Changeset 391 for python/trunk/Lib/xml/sax/saxutils.py

Timestamp:

Mar 19, 2014, 11:31:01 PM (11 years ago)

Author:

dmik

Message:

python: Merge vendor 2.7.6 to trunk.

Location:

python/trunk

Files:

• . (modified) (1 prop)
• Lib/xml/sax/saxutils.py (modified) (7 diffs)

python/trunk/Lib/xml/sax/saxutils.py

@@ -5 +5 @@
 
 import os, urlparse, urllib, types
+import io
+import sys
 import handler
 import xmlreader
@@ -12 +14 @@
 except AttributeError:
     _StringTypes = [types.StringType]
-
-# See whether the xmlcharrefreplace error handler is
-# supported
-try:
-    from codecs import xmlcharrefreplace_errors
-    _error_handling = "xmlcharrefreplace"
-    del xmlcharrefreplace_errors
-except ImportError:
-    _error_handling = "strict"
 
 def __dict_replace(s, d):
@@ -82 +75 @@
 
 
+def _gettextwriter(out, encoding):
+    if out is None:
+        import sys
+        out = sys.stdout
+
+    if isinstance(out, io.RawIOBase):
+        buffer = io.BufferedIOBase(out)
+        # Keep the original file open when the TextIOWrapper is
+        # destroyed
+        buffer.close = lambda: None
+    else:
+        # This is to handle passed objects that aren't in the
+        # IOBase hierarchy, but just have a write method
+        buffer = io.BufferedIOBase()
+        buffer.writable = lambda: True
+        buffer.write = out.write
+        try:
+            # TextIOWrapper uses this methods to determine
+            # if BOM (for UTF-16, etc) should be added
+            buffer.seekable = out.seekable
+            buffer.tell = out.tell
+        except AttributeError:
+            pass
+    # wrap a binary writer with TextIOWrapper
+    class UnbufferedTextIOWrapper(io.TextIOWrapper):
+        def write(self, s):
+            super(UnbufferedTextIOWrapper, self).write(s)
+            self.flush()
+    return UnbufferedTextIOWrapper(buffer, encoding=encoding,
+                                   errors='xmlcharrefreplace',
+                                   newline='\n')
+
 class XMLGenerator(handler.ContentHandler):
 
     def __init__(self, out=None, encoding="iso-8859-1"):
-        if out is None:
-            import sys
-            out = sys.stdout
         handler.ContentHandler.__init__(self)
-        self._out = out
+        out = _gettextwriter(out, encoding)
+        self._write = out.write
+        self._flush = out.flush
         self._ns_contexts = [{}] # contains uri -> prefix dicts
         self._current_context = self._ns_contexts[-1]
@@ -95 +119 @@
         self._encoding = encoding
 
-    def _write(self, text):
-        if isinstance(text, str):
-            self._out.write(text)
-        else:
-            self._out.write(text.encode(self._encoding, _error_handling))
-
     def _qname(self, name):
         """Builds a qualified name from a (ns_url, localname) pair"""
         if name[0]:
+            # Per http://www.w3.org/XML/1998/namespace, The 'xml' prefix is
+            # bound by definition to http://www.w3.org/XML/1998/namespace.  It
+            # does not need to be declared and will not usually be found in
+            # self._current_context.
+            if 'http://www.w3.org/XML/1998/namespace' == name[0]:
+                return 'xml:' + name[1]
             # The name is in a non-empty namespace
             prefix = self._current_context[name[0]]
@@ -115 +139 @@
 
     def startDocument(self):
-        self._write('<?xml version="1.0" encoding="%s"?>\n' %
+        self._write(u'<?xml version="1.0" encoding="%s"?>\n' %
                         self._encoding)
+
+    def endDocument(self):
+        self._flush()
 
     def startPrefixMapping(self, prefix, uri):
@@ -128 +155 @@
 
     def startElement(self, name, attrs):
-        self._write('<' + name)
+        self._write(u'<' + name)
         for (name, value) in attrs.items():
-            self._write(' %s=%s' % (name, quoteattr(value)))
-        self._write('>')
+            self._write(u' %s=%s' % (name, quoteattr(value)))
+        self._write(u'>')
 
     def endElement(self, name):
-        self._write('</%s>' % name)
+        self._write(u'</%s>' % name)
 
     def startElementNS(self, name, qname, attrs):
-        self._write('<' + self._qname(name))
+        self._write(u'<' + self._qname(name))
 
         for prefix, uri in self._undeclared_ns_maps:
             if prefix:
-                self._out.write(' xmlns:%s="%s"' % (prefix, uri))
+                self._write(u' xmlns:%s="%s"' % (prefix, uri))
             else:
-                self._out.write(' xmlns="%s"' % uri)
+                self._write(u' xmlns="%s"' % uri)
         self._undeclared_ns_maps = []
 
         for (name, value) in attrs.items():
-            self._write(' %s=%s' % (self._qname(name), quoteattr(value)))
-        self._write('>')
+            self._write(u' %s=%s' % (self._qname(name), quoteattr(value)))
+        self._write(u'>')
 
     def endElementNS(self, name, qname):
-        self._write('</%s>' % self._qname(name))
+        self._write(u'</%s>' % self._qname(name))
 
     def characters(self, content):
+        if not isinstance(content, unicode):
+            content = unicode(content, self._encoding)
         self._write(escape(content))
 
     def ignorableWhitespace(self, content):
+        if not isinstance(content, unicode):
+            content = unicode(content, self._encoding)
         self._write(content)
 
     def processingInstruction(self, target, data):
-        self._write('<?%s %s?>' % (target, data))
+        self._write(u'<?%s %s?>' % (target, data))
 
 
@@ -288 +319 @@
 
     if source.getByteStream() is None:
-        sysid = source.getSystemId()
-        basehead = os.path.dirname(os.path.normpath(base))
-        sysidfilename = os.path.join(basehead, sysid)
-        if os.path.isfile(sysidfilename):
+        try:
+            sysid = source.getSystemId()
+            basehead = os.path.dirname(os.path.normpath(base))
+            encoding = sys.getfilesystemencoding()
+            if isinstance(sysid, unicode):
+                if not isinstance(basehead, unicode):
+                    try:
+                        basehead = basehead.decode(encoding)
+                    except UnicodeDecodeError:
+                        sysid = sysid.encode(encoding)
+            else:
+                if isinstance(basehead, unicode):
+                    try:
+                        sysid = sysid.decode(encoding)
+                    except UnicodeDecodeError:
+                        basehead = basehead.encode(encoding)
+            sysidfilename = os.path.join(basehead, sysid)
+            isfile = os.path.isfile(sysidfilename)
+        except UnicodeError:
+            isfile = False
+        if isfile:
             source.setSystemId(sysidfilename)
             f = open(sysidfilename, "rb")
         else:
-            source.setSystemId(urlparse.urljoin(base, sysid))
+            source.setSystemId(urlparse.urljoin(base, source.getSystemId()))
             f = urllib.urlopen(source.getSystemId())