElementTreeTestCase.required_versions_cET, CET_VERSION)
+
 def test_suite():
     suite = unittest.TestSuite()
     if etree:

From scoder at codespeak.net  Tue Oct 13 09:15:10 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Tue, 13 Oct 2009 09:15:10 +0200 (CEST)
Subject: [Lxml-checkins] r68361 - in lxml/trunk: . src/lxml
Message-ID: <20091013071510.30A5B168058@codespeak.net>

Author: scoder
Date: Tue Oct 13 09:15:07 2009
New Revision: 68361

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/extensions.pxi
Log:
 r5273 at delle:  sbehnel | 2009-10-12 21:46:24 +0200
 code simplification


Modified: lxml/trunk/src/lxml/extensions.pxi
==============================================================================
--- lxml/trunk/src/lxml/extensions.pxi	(original)
+++ lxml/trunk/src/lxml/extensions.pxi	Tue Oct 13 09:15:07 2009
@@ -677,12 +677,12 @@
                                    xpath.xmlXPathParserContext* ctxt, int nargs):
     cdef _Document doc
     cdef xpath.xmlXPathObject* obj
-    cdef int i
     cdef list args
+    cdef int i
     doc = context._doc
     try:
         args = []
-        for i from 0 <= i < nargs:
+        for i in xrange(nargs):
             obj = xpath.valuePop(ctxt)
             o = _unwrapXPathObject(obj, doc, context._build_smart_strings)
             _freeXPathObject(obj)

From scoder at codespeak.net  Tue Oct 13 09:15:13 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Tue, 13 Oct 2009 09:15:13 +0200 (CEST)
Subject: [Lxml-checkins] r68362 - in lxml/trunk: . src/lxml/html
	src/lxml/html/tests
Message-ID: <20091013071513.E2976168058@codespeak.net>

Author: scoder
Date: Tue Oct 13 09:15:13 2009
New Revision: 68362

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/CHANGES.txt
   lxml/trunk/src/lxml/html/__init__.py
   lxml/trunk/src/lxml/html/tests/test_forms.txt
Log:
 r5274 at delle:  sbehnel | 2009-10-13 08:48:17 +0200
 let textarea.value represent the complete tag content


Modified: lxml/trunk/CHANGES.txt
==============================================================================
--- lxml/trunk/CHANGES.txt	(original)
+++ lxml/trunk/CHANGES.txt	Tue Oct 13 09:15:13 2009
@@ -22,6 +22,11 @@
 Bugs fixed
 ----------
 
+* The ``.value`` attribute of ``textarea`` elements in lxml.html did
+  not represent the complete raw value (including child tags etc.). It
+  now serialises the complete content on read and replaces the
+  complete content by a string on write.
+
 * Looking up and deleting attributes without a namespace could hit a
   namespaced attribute of the same name instead.
 

Modified: lxml/trunk/src/lxml/html/__init__.py
==============================================================================
--- lxml/trunk/src/lxml/html/__init__.py	(original)
+++ lxml/trunk/src/lxml/html/__init__.py	Tue Oct 13 09:15:13 2009
@@ -965,11 +965,21 @@
         """
         Get/set the value (which is the contents of this element)
         """
-        return self.text or ''
+        content = self.text or ''
+        if self.tag.startswith("{%s}" % XHTML_NAMESPACE):
+            serialisation_method = 'xml'
+        else:
+            serialisation_method = 'html'
+        for el in self:
+            # it's rare that we actually get here, so let's not use ''.join()
+            content += etree.tostring(el, method=serialisation_method)
+        return content
     def _value__set(self, value):
+        del self[:]
         self.text = value
     def _value__del(self):
         self.text = ''
+        del self[:]
     value = property(_value__get, _value__set, _value__del, doc=_value__get.__doc__)
 
 HtmlElementClassLookup._default_element_classes['textarea'] = TextareaElement

Modified: lxml/trunk/src/lxml/html/tests/test_forms.txt
==============================================================================
--- lxml/trunk/src/lxml/html/tests/test_forms.txt	(original)
+++ lxml/trunk/src/lxml/html/tests/test_forms.txt	Tue Oct 13 09:15:13 2009
@@ -166,3 +166,21 @@
 [('foo', 'bar')]
 >>> tree.forms[0].fields.values()
 ['bar']
+
+>>> tree = lxml.html.fromstring('''
+... <html><body>
+...  <form>
+...   <textarea name="foo">some <b>text<br>content</b> with tags</textarea>
+...  </form>
+... </body></html>
+... ''')
+>>> tree.forms[0].fields.keys()
+['foo']
+>>> ta = tree.forms[0].inputs['foo']
+>>> print(ta.value)
+some <b>text<br>content</b> with tags
+>>> ta.value = 'abc<br>def'
+>>> print(ta.value)
+abc<br>def
+>>> len(ta)
+0

From scoder at codespeak.net  Fri Oct 16 23:38:48 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 16 Oct 2009 23:38:48 +0200 (CEST)
Subject: [Lxml-checkins] r68550 - in lxml/trunk: . src/lxml
Message-ID: <20091016213848.B3629168067@codespeak.net>

Author: scoder
Date: Fri Oct 16 23:38:48 2009
New Revision: 68550

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/lxml.etree.pyx
Log:
 r5279 at delle:  sbehnel | 2009-10-16 23:30:29 +0200
 tiny code fix


Modified: lxml/trunk/src/lxml/lxml.etree.pyx
==============================================================================
--- lxml/trunk/src/lxml/lxml.etree.pyx	(original)
+++ lxml/trunk/src/lxml/lxml.etree.pyx	Fri Oct 16 23:38:48 2009
@@ -1606,11 +1606,11 @@
             _copyNonElementSiblings(self._context_node._c_node, root._c_node)
             doc = root._doc
             c_doc = self._context_node._doc._c_doc
-            if c_doc.intSubset and not doc._c_doc.intSubset:
+            if c_doc.intSubset is not NULL and doc._c_doc.intSubset is NULL:
                 doc._c_doc.intSubset = tree.xmlCopyDtd(c_doc.intSubset)
                 if doc._c_doc.intSubset is NULL:
                     python.PyErr_NoMemory()
-            if c_doc.extSubset and not doc._c_doc.extSubset:
+            if c_doc.extSubset is not NULL and not doc._c_doc.extSubset is NULL:
                 doc._c_doc.extSubset = tree.xmlCopyDtd(c_doc.extSubset)
                 if doc._c_doc.extSubset is NULL:
                     python.PyErr_NoMemory()

From scoder at codespeak.net  Fri Oct 16 23:38:54 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 16 Oct 2009 23:38:54 +0200 (CEST)
Subject: [Lxml-checkins] r68551 - in lxml/trunk: . src/lxml
Message-ID: <20091016213854.58F88168076@codespeak.net>

Author: scoder
Date: Fri Oct 16 23:38:52 2009
New Revision: 68551

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/CHANGES.txt
   lxml/trunk/src/lxml/parser.pxi
Log:
 r5280 at delle:  sbehnel | 2009-10-16 23:38:38 +0200
 fix bug 410916: entity non-resolution in feed parser


Modified: lxml/trunk/CHANGES.txt
==============================================================================
--- lxml/trunk/CHANGES.txt	(original)
+++ lxml/trunk/CHANGES.txt	Fri Oct 16 23:38:52 2009
@@ -22,6 +22,9 @@
 Bugs fixed
 ----------
 
+* The ``resolve_entities`` option did not work in the incremental feed
+  parser.
+
 * The ``.value`` attribute of ``textarea`` elements in lxml.html did
   not represent the complete raw value (including child tags etc.). It
   now serialises the complete content on read and replaces the

Modified: lxml/trunk/src/lxml/parser.pxi
==============================================================================
--- lxml/trunk/src/lxml/parser.pxi	(original)
+++ lxml/trunk/src/lxml/parser.pxi	Fri Oct 16 23:38:52 2009
@@ -1094,6 +1094,15 @@
             py_buffer_len -= buffer_len
             c_data += buffer_len
 
+            if error and not pctxt.replaceEntities and not pctxt.validate:
+                # in this mode, we ignore errors about undefined entities
+                for entry in context._error_log.filter_from_errors():
+                    if entry.type != ErrorTypes.WAR_UNDECLARED_ENTITY and \
+                           entry.type != ErrorTypes.ERR_UNDECLARED_ENTITY:
+                        break
+                else:
+                    error = 0
+
         if not recover and (error or not pctxt.wellFormed):
             self._feed_parser_running = 0
             try:

From scoder at codespeak.net  Fri Oct 16 23:43:18 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 16 Oct 2009 23:43:18 +0200 (CEST)
Subject: [Lxml-checkins] r68552 - in lxml/branch/lxml-2.2: . src/lxml
Message-ID: <20091016214318.DBAA4168067@codespeak.net>

Author: scoder
Date: Fri Oct 16 23:43:17 2009
New Revision: 68552

Modified:
   lxml/branch/lxml-2.2/   (props changed)
   lxml/branch/lxml-2.2/CHANGES.txt
   lxml/branch/lxml-2.2/src/lxml/parser.pxi
Log:
trunk merge: fix for bug 410916 - unresolved entities in feed parser

Modified: lxml/branch/lxml-2.2/CHANGES.txt
==============================================================================
--- lxml/branch/lxml-2.2/CHANGES.txt	(original)
+++ lxml/branch/lxml-2.2/CHANGES.txt	Fri Oct 16 23:43:17 2009
@@ -2,7 +2,7 @@
 lxml changelog
 ==============
 
-2.2.2 (2009-10-??)
+2.2.3 (2009-10-??)
 ==================
 
 Features added
@@ -11,6 +11,9 @@
 Bugs fixed
 ----------
 
+* The ``resolve_entities`` option did not work in the incremental feed
+  parser.
+
 * Looking up and deleting attributes without a namespace could hit a
   namespaced attribute of the same name instead.
 

Modified: lxml/branch/lxml-2.2/src/lxml/parser.pxi
==============================================================================
--- lxml/branch/lxml-2.2/src/lxml/parser.pxi	(original)
+++ lxml/branch/lxml-2.2/src/lxml/parser.pxi	Fri Oct 16 23:43:17 2009
@@ -1094,6 +1094,15 @@
             py_buffer_len -= buffer_len
             c_data += buffer_len
 
+            if error and not pctxt.replaceEntities and not pctxt.validate:
+                # in this mode, we ignore errors about undefined entities
+                for entry in context._error_log.filter_from_errors():
+                    if entry.type != ErrorTypes.WAR_UNDECLARED_ENTITY and \
+                           entry.type != ErrorTypes.ERR_UNDECLARED_ENTITY:
+                        break
+                else:
+                    error = 0
+
         if not recover and (error or not pctxt.wellFormed):
             self._feed_parser_running = 0
             try:

From scoder at codespeak.net  Sat Oct 17 01:18:39 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 01:18:39 +0200 (CEST)
Subject: [Lxml-checkins] r68557 - in lxml/trunk: . src/lxml
Message-ID: <20091016231839.9A7B0168041@codespeak.net>

Author: scoder
Date: Sat Oct 17 01:18:39 2009
New Revision: 68557

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/apihelpers.pxi
   lxml/trunk/src/lxml/docloader.pxi
   lxml/trunk/src/lxml/iterparse.pxi
   lxml/trunk/src/lxml/lxml.objectify.pyx
   lxml/trunk/src/lxml/objectpath.pxi
   lxml/trunk/src/lxml/parser.pxi
   lxml/trunk/src/lxml/python.pxd
Log:
 r5283 at delle:  sbehnel | 2009-10-17 01:16:46 +0200
 fix PyBytes/PyString usage


Modified: lxml/trunk/src/lxml/apihelpers.pxi
==============================================================================
--- lxml/trunk/src/lxml/apihelpers.pxi	(original)
+++ lxml/trunk/src/lxml/apihelpers.pxi	Sat Oct 17 01:18:39 2009
@@ -1233,12 +1233,12 @@
 
 cdef object _utf8(object s):
     cdef int invalid
-    if python.PyString_CheckExact(s):
+    if python.PyBytes_CheckExact(s):
         invalid = check_string_utf8(s)
     elif python.PyUnicode_CheckExact(s) or python.PyUnicode_Check(s):
         s = python.PyUnicode_AsUTF8String(s)
         invalid = check_string_utf8(s) == -1
-    elif python.PyString_Check(s):
+    elif python.PyBytes_Check(s):
         invalid = check_string_utf8(s)
     else:
         raise TypeError, u"Argument must be string or unicode."
@@ -1275,7 +1275,7 @@
     """
     if filename is None:
         return None
-    elif python.PyString_Check(filename):
+    elif python.PyBytes_Check(filename):
         return filename
     elif python.PyUnicode_Check(filename):
         filename8 = python.PyUnicode_AsEncodedString(
@@ -1313,7 +1313,7 @@
     cdef char* c_filename
     if filename is None:
         return None
-    elif python.PyString_Check(filename):
+    elif python.PyBytes_Check(filename):
         if not check_string_utf8(filename):
             # plain ASCII!
             return filename

Modified: lxml/trunk/src/lxml/docloader.pxi
==============================================================================
--- lxml/trunk/src/lxml/docloader.pxi	(original)
+++ lxml/trunk/src/lxml/docloader.pxi	Sat Oct 17 01:18:39 2009
@@ -49,7 +49,7 @@
         cdef _InputDocument doc_ref
         if python.PyUnicode_Check(string):
             string = python.PyUnicode_AsUTF8String(string)
-        elif not python.PyString_Check(string):
+        elif not python.PyBytes_Check(string):
             raise TypeError, "argument must be a byte string or unicode string"
         doc_ref = _InputDocument()
         doc_ref._type = PARSER_DATA_STRING

Modified: lxml/trunk/src/lxml/iterparse.pxi
==============================================================================
--- lxml/trunk/src/lxml/iterparse.pxi	(original)
+++ lxml/trunk/src/lxml/iterparse.pxi	Sat Oct 17 01:18:39 2009
@@ -479,7 +479,7 @@
         while python.PyList_GET_SIZE(context._events) == 0:
             if c_stream is NULL:
                 data = self._source.read(__ITERPARSE_CHUNK_SIZE)
-                if not python.PyString_Check(data):
+                if not python.PyBytes_Check(data):
                     self._source = None
                     raise TypeError, u"reading file objects must return plain strings"
                 c_data_len = python.PyString_GET_SIZE(data)

Modified: lxml/trunk/src/lxml/lxml.objectify.pyx
==============================================================================
--- lxml/trunk/src/lxml/lxml.objectify.pyx	(original)
+++ lxml/trunk/src/lxml/lxml.objectify.pyx	Sat Oct 17 01:18:39 2009
@@ -949,7 +949,7 @@
     cdef object _type
     cdef list _schema_types
     def __init__(self, name, type_check, type_class, stringify=None):
-        if python.PyString_Check(name):
+        if python.PyBytes_Check(name):
             name = python.PyUnicode_FromEncodedObject(name, 'ASCII', NULL)
         elif not python.PyUnicode_Check(name):
             raise TypeError, u"Type name must be a string"
@@ -1579,12 +1579,12 @@
     doc = element._doc
 
     if empty_type_name is not None:
-        if python.PyString_Check(empty_type_name):
+        if python.PyBytes_Check(empty_type_name):
             empty_type_name = python.PyUnicode_FromEncodedObject(
                 empty_type_name, "ASCII", NULL)
         dict_result = python.PyDict_GetItem(_SCHEMA_TYPE_DICT, empty_type_name)
     elif empty_pytype_name is not None:
-        if python.PyString_Check(empty_pytype_name):
+        if python.PyBytes_Check(empty_pytype_name):
             empty_pytype_name = python.PyUnicode_FromEncodedObject(
                 empty_pytype_name, "ASCII", NULL)
         dict_result = python.PyDict_GetItem(_PYTYPE_DICT, empty_pytype_name)

Modified: lxml/trunk/src/lxml/objectpath.pxi
==============================================================================
--- lxml/trunk/src/lxml/objectpath.pxi	(original)
+++ lxml/trunk/src/lxml/objectpath.pxi	Sat Oct 17 01:18:39 2009
@@ -93,7 +93,7 @@
     """
     cdef bint has_dot
     cdef list new_path = []
-    if python.PyString_Check(path):
+    if python.PyBytes_Check(path):
         path = python.PyUnicode_FromEncodedObject(path, 'ASCII', NULL)
     path = path.strip()
     if path == u'.':

Modified: lxml/trunk/src/lxml/parser.pxi
==============================================================================
--- lxml/trunk/src/lxml/parser.pxi	(original)
+++ lxml/trunk/src/lxml/parser.pxi	Sat Oct 17 01:18:39 2009
@@ -357,7 +357,7 @@
                 c_requested -= remaining
 
                 self._bytes = self._filelike.read(c_requested)
-                if not python.PyString_Check(self._bytes):
+                if not python.PyBytes_Check(self._bytes):
                     if python.PyUnicode_Check(self._bytes):
                         if self._encoding is None:
                             self._bytes = python.PyUnicode_AsUTF8String(self._bytes)
@@ -1043,7 +1043,7 @@
         cdef int buffer_len
         cdef int error
         cdef bint recover = self._parse_options & xmlparser.XML_PARSE_RECOVER
-        if python.PyString_Check(data):
+        if python.PyBytes_Check(data):
             if self._default_encoding is None:
                 c_encoding = NULL
             else:
@@ -1538,7 +1538,7 @@
         # pass native unicode only if libxml2 can handle it
         if _UNICODE_ENCODING is NULL:
             text = python.PyUnicode_AsUTF8String(text)
-    elif not python.PyString_Check(text):
+    elif not python.PyBytes_Check(text):
         raise ValueError, u"can only parse strings"
     if python.PyUnicode_Check(url):
         url = python.PyUnicode_AsUTF8String(url)

Modified: lxml/trunk/src/lxml/python.pxd
==============================================================================
--- lxml/trunk/src/lxml/python.pxd	(original)
+++ lxml/trunk/src/lxml/python.pxd	Sat Oct 17 01:18:39 2009
@@ -23,8 +23,8 @@
 
     cdef bint PyUnicode_Check(object obj)
     cdef bint PyUnicode_CheckExact(object obj)
-    cdef bint PyString_Check(object obj)
-    cdef bint PyString_CheckExact(object obj)
+    cdef bint PyBytes_Check(object obj)
+    cdef bint PyBytes_CheckExact(object obj)
 
     cdef object PyUnicode_FromEncodedObject(object s, char* encoding,
                                             char* errors)

From scoder at codespeak.net  Sat Oct 17 01:18:43 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 01:18:43 +0200 (CEST)
Subject: [Lxml-checkins] r68558 - in lxml/trunk: . src/lxml src/lxml/tests
Message-ID: <20091016231843.C7827168058@codespeak.net>

Author: scoder
Date: Sat Oct 17 01:18:43 2009
New Revision: 68558

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/extensions.pxi
   lxml/trunk/src/lxml/tests/test_xpathevaluator.py
   lxml/trunk/src/lxml/xpath.pxi
Log:
 r5284 at delle:  sbehnel | 2009-10-17 01:18:28 +0200
 ticket 446654: support strings in node-sets returned from extension functions


Modified: lxml/trunk/src/lxml/extensions.pxi
==============================================================================
--- lxml/trunk/src/lxml/extensions.pxi	(original)
+++ lxml/trunk/src/lxml/extensions.pxi	Sat Oct 17 01:18:43 2009
@@ -457,12 +457,15 @@
 ################################################################################
 # helper functions
 
-cdef xpath.xmlXPathObject* _wrapXPathObject(object obj) except NULL:
+cdef xpath.xmlXPathObject* _wrapXPathObject(object obj, _Document doc,
+                                            _BaseContext context) except NULL:
     cdef xpath.xmlNodeSet* resultSet
-    cdef _Element node
+    cdef _Element fake_node = None
+    cdef xmlNode* c_node
+
     if python.PyUnicode_Check(obj):
         obj = _utf8(obj)
-    if python.PyString_Check(obj):
+    if python.PyBytes_Check(obj):
         return xpath.xmlXPathNewCString(_cstr(obj))
     if python.PyBool_Check(obj):
         return xpath.xmlXPathNewBoolean(obj)
@@ -474,13 +477,42 @@
         resultSet = xpath.xmlXPathNodeSetCreate((<_Element>obj)._c_node)
     elif python.PySequence_Check(obj):
         resultSet = xpath.xmlXPathNodeSetCreate(NULL)
-        for element in obj:
-            if isinstance(element, _Element):
-                node = <_Element>element
-                xpath.xmlXPathNodeSetAdd(resultSet, node._c_node)
-            else:
-                xpath.xmlXPathFreeNodeSet(resultSet)
-                raise XPathResultError, u"This is not a node: %r" % element
+        try:
+            for value in obj:
+                if isinstance(value, _Element):
+                    if context is not None:
+                        context._hold(value)
+                    xpath.xmlXPathNodeSetAdd(resultSet, (<_Element>value)._c_node)
+                else:
+                    if context is None or doc is None:
+                        raise XPathResultError, \
+                              u"Non-Element values not supported at this point - got %r" % value
+                    # support strings by appending text nodes to an Element
+                    if python.PyUnicode_Check(value):
+                        value = _utf8(value)
+                    if python.PyBytes_Check(value):
+                        if fake_node is None:
+                            fake_node = _makeElement("text-root", NULL, doc, None,
+                                                     None, None, None, None, None)
+                            context._hold(fake_node)
+                        else:
+                            # append a comment node to keep the text nodes separate
+                            c_node = tree.xmlNewDocComment(doc._c_doc, "")
+                            if c_node is NULL:
+                                python.PyErr_NoMemory()
+                            tree.xmlAddChild(fake_node._c_node, c_node)
+                        context._hold(value)
+                        c_node = tree.xmlNewDocText(doc._c_doc, _cstr(value))
+                        if c_node is NULL:
+                            python.PyErr_NoMemory()
+                        tree.xmlAddChild(fake_node._c_node, c_node)
+                        xpath.xmlXPathNodeSetAdd(resultSet, c_node)
+                    else:
+                        raise XPathResultError, \
+                              u"This is not a supported node-set result: %r" % value
+        except:
+            xpath.xmlXPathFreeNodeSet(resultSet)
+            raise
     else:
         raise XPathResultError, u"Unknown return type: %s" % \
             python._fqtypename(obj)
@@ -613,7 +645,7 @@
     else:
         is_text = not (is_tail or is_attribute)
 
-    if python.PyString_CheckExact(string_value):
+    if python.PyBytes_CheckExact(string_value):
         result = _ElementStringResult(string_value)
         result._parent = parent
         result.is_attribute = is_attribute
@@ -691,7 +723,7 @@
 
         res = function(context, *args)
         # wrap result for XPath consumption
-        obj = _wrapXPathObject(res)
+        obj = _wrapXPathObject(res, doc, context)
         # prevent Python from deallocating elements handed to libxml2
         context._hold(res)
         xpath.valuePush(ctxt, obj)

Modified: lxml/trunk/src/lxml/tests/test_xpathevaluator.py
==============================================================================
--- lxml/trunk/src/lxml/tests/test_xpathevaluator.py	(original)
+++ lxml/trunk/src/lxml/tests/test_xpathevaluator.py	Sat Oct 17 01:18:43 2009
@@ -593,8 +593,14 @@
 def tag(elem):
     return elem.tag
 
+def tag_or_value(elem):
+    return getattr(elem, 'tag', elem)
+
 def stringTest(ctxt, s1):
     return "Hello "+s1
+
+def stringListTest(ctxt, s1):
+    return ["Hello "] + list(s1) +  ["!"]
     
 def floatTest(ctxt, f1):
     return f1+4
@@ -616,7 +622,7 @@
     return st1
 
 def resultTypesTest(ctxt):
-    return ["x","y"]
+    return [None,None]
 
 def resultTypesTest2(ctxt):
     return resultTypesTest
@@ -624,6 +630,7 @@
 uri = "http://www.example.com/"
 
 extension = {(None, 'stringTest'): stringTest,
+             (None, 'stringListTest'): stringListTest,
              (None, 'floatTest'): floatTest,
              (None, 'booleanTest'): booleanTest,
              (None, 'setTest'): setTest,
@@ -655,6 +662,8 @@
     ['tag']
     >>> list(map(tag, e("setTest2(/body/*)")))
     ['tag', 'section']
+    >>> list(map(tag_or_value, e("stringListTest(/body/tag)")))
+    ['Hello', 'tag', 'tag', 'tag', '!']
     >>> e("argsTest1('a',1.5,true(),/body/tag)")
     "a, 1.5, True, ['tag', 'tag', 'tag']"
     >>> list(map(tag, e("argsTest2(/body/tag, /body/section)")))
@@ -662,7 +671,7 @@
     >>> e("resultTypesTest()")
     Traceback (most recent call last):
     ...
-    XPathResultError: This is not a node: 'x'
+    XPathResultError: This is not a supported node-set result: None
     >>> try:
     ...     e("resultTypesTest2()")
     ... except etree.XPathResultError:

Modified: lxml/trunk/src/lxml/xpath.pxi
==============================================================================
--- lxml/trunk/src/lxml/xpath.pxi	(original)
+++ lxml/trunk/src/lxml/xpath.pxi	Sat Oct 17 01:18:43 2009
@@ -79,16 +79,16 @@
         for name, value in variable_dict.items():
             name_utf = self._to_utf(name)
             xpath.xmlXPathRegisterVariable(
-                self._xpathCtxt, _cstr(name_utf), _wrapXPathObject(value))
+                self._xpathCtxt, _cstr(name_utf), _wrapXPathObject(value, None, None))
 
     cdef registerVariable(self, name, value):
         name_utf = self._to_utf(name)
         xpath.xmlXPathRegisterVariable(
-            self._xpathCtxt, _cstr(name_utf), _wrapXPathObject(value))
+            self._xpathCtxt, _cstr(name_utf), _wrapXPathObject(value, None, None))
 
     cdef void _registerVariable(self, name_utf, value):
         xpath.xmlXPathRegisterVariable(
-            self._xpathCtxt, _cstr(name_utf), _wrapXPathObject(value))
+            self._xpathCtxt, _cstr(name_utf), _wrapXPathObject(value, None, None))
 
     cdef void _setupDict(self, xpath.xmlXPathContext* xpathCtxt):
         __GLOBAL_PARSER_CONTEXT.initXPathParserDict(xpathCtxt)

From scoder at codespeak.net  Sat Oct 17 01:20:52 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 01:20:52 +0200 (CEST)
Subject: [Lxml-checkins] r68559 - lxml/trunk
Message-ID: <20091016232052.20FDC16805A@codespeak.net>

Author: scoder
Date: Sat Oct 17 01:20:51 2009
New Revision: 68559

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/CHANGES.txt
Log:
 r5288 at delle:  sbehnel | 2009-10-17 01:20:44 +0200
 changelog


Modified: lxml/trunk/CHANGES.txt
==============================================================================
--- lxml/trunk/CHANGES.txt	(original)
+++ lxml/trunk/CHANGES.txt	Sat Oct 17 01:20:51 2009
@@ -8,6 +8,9 @@
 Features added
 --------------
 
+* Support for strings (in addition to Elements) in node-sets returned
+  by extension functions.
+
 * Forms that lack an ``action`` attribute default to the base URL of
   the document on submit.
 

From scoder at codespeak.net  Sat Oct 17 01:22:55 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 01:22:55 +0200 (CEST)
Subject: [Lxml-checkins] r68560 - in lxml/trunk: . src/lxml/tests
Message-ID: <20091016232255.AC94F16805C@codespeak.net>

Author: scoder
Date: Sat Oct 17 01:22:55 2009
New Revision: 68560

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/tests/test_xpathevaluator.py
Log:
 r5290 at delle:  sbehnel | 2009-10-17 01:22:51 +0200
 test fix


Modified: lxml/trunk/src/lxml/tests/test_xpathevaluator.py
==============================================================================
--- lxml/trunk/src/lxml/tests/test_xpathevaluator.py	(original)
+++ lxml/trunk/src/lxml/tests/test_xpathevaluator.py	Sat Oct 17 01:22:55 2009
@@ -663,7 +663,7 @@
     >>> list(map(tag, e("setTest2(/body/*)")))
     ['tag', 'section']
     >>> list(map(tag_or_value, e("stringListTest(/body/tag)")))
-    ['Hello', 'tag', 'tag', 'tag', '!']
+    ['Hello ', 'tag', 'tag', 'tag', '!']
     >>> e("argsTest1('a',1.5,true(),/body/tag)")
     "a, 1.5, True, ['tag', 'tag', 'tag']"
     >>> list(map(tag, e("argsTest2(/body/tag, /body/section)")))

From scoder at codespeak.net  Sat Oct 17 02:11:43 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 02:11:43 +0200 (CEST)
Subject: [Lxml-checkins] r68563 - in lxml/trunk: . src/lxml/html
	src/lxml/html/tests
Message-ID: <20091017001143.24E1016803B@codespeak.net>

Author: scoder
Date: Sat Oct 17 02:11:42 2009
New Revision: 68563

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/html/__init__.py
   lxml/trunk/src/lxml/html/tests/test_rewritelinks.txt
Log:
 r5292 at delle:  sbehnel | 2009-10-17 02:11:33 +0200
 fix bug 449926: reverse URL iteration inside of text content to simplify replacements


Modified: lxml/trunk/src/lxml/html/__init__.py
==============================================================================
--- lxml/trunk/src/lxml/html/__init__.py	(original)
+++ lxml/trunk/src/lxml/html/__init__.py	Sat Oct 17 02:11:42 2009
@@ -308,6 +308,13 @@
 
         Note: <base href> is *not* taken into account in any way.  The
         link you get is exactly the link in the document.
+
+        Note: multiple links inside of a single text string or
+        attribute value are returned in reversed order.  This makes it
+        possible to replace or delete them from the text string value
+        based on their reported text positions.  Otherwise, a
+        modification at one text position can change the positions of
+        links reported later on.
         """
         link_attrs = defs.link_attrs
         for el in self.iter():
@@ -347,15 +354,29 @@
                     ## http://www.w3.org/TR/html401/struct/objects.html#adef-valuetype
                     yield (el, 'value', el.get('value'), 0)
             if tag == 'style' and el.text:
-                for match in _css_url_re.finditer(el.text):
-                    url, start = _unquote_match(match.group(1), match.start(1))
-                    yield (el, None, url, start)
-                for match in _css_import_re.finditer(el.text):
-                    yield (el, None, match.group(1), match.start(1))
+                urls = [
+                    _unquote_match(match.group(1), match.start(1))
+                    for match in _css_url_re.finditer(el.text)
+                    ] + [
+                    (match.group(1), match.start(1))
+                    for match in _css_import_re.finditer(el.text)
+                    ]
+                if urls:
+                    # sort by start pos to bring both match sets back into order
+                    urls = [ (start, url) for (url, start) in urls ]
+                    urls.sort()
+                    # reverse the list to report correct positions despite
+                    # modifications
+                    urls.reverse()
+                    for start, url in urls:
+                        yield (el, None, url, start)
             if 'style' in attribs:
-                for match in _css_url_re.finditer(attribs['style']):
-                    url, start = _unquote_match(match.group(1), match.start(1))
-                    yield (el, 'style', url, start)
+                urls = list(_css_url_re.finditer(attribs['style']))
+                if urls:
+                    # return in reversed order to simplify in-place modifications
+                    for match in urls[::-1]:
+                        url, start = _unquote_match(match.group(1), match.start(1))
+                        yield (el, 'style', url, start)
 
     def rewrite_links(self, link_repl_func, resolve_base_href=True,
                       base_href=None):

Modified: lxml/trunk/src/lxml/html/tests/test_rewritelinks.txt
==============================================================================
--- lxml/trunk/src/lxml/html/tests/test_rewritelinks.txt	(original)
+++ lxml/trunk/src/lxml/html/tests/test_rewritelinks.txt	Sat Oct 17 02:11:42 2009
@@ -1,3 +1,8 @@
+
+Setup::
+
+    >>> import lxml.html
+
 We'll define a link translation function:
 
     >>> base_href = 'http://old/base/path.html'
@@ -118,8 +123,8 @@
     ...   </table>
     ...  </body></html>'''))
     link href="style.css"
-    style None="/bg.gif"@40
     style None="/other-styles.css"@69
+    style None="/bg.gif"@40
     script src="/js-funcs.js"
     a href="/test.html"
     a href="/other.html"
@@ -179,3 +184,48 @@
       </table>
      </body>
     </html>
+
+Check if we can replace multiple links inside of the same text string::
+
+    >>> html = lxml.html.fromstring ("""\
+    ... <html>
+    ...   <head>
+    ...      <title>Test</title>
+    ...      <style type='text/css'>
+    ...        .bg1 {
+    ...            background: url(images/bg1.png);
+    ...        }
+    ...        .bg2 {
+    ...            background: url(images/bg2.png);
+    ...        }
+    ...      </style>
+    ...   </head>
+    ...   <body>
+    ...      <p>Hi</p>
+    ...   </body>
+    ... </html>
+    ... """,
+    ... base_url = 'http://www.example.com/')
+
+    >>> html.make_links_absolute ()
+
+    >>> try: _unicode = unicode
+    ... except NameError: _unicode = str
+
+    >>> print(lxml.html.tostring (html, pretty_print = True, encoding=_unicode))
+    <html>
+      <head>
+        <title>Test</title>
+        <style type="text/css">
+          .bg1 {
+            background: url(http://www.example.com/images/bg1.png);
+          }
+          .bg2 {
+            background: url(http://www.example.com/images/bg2.png);
+          }
+        </style>
+      </head>
+      <body>
+        <p>Hi</p>
+      </body>
+    </html>

From scoder at codespeak.net  Sat Oct 17 03:34:13 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 03:34:13 +0200 (CEST)
Subject: [Lxml-checkins] r68565 - lxml/trunk
Message-ID: <20091017013413.9D72816800D@codespeak.net>

Author: scoder
Date: Sat Oct 17 03:34:13 2009
New Revision: 68565

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/CHANGES.txt
Log:
 r5294 at delle:  sbehnel | 2009-10-17 02:19:50 +0200
 changelog


Modified: lxml/trunk/CHANGES.txt
==============================================================================
--- lxml/trunk/CHANGES.txt	(original)
+++ lxml/trunk/CHANGES.txt	Sat Oct 17 03:34:13 2009
@@ -25,6 +25,14 @@
 Bugs fixed
 ----------
 
+* Rewriting multiple links inside of HTML text content could end up
+  replacing unrelated content as replacements could impact the
+  reported position of subsequent matches.  Modifications are now
+  simplified by letting the ``iterlinks()`` generator in ``lxml.html``
+  return links in reversed order if they appear inside the same text
+  node.  Thus, replacements and link-internal modifications no longer
+  change the position of links reported afterwards.
+
 * The ``resolve_entities`` option did not work in the incremental feed
   parser.
 

From scoder at codespeak.net  Sat Oct 17 03:34:16 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 03:34:16 +0200 (CEST)
Subject: [Lxml-checkins] r68566 - in lxml/trunk: . src/lxml
Message-ID: <20091017013416.D54D4168015@codespeak.net>

Author: scoder
Date: Sat Oct 17 03:34:16 2009
New Revision: 68566

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/extensions.pxi
Log:
 r5295 at delle:  sbehnel | 2009-10-17 03:23:53 +0200
 Py3 fix


Modified: lxml/trunk/src/lxml/extensions.pxi
==============================================================================
--- lxml/trunk/src/lxml/extensions.pxi	(original)
+++ lxml/trunk/src/lxml/extensions.pxi	Sat Oct 17 03:34:16 2009
@@ -714,7 +714,7 @@
     doc = context._doc
     try:
         args = []
-        for i in xrange(nargs):
+        for i in range(nargs):
             obj = xpath.valuePop(ctxt)
             o = _unwrapXPathObject(obj, doc, context._build_smart_strings)
             _freeXPathObject(obj)

From scoder at codespeak.net  Sat Oct 17 03:34:20 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Sat, 17 Oct 2009 03:34:20 +0200 (CEST)
Subject: [Lxml-checkins] r68567 - in lxml/trunk: . src/lxml
Message-ID: <20091017013420.72504168013@codespeak.net>

Author: scoder
Date: Sat Oct 17 03:34:19 2009
New Revision: 68567

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/CHANGES.txt
   lxml/trunk/src/lxml/lxml.etree.pyx
   lxml/trunk/src/lxml/python.pxd
Log:
 r5296 at delle:  sbehnel | 2009-10-17 03:34:05 +0200
 create _Attrib instance on each .attrib request instead of keeping a cyclic reference to it (alternative weak-ref implementation is actually slower)


Modified: lxml/trunk/CHANGES.txt
==============================================================================
--- lxml/trunk/CHANGES.txt	(original)
+++ lxml/trunk/CHANGES.txt	Sat Oct 17 03:34:19 2009
@@ -88,6 +88,10 @@
 Other changes
 -------------
 
+* ``Element.attrib`` no longer uses a cyclic reference back to its
+  Element object.  It therefore no longer requires the garbage
+  collector to clean up.
+
 * Static builds include libiconv, in addition to libxml2 and libxslt.
 
 

Modified: lxml/trunk/src/lxml/lxml.etree.pyx
==============================================================================
--- lxml/trunk/src/lxml/lxml.etree.pyx	(original)
+++ lxml/trunk/src/lxml/lxml.etree.pyx	Sat Oct 17 03:34:19 2009
@@ -528,7 +528,6 @@
     cdef _Document _doc
     cdef xmlNode* _c_node
     cdef object _tag
-    cdef object _attrib
 
     def _init(self):
         u"""_init(self)
@@ -814,9 +813,15 @@
         keys(), values() and items() to access element attributes.
         """
         def __get__(self):
-            if self._attrib is None:
-                self._attrib = _Attrib(self)
-            return self._attrib
+            return _Attrib(self)
+            ## cdef python.PyObject* ref
+            ## if self._attrib is not None:
+            ##     ref = python.PyWeakref_GET_OBJECT(self._attrib)
+            ##     if ref is not <python.PyObject*>None:
+            ##         return <object>ref
+            ## attrib = _Attrib(self)
+            ## self._attrib = python.PyWeakref_NewRef(attrib, NULL)
+            ## return attrib
 
     property text:
         u"""Text before the first subelement. This is either a string or 
@@ -1948,7 +1953,7 @@
     u"""A dict-like proxy for the ``Element.attrib`` property.
     """
     cdef _Element _element
-    def __init__(self, _Element element not None):
+    def __cinit__(self, _Element element not None):
         self._element = element
 
     # MANIPULATORS
@@ -2064,6 +2069,12 @@
             other = dict(other)
         return python.PyObject_RichCompare(one, other, op)
 
+
+cdef extern from "etree_defs.h":
+    # macro call to 't->tp_new()' for fast instantiation
+    cdef _Attrib NEW_ATTRIB "PY_NEW" (object t)
+
+
 cdef class _AttribIterator:
     u"""Attribute iterator - for internal use only!
     """

Modified: lxml/trunk/src/lxml/python.pxd
==============================================================================
--- lxml/trunk/src/lxml/python.pxd	(original)
+++ lxml/trunk/src/lxml/python.pxd	Sat Oct 17 03:34:19 2009
@@ -90,6 +90,9 @@
     cdef object PyObject_RichCompare(object o1, object o2, int op)
     cdef int PyObject_RichCompareBool(object o1, object o2, int op)
 
+#    object PyWeakref_NewRef(object ob, PyObject* callback)
+#    PyObject* PyWeakref_GET_OBJECT(object ref)
+
     cdef void* PyMem_Malloc(size_t size)
     cdef void* PyMem_Realloc(void* p, size_t size)
     cdef void PyMem_Free(void* p)

From scoder at codespeak.net  Fri Oct 30 14:38:03 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 30 Oct 2009 14:38:03 +0100 (CET)
Subject: [Lxml-checkins] r68859 - lxml/branch/lxml-2.2/doc
Message-ID: <20091030133803.43671168495@codespeak.net>

Author: scoder
Date: Fri Oct 30 14:38:02 2009
New Revision: 68859

Modified:
   lxml/branch/lxml-2.2/doc/elementsoup.txt
Log:
doc typo

Modified: lxml/branch/lxml-2.2/doc/elementsoup.txt
==============================================================================
--- lxml/branch/lxml-2.2/doc/elementsoup.txt	(original)
+++ lxml/branch/lxml-2.2/doc/elementsoup.txt	Fri Oct 30 14:38:02 2009
@@ -182,7 +182,7 @@
 
 If you prefer a 'real' (and fast) HTML parser instead of the regular
 expression based one in BeautifulSoup, you can still benefit from
-BeautifulSoup's _`support for encoding detection` in the
+BeautifulSoup's `support for encoding detection`_ in the
 ``UnicodeDammit`` class.
 
 .. sourcecode:: pycon

From scoder at codespeak.net  Fri Oct 30 15:02:05 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 30 Oct 2009 15:02:05 +0100 (CET)
Subject: [Lxml-checkins] r68861 - in lxml/branch/lxml-2.2: . doc
Message-ID: <20091030140205.93EDA1684A9@codespeak.net>

Author: scoder
Date: Fri Oct 30 15:02:05 2009
New Revision: 68861

Modified:
   lxml/branch/lxml-2.2/CHANGES.txt
   lxml/branch/lxml-2.2/doc/main.txt
Log:
prepare release of 2.2.3

Modified: lxml/branch/lxml-2.2/CHANGES.txt
==============================================================================
--- lxml/branch/lxml-2.2/CHANGES.txt	(original)
+++ lxml/branch/lxml-2.2/CHANGES.txt	Fri Oct 30 15:02:05 2009
@@ -2,7 +2,7 @@
 lxml changelog
 ==============
 
-2.2.3 (2009-10-??)
+2.2.3 (2009-10-30)
 ==================
 
 Features added

Modified: lxml/branch/lxml-2.2/doc/main.txt
==============================================================================
--- lxml/branch/lxml-2.2/doc/main.txt	(original)
+++ lxml/branch/lxml-2.2/doc/main.txt	Fri Oct 30 15:02:05 2009
@@ -147,8 +147,8 @@
 source release.  If you can't wait, consider trying a less recent
 release version first.
 
-The latest version is `lxml 2.2.2`_, released 2009-06-21
-(`changes for 2.2.2`_).  `Older versions`_ are listed below.
+The latest version is `lxml 2.2.3`_, released 2009-10-30
+(`changes for 2.2.3`_).  `Older versions`_ are listed below.
 
 Please take a look at the `installation instructions`_!
 
@@ -221,7 +221,9 @@
 <http://codespeak.net/lxml/2.1/>`_ and the `current in-development
 version <http://codespeak.net/lxml/dev/>`_.
 
-.. _`PDF documentation`: lxmldoc-2.2.2.pdf
+.. _`PDF documentation`: lxmldoc-2.2.3.pdf
+
+* `lxml 2.2.2`_, released 2009-06-21 (`changes for 2.2.2`_)
 
 * `lxml 2.2.1`_, released 2009-06-02 (`changes for 2.2.1`_)
 
@@ -323,6 +325,7 @@
 
 * `lxml 0.5`_,   released 2005-04-08
 
+.. _`lxml 2.2.3`: lxml-2.2.3.tgz
 .. _`lxml 2.2.2`: lxml-2.2.2.tgz
 .. _`lxml 2.2.1`: lxml-2.2.1.tgz
 .. _`lxml 2.2`: lxml-2.2.tgz
@@ -374,6 +377,7 @@
 .. _`lxml 0.5.1`: lxml-0.5.1.tgz
 .. _`lxml 0.5`: lxml-0.5.tgz
 
+.. _`changes for 2.2.3`: changes-2.2.3.html
 .. _`changes for 2.2.2`: changes-2.2.2.html
 .. _`changes for 2.2.1`: changes-2.2.1.html
 .. _`changes for 2.2`: changes-2.2.html

From scoder at codespeak.net  Fri Oct 30 15:10:34 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 30 Oct 2009 15:10:34 +0100 (CET)
Subject: [Lxml-checkins] r68862 - lxml/branch/lxml-2.2
Message-ID: <20091030141034.387BA1684AB@codespeak.net>

Author: scoder
Date: Fri Oct 30 15:10:33 2009
New Revision: 68862

Modified:
   lxml/branch/lxml-2.2/CHANGES.txt
Log:
typos

Modified: lxml/branch/lxml-2.2/CHANGES.txt
==============================================================================
--- lxml/branch/lxml-2.2/CHANGES.txt	(original)
+++ lxml/branch/lxml-2.2/CHANGES.txt	Fri Oct 30 15:10:33 2009
@@ -23,8 +23,9 @@
 * Modifying trees that contain parsed entity references could result
   in an infinite loop.
 
-* ObjectifiedElement.__setattr__ created an empty-string child element when the
-  attribute value was rejected as a non-unicode/non-ascii string
+* ``ObjectifiedElement.__setattr__`` created an empty-string child
+  element when the attribute value was rejected as a
+  non-unicode/non-ascii string
 
 * Syntax errors in ``lxml.cssselect`` could result in misleading error
   messages.
@@ -50,7 +51,7 @@
 * The ``ElementMaker`` in lxml.objectify no longer defines the default
   namespaces when annotation is disabled.
 
-* Feed parser failed to honout the 'recover' option on parse errors.
+* Feed parser failed to honour the 'recover' option on parse errors.
 
 * Diverting the error logging to Python's logging system was broken.
 

From scoder at codespeak.net  Fri Oct 30 15:11:35 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 30 Oct 2009 15:11:35 +0100 (CET)
Subject: [Lxml-checkins] r68863 - lxml/tag/lxml-2.2.3
Message-ID: <20091030141135.DF0E31684A9@codespeak.net>

Author: scoder
Date: Fri Oct 30 15:11:35 2009
New Revision: 68863

Added:
   lxml/tag/lxml-2.2.3/
      - copied from r68862, lxml/branch/lxml-2.2/
Log:
tag for 2.2.3

From scoder at codespeak.net  Fri Oct 30 15:13:09 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 30 Oct 2009 15:13:09 +0100 (CET)
Subject: [Lxml-checkins] r68864 - in lxml/trunk: . src/lxml
Message-ID: <20091030141309.D21B81684A9@codespeak.net>

Author: scoder
Date: Fri Oct 30 15:13:09 2009
New Revision: 68864

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/src/lxml/lxml.etree.pyx
Log:
 r5300 at delle:  sbehnel | 2009-10-17 09:33:41 +0200
 work-around for Py3.1


Modified: lxml/trunk/src/lxml/lxml.etree.pyx
==============================================================================
--- lxml/trunk/src/lxml/lxml.etree.pyx	(original)
+++ lxml/trunk/src/lxml/lxml.etree.pyx	Fri Oct 30 15:13:09 2009
@@ -985,7 +985,7 @@
         return _countElements(self._c_node.children)
 
     def __nonzero__(self):
-        u"__nonzero__(self)"
+        #u"__nonzero__(self)" # currently fails in Py3.1
         import warnings
         warnings.warn(
             u"The behavior of this method will change in future versions. "

From scoder at codespeak.net  Fri Oct 30 15:13:14 2009
From: scoder at codespeak.net (scoder at codespeak.net)
Date: Fri, 30 Oct 2009 15:13:14 +0100 (CET)
Subject: [Lxml-checkins] r68865 - in lxml/trunk: . doc
Message-ID: <20091030141314.C27111684A9@codespeak.net>

Author: scoder
Date: Fri Oct 30 15:13:13 2009
New Revision: 68865

Modified:
   lxml/trunk/   (props changed)
   lxml/trunk/CHANGES.txt
   lxml/trunk/doc/elementsoup.txt
   lxml/trunk/doc/main.txt
Log:
 r5301 at delle:  sbehnel | 2009-10-30 15:12:46 +0100
 integrated 2.2 branch release changes


Modified: lxml/trunk/CHANGES.txt
==============================================================================
--- lxml/trunk/CHANGES.txt	(original)
+++ lxml/trunk/CHANGES.txt	Fri Oct 30 15:13:13 2009
@@ -33,14 +33,37 @@
   node.  Thus, replacements and link-internal modifications no longer
   change the position of links reported afterwards.
 
-* The ``resolve_entities`` option did not work in the incremental feed
-  parser.
-
 * The ``.value`` attribute of ``textarea`` elements in lxml.html did
   not represent the complete raw value (including child tags etc.). It
   now serialises the complete content on read and replaces the
   complete content by a string on write.
 
+* Target parser didn't call ``.close()`` on the target object if
+  parsing failed.  Now it is guaranteed that ``.close()`` will be
+  called after parsing, regardless of the outcome.
+
+Other changes
+-------------
+
+* ``Element.attrib`` no longer uses a cyclic reference back to its
+  Element object.  It therefore no longer requires the garbage
+  collector to clean up.
+
+* Static builds include libiconv, in addition to libxml2 and libxslt.
+
+
+2.2.3 (2009-10-30)
+==================
+
+Features added
+--------------
+
+Bugs fixed
+----------
+
+* The ``resolve_entities`` option did not work in the incremental feed
+  parser.
+
 * Looking up and deleting attributes without a namespace could hit a
   namespaced attribute of the same name instead.
 
@@ -79,21 +102,11 @@
 
 * Feed parser failed to honout the 'recover' option on parse errors.
 
-* Target parser didn't call ``.close()`` on the target object if
-  parsing failed.  Now it is guaranteed that ``.close()`` will be
-  called after parsing, regardless of the outcome.
-
 * Diverting the error logging to Python's logging system was broken.
 
 Other changes
 -------------
 
-* ``Element.attrib`` no longer uses a cyclic reference back to its
-  Element object.  It therefore no longer requires the garbage
-  collector to clean up.
-
-* Static builds include libiconv, in addition to libxml2 and libxslt.
-
 
 2.2.2 (2009-06-21)
 ==================

Modified: lxml/trunk/doc/elementsoup.txt
==============================================================================
--- lxml/trunk/doc/elementsoup.txt	(original)
+++ lxml/trunk/doc/elementsoup.txt	Fri Oct 30 15:13:13 2009
@@ -182,7 +182,7 @@
 
 If you prefer a 'real' (and fast) HTML parser instead of the regular
 expression based one in BeautifulSoup, you can still benefit from
-BeautifulSoup's _`support for encoding detection` in the
+BeautifulSoup's `support for encoding detection`_ in the
 ``UnicodeDammit`` class.
 
 .. sourcecode:: pycon

Modified: lxml/trunk/doc/main.txt
==============================================================================
--- lxml/trunk/doc/main.txt	(original)
+++ lxml/trunk/doc/main.txt	Fri Oct 30 15:13:13 2009
@@ -221,7 +221,11 @@
 <http://codespeak.net/lxml/2.1/>`_ and the `current in-development
 version <http://codespeak.net/lxml/dev/>`_.
 
-.. _`PDF documentation`: lxmldoc-2.2.2.pdf
+.. _`PDF documentation`: lxmldoc-2.2.3.pdf
+
+* `lxml 2.2.3`_, released 2009-10-30 (`changes for 2.2.3`_)
+
+* `lxml 2.2.2`_, released 2009-06-21 (`changes for 2.2.2`_)
 
 * `lxml 2.2.1`_, released 2009-06-02 (`changes for 2.2.1`_)
 
@@ -323,6 +327,7 @@
 
 * `lxml 0.5`_,   released 2005-04-08
 
+.. _`lxml 2.2.3`: lxml-2.2.3.tgz
 .. _`lxml 2.2.2`: lxml-2.2.2.tgz
 .. _`lxml 2.2.1`: lxml-2.2.1.tgz
 .. _`lxml 2.2`: lxml-2.2.tgz
@@ -374,6 +379,7 @@
 .. _`lxml 0.5.1`: lxml-0.5.1.tgz
 .. _`lxml 0.5`: lxml-0.5.tgz
 
+.. _`changes for 2.2.3`: changes-2.2.2.html
 .. _`changes for 2.2.2`: changes-2.2.2.html
 .. _`changes for 2.2.1`: changes-2.2.1.html
 .. _`changes for 2.2`: changes-2.2.html