# HG changeset patch
# User paulb
# Date 1174160812 0
# Node ID b2a2555d645c6a58ca137d510124134dbdf08791
# Parent  4cc51eb695eaf01f1c1f7a24c64edf23f032fa48
[project @ 2007-03-17 19:46:48 by paulb]
Changed the parse functions to expose an HTML encoding parameter for
overriding libxml2's encoding detection mechanisms.

diff -r 4cc51eb695ea -r b2a2555d645c libxml2dom/__init__.py
--- a/libxml2dom/__init__.py	Sat Mar 17 19:45:57 2007 +0000
+++ b/libxml2dom/__init__.py	Sat Mar 17 19:46:52 2007 +0000
@@ -502,14 +502,15 @@
 def createDocument(namespaceURI, localName, doctype):
     return default_impl.createDocument(namespaceURI, localName, doctype)
 
-def parse(stream_or_string, html=0, impl=None):
+def parse(stream_or_string, html=0, htmlencoding=None, impl=None):
 
     """
     Parse the given 'stream_or_string', where the supplied object can either be
     a stream (such as a file or stream object), or a string (containing the
     filename of a document). If the optional 'html' parameter is set to a true
     value, the content to be parsed will be treated as being HTML rather than
-    XML.
+    XML. If the optional 'htmlencoding' is specified, HTML parsing will be
+    performed with the document encoding assumed to that specified.
 
     A document object is returned by this function.
     """
@@ -518,42 +519,48 @@
 
     if hasattr(stream_or_string, "read"):
         stream = stream_or_string
-        return parseString(stream.read(), html, impl)
+        return parseString(stream.read(), html, htmlencoding, impl)
     else:
-        return parseFile(stream_or_string, html, impl)
+        return parseFile(stream_or_string, html, htmlencoding, impl)
 
-def parseFile(filename, html=0, impl=None):
+def parseFile(filename, html=0, htmlencoding=None, impl=None):
 
     """
     Parse the file having the given 'filename'. If the optional 'html' parameter
     is set to a true value, the content to be parsed will be treated as being
-    HTML rather than XML.
+    HTML rather than XML. If the optional 'htmlencoding' is specified, HTML
+    parsing will be performed with the document encoding assumed to be that
+    specified.
 
     A document object is returned by this function.
     """
 
     impl = impl or default_impl
-    return impl.adoptDocument(Node_parseFile(filename, html))
+    return impl.adoptDocument(Node_parseFile(filename, html, htmlencoding))
 
-def parseString(s, html=0, impl=None):
+def parseString(s, html=0, htmlencoding=None, impl=None):
 
     """
     Parse the content of the given string 's'. If the optional 'html' parameter
     is set to a true value, the content to be parsed will be treated as being
-    HTML rather than XML.
+    HTML rather than XML. If the optional 'htmlencoding' is specified, HTML
+    parsing will be performed with the document encoding assumed to be that
+    specified.
 
     A document object is returned by this function.
     """
 
     impl = impl or default_impl
-    return impl.adoptDocument(Node_parseString(s, html))
+    return impl.adoptDocument(Node_parseString(s, html, htmlencoding))
 
-def parseURI(uri, html=0, impl=None):
+def parseURI(uri, html=0, htmlencoding=None, impl=None):
 
     """
     Parse the content found at the given 'uri'. If the optional 'html' parameter
     is set to a true value, the content to be parsed will be treated as being
-    HTML rather than XML.
+    HTML rather than XML. If the optional 'htmlencoding' is specified, HTML
+    parsing will be performed with the document encoding assumed to be that
+    specified.
 
     XML documents are retrieved using libxml2's own network capabilities; HTML
     documents are retrieved using the urllib module provided by Python. To
@@ -572,12 +579,12 @@
     if html:
         f = urllib.urlopen(uri)
         try:
-            return parse(f, html, impl)
+            return parse(f, html, htmlencoding, impl)
         finally:
             f.close()
     else:
         impl = impl or default_impl
-        return impl.adoptDocument(Node_parseURI(uri, html))
+        return impl.adoptDocument(Node_parseURI(uri, html, htmlencoding))
 
 def toString(node, encoding=None, prettyprint=0):
 
diff -r 4cc51eb695ea -r b2a2555d645c libxml2dom/macrolib/macrolib.py
--- a/libxml2dom/macrolib/macrolib.py	Sat Mar 17 19:45:57 2007 +0000
+++ b/libxml2dom/macrolib/macrolib.py	Sat Mar 17 19:46:52 2007 +0000
@@ -501,14 +501,14 @@
         libxml2mod.xmlCreateIntSubset(d, doctype.localName, doctype.publicId, doctype.systemId)
     return d
 
-def parse(stream_or_string, html=0):
+def parse(stream_or_string, html=0, htmlencoding=None):
     if hasattr(stream_or_string, "read"):
         stream = stream_or_string
-        return parseString(stream.read(), html)
+        return parseString(stream.read(), html, htmlencoding)
     else:
-        return parseFile(stream_or_string, html)
+        return parseFile(stream_or_string, html, htmlencoding)
 
-def parseFile(s, html=0):
+def parseFile(s, html=0, htmlencoding=None):
     # NOTE: Switching off validation and remote DTD resolution.
     if not html:
         context = libxml2mod.xmlCreateFileParserCtxt(s)
@@ -518,9 +518,9 @@
         libxml2mod.xmlParseDocument(context)
         return libxml2mod.xmlParserGetDoc(context)
     else:
-        return libxml2mod.htmlReadFile(s, None, HTML_PARSE_NOERROR | HTML_PARSE_NOWARNING | HTML_PARSE_NONET)
+        return libxml2mod.htmlReadFile(s, htmlencoding, HTML_PARSE_NOERROR | HTML_PARSE_NOWARNING | HTML_PARSE_NONET)
 
-def parseString(s, html=0):
+def parseString(s, html=0, htmlencoding=None):
     # NOTE: Switching off validation and remote DTD resolution.
     if not html:
         context = libxml2mod.xmlCreateMemoryParserCtxt(s, len(s))
@@ -532,10 +532,10 @@
     else:
         # NOTE: URL given as None.
         html_url = None
-        return libxml2mod.htmlReadMemory(s, len(s), html_url, None,
+        return libxml2mod.htmlReadMemory(s, len(s), html_url, htmlencoding,
             HTML_PARSE_NOERROR | HTML_PARSE_NOWARNING | HTML_PARSE_NONET)
 
-def parseURI(uri, html=0):
+def parseURI(uri, html=0, htmlencoding=None):
     # NOTE: Switching off validation and remote DTD resolution.
     if not html:
         context = libxml2mod.xmlCreateURLParserCtxt(uri, 0)
diff -r 4cc51eb695ea -r b2a2555d645c libxml2dom/svg.py
--- a/libxml2dom/svg.py	Sat Mar 17 19:45:57 2007 +0000
+++ b/libxml2dom/svg.py	Sat Mar 17 19:46:52 2007 +0000
@@ -98,17 +98,17 @@
 
 # Convenience functions.
 
-def parse(stream_or_string, html=0):
-    return libxml2dom.parse(stream_or_string, html, default_impl)
+def parse(stream_or_string, html=0, htmlencoding=None):
+    return libxml2dom.parse(stream_or_string, html, htmlencoding, default_impl)
 
-def parseFile(filename, html=0):
-    return libxml2dom.parseFile(filename, html, default_impl)
+def parseFile(filename, html=0, htmlencoding=None):
+    return libxml2dom.parseFile(filename, html, htmlencoding, default_impl)
 
-def parseString(s, html=0):
-    return libxml2dom.parseString(s, html, default_impl)
+def parseString(s, html=0, htmlencoding=None):
+    return libxml2dom.parseString(s, html, htmlencoding, default_impl)
 
-def parseURI(uri, html=0):
-    return libxml2dom.parseURI(uri, html, default_impl)
+def parseURI(uri, html=0, htmlencoding=None):
+    return libxml2dom.parseURI(uri, html, htmlencoding, default_impl)
 
 # Single instance of the implementation.