[libxml2]Add libxml2 library

libxml2/python/.gitignore

@@ -0,0 +1,7 @@
+/gen_prog
+/libxml2-export.c
+/libxml2-py.c
+/libxml2-py.h
+/libxml2.py
+/setup.py
+/tests/tmp.xml

libxml2/python/Makefile.am

@@ -0,0 +1,48 @@
+# Makefile for libxml2 python library
+
+# We use a rule with multiple output files which creates problems with
+# parallel builds.
+.NOTPARALLEL:
+
+SUBDIRS = . tests
+
+EXTRA_DIST = 			\
+	generator.py		\
+	libxml.py		\
+	pyproject.toml		\
+	meson.build
+
+if WITH_PYTHON
+AM_CPPFLAGS = \
+	-I$(top_builddir)/include \
+	-I$(top_srcdir)/include	\
+	$(PYTHON_CFLAGS)
+
+pyexec_LTLIBRARIES = libxml2mod.la
+
+libxml2mod_la_SOURCES = libxml.c libxml_wrap.h types.c
+nodist_libxml2mod_la_SOURCES = libxml2-py.h libxml2-py.c
+libxml2mod_la_LDFLAGS = $(AM_LDFLAGS) $(PYTHON_LDFLAGS) -module -avoid-version
+libxml2mod_la_LIBADD = $(top_builddir)/libxml2.la $(PYTHON_LIBS)
+
+BUILT_SOURCES = libxml2-export.c libxml2-py.h libxml2-py.c
+
+python_PYTHON = drv_libxml2.py
+nodist_python_PYTHON = libxml2.py
+
+API_DESC = ../doc/html.stamp
+GENERATED = libxml2.py $(BUILT_SOURCES)
+CLEANFILES = $(GENERATED)
+
+all-local: libxml2.py
+
+$(GENERATED): $(srcdir)/generator.py $(API_DESC)
+	$(PYTHON) $(srcdir)/generator.py $(builddir)
+
+# libxml.c #includes libxml2-export.c
+libxml.$(OBJEXT): libxml2-export.c
+
+clean-local:
+	rm -rf __pycache__ *.pyc
+
+endif

libxml2/python/README

@@ -0,0 +1,37 @@
+		Module libxml2-python
+		=====================
+
+DEPRECATION WARNING: It is planned to remove the libxml2 Python
+bindings from the libxml2 repo.
+
+This is the libxml2 python module, providing access to the
+libxml2 and libxslt (if available) libraries. For general
+informationss on those XML and XSLT libraries check their 
+web pages at:
+    https://gitlab.gnome.org/GNOME/libxml2/-/wikis/home
+    and
+    https://gitlab.gnome.org/GNOME/libxslt/-/wikis/home
+
+The latest version of the sources for this module and the
+associated libraries can be found at:
+    https://gitlab.gnome.org/GNOME/libxml2/-/releases
+
+Binaries packages of the libxml2 and libxslt libraries can
+be found either on the FTP site for Linux, from external
+sources linked from the web pages, or as part of your set of
+packages provided with your operating system.
+
+NOTE:
+this module distribution is not the primary distribution
+of the libxml2 and libxslt Python binding code, but as 
+the Python way of packaging those for non-Linux systems.
+The main sources are the libxml2 and libxslt tar.gz found on
+the site. One side effect is that the official RPM packages for
+those modules are not generated from the libxml2-python
+distributions but as part of the normal RPM packaging of
+those two libraries.
+The RPM packages can be found at:
+    http://rpmfind.net/linux/rpm2html/search.php?query=libxml2-python
+    http://rpmfind.net/linux/rpm2html/search.php?query=libxslt-python
+
+Daniel Veillard

libxml2/python/drv_libxml2.py

@@ -0,0 +1,382 @@
+# -*- coding: iso-8859-1 -*-
+""" A SAX2 driver for libxml2, on top of it's XmlReader API
+
+USAGE
+    # put this file (drv_libxml2.py) in PYTHONPATH
+    import xml.sax
+    reader = xml.sax.make_parser(["drv_libxml2"])
+    # ...and the rest is standard python sax.
+
+CAVEATS
+    - Lexical handlers are supported, except for start/endEntity
+      (waiting for XmlReader.ResolveEntity) and start/endDTD
+    - Error callbacks are not exactly synchronous, they tend
+      to be invoked before the corresponding content callback,
+      because the underlying reader interface parses
+      data by chunks of 512 bytes
+    
+TODO
+    - search for TODO
+    - some ErrorHandler events (warning)
+    - some ContentHandler events (setDocumentLocator, skippedEntity)
+    - EntityResolver (using libxml2.?)
+    - DTDHandler (if/when libxml2 exposes such node types)
+    - DeclHandler (if/when libxml2 exposes such node types)
+    - property_xml_string?
+    - feature_string_interning?
+    - Incremental parser
+    - additional performance tuning:
+      - one might cache callbacks to avoid some name lookups
+      - one might implement a smarter way to pass attributes to startElement
+        (some kind of lazy evaluation?)
+      - there might be room for improvement in start/endPrefixMapping
+      - other?
+
+"""
+
+__author__  = "Stéphane Bidoul <sbi@skynet.be>"
+__version__ = "0.3"
+
+import sys
+import codecs
+
+if sys.version_info[0] < 3:
+    __author__  = codecs.unicode_escape_decode(__author__)[0]
+
+    StringTypes = (str, unicode)
+    # libxml2 returns strings as UTF8
+    _decoder = codecs.lookup("utf8")[1]
+    def _d(s):
+        if s is None:
+            return s
+        else:
+            return _decoder(s)[0]
+else:
+    StringTypes = str
+    # s is Unicode `str` already
+    def _d(s):
+        return s
+
+from xml.sax._exceptions import *
+from xml.sax import xmlreader, saxutils
+from xml.sax.handler import \
+     feature_namespaces, \
+     feature_namespace_prefixes, \
+     feature_string_interning, \
+     feature_validation, \
+     feature_external_ges, \
+     feature_external_pes, \
+     property_lexical_handler, \
+     property_declaration_handler, \
+     property_dom_node, \
+     property_xml_string
+
+try:
+    import libxml2
+except ImportError:
+    raise SAXReaderNotAvailable("libxml2 not available: " \
+                                "import error was: %s" % sys.exc_info()[1])
+
+class Locator(xmlreader.Locator):
+    """SAX Locator adapter for libxml2.xmlTextReaderLocator"""
+
+    def __init__(self,locator):
+        self.__locator = locator
+
+    def getColumnNumber(self):
+        "Return the column number where the current event ends."
+        return -1
+
+    def getLineNumber(self):
+        "Return the line number where the current event ends."
+        return self.__locator.LineNumber()
+
+    def getPublicId(self):
+        "Return the public identifier for the current event."
+        return None
+
+    def getSystemId(self):
+        "Return the system identifier for the current event."
+        return self.__locator.BaseURI()
+
+class LibXml2Reader(xmlreader.XMLReader):
+
+    def __init__(self):
+        xmlreader.XMLReader.__init__(self)
+        # features
+        self.__ns = 0
+        self.__nspfx = 0
+        self.__validate = 0
+        self.__extparams = 1
+        # parsing flag
+        self.__parsing = 0
+        # additional handlers
+        self.__lex_handler = None
+        self.__decl_handler = None
+        # error messages accumulator
+        self.__errors = None
+
+    def _errorHandler(self,arg,msg,severity,locator):
+        if self.__errors is None:
+            self.__errors = []
+        self.__errors.append((severity,
+                              SAXParseException(msg,None,
+                                                Locator(locator))))
+
+    def _reportErrors(self,fatal):
+        for severity,exception in self.__errors:
+            if severity in (libxml2.PARSER_SEVERITY_VALIDITY_WARNING,
+                            libxml2.PARSER_SEVERITY_WARNING):
+                self._err_handler.warning(exception)
+            else:
+                # when fatal is set, the parse will stop;
+                # we consider that the last error reported
+                # is the fatal one.
+                if fatal and exception is self.__errors[-1][1]:
+                    self._err_handler.fatalError(exception)
+                else:
+                    self._err_handler.error(exception)
+        self.__errors = None
+
+    def parse(self, source):
+        self.__parsing = 1
+        try:
+            # prepare source and create reader
+            if isinstance(source, StringTypes):
+                reader = libxml2.newTextReaderFilename(source)
+            else:
+                source = saxutils.prepare_input_source(source)
+                stream = source.getCharacterStream()
+                if stream is None:
+                    stream = source.getByteStream()
+                input = libxml2.inputBuffer(stream)
+                reader = input.newTextReader(source.getSystemId())
+            reader.SetErrorHandler(self._errorHandler,None)
+            # configure reader
+            if self.__extparams:
+                reader.SetParserProp(libxml2.PARSER_LOADDTD,1)
+                reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS,1)
+                reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)
+                reader.SetParserProp(libxml2.PARSER_VALIDATE,self.__validate)
+            else:
+                reader.SetParserProp(libxml2.PARSER_LOADDTD, 0)
+            # we reuse attribute maps (for a slight performance gain)
+            if self.__ns:
+                attributesNSImpl = xmlreader.AttributesNSImpl({},{})
+            else:
+                attributesImpl = xmlreader.AttributesImpl({})
+            # prefixes to pop (for endPrefixMapping)
+            prefixes = []
+            # start loop
+            self._cont_handler.startDocument()
+            while 1:
+                r = reader.Read()
+                # check for errors
+                if r == 1:
+                    if not self.__errors is None:
+                        self._reportErrors(0)
+                elif r == 0:
+                    if not self.__errors is None:
+                        self._reportErrors(0)
+                    break # end of parse
+                else:
+                    if not self.__errors is None:
+                        self._reportErrors(1)
+                    else:
+                        self._err_handler.fatalError(\
+                            SAXException("Read failed (no details available)"))
+                    break # fatal parse error
+                # get node type
+                nodeType = reader.NodeType()
+                # Element
+                if nodeType == 1: 
+                    if self.__ns:
+                        eltName = (_d(reader.NamespaceUri()),\
+                                   _d(reader.LocalName()))
+                        eltQName = _d(reader.Name())
+                        attributesNSImpl._attrs = attrs = {}
+                        attributesNSImpl._qnames = qnames = {}
+                        newPrefixes = []
+                        while reader.MoveToNextAttribute():
+                            qname = _d(reader.Name())
+                            value = _d(reader.Value())
+                            if qname.startswith("xmlns"):
+                                if len(qname) > 5:
+                                    newPrefix = qname[6:]
+                                else:
+                                    newPrefix = None
+                                newPrefixes.append(newPrefix)
+                                self._cont_handler.startPrefixMapping(\
+                                    newPrefix,value)
+                                if not self.__nspfx:
+                                    continue # don't report xmlns attribute
+                            attName = (_d(reader.NamespaceUri()),
+                                       _d(reader.LocalName()))
+                            qnames[attName] = qname
+                            attrs[attName] = value
+                        reader.MoveToElement()
+                        self._cont_handler.startElementNS( \
+                            eltName,eltQName,attributesNSImpl) 
+                        if reader.IsEmptyElement():
+                            self._cont_handler.endElementNS(eltName,eltQName)
+                            for newPrefix in newPrefixes:
+                                self._cont_handler.endPrefixMapping(newPrefix)
+                        else:
+                            prefixes.append(newPrefixes)
+                    else:
+                        eltName = _d(reader.Name())
+                        attributesImpl._attrs = attrs = {}
+                        while reader.MoveToNextAttribute():
+                            attName = _d(reader.Name())
+                            attrs[attName] = _d(reader.Value())
+                        reader.MoveToElement()
+                        self._cont_handler.startElement( \
+                            eltName,attributesImpl)
+                        if reader.IsEmptyElement():
+                            self._cont_handler.endElement(eltName)
+                # EndElement
+                elif nodeType == 15: 
+                    if self.__ns:
+                        self._cont_handler.endElementNS( \
+                             (_d(reader.NamespaceUri()),_d(reader.LocalName())),
+                             _d(reader.Name()))
+                        for prefix in prefixes.pop():
+                            self._cont_handler.endPrefixMapping(prefix)
+                    else:
+                        self._cont_handler.endElement(_d(reader.Name()))
+                # Text
+                elif nodeType == 3: 
+                    self._cont_handler.characters(_d(reader.Value()))
+                # Whitespace
+                elif nodeType == 13: 
+                    self._cont_handler.ignorableWhitespace(_d(reader.Value()))
+                # SignificantWhitespace
+                elif nodeType == 14:
+                    self._cont_handler.characters(_d(reader.Value()))
+                # CDATA
+                elif nodeType == 4:
+                    if not self.__lex_handler is None:
+                        self.__lex_handler.startCDATA()
+                    self._cont_handler.characters(_d(reader.Value()))
+                    if not self.__lex_handler is None:
+                        self.__lex_handler.endCDATA()
+                # EntityReference
+                elif nodeType == 5:
+                    if not self.__lex_handler is None:
+                        self.startEntity(_d(reader.Name()))
+                    reader.ResolveEntity()
+                # EndEntity
+                elif nodeType == 16:
+                    if not self.__lex_handler is None:
+                        self.endEntity(_d(reader.Name()))
+                # ProcessingInstruction
+                elif nodeType == 7: 
+                    self._cont_handler.processingInstruction( \
+                        _d(reader.Name()),_d(reader.Value()))
+                # Comment
+                elif nodeType == 8:
+                    if not self.__lex_handler is None:
+                        self.__lex_handler.comment(_d(reader.Value()))
+                # DocumentType
+                elif nodeType == 10:
+                    #if not self.__lex_handler is None:
+                    #    self.__lex_handler.startDTD()
+                    pass # TODO (how to detect endDTD? on first non-dtd event?)
+                # XmlDeclaration
+                elif nodeType == 17:
+                    pass # TODO
+                # Entity
+                elif nodeType == 6:
+                    pass # TODO (entity decl)
+                # Notation (decl)
+                elif nodeType == 12:
+                    pass # TODO
+                # Attribute (never in this loop)
+                #elif nodeType == 2: 
+                #    pass
+                # Document (not exposed)
+                #elif nodeType == 9: 
+                #    pass
+                # DocumentFragment (never returned by XmlReader)
+                #elif nodeType == 11:
+                #    pass
+                # None
+                #elif nodeType == 0:
+                #    pass
+                # -
+                else:
+                    raise SAXException("Unexpected node type %d" % nodeType)
+            if r == 0:
+                self._cont_handler.endDocument()
+            reader.Close()
+        finally:
+            self.__parsing = 0
+
+    def setDTDHandler(self, handler):
+        # TODO (when supported, the inherited method works just fine)
+        raise SAXNotSupportedException("DTDHandler not supported")
+
+    def setEntityResolver(self, resolver):
+        # TODO (when supported, the inherited method works just fine)
+        raise SAXNotSupportedException("EntityResolver not supported")
+
+    def getFeature(self, name):
+        if name == feature_namespaces:
+            return self.__ns
+        elif name == feature_namespace_prefixes:
+            return self.__nspfx
+        elif name == feature_validation:
+            return self.__validate
+        elif name == feature_external_ges:
+            return 1 # TODO (does that relate to PARSER_LOADDTD)?
+        elif name == feature_external_pes:
+            return self.__extparams
+        else:
+            raise SAXNotRecognizedException("Feature '%s' not recognized" % \
+                                            name)
+
+    def setFeature(self, name, state):
+        if self.__parsing:
+            raise SAXNotSupportedException("Cannot set feature %s " \
+                                           "while parsing" % name)
+        if name == feature_namespaces:
+            self.__ns = state
+        elif name == feature_namespace_prefixes:
+            self.__nspfx = state
+        elif name == feature_validation:
+            self.__validate = state
+        elif name == feature_external_ges:
+            if state == 0:
+                # TODO (does that relate to PARSER_LOADDTD)?
+                raise SAXNotSupportedException("Feature '%s' not supported" % \
+                                               name)
+        elif name == feature_external_pes:
+            self.__extparams = state
+        else:
+            raise SAXNotRecognizedException("Feature '%s' not recognized" % \
+                                            name)
+
+    def getProperty(self, name):
+        if name == property_lexical_handler:
+            return self.__lex_handler
+        elif name == property_declaration_handler:
+            return self.__decl_handler
+        else:
+            raise SAXNotRecognizedException("Property '%s' not recognized" % \
+                                            name)
+
+    def setProperty(self, name, value):     
+        if name == property_lexical_handler:
+            self.__lex_handler = value
+        elif name == property_declaration_handler:
+            # TODO: remove if/when libxml2 supports dtd events
+            raise SAXNotSupportedException("Property '%s' not supported" % \
+                                           name)
+            self.__decl_handler = value
+        else:
+            raise SAXNotRecognizedException("Property '%s' not recognized" % \
+                                            name)
+
+def create_parser():
+    return LibXml2Reader()
+

libxml2/python/libxml.py

@@ -0,0 +1,794 @@
+import libxml2mod
+import types
+import sys
+
+# The root of all libxml2 errors.
+class libxmlError(Exception): pass
+
+# Type of the wrapper class for the C objects wrappers
+def checkWrapper(obj):
+    try:
+        n = type(_obj).__name__
+        if n != 'PyCObject' and n != 'PyCapsule':
+            return 1
+    except:
+        return 0
+    return 0
+
+#
+# id() is sometimes negative ...
+#
+def pos_id(o):
+    i = id(o)
+    if (i < 0):
+        return (sys.maxsize - i)
+    return i
+
+#
+# Errors raised by the wrappers when some tree handling failed.
+#
+class treeError(libxmlError):
+    def __init__(self, msg):
+        self.msg = msg
+    def __str__(self):
+        return self.msg
+
+class parserError(libxmlError):
+    def __init__(self, msg):
+        self.msg = msg
+    def __str__(self):
+        return self.msg
+
+class uriError(libxmlError):
+    def __init__(self, msg):
+        self.msg = msg
+    def __str__(self):
+        return self.msg
+
+class xpathError(libxmlError):
+    def __init__(self, msg):
+        self.msg = msg
+    def __str__(self):
+        return self.msg
+
+class ioWrapper:
+    def __init__(self, _obj):
+        self.__io = _obj
+        self._o = None
+
+    def io_close(self):
+        if self.__io == None:
+            return(-1)
+        self.__io.close()
+        self.__io = None
+        return(0)
+
+    def io_flush(self):
+        if self.__io == None:
+            return(-1)
+        self.__io.flush()
+        return(0)
+
+    def io_read(self, len = -1):
+        if self.__io == None:
+            return(-1)
+        try:
+            if len < 0:
+                ret = self.__io.read()
+            else:
+                ret = self.__io.read(len)
+        except Exception:
+            import sys
+            e = sys.exc_info()[1]
+            print("failed to read from Python:", type(e))
+            print("on IO:", self.__io)
+            self.__io == None
+            return(-1)
+
+        return(ret)
+
+    def io_write(self, str, len = -1):
+        if self.__io == None:
+            return(-1)
+        if len < 0:
+            return(self.__io.write(str))
+        return(self.__io.write(str, len))
+
+class ioReadWrapper(ioWrapper):
+    def __init__(self, _obj, enc = ""):
+        ioWrapper.__init__(self, _obj)
+        self._o = libxml2mod.xmlCreateInputBuffer(self, enc)
+
+    def __del__(self):
+        print("__del__")
+        self.io_close()
+        if self._o != None:
+            libxml2mod.xmlFreeParserInputBuffer(self._o)
+        self._o = None
+
+    def close(self):
+        self.io_close()
+        if self._o != None:
+            libxml2mod.xmlFreeParserInputBuffer(self._o)
+        self._o = None
+
+class ioWriteWrapper(ioWrapper):
+    def __init__(self, _obj, enc = ""):
+#        print "ioWriteWrapper.__init__", _obj
+        if type(_obj) == type(''):
+            print("write io from a string")
+            self.o = None
+        elif type(_obj).__name__ == 'PyCapsule':
+            file = libxml2mod.outputBufferGetPythonFile(_obj)
+            if file != None:
+                ioWrapper.__init__(self, file)
+            else:
+                ioWrapper.__init__(self, _obj)
+            self._o = _obj
+#        elif type(_obj) == types.InstanceType:
+#            print(("write io from instance of %s" % (_obj.__class__)))
+#            ioWrapper.__init__(self, _obj)
+#            self._o = libxml2mod.xmlCreateOutputBuffer(self, enc)
+        else:
+            file = libxml2mod.outputBufferGetPythonFile(_obj)
+            if file != None:
+                ioWrapper.__init__(self, file)
+            else:
+                ioWrapper.__init__(self, _obj)
+            self._o = _obj
+
+    def __del__(self):
+#        print "__del__"
+        self.io_close()
+        if self._o != None:
+            libxml2mod.xmlOutputBufferClose(self._o)
+        self._o = None
+
+    def flush(self):
+        self.io_flush()
+        if self._o != None:
+            libxml2mod.xmlOutputBufferClose(self._o)
+        self._o = None
+
+    def close(self):
+        self.io_flush()
+        if self._o != None:
+            libxml2mod.xmlOutputBufferClose(self._o)
+        self._o = None
+
+#
+# Example of a class to handle SAX events
+#
+class SAXCallback:
+    """Base class for SAX handlers"""
+    def startDocument(self):
+        """called at the start of the document"""
+        pass
+
+    def endDocument(self):
+        """called at the end of the document"""
+        pass
+
+    def startElement(self, tag, attrs):
+        """called at the start of every element, tag is the name of
+           the element, attrs is a dictionary of the element's attributes"""
+        pass
+
+    def endElement(self, tag):
+        """called at the start of every element, tag is the name of
+           the element"""
+        pass
+
+    def characters(self, data):
+        """called when character data have been read, data is the string
+           containing the data, multiple consecutive characters() callback
+           are possible."""
+        pass
+
+    def cdataBlock(self, data):
+        """called when CDATA section have been read, data is the string
+           containing the data, multiple consecutive cdataBlock() callback
+           are possible."""
+        pass
+
+    def reference(self, name):
+        """called when an entity reference has been found"""
+        pass
+
+    def ignorableWhitespace(self, data):
+        """called when potentially ignorable white spaces have been found"""
+        pass
+
+    def processingInstruction(self, target, data):
+        """called when a PI has been found, target contains the PI name and
+           data is the associated data in the PI"""
+        pass
+
+    def comment(self, content):
+        """called when a comment has been found, content contains the comment"""
+        pass
+
+    def externalSubset(self, name, externalID, systemID):
+        """called when a DOCTYPE declaration has been found, name is the
+           DTD name and externalID, systemID are the DTD public and system
+           identifier for that DTd if available"""
+        pass
+
+    def internalSubset(self, name, externalID, systemID):
+        """called when a DOCTYPE declaration has been found, name is the
+           DTD name and externalID, systemID are the DTD public and system
+           identifier for that DTD if available"""
+        pass
+
+    def entityDecl(self, name, type, externalID, systemID, content):
+        """called when an ENTITY declaration has been found, name is the
+           entity name and externalID, systemID are the entity public and
+           system identifier for that entity if available, type indicates
+           the entity type, and content reports it's string content"""
+        pass
+
+    def notationDecl(self, name, externalID, systemID):
+        """called when an NOTATION declaration has been found, name is the
+           notation name and externalID, systemID are the notation public and
+           system identifier for that notation if available"""
+        pass
+
+    def attributeDecl(self, elem, name, type, defi, defaultValue, nameList):
+        """called when an ATTRIBUTE definition has been found"""
+        pass
+
+    def elementDecl(self, name, type, content):
+        """called when an ELEMENT definition has been found"""
+        pass
+
+    def entityDecl(self, name, publicId, systemID, notationName):
+        """called when an unparsed ENTITY declaration has been found,
+           name is the entity name and publicId,, systemID are the entity
+           public and system identifier for that entity if available,
+           and notationName indicate the associated NOTATION"""
+        pass
+
+    def warning(self, msg):
+        #print msg
+        pass
+
+    def error(self, msg):
+        raise parserError(msg)
+
+    def fatalError(self, msg):
+        raise parserError(msg)
+
+#
+# This class is the ancestor of all the Node classes. It provides
+# the basic functionalities shared by all nodes (and handle
+# gracefylly the exception), like name, navigation in the tree,
+# doc reference, content access and serializing to a string or URI
+#
+class xmlCore:
+    def __init__(self, _obj=None):
+        if _obj != None: 
+            self._o = _obj;
+            return
+        self._o = None
+
+    def __eq__(self, other):
+        if other == None:
+            return False
+        ret = libxml2mod.compareNodesEqual(self._o, other._o)
+        if ret == None:
+            return False
+        return ret == True
+    def __ne__(self, other):
+        if other == None:
+            return True
+        ret = libxml2mod.compareNodesEqual(self._o, other._o)
+        return not ret
+    def __hash__(self):
+        ret = libxml2mod.nodeHash(self._o)
+        return ret
+
+    def __str__(self):
+        return self.serialize()
+    def get_parent(self):
+        ret = libxml2mod.parent(self._o)
+        if ret == None:
+            return None
+        return nodeWrap(ret)
+    def get_children(self):
+        ret = libxml2mod.children(self._o)
+        if ret == None:
+            return None
+        return nodeWrap(ret)
+    def get_last(self):
+        ret = libxml2mod.last(self._o)
+        if ret == None:
+            return None
+        return nodeWrap(ret)
+    def get_next(self):
+        ret = libxml2mod.next(self._o)
+        if ret == None:
+            return None
+        return nodeWrap(ret)
+    def get_properties(self):
+        ret = libxml2mod.properties(self._o)
+        if ret == None:
+            return None
+        return xmlAttr(_obj=ret)
+    def get_prev(self):
+        ret = libxml2mod.prev(self._o)
+        if ret == None:
+            return None
+        return nodeWrap(ret)
+    def get_content(self):
+        return libxml2mod.xmlNodeGetContent(self._o)
+    getContent = get_content  # why is this duplicate naming needed ?
+    def get_name(self):
+        return libxml2mod.name(self._o)
+    def get_type(self):
+        return libxml2mod.type(self._o)
+    def get_doc(self):
+        ret = libxml2mod.doc(self._o)
+        if ret == None:
+            if self.type in ["document_xml", "document_html"]:
+                return xmlDoc(_obj=self._o)
+            else:
+                return None
+        return xmlDoc(_obj=ret)
+    #
+    # Those are common attributes to nearly all type of nodes
+    # defined as python2 properties
+    # 
+    import sys
+    if float(sys.version[0:3]) < 2.2:
+        def __getattr__(self, attr):
+            if attr == "parent":
+                ret = libxml2mod.parent(self._o)
+                if ret == None:
+                    return None
+                return nodeWrap(ret)
+            elif attr == "properties":
+                ret = libxml2mod.properties(self._o)
+                if ret == None:
+                    return None
+                return xmlAttr(_obj=ret)
+            elif attr == "children":
+                ret = libxml2mod.children(self._o)
+                if ret == None:
+                    return None
+                return nodeWrap(ret)
+            elif attr == "last":
+                ret = libxml2mod.last(self._o)
+                if ret == None:
+                    return None
+                return nodeWrap(ret)
+            elif attr == "next":
+                ret = libxml2mod.next(self._o)
+                if ret == None:
+                    return None
+                return nodeWrap(ret)
+            elif attr == "prev":
+                ret = libxml2mod.prev(self._o)
+                if ret == None:
+                    return None
+                return nodeWrap(ret)
+            elif attr == "content":
+                return libxml2mod.xmlNodeGetContent(self._o)
+            elif attr == "name":
+                return libxml2mod.name(self._o)
+            elif attr == "type":
+                return libxml2mod.type(self._o)
+            elif attr == "doc":
+                ret = libxml2mod.doc(self._o)
+                if ret == None:
+                    if self.type == "document_xml" or self.type == "document_html":
+                        return xmlDoc(_obj=self._o)
+                    else:
+                        return None
+                return xmlDoc(_obj=ret)
+            raise AttributeError(attr)
+    else:
+        parent = property(get_parent, None, None, "Parent node")
+        children = property(get_children, None, None, "First child node")
+        last = property(get_last, None, None, "Last sibling node")
+        next = property(get_next, None, None, "Next sibling node")
+        prev = property(get_prev, None, None, "Previous sibling node")
+        properties = property(get_properties, None, None, "List of properties")
+        content = property(get_content, None, None, "Content of this node")
+        name = property(get_name, None, None, "Node name")
+        type = property(get_type, None, None, "Node type")
+        doc = property(get_doc, None, None, "The document this node belongs to")
+
+    #
+    # Serialization routines, the optional arguments have the following
+    # meaning:
+    #     encoding: string to ask saving in a specific encoding
+    #     indent: if 1 the serializer is asked to indent the output
+    #
+    def serialize(self, encoding = None, format = 0):
+        return libxml2mod.serializeNode(self._o, encoding, format)
+    def saveTo(self, file, encoding = None, format = 0):
+        return libxml2mod.saveNodeTo(self._o, file, encoding, format)
+            
+    #
+    # Canonicalization routines:
+    #
+    #   nodes: the node set (tuple or list) to be included in the
+    #     canonized image or None if all document nodes should be
+    #     included.
+    #   exclusive: the exclusive flag (0 - non-exclusive
+    #     canonicalization; otherwise - exclusive canonicalization)
+    #   prefixes: the list of inclusive namespace prefixes (strings),
+    #     or None if there is no inclusive namespaces (only for
+    #     exclusive canonicalization, ignored otherwise)
+    #   with_comments: include comments in the result (!=0) or not
+    #     (==0)
+    def c14nMemory(self,
+                   nodes=None,
+                   exclusive=0,
+                   prefixes=None,
+                   with_comments=0):
+        if nodes:
+            nodes = [n._o for n in nodes]
+        return libxml2mod.xmlC14NDocDumpMemory(
+            self.get_doc()._o,
+            nodes,
+            exclusive != 0,
+            prefixes,
+            with_comments != 0)
+    def c14nSaveTo(self,
+                   file,
+                   nodes=None,
+                   exclusive=0,
+                   prefixes=None,
+                   with_comments=0):
+        if nodes:
+            nodes = [n._o for n in nodes]
+        return libxml2mod.xmlC14NDocSaveTo(
+            self.get_doc()._o,
+            nodes,
+            exclusive != 0,
+            prefixes,
+            with_comments != 0,
+            file)
+
+    #
+    # Selecting nodes using XPath, a bit slow because the context
+    # is allocated/freed every time but convenient.
+    #
+    def xpathEval(self, expr):
+        doc = self.doc
+        if doc == None:
+            return None
+        ctxt = doc.xpathNewContext()
+        ctxt.setContextNode(self)
+        res = ctxt.xpathEval(expr)
+        ctxt.xpathFreeContext()
+        return res
+
+#    #
+#    # Selecting nodes using XPath, faster because the context
+#    # is allocated just once per xmlDoc.
+#    #
+#    # Removed: DV memleaks c.f. #126735
+#    #
+#    def xpathEval2(self, expr):
+#        doc = self.doc
+#        if doc == None:
+#            return None
+#        try:
+#            doc._ctxt.setContextNode(self)
+#        except:
+#            doc._ctxt = doc.xpathNewContext()
+#            doc._ctxt.setContextNode(self)
+#        res = doc._ctxt.xpathEval(expr)
+#        return res
+    def xpathEval2(self, expr):
+        return self.xpathEval(expr)
+
+    # Remove namespaces
+    def removeNsDef(self, href):
+        """
+        Remove a namespace definition from a node.  If href is None,
+        remove all of the ns definitions on that node.  The removed
+        namespaces are returned as a linked list.
+
+        Note: If any child nodes referred to the removed namespaces,
+        they will be left with dangling links.  You should call
+        renconciliateNs() to fix those pointers.
+
+        Note: This method does not free memory taken by the ns
+        definitions.  You will need to free it manually with the
+        freeNsList() method on the returns xmlNs object.
+        """
+
+        ret = libxml2mod.xmlNodeRemoveNsDef(self._o, href)
+        if ret is None:return None
+        __tmp = xmlNs(_obj=ret)
+        return __tmp
+
+    # support for python2 iterators
+    def walk_depth_first(self):
+        return xmlCoreDepthFirstItertor(self)
+    def walk_breadth_first(self):
+        return xmlCoreBreadthFirstItertor(self)
+    __iter__ = walk_depth_first
+
+    def free(self):
+        try:
+            self.doc._ctxt.xpathFreeContext()
+        except:
+            pass
+        libxml2mod.xmlFreeDoc(self._o)
+
+
+#
+# implements the depth-first iterator for libxml2 DOM tree
+#
+class xmlCoreDepthFirstItertor:
+    def __init__(self, node):
+        self.node = node
+        self.parents = []
+    def __iter__(self):
+        return self
+    def __next__(self):
+        while 1:
+            if self.node:
+                ret = self.node
+                self.parents.append(self.node)
+                self.node = self.node.children
+                return ret
+            try:
+                parent = self.parents.pop()
+            except IndexError:
+                raise StopIteration
+            self.node = parent.next
+    next = __next__
+
+#
+# implements the breadth-first iterator for libxml2 DOM tree
+#
+class xmlCoreBreadthFirstItertor:
+    def __init__(self, node):
+        self.node = node
+        self.parents = []
+    def __iter__(self):
+        return self
+    def __next__(self):
+        while 1:
+            if self.node:
+                ret = self.node
+                self.parents.append(self.node)
+                self.node = self.node.next
+                return ret
+            try:
+                parent = self.parents.pop()
+            except IndexError:
+                raise StopIteration
+            self.node = parent.children
+    next = __next__
+
+#
+# converters to present a nicer view of the XPath returns
+#
+def nodeWrap(o):
+    # TODO try to cast to the most appropriate node class
+    name = libxml2mod.type(o)
+    if name == "element" or name == "text":
+        return xmlNode(_obj=o)
+    if name == "attribute":
+        return xmlAttr(_obj=o)
+    if name[0:8] == "document":
+        return xmlDoc(_obj=o)
+    if name == "namespace":
+        return xmlNs(_obj=o)
+    if name == "elem_decl":
+        return xmlElement(_obj=o)
+    if name == "attribute_decl":
+        return xmlAttribute(_obj=o)
+    if name == "entity_decl":
+        return xmlEntity(_obj=o)
+    if name == "dtd":
+        return xmlDtd(_obj=o)
+    return xmlNode(_obj=o)
+
+def xpathObjectRet(o):
+    otype = type(o)
+    if otype == type([]):
+        ret = list(map(xpathObjectRet, o))
+        return ret
+    elif otype == type(()):
+        ret = list(map(xpathObjectRet, o))
+        return tuple(ret)
+    elif otype == type('') or otype == type(0) or otype == type(0.0):
+        return o
+    else:
+        return nodeWrap(o)
+
+#
+# register an XPath function
+#
+def registerXPathFunction(ctxt, name, ns_uri, f):
+    ret = libxml2mod.xmlRegisterXPathFunction(ctxt, name, ns_uri, f)
+
+#
+# For the xmlTextReader parser configuration
+#
+PARSER_LOADDTD=1
+PARSER_DEFAULTATTRS=2
+PARSER_VALIDATE=3
+PARSER_SUBST_ENTITIES=4
+
+#
+# For the error callback severities
+#
+PARSER_SEVERITY_VALIDITY_WARNING=1
+PARSER_SEVERITY_VALIDITY_ERROR=2
+PARSER_SEVERITY_WARNING=3
+PARSER_SEVERITY_ERROR=4
+
+#
+# register the libxml2 error handler
+#
+def registerErrorHandler(f, ctx):
+    """Register a Python written function to for error reporting.
+       The function is called back as f(ctx, error). """
+    import sys
+    if 'libxslt' not in sys.modules:
+        # normal behaviour when libxslt is not imported
+        ret = libxml2mod.xmlRegisterErrorHandler(f,ctx)
+    else:
+        # when libxslt is already imported, one must
+        # use libxst's error handler instead
+        import libxslt
+        ret = libxslt.registerErrorHandler(f,ctx)
+    return ret
+
+class parserCtxtCore:
+
+    def __init__(self, _obj=None):
+        if _obj != None: 
+            self._o = _obj;
+            return
+        self._o = None
+
+    def __del__(self):
+        if self._o != None:
+            libxml2mod.xmlFreeParserCtxt(self._o)
+        self._o = None
+
+    def setErrorHandler(self,f,arg):
+        """Register an error handler that will be called back as
+           f(arg,msg,severity,reserved).
+           
+           @reserved is currently always None."""
+        libxml2mod.xmlParserCtxtSetErrorHandler(self._o,f,arg)
+
+    def getErrorHandler(self):
+        """Return (f,arg) as previously registered with setErrorHandler
+           or (None,None)."""
+        return libxml2mod.xmlParserCtxtGetErrorHandler(self._o)
+
+    def addLocalCatalog(self, uri):
+        """Register a local catalog with the parser"""
+        return libxml2mod.addLocalCatalog(self._o, uri)
+    
+
+class ValidCtxtCore:
+
+    def __init__(self, *args, **kw):
+        pass
+
+    def setValidityErrorHandler(self, err_func, warn_func, arg=None):
+        """
+        Register error and warning handlers for DTD validation.
+        These will be called back as f(msg,arg)
+        """
+        libxml2mod.xmlSetValidErrors(self._o, err_func, warn_func, arg)
+    
+
+class SchemaValidCtxtCore:
+
+    def __init__(self, *args, **kw):
+        pass
+
+    def setValidityErrorHandler(self, err_func, warn_func, arg=None):
+        """
+        Register error and warning handlers for Schema validation.
+        These will be called back as f(msg,arg)
+        """
+        libxml2mod.xmlSchemaSetValidErrors(self._o, err_func, warn_func, arg)
+
+
+class relaxNgValidCtxtCore:
+
+    def __init__(self, *args, **kw):
+        pass
+
+    def setValidityErrorHandler(self, err_func, warn_func, arg=None):
+        """
+        Register error and warning handlers for RelaxNG validation.
+        These will be called back as f(msg,arg)
+        """
+        libxml2mod.xmlRelaxNGSetValidErrors(self._o, err_func, warn_func, arg)
+
+    
+def _xmlTextReaderErrorFunc(xxx_todo_changeme,msg,severity,locator):
+    """Intermediate callback to wrap the locator"""
+    (f,arg) = xxx_todo_changeme
+    return f(arg,msg,severity,xmlTextReaderLocator(locator))
+
+class xmlTextReaderCore:
+
+    def __init__(self, _obj=None):
+        self.input = None
+        if _obj != None:self._o = _obj;return
+        self._o = None
+
+    def __del__(self):
+        if self._o != None:
+            libxml2mod.xmlFreeTextReader(self._o)
+        self._o = None
+
+    def SetErrorHandler(self,f,arg):
+        """Register an error handler that will be called back as
+           f(arg,msg,severity,locator)."""
+        if f is None:
+            libxml2mod.xmlTextReaderSetErrorHandler(\
+                self._o,None,None)
+        else:
+            libxml2mod.xmlTextReaderSetErrorHandler(\
+                self._o,_xmlTextReaderErrorFunc,(f,arg))
+
+    def GetErrorHandler(self):
+        """Return (f,arg) as previously registered with setErrorHandler
+           or (None,None)."""
+        f,arg = libxml2mod.xmlTextReaderGetErrorHandler(self._o)
+        if f is None:
+            return None,None
+        else:
+            # assert f is _xmlTextReaderErrorFunc
+            return arg
+
+#
+# The cleanup now goes though a wrapper in libxml.c
+#
+def cleanupParser():
+    libxml2mod.xmlPythonCleanupParser()
+
+#
+# The interface to xmlRegisterInputCallbacks.
+# Since this API does not allow to pass a data object along with
+# match/open callbacks, it is necessary to maintain a list of all
+# Python callbacks.
+#
+__input_callbacks = []
+def registerInputCallback(func):
+    def findOpenCallback(URI):
+        for cb in reversed(__input_callbacks):
+            o = cb(URI)
+            if o is not None:
+                return o
+    libxml2mod.xmlRegisterInputCallback(findOpenCallback)
+    __input_callbacks.append(func)
+
+def popInputCallbacks():
+    # First pop python-level callbacks, when no more available - start
+    # popping built-in ones.
+    if len(__input_callbacks) > 0:
+        __input_callbacks.pop()
+    if len(__input_callbacks) == 0:
+        libxml2mod.xmlUnregisterInputCallback()
+
+#
+# Deprecated
+#
+def dumpMemory():
+    """DEPRECATED: This feature was removed."""
+
+# WARNING WARNING WARNING WARNING WARNING WARNING WARNING WARNING WARNING
+#
+# Everything before this line comes from libxml.py 
+# Everything after this line is automatically generated
+#
+# WARNING WARNING WARNING WARNING WARNING WARNING WARNING WARNING WARNING
+

libxml2/python/libxml_wrap.h

@@ -0,0 +1,317 @@
+#include <Python.h>
+#include <libxml/tree.h>
+#include <libxml/parser.h>
+#include <libxml/parserInternals.h>
+#include <libxml/catalog.h>
+#include <libxml/threads.h>
+#include <libxml/nanohttp.h>
+#include <libxml/uri.h>
+#include <libxml/xpath.h>
+#include <libxml/xpathInternals.h>
+#include <libxml/debugXML.h>
+#include <libxml/HTMLparser.h>
+#include <libxml/HTMLtree.h>
+#include <libxml/xinclude.h>
+#include <libxml/xpointer.h>
+#include <libxml/xmlregexp.h>
+#include <libxml/xmlautomata.h>
+#include <libxml/xmlreader.h>
+#include <libxml/xmlsave.h>
+#ifdef LIBXML_RELAXNG_ENABLED
+#include <libxml/relaxng.h>
+#endif
+#ifdef LIBXML_SCHEMAS_ENABLED
+#include <libxml/xmlschemas.h>
+#endif
+
+/*
+ * for older versions of Python, we don't use PyBytes, but keep PyString
+ * and don't use Capsule but CObjects
+ */
+#if PY_VERSION_HEX < 0x02070000
+#ifndef PyBytes_Check
+#define PyBytes_Check PyString_Check
+#define PyBytes_Size PyString_Size
+#define PyBytes_AsString PyString_AsString
+#define PyBytes_AS_STRING PyString_AS_STRING
+#define PyBytes_GET_SIZE PyString_GET_SIZE
+#endif
+#ifndef PyCapsule_New
+#define PyCapsule_New PyCObject_FromVoidPtrAndDesc
+#define PyCapsule_CheckExact PyCObject_Check
+#define PyCapsule_GetPointer(o, n) PyCObject_GetDesc((o))
+#endif
+#endif
+
+/**
+ * ATTRIBUTE_UNUSED:
+ *
+ * Macro used to signal to GCC unused function parameters
+ * Repeated here since the definition is not available when
+ * compiled outside the libxml2 build tree.
+ */
+#if defined(__GNUC__) || defined(__clang__)
+#ifdef ATTRIBUTE_UNUSED
+#undef ATTRIBUTE_UNUSED
+#endif
+#ifndef ATTRIBUTE_UNUSED
+#define ATTRIBUTE_UNUSED __attribute__ ((__unused__))
+#endif /* ATTRIBUTE_UNUSED */
+#else
+#define ATTRIBUTE_UNUSED
+#endif
+
+/*
+ * Macros to ignore deprecation warnings
+ */
+#if defined(__LCC__)
+  #define XML_IGNORE_DEPRECATION_WARNINGS _Pragma("diag_suppress 1215")
+  #define XML_POP_WARNINGS _Pragma("diag_default 1215")
+#elif defined(__clang__) || \
+      (defined(__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 406))
+  #define XML_IGNORE_DEPRECATION_WARNINGS \
+    _Pragma("GCC diagnostic push") \
+    _Pragma("GCC diagnostic ignored \"-Wdeprecated-declarations\"")
+  #define XML_POP_WARNINGS _Pragma("GCC diagnostic pop")
+#elif defined (_MSC_VER) && (_MSC_VER >= 1400)
+  #define XML_IGNORE_DEPRECATION_WARNINGS \
+    __pragma(warning(push)) \
+    __pragma(warning(disable : 4996))
+  #define XML_POP_WARNINGS __pragma(warning(pop))
+#else
+  #define XML_IGNORE_DEPRECATION_WARNINGS
+  #define XML_POP_WARNINGS
+#endif
+
+#define PyxmlNode_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyxmlNode_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlNodePtr obj;
+} PyxmlNode_Object;
+
+#ifdef LIBXML_XPATH_ENABLED
+#define PyxmlXPathContext_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyxmlXPathContext_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlXPathContextPtr obj;
+} PyxmlXPathContext_Object;
+
+#define PyxmlXPathParserContext_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyxmlXPathParserContext_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlXPathParserContextPtr obj;
+} PyxmlXPathParserContext_Object;
+#endif /* LIBXML_XPATH_ENABLED */
+
+#define PyparserCtxt_Get(v) (((v) == Py_None) ? NULL : \
+        (((PyparserCtxt_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlParserCtxtPtr obj;
+} PyparserCtxt_Object;
+
+#define PyValidCtxt_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyValidCtxt_Object *)(v))->obj))
+
+typedef struct {
+	PyObject_HEAD
+	xmlValidCtxtPtr obj;
+} PyValidCtxt_Object;
+
+#ifdef LIBXML_CATALOG_ENABLED
+#define Pycatalog_Get(v) (((v) == Py_None) ? NULL : \
+        (((Pycatalog_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlCatalogPtr obj;
+} Pycatalog_Object;
+#endif /* LIBXML_CATALOG_ENABLED */
+
+#ifdef LIBXML_REGEXP_ENABLED
+#define PyxmlReg_Get(v) (((v) == Py_None) ? NULL : \
+        (((PyxmlReg_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlRegexpPtr obj;
+} PyxmlReg_Object;
+#endif /* LIBXML_REGEXP_ENABLED */
+
+#ifdef LIBXML_READER_ENABLED
+#define PyxmlTextReader_Get(v) (((v) == Py_None) ? NULL : \
+        (((PyxmlTextReader_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlTextReaderPtr obj;
+} PyxmlTextReader_Object;
+
+#define PyxmlTextReaderLocator_Get(v) (((v) == Py_None) ? NULL : \
+        (((PyxmlTextReaderLocator_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlTextReaderLocatorPtr obj;
+} PyxmlTextReaderLocator_Object;
+#endif
+
+#define PyURI_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyURI_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlErrorPtr obj;
+} PyError_Object;
+
+#define PyError_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyError_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlOutputBufferPtr obj;
+} PyoutputBuffer_Object;
+
+#define PyoutputBuffer_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyoutputBuffer_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlParserInputBufferPtr obj;
+} PyinputBuffer_Object;
+
+#define PyinputBuffer_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyinputBuffer_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlURIPtr obj;
+} PyURI_Object;
+
+/* FILE * have their own internal representation */
+#if PY_MAJOR_VERSION >= 3
+FILE *libxml_PyFileGet(PyObject *f);
+void libxml_PyFileRelease(FILE *f);
+#define PyFile_Get(v) (((v) == Py_None) ? NULL : libxml_PyFileGet(v))
+#define PyFile_Release(f) libxml_PyFileRelease(f)
+#else
+#define PyFile_Get(v) (((v) == Py_None) ? NULL : \
+	(PyFile_Check(v) ? (PyFile_AsFile(v)) : stdout))
+#define PyFile_Release(f)
+#endif
+
+#ifdef LIBXML_RELAXNG_ENABLED
+typedef struct {
+    PyObject_HEAD
+    xmlRelaxNGPtr obj;
+} PyrelaxNgSchema_Object;
+
+#define PyrelaxNgSchema_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyrelaxNgSchema_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlRelaxNGParserCtxtPtr obj;
+} PyrelaxNgParserCtxt_Object;
+
+#define PyrelaxNgParserCtxt_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyrelaxNgParserCtxt_Object *)(v))->obj))
+
+typedef struct {
+    PyObject_HEAD
+    xmlRelaxNGValidCtxtPtr obj;
+} PyrelaxNgValidCtxt_Object;
+
+#define PyrelaxNgValidCtxt_Get(v) (((v) == Py_None) ? NULL : \
+	(((PyrelaxNgValidCtxt_Object *)(v))->obj))
+
+#endif /* LIBXML_RELAXNG_ENABLED */
+
+#ifdef LIBXML_SCHEMAS_ENABLED
+typedef struct {
+	PyObject_HEAD
+	xmlSchemaPtr obj;
+} PySchema_Object;
+
+#define PySchema_Get(v) (((v) == Py_None) ? NULL : \
+	(((PySchema_Object *)(v))->obj))
+
+typedef struct {
+	PyObject_HEAD
+	xmlSchemaParserCtxtPtr obj;
+} PySchemaParserCtxt_Object;
+
+#define PySchemaParserCtxt_Get(v) (((v) == Py_None) ? NULL : \
+	(((PySchemaParserCtxt_Object *)(v))->obj))
+
+typedef struct {
+	PyObject_HEAD
+	xmlSchemaValidCtxtPtr obj;
+} PySchemaValidCtxt_Object;
+
+#define PySchemaValidCtxt_Get(v) (((v) == Py_None) ? NULL : \
+	(((PySchemaValidCtxt_Object *)(v))->obj))
+
+#endif /* LIBXML_SCHEMAS_ENABLED */
+
+PyObject * libxml_intWrap(int val);
+PyObject * libxml_longWrap(long val);
+PyObject * libxml_xmlCharPtrWrap(xmlChar *str);
+PyObject * libxml_constxmlCharPtrWrap(const xmlChar *str);
+PyObject * libxml_charPtrWrap(char *str);
+PyObject * libxml_constcharPtrWrap(const char *str);
+PyObject * libxml_charPtrConstWrap(const char *str);
+PyObject * libxml_xmlCharPtrConstWrap(const xmlChar *str);
+PyObject * libxml_xmlDocPtrWrap(xmlDocPtr doc);
+PyObject * libxml_xmlNodePtrWrap(xmlNodePtr node);
+PyObject * libxml_xmlAttrPtrWrap(xmlAttrPtr attr);
+PyObject * libxml_xmlNsPtrWrap(xmlNsPtr ns);
+PyObject * libxml_xmlAttributePtrWrap(xmlAttributePtr ns);
+PyObject * libxml_xmlElementPtrWrap(xmlElementPtr ns);
+PyObject * libxml_doubleWrap(double val);
+PyObject * libxml_xmlParserCtxtPtrWrap(xmlParserCtxtPtr ctxt);
+#ifdef LIBXML_XPATH_ENABLED
+PyObject * libxml_xmlXPathContextPtrWrap(xmlXPathContextPtr ctxt);
+PyObject * libxml_xmlXPathParserContextPtrWrap(xmlXPathParserContextPtr ctxt);
+PyObject * libxml_xmlXPathObjectPtrWrap(xmlXPathObjectPtr obj);
+xmlXPathObjectPtr libxml_xmlXPathObjectPtrConvert(PyObject * obj);
+#endif
+PyObject * libxml_xmlValidCtxtPtrWrap(xmlValidCtxtPtr valid);
+#ifdef LIBXML_CATALOG_ENABLED
+PyObject * libxml_xmlCatalogPtrWrap(xmlCatalogPtr obj);
+#endif
+PyObject * libxml_xmlURIPtrWrap(xmlURIPtr uri);
+PyObject * libxml_xmlOutputBufferPtrWrap(xmlOutputBufferPtr buffer);
+PyObject * libxml_xmlParserInputBufferPtrWrap(xmlParserInputBufferPtr buffer);
+#ifdef LIBXML_REGEXP_ENABLED
+PyObject * libxml_xmlRegexpPtrWrap(xmlRegexpPtr regexp);
+#endif /* LIBXML_REGEXP_ENABLED */
+#ifdef LIBXML_READER_ENABLED
+PyObject * libxml_xmlTextReaderPtrWrap(xmlTextReaderPtr reader);
+PyObject * libxml_xmlTextReaderLocatorPtrWrap(xmlTextReaderLocatorPtr locator);
+#endif
+
+#ifdef LIBXML_RELAXNG_ENABLED
+PyObject * libxml_xmlRelaxNGPtrWrap(xmlRelaxNGPtr ctxt);
+PyObject * libxml_xmlRelaxNGParserCtxtPtrWrap(xmlRelaxNGParserCtxtPtr ctxt);
+PyObject * libxml_xmlRelaxNGValidCtxtPtrWrap(xmlRelaxNGValidCtxtPtr valid);
+#endif /* LIBXML_RELAXNG_ENABLED */
+#ifdef LIBXML_SCHEMAS_ENABLED
+PyObject * libxml_xmlSchemaPtrWrap(xmlSchemaPtr ctxt);
+PyObject * libxml_xmlSchemaParserCtxtPtrWrap(xmlSchemaParserCtxtPtr ctxt);
+PyObject * libxml_xmlSchemaValidCtxtPtrWrap(xmlSchemaValidCtxtPtr valid);
+#endif /* LIBXML_SCHEMAS_ENABLED */
+PyObject * libxml_xmlErrorPtrWrap(const xmlError *error);
+PyObject * libxml_xmlSchemaSetValidErrors(PyObject * self, PyObject * args);
+PyObject * libxml_xmlRegisterInputCallback(PyObject *self, PyObject *args);
+PyObject * libxml_xmlUnregisterInputCallback(PyObject *self, PyObject *args);
+PyObject * libxml_xmlNodeRemoveNsDef(PyObject * self, PyObject * args);
+
+int libxml_deprecationWarning(const char *func);

libxml2/python/meson.build

@@ -0,0 +1,113 @@
+
+pymod = import('python')
+py = pymod.find_installation('python3', required: false)
+
+if py.found() == true
+    message('Installing python modules to', py.get_install_dir())
+
+    pygenerated = custom_target(
+        'Python generated files',
+        input: doxygen_docs[1],
+        output: [
+            'libxml2-py.h',
+            'libxml2-export.c',
+            'libxml2-py.c',
+            'libxml2.py',
+        ],
+        command: [ py, files('generator.py'), meson.current_build_dir() ],
+        install: true,
+        install_dir: [ false, false, false, py.get_install_dir() ],
+    )
+    pygenerated_dep = declare_dependency(
+        sources : [pygenerated[0]],
+    )
+
+    libxml2mod_src = [
+        files('libxml.c', 'types.c'),
+        pygenerated[2],
+    ]
+
+    py.extension_module(
+        'libxml2mod',
+        libxml2mod_src,
+        dependencies: [
+            py.dependency(),
+            xml_dep,
+            pygenerated_dep,
+        ],
+        include_directories: [config_dir, '.'],
+        install: true,
+    )
+
+    py.install_sources(files('drv_libxml2.py'))
+
+    setup_py = configuration_data()
+    setup_py.set('prefix', get_option('prefix'))
+    setup_py.set('LIBXML_VERSION', meson.project_version())
+    setup_py.set('WITH_ICONV', want_iconv.to_int())
+    setup_py.set('WITH_ICU', want_icu.to_int())
+    setup_py.set('WITH_ZLIB', want_zlib.to_int())
+    setup_py.set('WITH_THREADS', want_threads.to_int())
+    configure_file(
+        input: 'setup.py.in',
+        output: 'setup.py',
+        configuration: setup_py,
+    )
+
+    python_tests = [
+        'attribs.py',
+        'build.py',
+        'compareNodes.py',
+        'ctxterror.py',
+        'cutnpaste.py',
+        'dtdvalid.py',
+        'error.py',
+        'inbuf.py',
+        'indexes.py',
+        'input_callback.py',
+        'nsdel.py',
+        'outbuf.py',
+        'push.py',
+        'pushSAX.py',
+        'pushSAXhtml.py',
+        'reader.py',
+        'reader2.py',
+        'reader3.py',
+        'reader4.py',
+        'reader5.py',
+        'reader6.py',
+        'reader7.py',
+        'reader8.py',
+        'readererr.py',
+        'readernext.py',
+        'regexp.py',
+        'relaxng.py',
+        'resolver.py',
+        'schema.py',
+        'serialize.py',
+        'sync.py',
+        'thread2.py',
+        'tst.py',
+        'tstLastError.py',
+        'tstURI.py',
+        'tstmem.py',
+        'tstxpath.py',
+        'unicode.py',
+        'validDTD.py',
+        'validRNG.py',
+        'validSchemas.py',
+        'validate.py',
+        'walker.py',
+        'xpath.py',
+        'xpathext.py',
+        'xpathleak.py',
+        'xpathns.py',
+        'xpathret.py',
+    ]
+
+    foreach file : python_tests
+        test(file, py, args: [ file ],
+             workdir: meson.current_source_dir() / 'tests',
+             env: { 'PYTHONPATH': meson.current_build_dir() })
+    endforeach
+endif

libxml2/python/pyproject.toml

@@ -0,0 +1,3 @@
+[build-system]
+requires = ["setuptools"]
+build-backend = "setuptools.build_meta"

libxml2/python/setup.py.in

@@ -0,0 +1,255 @@
+#!/usr/bin/env python3
+#
+# Setup script for libxml2 and libxslt if found
+#
+import sys, os
+
+try:
+    from setuptools import setup, Extension
+except ImportError:
+    try:
+        # Using distutils, for python < 3.12
+        from distutils.core import setup, Extension
+    except ImportError:
+        # distutils is not present in python 3.12 and greater
+        print("setuptools is required for python >= 3.12")
+        sys.exit(1)
+
+# Below ROOT, we expect to find include, include/libxml2, lib and bin.
+# On *nix, it is not needed (but should not harm),
+# on Windows, it is set by configure.js.
+ROOT = r'@prefix@'
+
+# Thread-enabled libxml2
+with_threads = @WITH_THREADS@
+
+# Features of libxml2 requiring external DLLs
+with_iconv = @WITH_ICONV@
+with_zlib = @WITH_ZLIB@
+with_icu = @WITH_ICU@
+
+icu_series = 69
+
+if icu_series is not None:
+    icu_series_s = str(icu_series)
+else:
+    icu_series_s = ''
+
+# If bundling DLLs, check the following to ensure things are correct
+# (Check the value of `icu_series` above as well)
+iconv_dll = 'iconv.dll'
+zlib_dll = 'zlib1.dll'
+icu_dlls = ['icuuc%s.dll' % icu_series_s, 'icudt%s.dll' % icu_series_s]
+
+# If this flag is set (windows only),
+# a private copy of the dlls are included in the package.
+# If this flag is not set, the libxml2 and libxslt
+# dlls must be found somewhere in the PATH at runtime.
+WITHDLLS = 1 and sys.platform.startswith('win')
+
+if WITHDLLS:
+    def altImport(s):
+        s = s.replace("import libxml2mod","from libxmlmods import libxml2mod")
+        return s
+
+def missing(file):
+    if os.access(file, os.R_OK) == 0:
+        return 1
+    return 0
+
+try:
+    HOME = os.environ['HOME']
+except:
+    HOME="C:"
+
+if sys.platform.startswith('win'):
+    libraryPrefix = 'lib'
+    platformLibs = []
+else:
+    libraryPrefix = ''
+    platformLibs = ["m","z"]
+
+# those are examined to find
+# - libxml2/libxml/tree.h
+# - libxslt/xsltconfig.h
+includes_dir = [
+"/usr/include",
+"/usr/local/include",
+"/opt/include",
+os.path.join(ROOT,'include'),
+HOME
+];
+
+xml_includes=""
+for dir in includes_dir:
+    if not missing(dir + "/libxml2/libxml/tree.h"):
+        xml_includes=dir + "/libxml2"
+        break;
+
+if xml_includes == "":
+    print("failed to find headers for libxml2: update includes_dir")
+    sys.exit(1)
+
+# those are added in the linker search path for libraries
+libdirs = [
+os.path.join(ROOT,'lib'),
+]
+
+xml_files = ["libxml.c", "libxml.py", "libxml_wrap.h", "types.c",
+             "xmlgenerator.py", "README", "TODO", "drv_libxml2.py"]
+
+xslt_files = ["libxslt-api.xml", "libxslt-python-api.xml",
+             "libxslt.c", "libxsl.py", "libxslt_wrap.h",
+             "xsltgenerator.py"]
+
+if missing("libxml2-py.c") or missing("libxml2.py"):
+    try:
+        try:
+            import xmlgenerator
+        except:
+            import generator
+    except:
+        print("failed to find and generate stubs for libxml2, aborting ...")
+        print(sys.exc_info()[0], sys.exc_info()[1])
+        sys.exit(1)
+
+with_xslt=0
+if missing("libxslt-py.c") or missing("libxslt.py"):
+    if missing("xsltgenerator.py") or missing("libxslt-api.xml"):
+        print("libxslt stub generator not found, libxslt not built")
+    else:
+        try:
+            import xsltgenerator
+        except:
+            print("failed to generate stubs for libxslt, aborting ...")
+            print(sys.exc_info()[0], sys.exc_info()[1])
+        else:
+            head = open("libxsl.py", "r")
+            generated = open("libxsltclass.py", "r")
+            result = open("libxslt.py", "w")
+            for line in head.readlines():
+                if WITHDLLS:
+                    result.write(altImport(line))
+                else:
+                    result.write(line)
+            for line in generated.readlines():
+                result.write(line)
+            head.close()
+            generated.close()
+            result.close()
+            with_xslt=1
+else:
+    with_xslt=1
+
+if with_xslt == 1:
+    xslt_includes=""
+    for dir in includes_dir:
+        if not missing(dir + "/libxslt/xsltconfig.h"):
+            xslt_includes=dir + "/libxslt"
+            break;
+
+    if xslt_includes == "":
+        print("failed to find headers for libxslt: update includes_dir")
+        with_xslt = 0
+
+if WITHDLLS:
+    # libxml dlls (expected in ROOT/bin)
+    dlls = [ 'libxml2.dll' ]
+
+    if with_zlib == 1:
+        dlls.append(zlib_dll)
+    if with_iconv == 1:
+        dlls.append(iconv_dll)
+    if with_icu == 1:
+        dlls += icu_dlls
+    if with_xslt == 1:
+        dlls += ['libxslt.dll','libexslt.dll']
+
+    packaged_dlls = [os.path.join(ROOT,'bin',dll) for dll in dlls]
+
+    # create __init__.py for the libxmlmods package
+    if not os.path.exists("libxmlmods"):
+        os.mkdir("libxmlmods")
+        open("libxmlmods/__init__.py","w").close()
+
+    packaged_dlls = [os.path.join(ROOT,'bin',dll) for dll in dlls]
+
+descr = "libxml2 package"
+modules = [ 'libxml2', 'drv_libxml2' ]
+if WITHDLLS:
+    modules.append('libxmlmods.__init__')
+c_files = ['libxml2-py.c', 'libxml.c', 'types.c' ]
+includes= [xml_includes]
+libs    = [libraryPrefix + "xml2"] + platformLibs
+macros  = []
+if with_threads:
+    macros.append(('_REENTRANT','1'))
+if with_xslt == 1:
+    descr = "libxml2 and libxslt package"
+    if not sys.platform.startswith('win'):
+        #
+        # We are gonna build 2 identical shared libs with merge initializing
+        # both libxml2mod and libxsltmod
+        #
+        c_files = c_files + ['libxslt-py.c', 'libxslt.c']
+        xslt_c_files = c_files
+        macros.append(('MERGED_MODULES', '1'))
+    else:
+        #
+        # On windows the MERGED_MODULE option is not needed
+        # (and does not work)
+        #
+        xslt_c_files = ['libxslt-py.c', 'libxslt.c', 'types.c']
+    libs.insert(0, libraryPrefix + 'exslt')
+    libs.insert(0, libraryPrefix + 'xslt')
+    includes.append(xslt_includes)
+    modules.append('libxslt')
+
+
+extens=[Extension('libxml2mod', c_files, include_dirs=includes,
+                  library_dirs=libdirs,
+                  libraries=libs, define_macros=macros)]
+if with_xslt == 1:
+    extens.append(Extension('libxsltmod', xslt_c_files, include_dirs=includes,
+                            library_dirs=libdirs,
+                            libraries=libs, define_macros=macros))
+
+if missing("MANIFEST"):
+
+    manifest = open("MANIFEST", "w")
+    manifest.write("setup.py\n")
+    for file in xml_files:
+        manifest.write(file + "\n")
+    if with_xslt == 1:
+        for file in xslt_files:
+            manifest.write(file + "\n")
+    manifest.close()
+
+if WITHDLLS:
+    ext_package = "libxmlmods"
+    if sys.version >= "2.2":
+        base = "lib/site-packages/"
+    else:
+        base = ""
+    data_files = [(base+"libxmlmods",packaged_dlls)]
+else:
+    ext_package = None
+    data_files = []
+
+setup (name = "libxml2-python",
+       # On *nix, the version number is created from setup.py.in
+       # On windows, it is set by configure.js
+       version = "@LIBXML_VERSION@",
+       description = descr,
+       author = "Daniel Veillard",
+       author_email = "veillard@redhat.com",
+       url = "https://gitlab.gnome.org/GNOME/libxml2",
+       license="MIT License",
+       py_modules=modules,
+       ext_modules=extens,
+       ext_package=ext_package,
+       data_files=data_files,
+       )
+
+sys.exit(0)
+

libxml2/python/tests/Makefile.am

@@ -0,0 +1,77 @@
+PYTESTS=	\
+    build.py	\
+    attribs.py	\
+    tst.py	\
+    tstxpath.py	\
+    xpathext.py	\
+    push.py	\
+    pushSAX.py	\
+    pushSAXhtml.py \
+    error.py	\
+    serialize.py\
+    validate.py	\
+    tstURI.py	\
+    cutnpaste.py\
+    xpathret.py	\
+    xpath.py	\
+    outbuf.py	\
+    inbuf.py	\
+    input_callback.py \
+    resolver.py \
+    regexp.py	\
+    reader.py	\
+    reader2.py	\
+    reader3.py	\
+    reader4.py	\
+    reader5.py	\
+    reader6.py	\
+    reader7.py	\
+    reader8.py	\
+    readernext.py	\
+    walker.py	\
+    nsdel.py	\
+    ctxterror.py\
+    readererr.py\
+    relaxng.py	\
+    schema.py	\
+    thread2.py \
+    sync.py \
+    tstLastError.py \
+    indexes.py \
+    dtdvalid.py \
+    tstmem.py	\
+    validDTD.py	\
+    validSchemas.py \
+    validRNG.py \
+    compareNodes.py \
+    xpathns.py \
+    xpathleak.py \
+    unicode.py
+
+XMLS=		\
+    tst.xml	\
+    valid.xml	\
+    invalid.xml	\
+    test.dtd
+
+EXTRA_DIST = $(PYTESTS) $(XMLS) setup_test.py
+
+CLEANFILES = core tmp.xml *.pyc
+
+if WITH_PYTHON
+check-local:
+	@for f in $(XMLS) ; do test -f $$f || $(LN_S) $(srcdir)/$$f . ; done
+	@echo "## running Python regression tests"
+	@(export PYTHONPATH="..:../.libs:$(srcdir)/..:$$PYTHONPATH" ; \
+	  export LD_LIBRARY_PATH="$(top_builddir)/.libs:$$LD_LIBRARY_PATH" ; \
+	  export DYLD_LIBRARY_PATH="$(top_builddir)/.libs:$$DYLD_LIBRARY_PATH" ; \
+	  export PATH="$(top_builddir)/.libs:$$PATH" ; \
+	  for test in $(PYTESTS) ; do \
+	    log=`$(PYTHON) $(srcdir)/$$test` ; \
+	    if [ "$$?" -ne 0 ] ; then \
+	      echo "-- $$test" ; \
+	      echo "$$log" ; \
+	      exit 1 ; \
+	    fi ; \
+	  done)
+endif

libxml2/python/tests/attribs.py

@@ -0,0 +1,35 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+#
+# Testing XML document serialization
+#
+doc = libxml2.readDoc(
+"""<?xml version="1.0" encoding="iso-8859-1"?>
+<!DOCTYPE test [
+<!ELEMENT test (#PCDATA) >
+<!ATTLIST test xmlns:abc CDATA #FIXED "http://abc.org" >
+<!ATTLIST test abc:attr CDATA #FIXED "def" >
+]>
+<test />
+""", None, None, libxml2.XML_PARSE_DTDATTR)
+elem = doc.getRootElement()
+attr = elem.hasNsProp('attr', 'http://abc.org')
+print(attr.serialize())
+if attr == None:
+    print("Failed to find defaulted attribute abc:attr")
+    sys.exit(1)
+
+doc.freeDoc()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/build.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+doc = libxml2.newDoc("1.0")
+comment = doc.newDocComment("This is a generated document")
+doc.addChild(comment)
+pi = libxml2.newPI("test", "PI content")
+doc.addChild(pi)
+root = doc.newChild(None, "doc", None)
+ns = root.newNs("http://example.com/doc", "my")
+root.setNs(ns)
+elem = root.newChild(None, "foo", "bar")
+elem.setBase("http://example.com/imgs")
+elem.setProp("img", "image.gif")
+doc.saveFile("tmp.xml")
+doc.freeDoc()
+
+doc = libxml2.parseFile("tmp.xml")
+comment = doc.children
+if comment.type != "comment" or \
+   comment.content != "This is a generated document":
+   print("error rereading comment")
+   sys.exit(1)
+pi = comment.next
+if pi.type != "pi" or pi.name != "test" or pi.content != "PI content":
+   print("error rereading PI")
+   sys.exit(1)
+root = pi.next
+if root.name != "doc":
+   print("error rereading root")
+   sys.exit(1)
+ns = root.ns()
+if ns.name != "my" or ns.content != "http://example.com/doc":
+   print("error rereading namespace")
+   sys.exit(1)
+elem = root.children
+if elem.name != "foo":
+   print("error rereading elem")
+   sys.exit(1)
+if elem.getBase(None) != "http://example.com/imgs":
+   print("error rereading base")
+   sys.exit(1)
+if elem.prop("img") != "image.gif":
+   print("error rereading property")
+   sys.exit(1)
+
+doc.freeDoc()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/compareNodes.py

@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+#
+# Testing XML Node comparison and Node hash-value
+#
+doc = libxml2.parseDoc("""<root><foo/></root>""")
+root = doc.getRootElement()
+
+# Create two different objects which point to foo
+foonode1 = root.children
+foonode2 = root.children
+
+# Now check that [in]equality tests work ok
+if not ( foonode1 == foonode2 ):
+    print("Error comparing nodes with ==, nodes should be equal but are unequal")
+    sys.exit(1)
+if not ( foonode1 != root ):
+    print("Error comparing nodes with ==, nodes should not be equal but are equal")
+    sys.exit(1)
+if not ( foonode1 != root ):
+    print("Error comparing nodes with !=, nodes should not be equal but are equal")
+if ( foonode1 != foonode2 ):
+    print("Error comparing nodes with !=, nodes should be equal but are unequal")
+
+# Next check that the hash function for the objects also works ok
+if not (hash(foonode1) == hash(foonode2)):
+    print("Error hash values for two equal nodes are different")
+    sys.exit(1)
+if not (hash(foonode1) != hash(root)):
+    print("Error hash values for two unequal nodes are not different")
+    sys.exit(1)
+if hash(foonode1) == hash(root):
+    print("Error hash values for two unequal nodes are equal")
+    sys.exit(1)
+
+# Basic tests successful
+doc.freeDoc()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/ctxterror.py

@@ -0,0 +1,56 @@
+#!/usr/bin/env python3
+#
+# This test exercise the redirection of error messages with a
+# functions defined in Python.
+#
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+expect="""--> (3) xmlns: URI foo is not absolute
+--> (4) Opening and ending tag mismatch: x line 1 and y
+"""
+
+err=""
+def callback(arg,msg,severity,reserved):
+    global err
+    err = err + "%s (%d) %s" % (arg,severity,msg)
+
+s = """<x xmlns="foo"></y>"""
+
+parserCtxt = libxml2.createPushParser(None,"",0,"test.xml")
+parserCtxt.setErrorHandler(callback, "-->")
+if parserCtxt.getErrorHandler() != (callback,"-->"):
+    print("getErrorHandler failed")
+    sys.exit(1)
+parserCtxt.parseChunk(s,len(s),1)
+doc = parserCtxt.doc()
+doc.freeDoc()
+parserCtxt = None
+
+if err != expect:
+    print("error")
+    print("received %s" %(err))
+    print("expected %s" %(expect))
+    sys.exit(1)
+
+i = 10000
+while i > 0:
+    parserCtxt = libxml2.createPushParser(None,"",0,"test.xml")
+    parserCtxt.setErrorHandler(callback, "-->")
+    parserCtxt.parseChunk(s,len(s),1)
+    doc = parserCtxt.doc()
+    doc.freeDoc()
+    parserCtxt = None
+    err = ""
+    i = i - 1
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/cutnpaste.py

@@ -0,0 +1,48 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+#
+# Testing XML document serialization
+#
+source = libxml2.parseDoc("""<?xml version="1.0"?>
+<root xmlns:foo="http://example.org/foo"
+      xmlns:bar="http://example.org/bar">
+<include xmlns="http://example.org/include">
+<fragment><foo:elem bar="tricky"/></fragment>
+</include>
+</root>
+""")
+
+target = libxml2.parseDoc("""<?xml version="1.0"?>
+<root xmlns:foobar="http://example.org/bar"/>""")
+
+fragment = source.xpathEval("//*[name()='fragment']")[0]
+dest = target.getRootElement()
+
+# do a cut and paste operation
+fragment.unlinkNode()
+dest.addChild(fragment)
+# do the namespace fixup
+dest.reconciliateNs(target)
+
+# The source tree can be freed at that point
+source.freeDoc()
+
+# check the resulting tree
+str = dest.serialize()
+if str != """<root xmlns:foobar="http://example.org/bar" xmlns:default="http://example.org/include" xmlns:foo="http://example.org/foo"><default:fragment><foo:elem bar="tricky"/></default:fragment></root>""":
+    print("reconciliateNs() failed")
+    sys.exit(1)
+target.freeDoc()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/dtdvalid.py

@@ -0,0 +1,32 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+dtd="""<!ELEMENT foo EMPTY>"""
+instance="""<?xml version="1.0"?>
+<foo></foo>"""
+
+dtd = libxml2.parseDTD(None, 'test.dtd')
+ctxt = libxml2.newValidCtxt()
+doc = libxml2.parseDoc(instance)
+ret = doc.validateDtd(ctxt, dtd)
+if ret != 1:
+    print("error doing DTD validation")
+    sys.exit(1)
+
+doc.freeDoc()
+dtd.freeDtd()
+del dtd
+del ctxt
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/error.py

@@ -0,0 +1,51 @@
+#!/usr/bin/env python3
+#
+# This test exercise the redirection of error messages with a
+# functions defined in Python.
+#
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+expect='--> I/O --> warning : --> failed to load "missing.xml": No such file or directory\n'
+err=""
+def callback(ctx, str):
+     global err
+
+     err = err + "%s %s" % (ctx, str)
+
+got_exc = 0
+libxml2.registerErrorHandler(callback, "-->")
+try:
+    doc = libxml2.parseFile("missing.xml")
+except libxml2.parserError:
+    got_exc = 1
+
+if got_exc == 0:
+    print("Failed to get a parser exception")
+    sys.exit(1)
+
+if err != expect:
+    print("error")
+    print("received %s" %(err))
+    print("expected %s" %(expect))
+    sys.exit(1)
+
+i = 10000
+while i > 0:
+    try:
+        doc = libxml2.parseFile("missing.xml")
+    except libxml2.parserError:
+        got_exc = 1
+    err = ""
+    i = i - 1
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/inbuf.py

@@ -0,0 +1,30 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+i = 0
+while i < 5000:
+    f = str_io("foobar")
+    buf = libxml2.inputBuffer(f)
+    i = i + 1
+
+del f
+del buf
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/indexes.py

@@ -0,0 +1,113 @@
+#!/usr/bin/env python3
+# -*- coding: ISO-8859-1 -*-
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+ctxt = None
+
+class callback:
+    def __init__(self, startd, starte, ende, delta, endd):
+        self.startd = startd
+        self.starte = starte
+        self.ende = ende
+        self.endd = endd
+        self.delta = delta
+        self.count = 0
+
+    def startDocument(self):
+        global ctxt
+        if ctxt.byteConsumed() != self.startd:
+            print("document start at wrong index: %d expecting %d\n" % (
+                  ctxt.byteConsumed(), self.startd))
+            sys.exit(1)
+
+    def endDocument(self):
+        global ctxt
+        expect = self.ende + self.delta * (self.count - 1) + self.endd
+        if ctxt.byteConsumed() != expect:
+            print("document end at wrong index: %d expecting %d\n" % (
+                  ctxt.byteConsumed(), expect))
+            sys.exit(1)
+
+    def startElement(self, tag, attrs):
+        global ctxt
+        if tag == "bar1":
+            expect = self.starte + self.delta * self.count
+            if ctxt.byteConsumed() != expect:
+                print("element start at wrong index: %d expecting %d\n" % (
+                   ctxt.byteConsumed(), expect))
+                sys.exit(1)
+            
+
+    def endElement(self, tag):
+        global ctxt
+        if tag == "bar1":
+            expect = self.ende + self.delta * self.count
+            if ctxt.byteConsumed() != expect:
+                print("element end at wrong index: %d expecting %d\n" % (
+                      ctxt.byteConsumed(), expect))
+                sys.exit(1)
+            self.count = self.count + 1
+
+    def characters(self, data):
+        pass
+
+#
+# First run a pure UTF-8 test
+#
+handler = callback(0, 13, 27, 198, 183)
+ctxt = libxml2.createPushParser(handler, "<foo>\n", 6, "test.xml")
+chunk = """  <bar1>chars1</bar1>
+  <bar2>chars2</bar2>
+  <bar3>chars3</bar3>
+  <bar4>chars4</bar4>
+  <bar5>chars5</bar5>
+  <bar6>&lt;s6</bar6>
+  <bar7>chars7</bar7>
+  <bar8>&#38;8</bar8>
+  <bar9>chars9</bar9>
+"""
+i = 0
+while i < 10000:
+    ctxt.parseChunk(chunk, len(chunk), 0)
+    i = i + 1
+chunk = "</foo>"
+ctxt.parseChunk(chunk, len(chunk), 1)
+ctxt=None
+
+#
+# Then run a test relying on ISO-Latin-1
+#
+handler = callback(43, 57, 71, 198, 183)
+chunk="""<?xml version="1.0" encoding="ISO-8859-1"?>
+<foo>
+"""
+ctxt = libxml2.createPushParser(handler, chunk, len(chunk), "test.xml")
+chunk = """  <bar1>chars1</bar1>
+  <bar2>chars2</bar2>
+  <bar3>chars3</bar3>
+  <bar4>chàrs4</bar4>
+  <bar5>chars5</bar5>
+  <bar6>&lt;s6</bar6>
+  <bar7>chars7</bar7>
+  <bar8>&#38;8</bar8>
+  <bar9>très 9</bar9>
+"""
+i = 0
+while i < 10000:
+    ctxt.parseChunk(chunk, len(chunk), 0)
+    i = i + 1
+chunk = "</foo>"
+ctxt.parseChunk(chunk, len(chunk), 1)
+ctxt=None
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/input_callback.py

@@ -0,0 +1,149 @@
+#!/usr/bin/env python3
+#
+# This tests custom input callbacks
+#
+import sys
+import setup_test
+import libxml2
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# We implement a new scheme, py://strings/ that will reference this dictionary
+pystrings = {
+    'catalogs/catalog.xml' :
+'''<?xml version="1.0" encoding="utf-8"?>
+<!DOCTYPE catalog PUBLIC "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd">
+<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog">
+  <rewriteSystem systemIdStartString="http://example.com/dtds/" rewritePrefix="../dtds/"/>
+</catalog>''',
+
+    'xml/sample.xml' :
+'''<?xml version="1.0" encoding="utf-8"?>
+<!DOCTYPE root SYSTEM "http://example.com/dtds/sample.dtd">
+<root>&sample.entity;</root>''',
+
+    'dtds/sample.dtd' :
+'''
+<!ELEMENT root (#PCDATA)>
+<!ENTITY sample.entity "replacement text">'''
+}
+
+prefix = "py://strings/"
+startURL = prefix + "xml/sample.xml"
+catURL = prefix + "catalogs/catalog.xml"
+
+def my_input_cb(URI):
+    if not(URI.startswith(prefix)):
+        return None
+    path = URI[len(prefix):]
+    if path not in pystrings:
+        return None
+    return str_io(pystrings[path])
+
+
+def run_test(desc, docpath, catalog, exp_status="verified", exp_err=[], test_callback=None,
+        root_name="root", root_content="replacement text"):
+    opts = libxml2.XML_PARSE_DTDLOAD | libxml2.XML_PARSE_NONET | libxml2.XML_PARSE_COMPACT
+    actual_err = []
+
+    def my_global_error_cb(ctx, msg):
+        actual_err.append((-1, msg))
+    def my_ctx_error_cb(arg, msg, severity, reserved):
+        actual_err.append((severity, msg))
+
+    libxml2.registerErrorHandler(my_global_error_cb, None)
+    try:
+        parser = libxml2.createURLParserCtxt(docpath, opts)
+        parser.setErrorHandler(my_ctx_error_cb, None)
+        if catalog is not None:
+            parser.addLocalCatalog(catalog)
+        if test_callback is not None:
+            test_callback()
+        parser.parseDocument()
+        doc = parser.doc()
+        actual_status = "loaded"
+        e = doc.getRootElement()
+        if e.name == root_name and e.content == root_content:
+            actual_status = "verified"
+        doc.freeDoc()
+    except libxml2.parserError:
+        actual_status = "not loaded"
+
+    if actual_status != exp_status:
+        print("Test '%s' failed: expect status '%s', actual '%s'" % (desc, exp_status, actual_status))
+        sys.exit(1)
+    elif actual_err != exp_err:
+        print("Test '%s' failed" % desc)
+        print("Expect errors:")
+        for s,m in exp_err: print("  [%2d] '%s'" % (s,m))
+        print("Actual errors:")
+        for s,m in actual_err: print("  [%2d] '%s'" % (s,m))
+        sys.exit(1)
+
+
+# Check that we cannot read custom schema without custom callback
+run_test(desc="Loading entity without custom callback",
+        docpath=startURL, catalog=None,
+        exp_status="not loaded", exp_err=[
+            (-1, "I/O "),
+            (-1, "warning : "),
+            (-1, "failed to load \"py://strings/xml/sample.xml\": No such file or directory\n")
+            ])
+
+# Register handler and try to load the same entity
+libxml2.registerInputCallback(my_input_cb)
+run_test(desc="Loading entity with custom callback",
+        docpath=startURL, catalog=None,
+        exp_status="loaded", exp_err=[
+            (  4, 'failed to load "http://example.com/dtds/sample.dtd": Attempt to load network entity\n'),
+            (  4, "Entity 'sample.entity' not defined\n")
+            ])
+
+# Register a catalog (also accessible via pystr://) and retry
+run_test(desc="Loading entity with custom callback and catalog",
+        docpath=startURL, catalog=catURL)
+
+# Unregister custom callback when parser is already created
+run_test(desc="Loading entity and unregistering callback",
+        docpath=startURL, catalog=catURL,
+        test_callback=lambda: libxml2.popInputCallbacks(),
+        exp_status="loaded", exp_err=[
+            ( 3, "failed to load \"py://strings/dtds/sample.dtd\": No such file or directory\n"),
+            ( 4, "Entity 'sample.entity' not defined\n")
+            ])
+
+# Try to load the document again
+run_test(desc="Retry loading document after unregistering callback",
+        docpath=startURL, catalog=catURL,
+        exp_status="not loaded", exp_err=[
+            (-1, "I/O "),
+            (-1, "warning : "),
+            (-1, "failed to load \"py://strings/xml/sample.xml\": No such file or directory\n")
+            ])
+
+# But should be able to read standard I/O yet...
+run_test(desc="Loading using standard i/o after unregistering callback",
+        docpath="tst.xml", catalog=None,
+        root_name='doc', root_content='bar')
+
+# Now pop ALL input callbacks, should fail to load even standard I/O
+try:
+    while True:
+        libxml2.popInputCallbacks()
+except IndexError:
+    pass
+
+run_test(desc="Loading using standard i/o after unregistering all callbacks",
+        docpath="tst.xml", catalog=None,
+        exp_status="not loaded", exp_err=[
+            (-1, "I/O "),
+            (-1, "warning : "),
+            (-1, "failed to load \"tst.xml\": No such file or directory\n")
+            ])
+
+print("OK")
+sys.exit(0);

libxml2/python/tests/invalid.xml

@@ -0,0 +1,6 @@
+<!DOCTYPE doc [
+<!ELEMENT doc (a, b, a)>
+<!ELEMENT a EMPTY>
+<!ELEMENT b EMPTY>
+]>
+<doc><b/><a/><b/></doc>

libxml2/python/tests/nsdel.py

@@ -0,0 +1,62 @@
+#!/usr/bin/env python3
+#
+# this test exercise the XPath basic engine, parser, etc, and
+# allows to detect memory leaks
+#
+import sys
+import setup_test
+import libxml2
+
+instance="""<?xml version="1.0"?>
+<tag xmlns:foo='urn:foo' xmlns:bar='urn:bar' xmlns:baz='urn:baz' />"""
+
+def namespaceDefs(node):
+    n = node.nsDefs()
+    while n:
+        yield n
+        n = n.next
+
+def checkNamespaceDefs(node, count):
+    nsList = list(namespaceDefs(node))
+    #print nsList
+    if len(nsList) != count :
+        raise Exception("Error: saw %d namespace declarations.  Expected %d" % (len(nsList), count))
+    
+# Memory debug specific
+libxml2.debugMemory(1)
+
+# Remove single namespace
+doc = libxml2.parseDoc(instance)
+node = doc.getRootElement()
+checkNamespaceDefs(node, 3)
+ns = node.removeNsDef('urn:bar')
+checkNamespaceDefs(node, 2)
+ns.freeNsList()
+doc.freeDoc()
+
+# Remove all namespaces
+doc = libxml2.parseDoc(instance)
+node = doc.getRootElement()
+checkNamespaceDefs(node, 3)
+ns = node.removeNsDef(None)
+checkNamespaceDefs(node, 0)
+ns.freeNsList()
+doc.freeDoc()
+
+# Remove a namespace referred to by a child
+doc = libxml2.newDoc("1.0")
+root = doc.newChild(None, "root", None)
+namespace = root.newNs("http://example.com/sample", "s")
+child = root.newChild(namespace, "child", None)
+root.removeNsDef("http://example.com/sample")
+doc.reconciliateNs(root)
+namespace.freeNsList()
+doc.serialize() # This should not segfault
+doc.freeDoc()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/outbuf.py

@@ -0,0 +1,110 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+def testSimpleBufferWrites():
+    f = str_io()
+    buf = libxml2.createOutputBuffer(f, "ISO-8859-1")
+    buf.write(3, "foo")
+    buf.writeString("bar")
+    buf.close()
+
+    if f.getvalue() != "foobar":
+        print("Failed to save to StringIO")
+        sys.exit(1)
+
+def testSaveDocToBuffer():
+    """
+    Regression test for bug #154294.
+    """
+    input = '<foo>Hello</foo>'
+    expected = '''\
+<?xml version="1.0" encoding="UTF-8"?>
+<foo>Hello</foo>
+'''
+    f = str_io()
+    buf = libxml2.createOutputBuffer(f, 'UTF-8')
+    doc = libxml2.parseDoc(input)
+    doc.saveFileTo(buf, 'UTF-8')
+    doc.freeDoc()
+    if f.getvalue() != expected:
+        print('xmlDoc.saveFileTo() call failed.')
+        print('     got: %s' % repr(f.getvalue()))
+        print('expected: %s' % repr(expected))
+        sys.exit(1)
+
+def testSaveFormattedDocToBuffer():
+    input = '<outer><inner>Some text</inner><inner/></outer>'
+    # The formatted and non-formatted versions of the output.
+    expected = ('''\
+<?xml version="1.0" encoding="UTF-8"?>
+<outer><inner>Some text</inner><inner/></outer>
+''', '''\
+<?xml version="1.0" encoding="UTF-8"?>
+<outer>
+  <inner>Some text</inner>
+  <inner/>
+</outer>
+''')
+    doc = libxml2.parseDoc(input)
+    for i in (0, 1):
+        f = str_io()
+        buf = libxml2.createOutputBuffer(f, 'UTF-8')
+        doc.saveFormatFileTo(buf, 'UTF-8', i)
+        if f.getvalue() != expected[i]:
+            print('xmlDoc.saveFormatFileTo() call failed.')
+            print('     got: %s' % repr(f.getvalue()))
+            print('expected: %s' % repr(expected[i]))
+            sys.exit(1)
+    doc.freeDoc()
+
+def testSaveIntoOutputBuffer():
+    """
+    Similar to the previous two tests, except this time we invoke the save
+    methods on the output buffer object and pass in an XML node object.
+    """
+    input = '<foo>Hello</foo>'
+    expected = '''\
+<?xml version="1.0" encoding="UTF-8"?>
+<foo>Hello</foo>
+'''
+    f = str_io()
+    doc = libxml2.parseDoc(input)
+    buf = libxml2.createOutputBuffer(f, 'UTF-8')
+    buf.saveFileTo(doc, 'UTF-8')
+    if f.getvalue() != expected:
+        print('outputBuffer.saveFileTo() call failed.')
+        print('     got: %s' % repr(f.getvalue()))
+        print('expected: %s' % repr(expected))
+        sys.exit(1)
+    f = str_io()
+    buf = libxml2.createOutputBuffer(f, 'UTF-8')
+    buf.saveFormatFileTo(doc, 'UTF-8', 1)
+    if f.getvalue() != expected:
+        print('outputBuffer.saveFormatFileTo() call failed.')
+        print('     got: %s' % repr(f.getvalue()))
+        print('expected: %s' % repr(expected))
+        sys.exit(1)
+    doc.freeDoc()
+
+if __name__ == '__main__':
+    # Memory debug specific
+    libxml2.debugMemory(1)
+
+    testSimpleBufferWrites()
+    testSaveDocToBuffer()
+    testSaveFormattedDocToBuffer()
+    testSaveIntoOutputBuffer()
+
+    libxml2.cleanupParser()
+    if libxml2.debugMemory(1) == 0:
+        print("OK")
+    else:
+        print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/push.py

@@ -0,0 +1,35 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+ctxt = libxml2.createPushParser(None, "<foo", 4, "test.xml")
+ctxt.parseChunk("/>", 2, 1)
+doc = ctxt.doc()
+ctxt=None
+if doc.name != "test.xml":
+    print("document name error")
+    sys.exit(1)
+root = doc.children
+if root.name != "foo":
+    print("root element name error")
+    sys.exit(1)
+doc.freeDoc()
+i = 10000
+while i > 0:
+    ctxt = libxml2.createPushParser(None, "<foo", 4, "test.xml")
+    ctxt.parseChunk("/>", 2, 1)
+    doc = ctxt.doc()
+    doc.freeDoc()
+    i = i -1
+ctxt=None
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/pushSAX.py

@@ -0,0 +1,64 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+log = ""
+
+class callback:
+    def startDocument(self):
+        global log
+        log = log + "startDocument:"
+
+    def endDocument(self):
+        global log
+        log = log + "endDocument:"
+
+    def startElement(self, tag, attrs):
+        global log
+        log = log + "startElement %s %s:" % (tag, attrs)
+
+    def endElement(self, tag):
+        global log
+        log = log + "endElement %s:" % (tag)
+
+    def characters(self, data):
+        global log
+        log = log + "characters: %s:" % (data)
+
+    def warning(self, msg):
+        global log
+        log = log + "warning: %s:" % (msg)
+
+    def error(self, msg):
+        global log
+        log = log + "error: %s:" % (msg)
+
+    def fatalError(self, msg):
+        global log
+        log = log + "fatalError: %s:" % (msg)
+
+handler = callback()
+
+ctxt = libxml2.createPushParser(handler, "<foo", 4, "test.xml")
+chunk = " url='tst'>b"
+ctxt.parseChunk(chunk, len(chunk), 0)
+chunk = "ar</foo>"
+ctxt.parseChunk(chunk, len(chunk), 1)
+ctxt=None
+
+reference = "startDocument:startElement foo {'url': 'tst'}:characters: bar:endElement foo:endDocument:"
+if log != reference:
+    print("Error got: %s" % log)
+    print("Exprected: %s" % reference)
+    sys.exit(1)
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/pushSAXhtml.py

@@ -0,0 +1,64 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+log = ""
+
+class callback:
+    def startDocument(self):
+        global log
+        log = log + "startDocument:"
+
+    def endDocument(self):
+        global log
+        log = log + "endDocument:"
+
+    def startElement(self, tag, attrs):
+        global log
+        log = log + "startElement %s %s:" % (tag, attrs)
+
+    def endElement(self, tag):
+        global log
+        log = log + "endElement %s:" % (tag)
+
+    def characters(self, data):
+        global log
+        log = log + "characters: %s:" % (data)
+
+    def warning(self, msg):
+        global log
+        log = log + "warning: %s:" % (msg)
+
+    def error(self, msg):
+        global log
+        log = log + "error: %s:" % (msg)
+
+    def fatalError(self, msg):
+        global log
+        log = log + "fatalError: %s:" % (msg)
+
+handler = callback()
+
+ctxt = libxml2.htmlCreatePushParser(handler, "<foo", 4, "test.xml")
+chunk = " url='tst'>b"
+ctxt.htmlParseChunk(chunk, len(chunk), 0)
+chunk = "ar</foo>"
+ctxt.htmlParseChunk(chunk, len(chunk), 1)
+ctxt=None
+
+reference = """startDocument:startElement html None:startElement body None:startElement foo {'url': 'tst'}:characters: bar:endElement foo:endElement body:endElement html:endDocument:"""
+if log != reference:
+    print("Error got: %s" % log)
+    print("Exprected: %s" % reference)
+    sys.exit(1)
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader.py

@@ -0,0 +1,446 @@
+#!/usr/bin/env python3
+# -*- coding: ISO-8859-1 -*-
+#
+# this tests the basic APIs of the XmlTextReader interface
+#
+import setup_test
+import libxml2
+import sys
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+f = str_io("""<a><b b1="b1"/><c>content of c</c></a>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test1")
+ret = reader.Read()
+if ret != 1:
+    print("test1: Error reading to first element")
+    sys.exit(1)
+if reader.Name() != "a" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 1 or reader.HasAttributes() != 0:
+    print("test1: Error reading the first element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test1: Error reading to second element")
+    sys.exit(1)
+if reader.Name() != "b" or reader.IsEmptyElement() != 1 or \
+   reader.NodeType() != 1 or reader.HasAttributes() != 1:
+    print("test1: Error reading the second element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test1: Error reading to third element")
+    sys.exit(1)
+if reader.Name() != "c" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 1 or reader.HasAttributes() != 0:
+    print("test1: Error reading the third element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test1: Error reading to text node")
+    sys.exit(1)
+if reader.Name() != "#text" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 3 or reader.HasAttributes() != 0 or \
+   reader.Value() != "content of c":
+    print("test1: Error reading the text node")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test1: Error reading to end of third element")
+    sys.exit(1)
+if reader.Name() != "c" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 15 or reader.HasAttributes() != 0:
+    print("test1: Error reading the end of third element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test1: Error reading to end of first element")
+    sys.exit(1)
+if reader.Name() != "a" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 15 or reader.HasAttributes() != 0:
+    print("test1: Error reading the end of first element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 0:
+    print("test1: Error reading to end of document")
+    sys.exit(1)
+
+#
+# example from the XmlTextReader docs
+#
+f = str_io("""<test xmlns:dt="urn:datatypes" dt:type="int"/>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test2")
+
+ret = reader.Read()
+if ret != 1:
+    print("Error reading test element")
+    sys.exit(1)
+if reader.GetAttributeNo(0) != "urn:datatypes" or \
+   reader.GetAttributeNo(1) != "int" or \
+   reader.GetAttributeNs("type", "urn:datatypes") != "int" or \
+   reader.GetAttribute("dt:type") != "int":
+    print("error reading test attributes")
+    sys.exit(1)
+
+#
+# example from the XmlTextReader docs
+#
+f = str_io("""<root xmlns:a="urn:456">
+<item>
+<ref href="a:b"/>
+</item>
+</root>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test3")
+
+ret = reader.Read()
+while ret == 1:
+    if reader.Name() == "ref":
+        if reader.LookupNamespace("a") != "urn:456":
+            print("error resolving namespace prefix")
+            sys.exit(1)
+        break
+    ret = reader.Read()
+if ret != 1:
+    print("Error finding the ref element")
+    sys.exit(1)
+
+#
+# Home made example for the various attribute access functions
+#
+f = str_io("""<testattr xmlns="urn:1" xmlns:a="urn:2" b="b" a:b="a:b"/>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test4")
+ret = reader.Read()
+if ret != 1:
+    print("Error reading the testattr element")
+    sys.exit(1)
+#
+# Attribute exploration by index
+#
+if reader.MoveToAttributeNo(0) != 1:
+    print("Failed moveToAttribute(0)")
+    sys.exit(1)
+if reader.Value() != "urn:1":
+    print("Failed to read attribute(0)")
+    sys.exit(1)
+if reader.Name() != "xmlns":
+    print("Failed to read attribute(0) name")
+    sys.exit(1)
+if reader.MoveToAttributeNo(1) != 1:
+    print("Failed moveToAttribute(1)")
+    sys.exit(1)
+if reader.Value() != "urn:2":
+    print("Failed to read attribute(1)")
+    sys.exit(1)
+if reader.Name() != "xmlns:a":
+    print("Failed to read attribute(1) name")
+    sys.exit(1)
+if reader.MoveToAttributeNo(2) != 1:
+    print("Failed moveToAttribute(2)")
+    sys.exit(1)
+if reader.Value() != "b":
+    print("Failed to read attribute(2)")
+    sys.exit(1)
+if reader.Name() != "b":
+    print("Failed to read attribute(2) name")
+    sys.exit(1)
+if reader.MoveToAttributeNo(3) != 1:
+    print("Failed moveToAttribute(3)")
+    sys.exit(1)
+if reader.Value() != "a:b":
+    print("Failed to read attribute(3)")
+    sys.exit(1)
+if reader.Name() != "a:b":
+    print("Failed to read attribute(3) name")
+    sys.exit(1)
+#
+# Attribute exploration by name
+#
+if reader.MoveToAttribute("xmlns") != 1:
+    print("Failed moveToAttribute('xmlns')")
+    sys.exit(1)
+if reader.Value() != "urn:1":
+    print("Failed to read attribute('xmlns')")
+    sys.exit(1)
+if reader.MoveToAttribute("xmlns:a") != 1:
+    print("Failed moveToAttribute('xmlns')")
+    sys.exit(1)
+if reader.Value() != "urn:2":
+    print("Failed to read attribute('xmlns:a')")
+    sys.exit(1)
+if reader.MoveToAttribute("b") != 1:
+    print("Failed moveToAttribute('b')")
+    sys.exit(1)
+if reader.Value() != "b":
+    print("Failed to read attribute('b')")
+    sys.exit(1)
+if reader.MoveToAttribute("a:b") != 1:
+    print("Failed moveToAttribute('a:b')")
+    sys.exit(1)
+if reader.Value() != "a:b":
+    print("Failed to read attribute('a:b')")
+    sys.exit(1)
+if reader.MoveToAttributeNs("b", "urn:2") != 1:
+    print("Failed moveToAttribute('b', 'urn:2')")
+    sys.exit(1)
+if reader.Value() != "a:b":
+    print("Failed to read attribute('b', 'urn:2')")
+    sys.exit(1)
+#
+# Go back and read in sequence
+#
+if reader.MoveToElement() != 1:
+    print("Failed to move back to element")
+    sys.exit(1)
+if reader.MoveToFirstAttribute() != 1:
+    print("Failed to move to first attribute")
+    sys.exit(1)
+if reader.Value() != "urn:1":
+    print("Failed to read attribute(0)")
+    sys.exit(1)
+if reader.Name() != "xmlns":
+    print("Failed to read attribute(0) name")
+    sys.exit(1)
+if reader.MoveToNextAttribute() != 1:
+    print("Failed to move to next attribute")
+    sys.exit(1)
+if reader.Value() != "urn:2":
+    print("Failed to read attribute(1)")
+    sys.exit(1)
+if reader.Name() != "xmlns:a":
+    print("Failed to read attribute(1) name")
+    sys.exit(1)
+if reader.MoveToNextAttribute() != 1:
+    print("Failed to move to next attribute")
+    sys.exit(1)
+if reader.Value() != "b":
+    print("Failed to read attribute(2)")
+    sys.exit(1)
+if reader.Name() != "b":
+    print("Failed to read attribute(2) name")
+    sys.exit(1)
+if reader.MoveToNextAttribute() != 1:
+    print("Failed to move to next attribute")
+    sys.exit(1)
+if reader.Value() != "a:b":
+    print("Failed to read attribute(3)")
+    sys.exit(1)
+if reader.Name() != "a:b":
+    print("Failed to read attribute(3) name")
+    sys.exit(1)
+if reader.MoveToNextAttribute() != 0:
+    print("Failed to detect last attribute")
+    sys.exit(1)
+
+    
+#
+# a couple of tests for namespace nodes
+#
+f = str_io("""<a xmlns="http://example.com/foo"/>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test6")
+ret = reader.Read()
+if ret != 1:
+    print("test6: failed to Read()")
+    sys.exit(1)
+ret = reader.MoveToFirstAttribute()
+if ret != 1:
+    print("test6: failed to MoveToFirstAttribute()")
+    sys.exit(1)
+if reader.NamespaceUri() != "http://www.w3.org/2000/xmlns/" or \
+   reader.LocalName() != "xmlns" or reader.Name() != "xmlns" or \
+   reader.Value() != "http://example.com/foo" or reader.NodeType() != 2:
+    print("test6: failed to read the namespace node")
+    sys.exit(1)
+
+f = str_io("""<a xmlns:prefix="http://example.com/foo"/>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test7")
+ret = reader.Read()
+if ret != 1:
+    print("test7: failed to Read()")
+    sys.exit(1)
+ret = reader.MoveToFirstAttribute()
+if ret != 1:
+    print("test7: failed to MoveToFirstAttribute()")
+    sys.exit(1)
+if reader.NamespaceUri() != "http://www.w3.org/2000/xmlns/" or \
+   reader.LocalName() != "prefix" or reader.Name() != "xmlns:prefix" or \
+   reader.Value() != "http://example.com/foo" or reader.NodeType() != 2:
+    print("test7: failed to read the namespace node")
+    sys.exit(1)
+
+#
+# Test for a limit case:
+#
+f = str_io("""<a/>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test8")
+ret = reader.Read()
+if ret != 1:
+    print("test8: failed to read the node")
+    sys.exit(1)
+if reader.Name() != "a" or reader.IsEmptyElement() != 1:
+    print("test8: failed to analyze the node")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 0:
+    print("test8: failed to detect the EOF")
+    sys.exit(1)
+
+#
+# Another test provided by Stéphane Bidoul and checked with C#
+#
+def tst_reader(s):
+    f = str_io(s)
+    input = libxml2.inputBuffer(f)
+    reader = input.newTextReader("tst")
+    res = ""
+    while reader.Read():
+        res=res + "%s (%s) [%s] %d %d\n" % (reader.NodeType(),reader.Name(),
+                                      reader.Value(), reader.IsEmptyElement(),
+                                      reader.Depth())
+        if reader.NodeType() == 1: # Element
+            while reader.MoveToNextAttribute():
+                res = res + "-- %s (%s) [%s] %d %d\n" % (reader.NodeType(),
+                                       reader.Name(),reader.Value(),
+                                       reader.IsEmptyElement(), reader.Depth())
+    return res
+    
+doc="""<a><b b1="b1"/><c>content of c</c></a>"""
+expect="""1 (a) [None] 0 0
+1 (b) [None] 1 1
+-- 2 (b1) [b1] 0 2
+1 (c) [None] 0 1
+3 (#text) [content of c] 0 2
+15 (c) [None] 0 1
+15 (a) [None] 0 0
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test5 failed")
+    print(res)
+    sys.exit(1)
+
+doc="""<test><b/><c/></test>"""
+expect="""1 (test) [None] 0 0
+1 (b) [None] 1 1
+1 (c) [None] 1 1
+15 (test) [None] 0 0
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test9 failed")
+    print(res)
+    sys.exit(1)
+
+doc="""<a><b>bbb</b><c>ccc</c></a>"""
+expect="""1 (a) [None] 0 0
+1 (b) [None] 0 1
+3 (#text) [bbb] 0 2
+15 (b) [None] 0 1
+1 (c) [None] 0 1
+3 (#text) [ccc] 0 2
+15 (c) [None] 0 1
+15 (a) [None] 0 0
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test10 failed")
+    print(res)
+    sys.exit(1)
+
+doc="""<test a="a"/>"""
+expect="""1 (test) [None] 1 0
+-- 2 (a) [a] 0 1
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test11 failed")
+    print(res)
+    sys.exit(1)
+
+doc="""<test><a>aaa</a><b/></test>"""
+expect="""1 (test) [None] 0 0
+1 (a) [None] 0 1
+3 (#text) [aaa] 0 2
+15 (a) [None] 0 1
+1 (b) [None] 1 1
+15 (test) [None] 0 0
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test12 failed")
+    print(res)
+    sys.exit(1)
+
+doc="""<test><p></p></test>"""
+expect="""1 (test) [None] 0 0
+1 (p) [None] 0 1
+15 (p) [None] 0 1
+15 (test) [None] 0 0
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test13 failed")
+    print(res)
+    sys.exit(1)
+
+doc="""<p></p>"""
+expect="""1 (p) [None] 0 0
+15 (p) [None] 0 0
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test14 failed")
+    print(res)
+    sys.exit(1)
+
+#
+# test from bug #108801 
+#
+doc="""<?xml version="1.0" standalone="no"?>
+<!DOCTYPE article PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
+                  "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" [
+]>
+
+<article>
+xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+</article>
+"""
+expect="""10 (article) [None] 0 0
+1 (article) [None] 0 0
+3 (#text) [
+xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+] 0 1
+15 (article) [None] 0 0
+"""
+res = tst_reader(doc)
+if res != expect:
+    print("test15 failed")
+    print(res)
+    sys.exit(1)
+
+#
+# cleanup for memory allocation counting
+#
+del f
+del input
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader2.py

@@ -0,0 +1,337 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# this tests the DTD validation with the XmlTextReader interface
+#
+import sys
+import glob
+import os
+import setup_test
+import libxml2
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+err = ""
+basedir = os.path.dirname(os.path.realpath(__file__))
+dir_prefix = os.path.realpath(os.path.join(basedir, "..", "..", "test", "valid"))
+
+# This dictionary reflects the contents of the files
+# ../../test/valid/*.xml.err that are not empty, except that
+# the file paths in the messages start with ../../test/
+
+expect = {
+    '766956':
+"""{0}/dtds/766956.dtd:2: parser error : PEReference: expecting ';'
+%ä%ent;
+   ^
+{0}/dtds/766956.dtd:2: parser error : Content error in the external subset
+%ä%ent;
+        ^
+Entity: line 1: 
+value
+^
+""".format(dir_prefix),
+    '781333':
+"""{0}/781333.xml:4: element a: validity error : Element a content does not follow the DTD, expecting ( ..., got 
+<a/>
+    ^
+{0}/781333.xml:5: element a: validity error : Element a content does not follow the DTD, Expecting more children
+
+^
+""".format(dir_prefix),
+    'cond_sect2':
+"""{0}/dtds/cond_sect2.dtd:15: parser error : Parameter entity must match extSubsetDecl
+    %ent;
+         ^
+Entity: line 1: 
+]]>
+^
+{0}/dtds/cond_sect2.dtd:15: parser error : Content error in the external subset
+    %ent;
+         ^
+Entity: line 1: 
+]]>
+^
+""".format(dir_prefix),
+    'rss':
+"""{0}/rss.xml:177: element rss: validity error : Element rss does not carry attribute version
+</rss>
+      ^
+""".format(dir_prefix),
+    't8':
+"""{0}/t8.xml:6: parser error : Content error in the internal subset
+%defroot; %defmiddle; %deftest;
+         ^
+Entity: line 1: 
+&lt;!ELEMENT root (middle) >
+^
+""".format(dir_prefix),
+    't8a':
+"""{0}/t8a.xml:6: parser error : Content error in the internal subset
+%defroot;%defmiddle;%deftest;
+         ^
+Entity: line 1: 
+&lt;!ELEMENT root (middle) >
+^
+""".format(dir_prefix),
+    'xlink':
+"""{0}/xlink.xml:450: element termdef: validity error : ID dt-arc already defined
+	<p><termdef id="dt-arc" term="Arc">An <ter
+	                                  ^
+validity error : attribute def line 199 references an unknown ID "dt-xlg"
+""".format(dir_prefix),
+}
+
+# Add prefix_dir and extension to the keys
+expect = {os.path.join(dir_prefix, key + ".xml"): val for key, val in expect.items()}
+
+def callback(ctx, str):
+    global err
+    err = err + "%s" % (str)
+libxml2.registerErrorHandler(callback, "")
+
+parsing_error_files = ["766956", "cond_sect2", "t8", "t8a", "pe-in-text-decl"]
+expect_parsing_error = [os.path.join(dir_prefix, f + ".xml") for f in parsing_error_files]
+
+valid_files = glob.glob(os.path.join(dir_prefix, "*.x*"))
+assert valid_files, "found no valid files in '{}'".format(dir_prefix)
+valid_files.sort()
+failures = 0
+for file in valid_files:
+    err = ""
+    reader = libxml2.newTextReaderFilename(file)
+    #print "%s:" % (file)
+    reader.SetParserProp(libxml2.PARSER_VALIDATE, 1)
+    ret = reader.Read()
+    while ret == 1:
+        ret = reader.Read()
+    if ret != 0 and file not in expect_parsing_error:
+        print("Error parsing and validating %s" % (file))
+        #sys.exit(1)
+    if file in expect and err != expect[file]:
+        failures += 1
+        print("Error: ", err)
+        if file in expect:
+            print("Expected: ", expect[file])
+
+if failures:
+    print("Failed %d tests" % failures)
+    sys.exit(1)
+
+#
+# another separate test based on Stephane Bidoul one
+#
+s = """
+<!DOCTYPE test [
+<!ELEMENT test (x,b)>
+<!ELEMENT x (c)>
+<!ELEMENT b (#PCDATA)>
+<!ELEMENT c (#PCDATA)>
+<!ENTITY x "<x><c>xxx</c></x>">
+]>
+<test>
+    &x;
+    <b>bbb</b>
+</test>
+"""
+expect="""10,test
+1,test
+14,#text
+1,x
+1,c
+3,#text
+15,c
+15,x
+14,#text
+1,b
+3,#text
+15,b
+14,#text
+15,test
+"""
+res=""
+err=""
+
+input = libxml2.inputBuffer(str_io(s))
+reader = input.newTextReader("test2")
+reader.SetParserProp(libxml2.PARSER_LOADDTD,1)
+reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS,1)
+reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)
+reader.SetParserProp(libxml2.PARSER_VALIDATE,1)
+while reader.Read() == 1:
+    res = res + "%s,%s\n" % (reader.NodeType(),reader.Name())
+
+if res != expect:
+    print("test2 failed: unexpected output")
+    print(res)
+    sys.exit(1)
+if err != "":
+    print("test2 failed: validation error found")
+    print(err)
+    sys.exit(1)
+
+#
+# Another test for external entity parsing and validation
+#
+
+s = """<!DOCTYPE test [
+<!ELEMENT test (x)>
+<!ELEMENT x (#PCDATA)>
+<!ENTITY e SYSTEM "tst.ent">
+]>
+<test>
+  &e;
+</test>
+"""
+tst_ent = """<x>hello</x>"""
+expect="""10 test
+1 test
+14 #text
+1 x
+3 #text
+15 x
+14 #text
+15 test
+"""
+res=""
+
+def myResolver(URL, ID, ctxt):
+    if URL == "tst.ent":
+        return(str_io(tst_ent))
+    return None
+
+libxml2.setEntityLoader(myResolver)
+
+input = libxml2.inputBuffer(str_io(s))
+reader = input.newTextReader("test3")
+reader.SetParserProp(libxml2.PARSER_LOADDTD,1)
+reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS,1)
+reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)
+reader.SetParserProp(libxml2.PARSER_VALIDATE,1)
+while reader.Read() == 1:
+    res = res + "%s %s\n" % (reader.NodeType(),reader.Name())
+
+if res != expect:
+    print("test3 failed: unexpected output")
+    print(res)
+    sys.exit(1)
+if err != "":
+    print("test3 failed: validation error found")
+    print(err)
+    sys.exit(1)
+
+#
+# Another test for recursive entity parsing, validation, and replacement of
+# entities, making sure the entity ref node doesn't show up in that case
+#
+
+s = """<!DOCTYPE test [
+<!ELEMENT test (x, x)>
+<!ELEMENT x (y)>
+<!ELEMENT y (#PCDATA)>
+<!ENTITY x "<x>&y;</x>">
+<!ENTITY y "<y>yyy</y>">
+]>
+<test>
+  &x;
+  &x;
+</test>"""
+expect="""10 test 0
+1 test 0
+14 #text 1
+1 x 1
+1 y 2
+3 #text 3
+15 y 2
+15 x 1
+14 #text 1
+1 x 1
+1 y 2
+3 #text 3
+15 y 2
+15 x 1
+14 #text 1
+15 test 0
+"""
+res=""
+err=""
+
+input = libxml2.inputBuffer(str_io(s))
+reader = input.newTextReader("test4")
+reader.SetParserProp(libxml2.PARSER_LOADDTD,1)
+reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS,1)
+reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)
+reader.SetParserProp(libxml2.PARSER_VALIDATE,1)
+while reader.Read() == 1:
+    res = res + "%s %s %d\n" % (reader.NodeType(),reader.Name(),reader.Depth())
+
+if res != expect:
+    print("test4 failed: unexpected output")
+    print(res)
+    sys.exit(1)
+if err != "":
+    print("test4 failed: validation error found")
+    print(err)
+    sys.exit(1)
+
+#
+# The same test but without entity substitution this time
+#
+
+s = """<!DOCTYPE test [
+<!ELEMENT test (x, x)>
+<!ELEMENT x (y)>
+<!ELEMENT y (#PCDATA)>
+<!ENTITY x "<x>&y;</x>">
+<!ENTITY y "<y>yyy</y>">
+]>
+<test>
+  &x;
+  &x;
+</test>"""
+expect="""10 test 0
+1 test 0
+14 #text 1
+5 x 1
+14 #text 1
+5 x 1
+14 #text 1
+15 test 0
+"""
+res=""
+err=""
+
+input = libxml2.inputBuffer(str_io(s))
+reader = input.newTextReader("test5")
+reader.SetParserProp(libxml2.PARSER_VALIDATE,1)
+while reader.Read() == 1:
+    res = res + "%s %s %d\n" % (reader.NodeType(),reader.Name(),reader.Depth())
+
+if res != expect:
+    print("test5 failed: unexpected output")
+    print(res)
+    sys.exit(1)
+if err != "":
+    print("test5 failed: validation error found")
+    print(err)
+    sys.exit(1)
+
+#
+# cleanup
+#
+del input
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader3.py

@@ -0,0 +1,160 @@
+#!/usr/bin/env python3
+#
+# this tests the entities substitutions with the XmlTextReader interface
+#
+import sys
+import setup_test
+import libxml2
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+docstr="""<?xml version='1.0'?>
+<!DOCTYPE doc [
+<!ENTITY tst "<p>test</p>">
+]>
+<doc>&tst;</doc>"""
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+#
+# First test, normal don't substitute entities.
+#
+f = str_io(docstr)
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test_noent")
+ret = reader.Read()
+if ret != 1:
+    print("Error reading to root")
+    sys.exit(1)
+if reader.Name() == "doc" or reader.NodeType() == 10:
+    ret = reader.Read()
+if ret != 1:
+    print("Error reading to root")
+    sys.exit(1)
+if reader.Name() != "doc" or reader.NodeType() != 1:
+    print("test_normal: Error reading the root element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_normal: Error reading to the entity")
+    sys.exit(1)
+if reader.Name() != "tst" or reader.NodeType() != 5:
+    print("test_normal: Error reading the entity")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_normal: Error reading to the end of root")
+    sys.exit(1)
+if reader.Name() != "doc" or reader.NodeType() != 15:
+    print("test_normal: Error reading the end of the root element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 0:
+    print("test_normal: Error detecting the end")
+    sys.exit(1)
+
+#
+# Second test, completely substitute the entities.
+#
+f = str_io(docstr)
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test_noent")
+reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES, 1)
+ret = reader.Read()
+if ret != 1:
+    print("Error reading to root")
+    sys.exit(1)
+if reader.Name() == "doc" or reader.NodeType() == 10:
+    ret = reader.Read()
+if ret != 1:
+    print("Error reading to root")
+    sys.exit(1)
+if reader.Name() != "doc" or reader.NodeType() != 1:
+    print("test_noent: Error reading the root element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_noent: Error reading to the entity content")
+    sys.exit(1)
+if reader.Name() != "p" or reader.NodeType() != 1:
+    print("test_noent: Error reading the p element from entity")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_noent: Error reading to the text node")
+    sys.exit(1)
+if reader.NodeType() != 3 or reader.Value() != "test":
+    print("test_noent: Error reading the text node")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_noent: Error reading to the end of p element")
+    sys.exit(1)
+if reader.Name() != "p" or reader.NodeType() != 15:
+    print("test_noent: Error reading the end of the p element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_noent: Error reading to the end of root")
+    sys.exit(1)
+if reader.Name() != "doc" or reader.NodeType() != 15:
+    print("test_noent: Error reading the end of the root element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 0:
+    print("test_noent: Error detecting the end")
+    sys.exit(1)
+
+#
+# third test, crazy stuff about empty element in external parsed entities
+#
+s = """<!DOCTYPE struct [
+<!ENTITY simplestruct2.ent SYSTEM "simplestruct2.ent">
+]>
+<struct>&simplestruct2.ent;</struct>
+"""
+expect="""10 struct 0 0
+1 struct 0 0
+1 descr 1 1
+15 struct 0 0
+"""
+res=""
+simplestruct2_ent="""<descr/>"""
+
+def myResolver(URL, ID, ctxt):
+    if URL == "simplestruct2.ent":
+        return(str_io(simplestruct2_ent))
+    return None
+
+libxml2.setEntityLoader(myResolver)
+
+input = libxml2.inputBuffer(str_io(s))
+reader = input.newTextReader("test3")
+reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)
+while reader.Read() == 1:
+    res = res + "%s %s %d %d\n" % (reader.NodeType(),reader.Name(),
+                                   reader.Depth(),reader.IsEmptyElement())
+
+if res != expect:
+    print("test3 failed: unexpected output")
+    print(res)
+    sys.exit(1)
+
+#
+# cleanup
+#
+del f
+del input
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader4.py

@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+#
+# this tests the basic APIs of the XmlTextReader interface
+#
+import setup_test
+import libxml2
+import sys
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+def tst_reader(s):
+    f = str_io(s)
+    input = libxml2.inputBuffer(f)
+    reader = input.newTextReader("tst")
+    res = ""
+    while reader.Read():
+        res=res + "%s (%s) [%s] %d\n" % (reader.NodeType(),reader.Name(),
+				      reader.Value(), reader.IsEmptyElement())
+        if reader.NodeType() == 1: # Element
+            while reader.MoveToNextAttribute():
+                res = res + "-- %s (%s) [%s]\n" % (reader.NodeType(),
+						   reader.Name(),reader.Value())
+    return res
+    
+expect="""1 (test) [None] 0
+1 (b) [None] 1
+1 (c) [None] 1
+15 (test) [None] 0
+"""
+
+res = tst_reader("""<test><b/><c/></test>""")
+
+if res != expect:
+    print("Did not get the expected error message:")
+    print(res)
+    sys.exit(1)
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader5.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+#
+# this tests the Expand() API of the xmlTextReader interface
+# this extract the Dragon bibliography entries from the XML specification
+#
+import setup_test
+import libxml2
+import os
+import sys
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+expect="""<bibl id="Aho" key="Aho/Ullman">Aho, Alfred V., 
+Ravi Sethi, and Jeffrey D. Ullman.
+<emph>Compilers:  Principles, Techniques, and Tools</emph>.
+Reading:  Addison-Wesley, 1986, rpt. corr. 1988.</bibl>"""
+
+basedir = os.path.dirname(os.path.realpath(__file__))
+f = open(os.path.join(basedir, '../../test/valid/REC-xml-19980210.xml'), 'rb')
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("REC")
+res=""
+while reader.Read() > 0:
+    while reader.Name() == 'bibl':
+        node = reader.Expand()            # expand the subtree
+        if node.xpathEval("@id = 'Aho'"): # use XPath on it
+            res = res + node.serialize()
+        if reader.Next() != 1:            # skip the subtree
+            break;
+
+if res != expect:
+    print("Error: didn't get the expected output")
+    print("got '%s'" % (res))
+    print("expected '%s'" % (expect))
+    
+
+#
+# cleanup
+#
+del input
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader6.py

@@ -0,0 +1,128 @@
+#!/usr/bin/env python3
+#
+# this tests the entities substitutions with the XmlTextReader interface
+#
+import sys
+import setup_test
+import libxml2
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+schema="""<element name="foo" xmlns="http://relaxng.org/ns/structure/1.0"
+         datatypeLibrary="http://www.w3.org/2001/XMLSchema-datatypes">
+  <oneOrMore>
+    <element name="label">
+      <text/>
+    </element>
+    <optional>
+      <element name="opt">
+        <empty/>
+      </element>
+    </optional>
+    <element name="item">
+      <data type="byte"/>
+    </element>
+  </oneOrMore>
+</element>
+"""
+# Memory debug specific
+libxml2.debugMemory(1)
+
+#
+# Parse the Relax NG Schemas
+# 
+rngp = libxml2.relaxNGNewMemParserCtxt(schema, len(schema))
+rngs = rngp.relaxNGParse()
+del rngp
+
+#
+# Parse and validate the correct document
+#
+docstr="""<foo>
+<label>some text</label>
+<item>100</item>
+</foo>"""
+
+f = str_io(docstr)
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("correct")
+reader.RelaxNGSetSchema(rngs)
+ret = reader.Read()
+while ret == 1:
+    ret = reader.Read()
+
+if ret != 0:
+    print("Error parsing the document")
+    sys.exit(1)
+
+if reader.IsValid() != 1:
+    print("Document failed to validate")
+    sys.exit(1)
+
+#
+# Parse and validate the incorrect document
+#
+docstr="""<foo>
+<label>some text</label>
+<item>1000</item>
+</foo>"""
+
+err=""
+# RNG errors are not as good as before , TODO
+#expect="""RNG validity error: file error line 3 element text
+#Type byte doesn't allow value '1000'
+#RNG validity error: file error line 3 element text
+#Error validating datatype byte
+#RNG validity error: file error line 3 element text
+#Element item failed to validate content
+#"""
+expect="""Type byte doesn't allow value '1000'
+Error validating datatype byte
+Element item failed to validate content
+"""
+
+def callback(ctx, str):
+    global err
+    err = err + "%s" % (str)
+libxml2.registerErrorHandler(callback, "")
+
+f = str_io(docstr)
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("error")
+reader.RelaxNGSetSchema(rngs)
+ret = reader.Read()
+while ret == 1:
+    ret = reader.Read()
+
+if ret != 0:
+    print("Error parsing the document")
+    sys.exit(1)
+
+if reader.IsValid() != 0:
+    print("Document failed to detect the validation error")
+    sys.exit(1)
+
+if err != expect:
+    print("Did not get the expected error message:")
+    print(err)
+    sys.exit(1)
+
+#
+# cleanup
+#
+del f
+del input
+del reader
+del rngs
+libxml2.relaxNGCleanupTypes()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader7.py

@@ -0,0 +1,101 @@
+#!/usr/bin/env python3
+#
+# this tests the entities substitutions with the XmlTextReader interface
+#
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+result = ""
+def processNode(reader):
+    global result
+
+    result = result + "%d %d %s %d\n" % (reader.Depth(), reader.NodeType(),
+			   reader.Name(), reader.IsEmptyElement())
+
+#
+# Parse a document testing the readerForxxx API
+#
+docstr="""<foo>
+<label>some text</label>
+<item>100</item>
+</foo>"""
+expect="""0 1 foo 0
+1 14 #text 0
+1 1 label 0
+2 3 #text 0
+1 15 label 0
+1 14 #text 0
+1 1 item 0
+2 3 #text 0
+1 15 item 0
+1 14 #text 0
+0 15 foo 0
+"""
+result = ""
+
+reader = libxml2.readerForDoc(docstr, "test1", None, 0)
+ret = reader.Read()
+while ret == 1:
+    processNode(reader)
+    ret = reader.Read()
+
+if ret != 0:
+    print("Error parsing the document test1")
+    sys.exit(1)
+
+if result != expect:
+    print("Unexpected result for test1")
+    print(result)
+    sys.exit(1)
+
+#
+# Reuse the reader for another document testing the ReaderNewxxx API
+#
+docstr="""<foo>
+<label>some text</label>
+<item>1000</item>
+</foo>"""
+expect="""0 1 foo 0
+1 14 #text 0
+1 1 label 0
+2 3 #text 0
+1 15 label 0
+1 14 #text 0
+1 1 item 0
+2 3 #text 0
+1 15 item 0
+1 14 #text 0
+0 15 foo 0
+"""
+result = ""
+
+reader.NewDoc(docstr, "test2", None, 0)
+ret = reader.Read()
+while ret == 1:
+    processNode(reader)
+    ret = reader.Read()
+
+if ret != 0:
+    print("Error parsing the document test2")
+    sys.exit(1)
+
+if result != expect:
+    print("Unexpected result for test2")
+    print(result)
+    sys.exit(1)
+
+#
+# cleanup
+#
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/reader8.py

@@ -0,0 +1,36 @@
+#!/usr/bin/env python3
+#
+# this tests the entities substitutions with the XmlTextReader interface
+#
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+#
+# Parse a document testing the Close() API
+#
+docstr="""<foo>
+<label>some text</label>
+<item>100</item>
+</foo>"""
+
+reader = libxml2.readerForDoc(docstr, "test1", None, 0)
+ret = reader.Read()
+ret = reader.Read()
+ret = reader.Close()
+
+if ret != 0:
+    print("Error closing the document test1")
+    sys.exit(1)
+
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/readererr.py

@@ -0,0 +1,56 @@
+#!/usr/bin/env python3
+#
+# this tests the basic APIs of the XmlTextReader interface
+#
+import setup_test
+import libxml2
+import sys
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+expect="""--> (3) test1:1:xmlns: URI foo is not absolute
+--> (4) test1:1:Opening and ending tag mismatch: c line 1 and a
+"""
+err=""
+def myErrorHandler(arg,msg,severity,locator):
+    global err
+    err = err + "%s (%d) %s:%d:%s" % (arg,severity,locator.BaseURI(),locator.LineNumber(),msg)
+
+f = str_io("""<a xmlns="foo"><b b1="b1"/><c>content of c</a>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test1")
+reader.SetErrorHandler(myErrorHandler,"-->")
+while reader.Read() == 1:
+    pass
+
+if err != expect:
+    print("error")
+    print("received %s" %(err))
+    print("expected %s" %(expect))
+    sys.exit(1)
+
+reader.SetErrorHandler(None,None)
+if reader.GetErrorHandler() != (None,None):
+    print("GetErrorHandler failed")
+    sys.exit(1)
+
+#
+# cleanup for memory allocation counting
+#
+del f
+del input
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/readernext.py

@@ -0,0 +1,86 @@
+#!/usr/bin/env python3
+# -*- coding: ISO-8859-1 -*-
+#
+# this tests the next API of the XmlTextReader interface
+#
+import setup_test
+import libxml2
+import sys
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+f = str_io("""<a><b><c /></b><d>content of d</d></a>""")
+input = libxml2.inputBuffer(f)
+reader = input.newTextReader("test_next")
+ret = reader.Read()
+if ret != 1:
+    print("test_next: Error reading to first element")
+    sys.exit(1)
+if reader.Name() != "a" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 1 or reader.HasAttributes() != 0:
+    print("test_next: Error reading the first element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_next: Error reading to second element")
+    sys.exit(1)
+if reader.Name() != "b" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 1 or reader.HasAttributes() != 0:
+    print("test_next: Error reading the second element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_next: Error reading to third element")
+    sys.exit(1)
+if reader.Name() != "c" or reader.NodeType() != 1 or \
+   reader.HasAttributes() != 0:
+    print("test_next: Error reading the third element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 1:
+    print("test_next: Error reading to end of third element")
+    sys.exit(1)
+if reader.Name() != "b" or reader.NodeType() != 15:
+    print("test_next: Error reading to end of second element")
+    sys.exit(1)
+ret = reader.Next()
+if ret != 1:
+    print("test_next: Error moving to third element")
+    sys.exit(1)
+if reader.Name() != "d" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 1 or reader.HasAttributes() != 0:
+    print("test_next: Error reading third element")
+    sys.exit(1)
+ret = reader.Next()
+if ret != 1:
+    print("test_next: Error reading to end of first element")
+    sys.exit(1)
+if reader.Name() != "a" or reader.IsEmptyElement() != 0 or \
+   reader.NodeType() != 15 or reader.HasAttributes() != 0:
+    print("test_next: Error reading the end of first element")
+    sys.exit(1)
+ret = reader.Read()
+if ret != 0:
+    print("test_next: Error reading to end of document")
+    sys.exit(1)
+
+#
+# cleanup for memory allocation counting
+#
+del f
+del input
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/regexp.py

@@ -0,0 +1,43 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+
+def check_regex(regex, inside, outside):
+    re = libxml2.regexpCompile(regex)
+    for c in inside:
+        if re.regexpExec(c) != 1:
+            print("error checking inside", repr(regex), repr(c))
+            sys.exit(1)
+    for c in outside:
+        if re.regexpExec(c) != 0:
+            print("error checking outside", repr(regex), repr(c))
+            sys.exit(1)
+    if re.regexpIsDeterminist() != 1:
+        print("error checking determinism")
+        sys.exit(1)
+    del re
+
+
+check_regex("a|b", "ab", ["ab", ""])
+# https://gitlab.gnome.org/GNOME/libxml2/-/issues/1086
+check_regex(r"[\t-\r]", "\t\n\r", ["a", "(", ""])
+check_regex("[\\t-\\r]", "\t\n\r", ["a", "(", ""])
+check_regex(r"[\[-\]]", "[]\\", "Z^")
+check_regex(r"[\*-/]", "*+,-./", "()0")
+# Non-standard escape sequences
+check_regex(r"[\u0041-\u005A]", "ABCMYZ", ["a", "0", "[", ""])
+check_regex(r"[\u0009\u0061-\u007A]", "\tabcmyz", ["A", "0", "{", "", "\n"])
+check_regex(r"[\!-\%]", '!"#$%', ["&", " ", "a", ""])
+check_regex(r"[\:-\>]", ':;<=>',  ["?", "9", "@", ""])
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/relaxng.py

@@ -0,0 +1,48 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+schema="""<?xml version="1.0"?>
+<element name="foo"
+         xmlns="http://relaxng.org/ns/structure/1.0"
+         xmlns:a="http://relaxng.org/ns/annotation/1.0"
+         xmlns:ex1="http://www.example.com/n1"
+         xmlns:ex2="http://www.example.com/n2">
+  <a:documentation>A foo element.</a:documentation>
+  <element name="ex1:bar1">
+    <empty/>
+  </element>
+  <element name="ex2:bar2">
+    <empty/>
+  </element>
+</element>
+"""
+instance="""<?xml version="1.0"?>
+<foo><pre1:bar1 xmlns:pre1="http://www.example.com/n1"/><pre2:bar2 xmlns:pre2="http://www.example.com/n2"/></foo>"""
+
+rngp = libxml2.relaxNGNewMemParserCtxt(schema, len(schema))
+rngs = rngp.relaxNGParse()
+ctxt = rngs.relaxNGNewValidCtxt()
+doc = libxml2.parseDoc(instance)
+ret = doc.relaxNGValidateDoc(ctxt)
+if ret != 0:
+    print("error doing RelaxNG validation")
+    sys.exit(1)
+
+doc.freeDoc()
+del rngp
+del rngs
+del ctxt
+libxml2.relaxNGCleanupTypes()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/resolver.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+try:
+    import StringIO
+    str_io = StringIO.StringIO
+except:
+    import io
+    str_io = io.StringIO
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+def myResolver(URL, ID, ctxt):
+    return(str_io("<foo/>"))
+
+libxml2.setEntityLoader(myResolver)
+
+doc = libxml2.parseFile("doesnotexist.xml")
+root = doc.children
+if root.name != "foo":
+    print("root element name error")
+    sys.exit(1)
+doc.freeDoc()
+
+i = 0
+while i < 5000:
+    doc = libxml2.parseFile("doesnotexist.xml")
+    root = doc.children
+    if root.name != "foo":
+        print("root element name error")
+        sys.exit(1)
+    doc.freeDoc()
+    i = i + 1
+
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/schema.py

@@ -0,0 +1,52 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+schema="""<?xml version="1.0" encoding="iso-8859-1"?>
+<schema xmlns = "http://www.w3.org/2001/XMLSchema">
+	<element name = "Customer">
+		<complexType>
+			<sequence>
+				<element name = "FirstName" type = "string" />
+				<element name = "MiddleInitial" type = "string" />
+				<element name = "LastName" type = "string" />
+			</sequence>
+			<attribute name = "customerID" type = "integer" />
+		</complexType>
+	</element>
+</schema>"""
+
+instance="""<?xml version="1.0" encoding="iso-8859-1"?>
+<Customer customerID = "24332">
+	<FirstName>Raymond</FirstName>
+	<MiddleInitial>G</MiddleInitial>
+	<LastName>Bayliss</LastName>
+</Customer>	
+"""
+
+ctxt_parser = libxml2.schemaNewMemParserCtxt(schema, len(schema))
+ctxt_schema = ctxt_parser.schemaParse()
+ctxt_valid  = ctxt_schema.schemaNewValidCtxt()
+doc = libxml2.parseDoc(instance)
+ret = doc.schemaValidateDoc(ctxt_valid)
+if ret != 0:
+    print("error doing schema validation")
+    sys.exit(1)
+
+doc.freeDoc()
+del ctxt_parser
+del ctxt_schema
+del ctxt_valid
+libxml2.schemaCleanupTypes()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/serialize.py

@@ -0,0 +1,144 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+#
+# Testing XML document serialization
+#
+doc = libxml2.parseDoc("""<root><foo>hello</foo></root>""")
+str = doc.serialize()
+if str != """<?xml version="1.0"?>
+<root><foo>hello</foo></root>
+""":
+   print("error serializing XML document 1")
+   sys.exit(1)
+str = doc.serialize("iso-8859-1")
+if str != """<?xml version="1.0" encoding="iso-8859-1"?>
+<root><foo>hello</foo></root>
+""":
+   print("error serializing XML document 2")
+   sys.exit(1)
+str = doc.serialize(format=1)
+if str != """<?xml version="1.0"?>
+<root>
+  <foo>hello</foo>
+</root>
+""":
+   print("error serializing XML document 3")
+   sys.exit(1)
+str = doc.serialize("iso-8859-1", 1)
+if str != """<?xml version="1.0" encoding="iso-8859-1"?>
+<root>
+  <foo>hello</foo>
+</root>
+""":
+   print("error serializing XML document 4")
+   sys.exit(1)
+
+#
+# Test serializing a subnode
+#
+root = doc.getRootElement()
+str = root.serialize()
+if str != """<root><foo>hello</foo></root>""":
+   print("error serializing XML root 1")
+   sys.exit(1)
+str = root.serialize("iso-8859-1")
+if str != """<root><foo>hello</foo></root>""":
+   print("error serializing XML root 2")
+   sys.exit(1)
+str = root.serialize(format=1)
+if str != """<root>
+  <foo>hello</foo>
+</root>""":
+   print("error serializing XML root 3")
+   sys.exit(1)
+str = root.serialize("iso-8859-1", 1)
+if str != """<root>
+  <foo>hello</foo>
+</root>""":
+   print("error serializing XML root 4")
+   sys.exit(1)
+doc.freeDoc()
+
+#
+# Testing HTML document serialization
+#
+doc = libxml2.htmlParseDoc("""<html><head><title>Hello</title><body><p>hello</body></html>""", None)
+str = doc.serialize()
+if str != """<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
+<html><head><title>Hello</title></head><body><p>hello</p></body></html>
+""":
+   print("error serializing HTML document 1")
+   sys.exit(1)
+str = doc.serialize("ISO-8859-1")
+if str != """<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
+<html><head><meta charset="ISO-8859-1"><title>Hello</title></head><body><p>hello</p></body></html>
+""":
+   print("error serializing HTML document 2")
+   sys.exit(1)
+str = doc.serialize(format=1)
+if str != """<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
+<html>
+<head><title>Hello</title></head>
+<body><p>hello</p></body>
+</html>
+""":
+   print("error serializing HTML document 3")
+   sys.exit(1)
+str = doc.serialize("iso-8859-1", 1)
+if str != """<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
+<html>
+<head>
+<meta charset="iso-8859-1">
+<title>Hello</title>
+</head>
+<body><p>hello</p></body>
+</html>
+""":
+   print("error serializing HTML document 4", str)
+   sys.exit(1)
+
+#
+# Test serializing a subnode
+#
+doc.htmlSetMetaEncoding(None)
+root = doc.getRootElement()
+str = root.serialize()
+if str != """<html><head><title>Hello</title></head><body><p>hello</p></body></html>""":
+   print("error serializing HTML root 1")
+   sys.exit(1)
+str = root.serialize("ISO-8859-1")
+if str != """<html><head><meta charset="ISO-8859-1"><title>Hello</title></head><body><p>hello</p></body></html>""":
+   print("error serializing HTML root 2")
+   sys.exit(1)
+str = root.serialize(format=1)
+if str != """<html>
+<head><title>Hello</title></head>
+<body><p>hello</p></body>
+</html>""":
+   print("error serializing HTML root 3")
+   sys.exit(1)
+str = root.serialize("iso-8859-1", 1)
+if str != """<html>
+<head>
+<meta charset="iso-8859-1">
+<title>Hello</title>
+</head>
+<body><p>hello</p></body>
+</html>""":
+   print("error serializing HTML root 4")
+   sys.exit(1)
+
+doc.freeDoc()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/setup_test.py

@@ -0,0 +1,5 @@
+import os
+import sys
+
+if hasattr(os, 'add_dll_directory'):
+    os.add_dll_directory(os.path.join(os.getcwd(), '..', '..', '.libs'))

libxml2/python/tests/sync.py

@@ -0,0 +1,138 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+log = ""
+
+class callback:
+    def startDocument(self):
+        global log
+        log = log + "startDocument:"
+
+    def endDocument(self):
+        global log
+        log = log + "endDocument:"
+
+    def startElement(self, tag, attrs):
+        global log
+        log = log + "startElement %s %s:" % (tag, attrs)
+
+    def endElement(self, tag):
+        global log
+        log = log + "endElement %s:" % (tag)
+
+    def characters(self, data):
+        global log
+        log = log + "characters: %s:" % (data)
+
+    def warning(self, msg):
+        global log
+        log = log + "warning: %s:" % (msg)
+
+    def error(self, msg):
+        global log
+        log = log + "error: %s:" % (msg)
+
+    def fatalError(self, msg):
+        global log
+        log = log + "fatalError: %s:" % (msg)
+
+handler = callback()
+
+log=""
+chunk="""<foo><bar2/>"""
+ctxt = libxml2.createPushParser(handler, None, 0, "test.xml")
+ctxt.parseChunk(chunk, len(chunk), 0)
+ctxt=None
+
+reference = "startDocument:startElement foo None:startElement bar2 None:endElement bar2:"
+if log != reference:
+    print("Error got: %s" % log)
+    print("Expected: %s" % reference)
+    sys.exit(1)
+
+log=""
+chunk="""<foo><bar2></bar2>"""
+ctxt = libxml2.createPushParser(handler, None, 0, "test.xml")
+ctxt.parseChunk(chunk, len(chunk), 0)
+ctxt=None
+
+reference = "startDocument:startElement foo None:startElement bar2 None:endElement bar2:"
+if log != reference:
+    print("Error got: %s" % log)
+    print("Expected: %s" % reference)
+    sys.exit(1)
+
+log=""
+chunk="""<foo><bar2>"""
+ctxt = libxml2.createPushParser(handler, None, 0, "test.xml")
+ctxt.parseChunk(chunk, len(chunk), 0)
+ctxt=None
+
+reference = "startDocument:startElement foo None:startElement bar2 None:"
+if log != reference:
+    print("Error got: %s" % log)
+    print("Expected: %s" % reference)
+    sys.exit(1)
+
+log=""
+chunk="""<foo><bar2 a="1" b='2' />"""
+ctxt = libxml2.createPushParser(handler, None, 0, "test.xml")
+ctxt.parseChunk(chunk, len(chunk), 0)
+ctxt=None
+
+reference1 = "startDocument:startElement foo None:startElement bar2 {'a': '1', 'b': '2'}:endElement bar2:"
+reference2 = "startDocument:startElement foo None:startElement bar2 {'b': '2', 'a': '1'}:endElement bar2:"
+if log not in (reference1, reference2):
+    print("Error got: %s" % log)
+    print("Expected: %s" % reference)
+    sys.exit(1)
+
+log=""
+chunk="""<foo><bar2 a="1" b='2' >"""
+ctxt = libxml2.createPushParser(handler, None, 0, "test.xml")
+ctxt.parseChunk(chunk, len(chunk), 0)
+ctxt=None
+
+reference1 = "startDocument:startElement foo None:startElement bar2 {'a': '1', 'b': '2'}:"
+reference2 = "startDocument:startElement foo None:startElement bar2 {'b': '2', 'a': '1'}:"
+if log not in (reference1, reference2):
+    print("Error got: %s" % log)
+    print("Expected: %s" % reference)
+    sys.exit(1)
+
+log=""
+chunk="""<foo><bar2 a="1" b='2' ></bar2>"""
+ctxt = libxml2.createPushParser(handler, None, 0, "test.xml")
+ctxt.parseChunk(chunk, len(chunk), 0)
+ctxt=None
+
+reference1 = "startDocument:startElement foo None:startElement bar2 {'a': '1', 'b': '2'}:endElement bar2:"
+reference2 = "startDocument:startElement foo None:startElement bar2 {'b': '2', 'a': '1'}:endElement bar2:"
+if log not in (reference1, reference2):
+    print("Error got: %s" % log)
+    print("Expected: %s" % reference)
+    sys.exit(1)
+
+log=""
+chunk="""<foo><bar2 a="b='1' />"""
+ctxt = libxml2.createPushParser(handler, None, 0, "test.xml")
+ctxt.parseChunk(chunk, len(chunk), 0)
+ctxt=None
+
+reference = "startDocument:startElement foo None:"
+if log != reference:
+    print("Error got: %s" % log)
+    print("Expected: %s" % reference)
+    sys.exit(1)
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/test.dtd

@@ -0,0 +1 @@
+<!ELEMENT foo EMPTY>

libxml2/python/tests/thread2.py

@@ -0,0 +1,105 @@
+#!/usr/bin/env python3
+import string, sys, time
+try:
+    from _thread import get_ident
+except:
+    from thread import get_ident
+from threading import Thread, Lock
+
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+THREADS_COUNT = 15
+
+failed = 0
+
+class ErrorHandler:
+
+    def __init__(self):
+        self.errors = []
+        self.lock = Lock()
+
+    def handler(self,ctx,str):
+        self.lock.acquire()
+        self.errors.append(str)
+        self.lock.release()
+
+def getPedanticParserDefault():
+    old = libxml2.pedanticParserDefault(0)
+    libxml2.pedanticParserDefault(old)
+    return old
+
+def test(expectedPedanticParserDefault):
+    time.sleep(1)
+    global failed
+    # check a per thread-global
+    if expectedPedanticParserDefault != getPedanticParserDefault():
+        failed = 1
+        print("FAILED to obtain correct value for " \
+              "pedanticParserDefault in thread %d" % get_ident())
+    # check the global error handler 
+    # (which is NOT per-thread in the python bindings)
+    try:
+        doc = libxml2.parseFile("bad.xml")
+    except:
+        pass
+    else:
+        assert "failed"
+
+# global error handler
+eh = ErrorHandler()
+libxml2.registerErrorHandler(eh.handler,"")
+
+# set on the main thread only
+libxml2.pedanticParserDefault(1)
+test(1)
+ec = len(eh.errors)
+if ec == 0:
+    print("FAILED: should have obtained errors")
+    sys.exit(1)
+
+ts = []
+for i in range(THREADS_COUNT):
+    # expect 0 for pedanticParserDefault because
+    # the new value has been set on the main thread only
+    ts.append(Thread(target=test,args=(0,)))
+for t in ts:
+    t.start()
+for t in ts:
+    t.join()
+
+if len(eh.errors) != ec+THREADS_COUNT*ec:
+    print("FAILED: did not obtain the correct number of errors")
+    sys.exit(1)
+
+# set pedanticParserDefault for future new threads
+libxml2.thrDefPedanticParserDefaultValue(1)
+ts = []
+for i in range(THREADS_COUNT):
+    # expect 1 for pedanticParserDefault
+    ts.append(Thread(target=test,args=(1,)))
+for t in ts:
+    t.start()
+for t in ts:
+    t.join()
+
+if len(eh.errors) != ec+THREADS_COUNT*ec*2:
+    print("FAILED: did not obtain the correct number of errors")
+    sys.exit(1)
+
+if failed:
+    print("FAILED")
+    sys.exit(1)
+
+# Memory debug specific
+libxml2.cleanupParser()
+# Note that this can leak memory on Windows if the global state
+# destructors weren't run yet. They should be called eventually,
+# so this leak should be harmless.
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/tst.py

@@ -0,0 +1,28 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+doc = libxml2.parseFile("tst.xml")
+if doc.name != "tst.xml":
+    print("doc.name failed")
+    sys.exit(1)
+root = doc.children
+if root.name != "doc":
+    print("root.name failed")
+    sys.exit(1)
+child = root.children
+if child.name != "foo":
+    print("child.name failed")
+    sys.exit(1)
+doc.freeDoc()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/tst.xml

@@ -0,0 +1 @@
+<doc><foo>bar</foo></doc>

libxml2/python/tests/tstLastError.py

@@ -0,0 +1,85 @@
+#!/usr/bin/env python3
+import sys, unittest
+
+import setup_test
+import libxml2
+
+class TestCase(unittest.TestCase):
+
+    def runTest(self):
+        self.test1()
+        self.test2()
+
+    def setUp(self):
+        libxml2.debugMemory(1)
+
+    def tearDown(self):
+        libxml2.cleanupParser()
+        if libxml2.debugMemory(1) != 0:
+            self.fail("Memory leak %d bytes" % (libxml2.debugMemory(1),))
+        else:
+            print("OK")
+
+    def failUnlessXmlError(self,f,args,exc,domain,code,message,level,file,line):
+        """Run function f, with arguments args and expect an exception exc;
+        when the exception is raised, check the libxml2.lastError for
+        expected values."""
+        # disable the default error handler
+        def noerr(ctx, str):
+            pass
+        # None is not acceptable as function.
+        libxml2.registerErrorHandler(noerr,None)
+        try:
+            f(*args)
+        except exc:
+            e = libxml2.lastError()
+            if e is None:
+                self.fail("lastError not set")
+            if 0:
+                print("domain = ",e.domain())
+                print("code = ",e.code())
+                print("message =",repr(e.message()))
+                print("level =",e.level())
+                print("file =",e.file())
+                print("line =",e.line())
+                print()
+            self.assertEqual(domain,e.domain())
+            self.assertEqual(code,e.code())
+            self.assertEqual(message,e.message())
+            self.assertEqual(level,e.level())
+            self.assertEqual(file,e.file())
+            self.assertEqual(line,e.line())
+        else:
+            self.fail("exception %s should have been raised" % exc)
+
+    def test1(self):
+        """Test readFile with a file that does not exist"""
+        self.failUnlessXmlError(libxml2.readFile,
+                        ("dummy.xml",None,0),
+                        libxml2.treeError,
+                        domain=libxml2.XML_FROM_IO,
+                        code=libxml2.XML_IO_ENOENT,
+                        message='failed to load "dummy.xml": No such file or directory\n',
+                        level=libxml2.XML_ERR_WARNING,
+                        file=None,
+                        line=0)
+
+    def test2(self):
+        """Test a well-formedness error: we get the last error only"""
+        s = "<x>\n<a>\n</x>"
+        self.failUnlessXmlError(libxml2.readMemory,
+                        (s,len(s),"dummy.xml",None,0),
+                        libxml2.treeError,
+                        domain=libxml2.XML_FROM_PARSER,
+                        code=libxml2.XML_ERR_TAG_NAME_MISMATCH,
+                        message='Opening and ending tag mismatch: a line 2 and x\n',
+                        level=libxml2.XML_ERR_FATAL,
+                        file='dummy.xml',
+                        line=3)
+
+if __name__ == "__main__":
+    test = TestCase()
+    test.setUp()
+    test.test1()
+    test.test2()
+    test.tearDown()

libxml2/python/tests/tstURI.py

@@ -0,0 +1,41 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+uri = libxml2.parseURI("http://example.org:8088/foo/bar?query=simple#fragid")
+if uri.scheme() != 'http':
+    print("Error parsing URI: wrong scheme")
+    sys.exit(1)
+if uri.server() != 'example.org':
+    print("Error parsing URI: wrong server")
+    sys.exit(1)
+if uri.port() != 8088:
+    print("Error parsing URI: wrong port")
+    sys.exit(1)
+if uri.path() != '/foo/bar':
+    print("Error parsing URI: wrong path")
+    sys.exit(1)
+if uri.query() != 'query=simple':
+    print("Error parsing URI: wrong query")
+    sys.exit(1)
+if uri.fragment() != 'fragid':
+    print("Error parsing URI: wrong query")
+    sys.exit(1)
+uri.setScheme("https")
+uri.setPort(223)
+uri.setFragment(None)
+result=uri.saveUri()
+if result != "https://example.org:223/foo/bar?query=simple":
+    print("Error modifying or saving the URI")
+uri = None
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/tstmem.py

@@ -0,0 +1,41 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+
+try:
+    import libxml2mod
+except ModuleNotFoundError:
+    from libxmlmods import libxml2mod
+
+import sys
+
+def error(msg, data):
+    pass
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+dtd="""<!ELEMENT foo EMPTY>"""
+instance="""<?xml version="1.0"?>
+<foo></foo>"""
+
+dtd = libxml2.parseDTD(None, 'test.dtd')
+ctxt = libxml2.newValidCtxt()
+libxml2mod.xmlSetValidErrors(ctxt._o, error, error)
+doc = libxml2.parseDoc(instance)
+ret = doc.validateDtd(ctxt, dtd)
+if ret != 1:
+    print("error doing DTD validation")
+    sys.exit(1)
+
+doc.freeDoc()
+dtd.freeDtd()
+del dtd
+del ctxt
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/tstxpath.py

@@ -0,0 +1,63 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+#memory debug specific
+libxml2.debugMemory(1)
+
+called = ""
+
+def foo(ctx, x):
+    global called
+
+    #
+    # test that access to the XPath evaluation contexts
+    #
+    pctxt = libxml2.xpathParserContext(_obj=ctx)
+    ctxt = pctxt.context()
+    called = ctxt.function()
+    return x + 1
+
+def bar(ctxt, x):
+    return "%d" % (x + 2)
+
+doc = libxml2.parseFile("tst.xml")
+ctxt = doc.xpathNewContext()
+res = ctxt.xpathEval("//*")
+if len(res) != 2:
+    print("xpath query: wrong node set size")
+    sys.exit(1)
+if res[0].name != "doc" or res[1].name != "foo":
+    print("xpath query: wrong node set value")
+    sys.exit(1)
+libxml2.registerXPathFunction(ctxt._o, "foo", None, foo)
+libxml2.registerXPathFunction(ctxt._o, "bar", None, bar)
+i = 10000
+while i > 0:
+    res = ctxt.xpathEval("foo(1)")
+    if res != 2:
+        print("xpath extension failure")
+        sys.exit(1)
+    i = i - 1
+i = 10000
+while i > 0:
+    res = ctxt.xpathEval("bar(1)")
+    if res != "3":
+        print("xpath extension failure got %s expecting '3'")
+        sys.exit(1)
+    i = i - 1
+doc.freeDoc()
+ctxt.xpathFreeContext()
+
+if called != "foo":
+    print("xpath function: failed to access the context")
+    print("xpath function: %s" % (called))
+    sys.exit(1)
+
+#memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/unicode.py

@@ -0,0 +1,61 @@
+from io import BytesIO, StringIO
+import libxml2
+import sys
+from xml.sax.handler import ContentHandler
+from xml.sax.xmlreader import InputSource
+import xml.sax
+
+
+# Test data: an XML file with a 100,000 Unicode smileys, which expand
+# into 400,000 bytes after UTF-8 encoding.
+SMILEY = '\U0001f600'
+TEXT = 100_000 * SMILEY
+XML_STRING = '<?xml version="1.0" encoding="UTF-8" ?>\n<root>' + TEXT + '</root>'
+XML_BYTES = XML_STRING.encode('utf-8')
+
+
+def RunTest(test_name, source):
+    expected = TEXT
+    received = ''
+
+    class TestHandler(ContentHandler):
+        def characters(self, content):
+            nonlocal received
+            received += content
+
+    reader = xml.sax.make_parser(['drv_libxml2'])
+    reader.setContentHandler(TestHandler())
+    reader.parse(source)
+    if expected != received:
+        print(test_name, 'failed!')
+        print('Expected length:', len(expected))
+        print('Received length:', len(received))
+        print('Expected text: (prefix only)', expected[:100])
+        print('Received text: (prefix only)', received[:100])
+        sys.exit(1)
+
+
+def TestBytesInput():
+    source = InputSource()
+    source.setByteStream(BytesIO(XML_BYTES))
+    RunTest('TestBytesInput', source)
+
+
+def TestStringInput():
+    source = InputSource()
+    source.setCharacterStream(StringIO(XML_STRING))
+    RunTest('TestStringInput', source)
+
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+TestBytesInput()
+TestStringInput()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/valid.xml

@@ -0,0 +1,4 @@
+<!DOCTYPE doc [
+<!ELEMENT doc EMPTY>
+]>
+<doc/>

libxml2/python/tests/validDTD.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+ARG = 'test string'
+
+class ErrorHandler:
+
+    def __init__(self):
+        self.errors = []
+
+    def handler(self, msg, data):
+        if data != ARG:
+            raise Exception("Error handler did not receive correct argument")
+        self.errors.append(msg)
+
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+dtd="""<!ELEMENT foo EMPTY>"""
+valid="""<?xml version="1.0"?>
+<foo></foo>"""
+
+invalid="""<?xml version="1.0"?>
+<foo><bar/></foo>"""
+
+dtd = libxml2.parseDTD(None, 'test.dtd')
+ctxt = libxml2.newValidCtxt()
+e = ErrorHandler()
+ctxt.setValidityErrorHandler(e.handler, e.handler, ARG)
+
+# Test valid document
+doc = libxml2.parseDoc(valid)
+ret = doc.validateDtd(ctxt, dtd)
+if ret != 1 or e.errors:
+    print("error doing DTD validation")
+    sys.exit(1)
+doc.freeDoc()
+
+# Test invalid document
+doc = libxml2.parseDoc(invalid)
+ret = doc.validateDtd(ctxt, dtd)
+if ret != 0 or not e.errors:
+    print("Error: document supposed to be invalid")
+doc.freeDoc()
+
+dtd.freeDtd()
+del dtd
+del ctxt
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/validRNG.py

@@ -0,0 +1,76 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+ARG = 'test string'
+
+class ErrorHandler:
+
+    def __init__(self):
+        self.errors = []
+
+    def handler(self, msg, data):
+        if data != ARG:
+            raise Exception("Error handler did not receive correct argument")
+        self.errors.append(msg)
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+schema="""<?xml version="1.0"?>
+<element name="foo"
+         xmlns="http://relaxng.org/ns/structure/1.0"
+         xmlns:a="http://relaxng.org/ns/annotation/1.0"
+         xmlns:ex1="http://www.example.com/n1"
+         xmlns:ex2="http://www.example.com/n2">
+  <a:documentation>A foo element.</a:documentation>
+  <element name="ex1:bar1">
+    <empty/>
+  </element>
+  <element name="ex2:bar2">
+    <empty/>
+  </element>
+</element>
+"""
+
+valid="""<?xml version="1.0"?>
+<foo><pre1:bar1 xmlns:pre1="http://www.example.com/n1"/><pre2:bar2 xmlns:pre2="http://www.example.com/n2"/></foo>"""
+
+invalid="""<?xml version="1.0"?>
+<foo><pre1:bar1 xmlns:pre1="http://www.example.com/n1">bad</pre1:bar1><pre2:bar2 xmlns:pre2="http://www.example.com/n2"/></foo>"""
+
+rngp = libxml2.relaxNGNewMemParserCtxt(schema, len(schema))
+rngs = rngp.relaxNGParse()
+ctxt = rngs.relaxNGNewValidCtxt()
+e = ErrorHandler()
+ctxt.setValidityErrorHandler(e.handler, e.handler, ARG)
+
+# Test valid document
+doc = libxml2.parseDoc(valid)
+ret = doc.relaxNGValidateDoc(ctxt)
+if ret != 0 or e.errors:
+    print("error doing RelaxNG validation")
+    sys.exit(1)
+doc.freeDoc()
+
+# Test invalid document
+doc = libxml2.parseDoc(invalid)
+ret = doc.relaxNGValidateDoc(ctxt)
+if ret == 0 or not e.errors:
+    print("Error: document supposed to be RelaxNG invalid")
+    sys.exit(1)
+doc.freeDoc()
+
+del rngp
+del rngs
+del ctxt
+libxml2.relaxNGCleanupTypes()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/validSchemas.py

@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+ARG = 'test string'
+
+class ErrorHandler:
+
+    def __init__(self):
+        self.errors = []
+
+    def handler(self, msg, data):
+        if data != ARG:
+            raise Exception("Error handler did not receive correct argument")
+        self.errors.append(msg)
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+schema="""<?xml version="1.0" encoding="iso-8859-1"?>
+<schema xmlns = "http://www.w3.org/2001/XMLSchema">
+	<element name = "Customer">
+		<complexType>
+			<sequence>
+				<element name = "FirstName" type = "string" />
+				<element name = "MiddleInitial" type = "string" />
+				<element name = "LastName" type = "string" />
+			</sequence>
+			<attribute name = "customerID" type = "integer" />
+		</complexType>
+	</element>
+</schema>"""
+
+valid="""<?xml version="1.0" encoding="iso-8859-1"?>
+<Customer customerID = "24332">
+	<FirstName>Raymond</FirstName>
+	<MiddleInitial>G</MiddleInitial>
+	<LastName>Bayliss</LastName>
+</Customer>	
+"""
+
+invalid="""<?xml version="1.0" encoding="iso-8859-1"?>
+<Customer customerID = "24332">
+	<MiddleInitial>G</MiddleInitial>
+	<LastName>Bayliss</LastName>
+</Customer>	
+"""
+
+e = ErrorHandler()
+ctxt_parser = libxml2.schemaNewMemParserCtxt(schema, len(schema))
+ctxt_schema = ctxt_parser.schemaParse()
+ctxt_valid  = ctxt_schema.schemaNewValidCtxt()
+ctxt_valid.setValidityErrorHandler(e.handler, e.handler, ARG)
+
+# Test valid document
+doc = libxml2.parseDoc(valid)
+ret = doc.schemaValidateDoc(ctxt_valid)
+if ret != 0 or e.errors:
+    print("error doing schema validation")
+    sys.exit(1)
+doc.freeDoc()
+
+# Test invalid document
+doc = libxml2.parseDoc(invalid)
+ret = doc.schemaValidateDoc(ctxt_valid)
+if ret == 0 or not e.errors:
+    print("Error: document supposer to be schema invalid")
+    sys.exit(1)
+doc.freeDoc()
+
+del ctxt_parser
+del ctxt_schema
+del ctxt_valid
+libxml2.schemaCleanupTypes()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
+

libxml2/python/tests/validate.py

@@ -0,0 +1,82 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+ctxt = libxml2.createFileParserCtxt("valid.xml")
+ctxt.validate(1)
+ctxt.parseDocument()
+doc = ctxt.doc()
+valid = ctxt.isValid()
+
+if doc.name != "valid.xml":
+    print("doc.name failed")
+    sys.exit(1)
+root = doc.children
+if root.name != "doc":
+    print("root.name failed")
+    sys.exit(1)
+if valid != 1:
+    print("validity check failed")
+    sys.exit(1)
+doc.freeDoc()
+
+i = 1000
+while i > 0:
+    ctxt = libxml2.createFileParserCtxt("valid.xml")
+    ctxt.validate(1)
+    ctxt.parseDocument()
+    doc = ctxt.doc()
+    valid = ctxt.isValid()
+    doc.freeDoc()
+    if valid != 1:
+        print("validity check failed")
+        sys.exit(1)
+    i = i - 1
+
+#deactivate error messages from the validation
+def noerr(ctx, str):
+    pass
+
+libxml2.registerErrorHandler(noerr, None)
+
+ctxt = libxml2.createFileParserCtxt("invalid.xml")
+ctxt.validate(1)
+ctxt.parseDocument()
+doc = ctxt.doc()
+valid = ctxt.isValid()
+if doc.name != "invalid.xml":
+    print("doc.name failed")
+    sys.exit(1)
+root = doc.children
+if root.name != "doc":
+    print("root.name failed")
+    sys.exit(1)
+if valid != 0:
+    print("validity check failed")
+    sys.exit(1)
+doc.freeDoc()
+
+i = 1000
+while i > 0:
+    ctxt = libxml2.createFileParserCtxt("invalid.xml")
+    ctxt.validate(1)
+    ctxt.parseDocument()
+    doc = ctxt.doc()
+    valid = ctxt.isValid()
+    doc.freeDoc()
+    if valid != 0:
+        print("validity check failed")
+        sys.exit(1)
+    i = i - 1
+del ctxt
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/walker.py

@@ -0,0 +1,144 @@
+#!/usr/bin/env python3
+#
+# this tests the entities substitutions with the XmlTextReader interface
+#
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+result = ""
+def processNode(reader):
+    global result
+
+    result = result + "%d %d %s %d\n" % (reader.Depth(), reader.NodeType(),
+			   reader.Name(), reader.IsEmptyElement())
+
+#
+# Parse a document testing the readerForxxx API
+#
+docstr="""<foo>
+<label>some text</label>
+<item>100</item>
+</foo>"""
+expect="""0 1 foo 0
+1 14 #text 0
+1 1 label 0
+2 3 #text 0
+1 15 label 0
+1 14 #text 0
+1 1 item 0
+2 3 #text 0
+1 15 item 0
+1 14 #text 0
+0 15 foo 0
+"""
+result = ""
+
+doc = libxml2.parseDoc(docstr)
+reader = doc.readerWalker();
+ret = reader.Read()
+while ret == 1:
+    processNode(reader)
+    ret = reader.Read()
+
+if ret != 0:
+    print("Error parsing the document test1")
+    sys.exit(1)
+
+if result != expect:
+    print("Unexpected result for test1")
+    print(result)
+    sys.exit(1)
+
+doc.freeDoc()
+
+#
+# Reuse the reader for another document testing the ReaderNewWalker API
+#
+docstr="""<foo>
+<label>some text</label>
+<item>1000</item>
+</foo>"""
+expect="""0 1 foo 0
+1 14 #text 0
+1 1 label 0
+2 3 #text 0
+1 15 label 0
+1 14 #text 0
+1 1 item 0
+2 3 #text 0
+1 15 item 0
+1 14 #text 0
+0 15 foo 0
+"""
+result = ""
+
+doc = libxml2.parseDoc(docstr)
+reader.NewWalker(doc)
+
+ret = reader.Read()
+while ret == 1:
+    processNode(reader)
+    ret = reader.Read()
+
+if ret != 0:
+    print("Error parsing the document test2")
+    sys.exit(1)
+
+if result != expect:
+    print("Unexpected result for test2")
+    print(result)
+    sys.exit(1)
+
+doc.freeDoc()
+
+#
+# Reuse the reader for another document testing the ReaderNewxxx API
+#
+docstr="""<foo>
+<label>some text</label>
+<item>1000</item>
+</foo>"""
+expect="""0 1 foo 0
+1 14 #text 0
+1 1 label 0
+2 3 #text 0
+1 15 label 0
+1 14 #text 0
+1 1 item 0
+2 3 #text 0
+1 15 item 0
+1 14 #text 0
+0 15 foo 0
+"""
+result = ""
+
+reader.NewDoc(docstr, "test3", None, 0)
+ret = reader.Read()
+while ret == 1:
+    processNode(reader)
+    ret = reader.Read()
+
+if ret != 0:
+    print("Error parsing the document test3")
+    sys.exit(1)
+
+if result != expect:
+    print("Unexpected result for test3")
+    print(result)
+    sys.exit(1)
+
+#
+# cleanup
+#
+del reader
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/xpath.py

@@ -0,0 +1,51 @@
+#!/usr/bin/env python3
+#
+# this test exercise the XPath basic engine, parser, etc, and
+# allows to detect memory leaks
+#
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+doc = libxml2.parseFile("tst.xml")
+if doc.name != "tst.xml":
+    print("doc.name error")
+    sys.exit(1);
+
+ctxt = doc.xpathNewContext()
+res = ctxt.xpathEval("//*")
+if len(res) != 2:
+    print("xpath query: wrong node set size")
+    sys.exit(1)
+if res[0].name != "doc" or res[1].name != "foo":
+    print("xpath query: wrong node set value")
+    sys.exit(1)
+ctxt.setContextNode(res[0])
+res = ctxt.xpathEval("foo")
+if len(res) != 1:
+    print("xpath query: wrong node set size")
+    sys.exit(1)
+if res[0].name != "foo":
+    print("xpath query: wrong node set value")
+    sys.exit(1)
+doc.freeDoc()
+ctxt.xpathFreeContext()
+i = 1000
+while i > 0:
+    doc = libxml2.parseFile("tst.xml")
+    ctxt = doc.xpathNewContext()
+    res = ctxt.xpathEval("//*")
+    doc.freeDoc()
+    ctxt.xpathFreeContext()
+    i = i -1
+del ctxt
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/xpathext.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+def foo(ctx, x):
+    return x + 1
+
+def bar(ctx, x):
+    return "%d" % (x + 2)
+
+doc = libxml2.parseFile("tst.xml")
+ctxt = doc.xpathNewContext()
+res = ctxt.xpathEval("//*")
+if len(res) != 2:
+    print("xpath query: wrong node set size")
+    sys.exit(1)
+if res[0].name != "doc" or res[1].name != "foo":
+    print("xpath query: wrong node set value")
+    sys.exit(1)
+
+libxml2.registerXPathFunction(ctxt._o, "foo", None, foo)
+libxml2.registerXPathFunction(ctxt._o, "bar", None, bar)
+i = 10000
+while i > 0:
+    res = ctxt.xpathEval("foo(1)")
+    if res != 2:
+        print("xpath extension failure")
+        sys.exit(1)
+    i = i - 1
+i = 10000
+while i > 0:
+    res = ctxt.xpathEval("bar(1)")
+    if res != "3":
+        print("xpath extension failure got %s expecting '3'")
+        sys.exit(1)
+    i = i - 1
+doc.freeDoc()
+ctxt.xpathFreeContext()
+
+# Memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/xpathleak.py

@@ -0,0 +1,55 @@
+#!/usr/bin/env python3
+import setup_test
+import libxml2
+import sys
+
+libxml2.debugMemory(True)
+
+expect="""--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+--> Invalid expression
+"""
+err=""
+def callback(ctx, str):
+     global err
+
+     err = err + "%s %s" % (ctx, str)
+
+libxml2.registerErrorHandler(callback, "-->")
+
+doc = libxml2.parseDoc("<fish/>")
+ctxt = doc.xpathNewContext()
+ctxt.setContextNode(doc)
+badexprs = (
+	":false()", "bad:()", "bad(:)", ":bad(:)", "bad:(:)", "bad:bad(:)",
+	"a:/b", "/c:/d", "//e:/f", "g://h"
+	)
+for expr in badexprs:
+	try:
+		ctxt.xpathEval(expr)
+	except libxml2.xpathError:
+	        pass
+	else:
+		print("Unexpectedly legal expression:", expr)
+ctxt.xpathFreeContext()
+doc.freeDoc()
+
+if err != expect:
+    print("error")
+    print("received %s" %(err))
+    print("expected %s" %(expect))
+    sys.exit(1)
+
+libxml2.cleanupParser()
+leakedbytes = libxml2.debugMemory(True)
+if leakedbytes == 0:
+	print("OK")
+else:
+	print("Memory leak", leakedbytes, "bytes")

libxml2/python/tests/xpathns.py

@@ -0,0 +1,29 @@
+#!/usr/bin/env python3
+#
+import setup_test
+import libxml2
+
+expect=' xmlns:a="urn:whatevar"'
+
+# Memory debug specific
+libxml2.debugMemory(1)
+
+d = libxml2.parseDoc("<a:a xmlns:a='urn:whatevar'/>")
+res=""
+for n in d.xpathEval("//namespace::*"):
+    res = res + n.serialize()
+del n
+d.freeDoc()
+
+if res != expect:
+    print("test5 failed: unexpected output")
+    print(res)
+del res
+del d
+# Memory debug specific
+libxml2.cleanupParser()
+
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/tests/xpathret.py

@@ -0,0 +1,57 @@
+#!/usr/bin/env python3
+import sys
+import setup_test
+import libxml2
+
+#memory debug specific
+libxml2.debugMemory(1)
+
+#
+# A document hosting the nodes returned from the extension function
+#
+mydoc = libxml2.newDoc("1.0")
+
+def foo(ctx, str):
+    global mydoc
+
+    #
+    # test returning a node set works as expected
+    #
+    parent = mydoc.newDocNode(None, 'p', None)
+    mydoc.addChild(parent)
+    node = mydoc.newDocText(str)
+    parent.addChild(node)
+    return [parent]
+
+doc = libxml2.parseFile("tst.xml")
+ctxt = doc.xpathNewContext()
+libxml2.registerXPathFunction(ctxt._o, "foo", None, foo)
+res = ctxt.xpathEval("foo('hello')")
+if type(res) != type([]):
+    print("Failed to return a nodeset")
+    sys.exit(1)
+if len(res) != 1:
+    print("Unexpected nodeset size")
+    sys.exit(1)
+node = res[0]
+if node.name != 'p':
+    print("Unexpected nodeset element result")
+    sys.exit(1)
+node = node.children
+if node.type != 'text':
+    print("Unexpected nodeset element children type")
+    sys.exit(1)
+if node.content != 'hello':
+    print("Unexpected nodeset element children content")
+    sys.exit(1)
+
+doc.freeDoc()
+mydoc.freeDoc()
+ctxt.xpathFreeContext()
+
+#memory debug specific
+libxml2.cleanupParser()
+if libxml2.debugMemory(1) == 0:
+    print("OK")
+else:
+    print("Memory leak %d bytes" % (libxml2.debugMemory(1)))

libxml2/python/types.c

@@ -0,0 +1,895 @@
+/*
+ * types.c: converter functions between the internal representation
+ *          and the Python objects
+ *
+ * See Copyright for the status of this software.
+ *
+ * daniel@veillard.com
+ */
+#include "libxml_wrap.h"
+#include <libxml/xpathInternals.h>
+#include <string.h>
+
+#if PY_MAJOR_VERSION >= 3
+#define PY_IMPORT_STRING_SIZE PyUnicode_FromStringAndSize
+#define PY_IMPORT_STRING PyUnicode_FromString
+#define PY_IMPORT_INT PyLong_FromLong
+#else
+#define PY_IMPORT_STRING_SIZE PyString_FromStringAndSize
+#define PY_IMPORT_STRING PyString_FromString
+#define PY_IMPORT_INT PyInt_FromLong
+#endif
+
+#if PY_MAJOR_VERSION >= 3
+#include <stdio.h>
+#include <stdint.h>
+
+#ifdef _WIN32
+
+#include <windows.h>
+#include <crtdbg.h>
+
+/* Taken from info on MSDN site, as we may not have the Windows WDK/DDK headers */
+typedef struct _IO_STATUS_BLOCK {
+  union {
+    NTSTATUS Status;
+    PVOID    Pointer;
+  } DUMMYUNIONNAME;
+  ULONG_PTR Information;
+} IO_STATUS_BLOCK;
+
+typedef struct _FILE_ACCESS_INFORMATION {
+  ACCESS_MASK AccessFlags;
+} FILE_ACCESS_INFORMATION;
+
+typedef NTSTATUS (*t_NtQueryInformationFile) (HANDLE           FileHandle,
+                                              IO_STATUS_BLOCK *IoStatusBlock,
+                                              PVOID            FileInformation,
+                                              ULONG            Length,
+                                              int              FileInformationClass); /* this is an Enum */
+
+#if (defined (_MSC_VER) && _MSC_VER >= 1400)
+/*
+ * This is the (empty) invalid parameter handler
+ * that is used for Visual C++ 2005 (and later) builds
+ * so that we can use this instead of the system automatically
+ * aborting the process.
+ *
+ * This is necessary as we use _get_oshandle() to check the validity
+ * of the file descriptors as we close them, so when an invalid file
+ * descriptor is passed into that function as we check on it, we get
+ * -1 as the result, instead of the gspawn helper program aborting.
+ *
+ * Please see http://msdn.microsoft.com/zh-tw/library/ks2530z6%28v=vs.80%29.aspx
+ * for an explanation on this.
+ */
+void
+myInvalidParameterHandler(const wchar_t *expression,
+                          const wchar_t *function,
+                          const wchar_t *file,
+                          unsigned int   line,
+                          uintptr_t      pReserved)
+{
+}
+#endif
+#else
+#include <unistd.h>
+#include <fcntl.h>
+#endif
+
+FILE *
+libxml_PyFileGet(PyObject *f) {
+    FILE *res;
+    const char *mode;
+    int fd = PyObject_AsFileDescriptor(f);
+
+#ifdef _WIN32
+    intptr_t w_fh = -1;
+    HMODULE hntdll = NULL;
+    IO_STATUS_BLOCK status_block;
+    FILE_ACCESS_INFORMATION ai;
+    t_NtQueryInformationFile NtQueryInformationFile;
+    BOOL is_read = FALSE;
+    BOOL is_write = FALSE;
+    BOOL is_append = FALSE;
+
+#if (defined (_MSC_VER) && _MSC_VER >= 1400)
+    /* set up our empty invalid parameter handler */
+    _invalid_parameter_handler oldHandler, newHandler;
+    newHandler = myInvalidParameterHandler;
+    oldHandler = _set_invalid_parameter_handler(newHandler);
+
+    /* Disable the message box for assertions. */
+    _CrtSetReportMode(_CRT_ASSERT, 0);
+#endif
+
+    w_fh = _get_osfhandle(fd);
+
+    if (w_fh == -1)
+        return(NULL);
+
+    hntdll = GetModuleHandleW(L"ntdll.dll");
+
+    if (hntdll == NULL)
+        return(NULL);
+    NtQueryInformationFile = (t_NtQueryInformationFile) (void (*)(void))
+        GetProcAddress(hntdll, "NtQueryInformationFile");
+
+    if (NtQueryInformationFile != NULL &&
+        (NtQueryInformationFile((HANDLE)w_fh,
+                               &status_block,
+                               &ai,
+                                sizeof(FILE_ACCESS_INFORMATION),
+                                8) == 0)) /* 8 means "FileAccessInformation" */
+        {
+            if (ai.AccessFlags & FILE_READ_DATA)
+                is_read = TRUE;
+            if (ai.AccessFlags & FILE_WRITE_DATA)
+                is_write = TRUE;
+            if (ai.AccessFlags & FILE_APPEND_DATA)
+                is_append = TRUE;
+
+            if (is_write) {
+                if (is_read) {
+                    if (is_append)
+                        mode = "a+";
+                    else
+                        mode = "rw";
+                } else {
+                    if (is_append)
+                        mode = "a";
+                    else
+                        mode = "w";
+                }
+            } else {
+                if (is_append)
+                    mode = "r+";
+                else
+                    mode = "r";
+            }
+        }
+
+    FreeLibrary(hntdll);
+
+    if (!is_write && !is_read) /* also happens if we did not load or run NtQueryInformationFile() successfully */
+        return(NULL);
+#else
+    int flags;
+
+    /*
+     * macOS returns O_RDWR for standard streams, but fails to write to
+     * stdout or stderr when opened with fdopen(dup_fd, "rw").
+     */
+    switch (fd) {
+        case STDIN_FILENO:
+            mode = "r";
+            break;
+        case STDOUT_FILENO:
+        case STDERR_FILENO:
+            mode = "w";
+            break;
+        default:
+            /*
+             * Get the flags on the fd to understand how it was opened
+             */
+            flags = fcntl(fd, F_GETFL, 0);
+            switch (flags & O_ACCMODE) {
+                case O_RDWR:
+                    if (flags & O_APPEND)
+                        mode = "a+";
+                    else
+                        mode = "rw";
+                    break;
+                case O_RDONLY:
+                    if (flags & O_APPEND)
+                        mode = "r+";
+                    else
+                        mode = "r";
+                    break;
+                case O_WRONLY:
+                    if (flags & O_APPEND)
+                        mode = "a";
+                    else
+                        mode = "w";
+                    break;
+                default:
+                    return(NULL);
+            }
+    }
+#endif
+
+    /*
+     * the FILE struct gets a new fd, so that it can be closed
+     * independently of the file descriptor given. The risk though is
+     * lack of sync. So at the python level sync must be implemented
+     * before and after a conversion took place. No way around it
+     * in the Python3 infrastructure !
+     * The duplicated fd and FILE * will be released in the subsequent
+     * call to libxml_PyFileRelease() which must be generated accordingly
+     */
+    fd = dup(fd);
+    if (fd == -1)
+        return(NULL);
+    res = fdopen(fd, mode);
+    if (res == NULL) {
+        close(fd);
+	return(NULL);
+    }
+    return(res);
+}
+
+void libxml_PyFileRelease(FILE *f) {
+    if (f != NULL)
+        fclose(f);
+}
+#endif
+
+PyObject *
+libxml_intWrap(int val)
+{
+    PyObject *ret;
+
+    ret = PY_IMPORT_INT((long) val);
+    return (ret);
+}
+
+PyObject *
+libxml_longWrap(long val)
+{
+    PyObject *ret;
+
+    ret = PyLong_FromLong(val);
+    return (ret);
+}
+
+PyObject *
+libxml_doubleWrap(double val)
+{
+    PyObject *ret;
+
+    ret = PyFloat_FromDouble((double) val);
+    return (ret);
+}
+
+PyObject *
+libxml_charPtrWrap(char *str)
+{
+    PyObject *ret;
+
+    if (str == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PY_IMPORT_STRING(str);
+    xmlFree(str);
+    return (ret);
+}
+
+PyObject *
+libxml_charPtrConstWrap(const char *str)
+{
+    PyObject *ret;
+
+    if (str == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PY_IMPORT_STRING(str);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlCharPtrWrap(xmlChar * str)
+{
+    PyObject *ret;
+
+    if (str == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PY_IMPORT_STRING((char *) str);
+    xmlFree(str);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlCharPtrConstWrap(const xmlChar * str)
+{
+    PyObject *ret;
+
+    if (str == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PY_IMPORT_STRING((char *) str);
+    return (ret);
+}
+
+PyObject *
+libxml_constcharPtrWrap(const char *str)
+{
+    PyObject *ret;
+
+    if (str == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PY_IMPORT_STRING(str);
+    return (ret);
+}
+
+PyObject *
+libxml_constxmlCharPtrWrap(const xmlChar * str)
+{
+    PyObject *ret;
+
+    if (str == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PY_IMPORT_STRING((char *) str);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlDocPtrWrap(xmlDocPtr doc)
+{
+    PyObject *ret;
+
+    if (doc == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    /* TODO: look at deallocation */
+    ret = PyCapsule_New((void *) doc, (char *) "xmlDocPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlNodePtrWrap(xmlNodePtr node)
+{
+    PyObject *ret;
+
+    if (node == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *) node, (char *) "xmlNodePtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlURIPtrWrap(xmlURIPtr uri)
+{
+    PyObject *ret;
+
+    if (uri == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *) uri, (char *) "xmlURIPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlNsPtrWrap(xmlNsPtr ns)
+{
+    PyObject *ret;
+
+    if (ns == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *) ns, (char *) "xmlNsPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlAttrPtrWrap(xmlAttrPtr attr)
+{
+    PyObject *ret;
+
+    if (attr == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *) attr, (char *) "xmlAttrPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlAttributePtrWrap(xmlAttributePtr attr)
+{
+    PyObject *ret;
+
+    if (attr == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *) attr, (char *) "xmlAttributePtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlElementPtrWrap(xmlElementPtr elem)
+{
+    PyObject *ret;
+
+    if (elem == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *) elem, (char *) "xmlElementPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlParserCtxtPtrWrap(xmlParserCtxtPtr ctxt)
+{
+    PyObject *ret;
+
+    if (ctxt == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+
+    ret = PyCapsule_New((void *) ctxt, (char *) "xmlParserCtxtPtr", NULL);
+    return (ret);
+}
+
+#ifdef LIBXML_XPATH_ENABLED
+PyObject *
+libxml_xmlXPathContextPtrWrap(xmlXPathContextPtr ctxt)
+{
+    PyObject *ret;
+
+    if (ctxt == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *) ctxt, (char *) "xmlXPathContextPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlXPathParserContextPtrWrap(xmlXPathParserContextPtr ctxt)
+{
+    PyObject *ret;
+
+    if (ctxt == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret = PyCapsule_New((void *)ctxt, (char *)"xmlXPathParserContextPtr", NULL);
+    return (ret);
+}
+
+/**
+ * libxml_xmlXPathDestructNsNode:
+ * cap: xmlNsPtr namespace node capsule object
+ *
+ * This function is called if and when a namespace node returned in
+ * an XPath node set is to be destroyed. That's the only kind of
+ * object returned in node set not directly linked to the original
+ * xmlDoc document, see xmlXPathNodeSetDupNs.
+ */
+#if PY_VERSION_HEX < 0x02070000
+static void
+libxml_xmlXPathDestructNsNode(void *cap, void *desc ATTRIBUTE_UNUSED)
+#else
+static void
+libxml_xmlXPathDestructNsNode(PyObject *cap)
+#endif
+{
+#if PY_VERSION_HEX < 0x02070000
+    xmlXPathNodeSetFreeNs((xmlNsPtr) cap);
+#else
+    xmlXPathNodeSetFreeNs((xmlNsPtr) PyCapsule_GetPointer(cap, "xmlNsPtr"));
+#endif
+}
+
+PyObject *
+libxml_xmlXPathObjectPtrWrap(xmlXPathObjectPtr obj)
+{
+    PyObject *ret;
+
+    if (obj == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    switch (obj->type) {
+        case XPATH_XSLT_TREE: {
+            if ((obj->nodesetval == NULL) ||
+		(obj->nodesetval->nodeNr == 0) ||
+		(obj->nodesetval->nodeTab == NULL)) {
+                ret = PyList_New(0);
+	    } else {
+		int i, len = 0;
+		xmlNodePtr node;
+
+		node = obj->nodesetval->nodeTab[0]->children;
+		while (node != NULL) {
+		    len++;
+		    node = node->next;
+		}
+		ret = PyList_New(len);
+		node = obj->nodesetval->nodeTab[0]->children;
+		for (i = 0;i < len;i++) {
+                    PyList_SetItem(ret, i, libxml_xmlNodePtrWrap(node));
+		    node = node->next;
+		}
+	    }
+	    /*
+	     * Return now, do not free the object passed down
+	     */
+	    return (ret);
+	}
+        case XPATH_NODESET:
+            if ((obj->nodesetval == NULL)
+                || (obj->nodesetval->nodeNr == 0)) {
+                ret = PyList_New(0);
+	    } else {
+                int i;
+                xmlNodePtr node;
+
+                ret = PyList_New(obj->nodesetval->nodeNr);
+                for (i = 0; i < obj->nodesetval->nodeNr; i++) {
+                    node = obj->nodesetval->nodeTab[i];
+                    if (node->type == XML_NAMESPACE_DECL) {
+		        PyObject *ns = PyCapsule_New((void *) node,
+                                     (char *) "xmlNsPtr",
+				     libxml_xmlXPathDestructNsNode);
+			PyList_SetItem(ret, i, ns);
+			/* make sure the xmlNsPtr is not destroyed now */
+			obj->nodesetval->nodeTab[i] = NULL;
+		    } else {
+			PyList_SetItem(ret, i, libxml_xmlNodePtrWrap(node));
+		    }
+                }
+            }
+            break;
+        case XPATH_BOOLEAN:
+            ret = PY_IMPORT_INT((long) obj->boolval);
+            break;
+        case XPATH_NUMBER:
+            ret = PyFloat_FromDouble(obj->floatval);
+            break;
+        case XPATH_STRING:
+	    ret = PY_IMPORT_STRING((char *) obj->stringval);
+            break;
+        default:
+            Py_INCREF(Py_None);
+            ret = Py_None;
+    }
+    xmlXPathFreeObject(obj);
+    return (ret);
+}
+
+xmlXPathObjectPtr
+libxml_xmlXPathObjectPtrConvert(PyObject *obj)
+{
+    xmlXPathObjectPtr ret = NULL;
+
+    if (obj == NULL) {
+        return (NULL);
+    }
+    if (PyFloat_Check (obj)) {
+        ret = xmlXPathNewFloat((double) PyFloat_AS_DOUBLE(obj));
+    } else if (PyLong_Check(obj)) {
+#ifdef PyLong_AS_LONG
+        ret = xmlXPathNewFloat((double) PyLong_AS_LONG(obj));
+#else
+        ret = xmlXPathNewFloat((double) PyInt_AS_LONG(obj));
+#endif
+#ifdef PyBool_Check
+    } else if (PyBool_Check (obj)) {
+
+        if (obj == Py_True) {
+          ret = xmlXPathNewBoolean(1);
+        }
+        else {
+          ret = xmlXPathNewBoolean(0);
+        }
+#endif
+    } else if (PyBytes_Check (obj)) {
+        xmlChar *str;
+
+        str = xmlStrndup((const xmlChar *) PyBytes_AS_STRING(obj),
+                         PyBytes_GET_SIZE(obj));
+        ret = xmlXPathWrapString(str);
+#ifdef PyUnicode_Check
+    } else if (PyUnicode_Check (obj)) {
+#if PY_VERSION_HEX >= 0x03030000
+        xmlChar *str;
+	const char *tmp;
+	Py_ssize_t size;
+
+	/* tmp doesn't need to be deallocated */
+        tmp = PyUnicode_AsUTF8AndSize(obj, &size);
+        str = xmlStrndup((const xmlChar *) tmp, (int) size);
+        ret = xmlXPathWrapString(str);
+#else
+        xmlChar *str = NULL;
+        PyObject *b;
+
+	b = PyUnicode_AsUTF8String(obj);
+	if (b != NULL) {
+	    str = xmlStrndup((const xmlChar *) PyBytes_AS_STRING(b),
+			     PyBytes_GET_SIZE(b));
+	    Py_DECREF(b);
+	}
+	ret = xmlXPathWrapString(str);
+#endif
+#endif
+    } else if (PyList_Check (obj)) {
+        int i;
+        PyObject *node;
+        xmlNodePtr cur;
+        xmlNodeSetPtr set;
+
+        set = xmlXPathNodeSetCreate(NULL);
+
+        for (i = 0; i < PyList_Size(obj); i++) {
+            node = PyList_GetItem(obj, i);
+            if ((node == NULL) || (node->ob_type == NULL))
+                continue;
+
+            cur = NULL;
+            if (PyCapsule_CheckExact(node)) {
+                cur = PyxmlNode_Get(node);
+            } else if ((PyObject_HasAttrString(node, (char *) "_o")) &&
+	               (PyObject_HasAttrString(node, (char *) "get_doc"))) {
+		PyObject *wrapper;
+
+		wrapper = PyObject_GetAttrString(node, (char *) "_o");
+		if (wrapper != NULL)
+		    cur = PyxmlNode_Get(wrapper);
+            } else {
+            }
+            if (cur != NULL) {
+                xmlXPathNodeSetAdd(set, cur);
+            }
+        }
+        ret = xmlXPathWrapNodeSet(set);
+    } else {
+    }
+    return (ret);
+}
+#endif /* LIBXML_XPATH_ENABLED */
+
+PyObject *
+libxml_xmlValidCtxtPtrWrap(xmlValidCtxtPtr valid)
+{
+	PyObject *ret;
+
+	if (valid == NULL) {
+		Py_INCREF(Py_None);
+		return (Py_None);
+	}
+
+	ret = 
+		PyCapsule_New((void *) valid,
+									 (char *) "xmlValidCtxtPtr", NULL);
+
+	return (ret);
+}
+
+#ifdef LIBXML_CATALOG_ENABLED
+PyObject *
+libxml_xmlCatalogPtrWrap(xmlCatalogPtr catal)
+{
+    PyObject *ret;
+
+    if (catal == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) catal,
+                                     (char *) "xmlCatalogPtr", NULL);
+    return (ret);
+}
+#endif /* LIBXML_CATALOG_ENABLED */
+
+PyObject *
+libxml_xmlOutputBufferPtrWrap(xmlOutputBufferPtr buffer)
+{
+    PyObject *ret;
+
+    if (buffer == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) buffer,
+                                     (char *) "xmlOutputBufferPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlParserInputBufferPtrWrap(xmlParserInputBufferPtr buffer)
+{
+    PyObject *ret;
+
+    if (buffer == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) buffer,
+                                     (char *) "xmlParserInputBufferPtr", NULL);
+    return (ret);
+}
+
+#ifdef LIBXML_REGEXP_ENABLED
+PyObject *
+libxml_xmlRegexpPtrWrap(xmlRegexpPtr regexp)
+{
+    PyObject *ret;
+
+    if (regexp == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) regexp,
+                                     (char *) "xmlRegexpPtr", NULL);
+    return (ret);
+}
+#endif /* LIBXML_REGEXP_ENABLED */
+
+#ifdef LIBXML_READER_ENABLED
+PyObject *
+libxml_xmlTextReaderPtrWrap(xmlTextReaderPtr reader)
+{
+    PyObject *ret;
+
+    if (reader == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) reader,
+                                     (char *) "xmlTextReaderPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlTextReaderLocatorPtrWrap(xmlTextReaderLocatorPtr locator)
+{
+    PyObject *ret;
+
+    if (locator == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) locator,
+                                     (char *) "xmlTextReaderLocatorPtr", NULL);
+    return (ret);
+}
+#endif /* LIBXML_READER_ENABLED */
+
+#ifdef LIBXML_RELAXNG_ENABLED
+PyObject *
+libxml_xmlRelaxNGPtrWrap(xmlRelaxNGPtr ctxt)
+{
+    PyObject *ret;
+
+    if (ctxt == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) ctxt,
+                                     (char *) "xmlRelaxNGPtr", NULL);
+    return (ret);
+}
+
+PyObject *
+libxml_xmlRelaxNGParserCtxtPtrWrap(xmlRelaxNGParserCtxtPtr ctxt)
+{
+    PyObject *ret;
+
+    if (ctxt == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) ctxt,
+                                     (char *) "xmlRelaxNGParserCtxtPtr", NULL);
+    return (ret);
+}
+PyObject *
+libxml_xmlRelaxNGValidCtxtPtrWrap(xmlRelaxNGValidCtxtPtr valid)
+{
+    PyObject *ret;
+
+    if (valid == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    ret =
+        PyCapsule_New((void *) valid,
+                                     (char *) "xmlRelaxNGValidCtxtPtr", NULL);
+    return (ret);
+}
+#endif /* LIBXML_RELAXNG_ENABLED */
+
+#ifdef LIBXML_SCHEMAS_ENABLED
+PyObject *
+libxml_xmlSchemaPtrWrap(xmlSchemaPtr ctxt)
+{
+	PyObject *ret;
+
+	if (ctxt == NULL) {
+		Py_INCREF(Py_None);
+		return (Py_None);
+	}
+	ret =
+		PyCapsule_New((void *) ctxt,
+									 (char *) "xmlSchemaPtr", NULL);
+	return (ret);
+}
+
+PyObject *
+libxml_xmlSchemaParserCtxtPtrWrap(xmlSchemaParserCtxtPtr ctxt)
+{
+	PyObject *ret;
+
+	if (ctxt == NULL) {
+		Py_INCREF(Py_None);
+		return (Py_None);
+	}
+	ret = 
+		PyCapsule_New((void *) ctxt,
+									 (char *) "xmlSchemaParserCtxtPtr", NULL);
+
+	return (ret);
+}
+
+PyObject *
+libxml_xmlSchemaValidCtxtPtrWrap(xmlSchemaValidCtxtPtr valid)
+{
+	PyObject *ret;
+	
+	if (valid == NULL) {
+		Py_INCREF(Py_None);
+		return (Py_None);
+	}
+
+	ret = 
+		PyCapsule_New((void *) valid,
+									 (char *) "xmlSchemaValidCtxtPtr", NULL);
+
+	return (ret);
+}
+#endif /* LIBXML_SCHEMAS_ENABLED */
+
+static void
+libxml_xmlDestructError(PyObject *cap) {
+    xmlErrorPtr err = (xmlErrorPtr) PyCapsule_GetPointer(cap, "xmlErrorPtr");
+    xmlResetError(err);
+    xmlFree(err);
+}
+
+PyObject *
+libxml_xmlErrorPtrWrap(const xmlError *error)
+{
+    PyObject *ret;
+    xmlErrorPtr copy;
+
+    if (error == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    copy = xmlMalloc(sizeof(*copy));
+    if (copy == NULL) {
+        Py_INCREF(Py_None);
+        return (Py_None);
+    }
+    memset(copy, 0, sizeof(*copy));
+    xmlCopyError(error, copy);
+    ret = PyCapsule_New(copy, "xmlErrorPtr", libxml_xmlDestructError);
+    return (ret);
+}