Replaced the ToString class with a tostring function.

The sleekxmpp.xmlstream.tostring and sleekxmpp.xmlstream.tostring26 packages
have been merged to sleekxmpp.xmlstream.tostring. The __init__.py file will
import the appropriate tostring function depending on the Python version.

The setup.py file has been updated with the package changes.

ElementBase is now a direct descendent of object and does not subclass ToString.

Stanza objects now return their XML contents for __repr__.

setup.py

@@ -42,12 +42,8 @@ packages     = [ 'sleekxmpp',
 				 'sleekxmpp/stanza',
 				 'sleekxmpp/xmlstream',
 				 'sleekxmpp/xmlstream/matcher',
-				 'sleekxmpp/xmlstream/handler' ]
+				 'sleekxmpp/xmlstream/handler',
-
+                 'sleekxmpp/xmlstream/tostring']
-if sys.version_info < (3, 0):
-	packages.append('sleekxmpp/xmlstream/tostring26')
-else:
-	packages.append('sleekxmpp/xmlstream/tostring')
 
 setup(
     name             = "sleekxmpp",

sleekxmpp/basexmpp.py

@@ -25,6 +25,7 @@ from . stanza.roster import Roster
 from . stanza.nick import Nick
 from . stanza.htmlim import HTMLIM
 from . stanza.error import Error
+from sleekxmpp.xmlstream.tostring import tostring
 
 import logging
 import threading
@@ -128,7 +129,7 @@ class basexmpp(object):
 		return "%x".upper() % self.id
 
 	def sendXML(self, data, mask=None, timeout=10):
-		return self.send(self.tostring(data), mask, timeout)
+		return self.send(tostring(data), mask, timeout)
 
 	def send(self, data, mask=None, timeout=10):
 		#logging.warning("Deprecated send used for \"%s\"" % (data,))

sleekxmpp/xmlstream/stanzabase.py

@@ -12,10 +12,7 @@ import weakref
 import copy
 from . jid import JID
 
-if sys.version_info < (3,0):
+from sleekxmpp.xmlstream.tostring import tostring
-        from . import tostring26 as tostring
-else:
-        from . import tostring
 
 xmltester = type(ET.Element('xml'))
 
@@ -29,7 +26,7 @@ def registerStanzaPlugin(stanza, plugin):
         stanza.plugin_tag_map[tag] = plugin
 
 
-class ElementBase(tostring.ToString):
+class ElementBase(object):
         name = 'stanza'
         plugin_attrib = 'plugin'
         namespace = 'jabber:client'
@@ -306,6 +303,12 @@ class ElementBase(tostring.ToString):
         def __copy__(self):
                 return self.__class__(xml=copy.deepcopy(self.xml), parent=self.parent)
 
+        def __str__(self):
+                return tostring(self.xml, xmlns='', stanza_ns=self.namespace)
+
+        def __repr__(self):
+                return self.__str__()
+
         #def __del__(self): #prevents garbage collection of reference cycle
         #       if self.parent is not None:
         #               self.parent.xml.remove(self.xml)
@@ -391,3 +394,5 @@ class StanzaBase(ElementBase):
         def __copy__(self):
                 return self.__class__(xml=copy.deepcopy(self.xml), stream=self.stream)
 
+        def __str__(self):
+                return tostring(self.xml, xmlns='', stanza_ns=self.namespace, stream=self.stream)

sleekxmpp/xmlstream/tostring/__init__.py

@@ -1,14 +1,19 @@
 """
+    SleekXMPP: The Sleek XMPP Library
+    Copyright (C) 2010  Nathanael C. Fritz
+    This file is part of SleekXMPP.
 
+    See the file LICENSE for copying permission.
 """
 
 import sys
 
+# Import the correct tostring and xml_escape functions based on the Python
+# version in order to properly handle Unicode.
 
-# Import the correct ToString class based on the Python version.
 if sys.version_info < (3, 0):
-    from sleekxmpp.xmlstream.tostring.tostring26 import ToString
+    from sleekxmpp.xmlstream.tostring.tostring26 import tostring, xml_escape
 else:
-    from sleekxmpp.xmlstream.tostring.tostring import ToString
+    from sleekxmpp.xmlstream.tostring.tostring import tostring, xml_escape
 
-__all__ = ['ToString']
+__all__ = ['tostring', 'xml_escape']

sleekxmpp/xmlstream/tostring/tostring.py

@@ -1,60 +1,91 @@
+"""
+    SleekXMPP: The Sleek XMPP Library
+    Copyright (C) 2010  Nathanael C. Fritz
+    This file is part of SleekXMPP.
 
-class ToString(object):
+    See the file LICENSE for copying permission.
-	def __str__(self, xml=None, xmlns='', stringbuffer=''):
+"""
-		if xml is None:
-			xml = self.xml
-		newoutput = [stringbuffer]
-		#TODO respect ET mapped namespaces
-		itag = xml.tag.split('}', 1)[-1]
-		if '}' in xml.tag:
-			ixmlns = xml.tag.split('}', 1)[0][1:]
-		else:
-			ixmlns = ''
-		nsbuffer = ''
-		if xmlns != ixmlns and ixmlns != '' and ixmlns != self.namespace:
-			if self.stream is not None and ixmlns in self.stream.namespace_map:
-				if self.stream.namespace_map[ixmlns] != '':
-					itag = "%s:%s" % (self.stream.namespace_map[ixmlns], itag)
-			else:
-				nsbuffer = """ xmlns="%s\"""" % ixmlns
-		if ixmlns not in ('', xmlns, self.namespace):
-			nsbuffer = """ xmlns="%s\"""" % ixmlns
-		newoutput.append("<%s" % itag)
-		newoutput.append(nsbuffer)
-		for attrib in xml.attrib:
-			if '{' not in attrib:
-				newoutput.append(""" %s="%s\"""" % (attrib, self.xmlesc(xml.attrib[attrib])))
-		if len(xml) or xml.text or xml.tail:
-			newoutput.append(">")
-			if xml.text:
-				newoutput.append(self.xmlesc(xml.text))
-			if len(xml):
-				for child in xml.getchildren():
-					newoutput.append(self.__str__(child, ixmlns))
-			newoutput.append("</%s>" % (itag, ))
-			if xml.tail:
-				newoutput.append(self.xmlesc(xml.tail))
-		elif xml.text:
-			newoutput.append(">%s</%s>" % (self.xmlesc(xml.text), itag))
-		else:
-			newoutput.append(" />")
-		return ''.join(newoutput)
 
-	def xmlesc(self, text):
+
-		text = list(text)
+def tostring(xml=None, xmlns='', stanza_ns='', stream=None, outbuffer=''):
-		cc = 0
+    """
-		matches = ('&', '<', '"', '>', "'")
+    Serialize an XML object to a Unicode string.
-		for c in text:
+
-			if c in matches:
+    Arguments:
-				if c == '&':
+        xml       -- The XML object to serialize. If the value is None,
-					text[cc] = '&amp;'
+                     then the XML object contained in this stanza
-				elif c == '<':
+                     object will be used.
-					text[cc] = '&lt;'
+        xmlns     -- Optional namespace of an element wrapping the XML
-				elif c == '>':
+                     object.
-					text[cc] = '&gt;'
+        stanza_ns -- The namespace of the stanza object that contains
-				elif c == "'":
+                     the XML object.
-					text[cc] = '&apos;'
+        stream    -- The XML stream that generated the XML object.
-				else:
+        outbuffer -- Optional buffer for storing serializations during
-					text[cc] = '&quot;'
+                     recursive calls.
-			cc += 1
+    """
-		return ''.join(text)
+    # Add previous results to the start of the output.
+    output = [outbuffer]
+
+    # Extract the element's tag name.
+    tag_name = xml.tag.split('}', 1)[-1]
+
+    # Extract the element's namespace if it is defined.
+    if '}' in xml.tag:
+        tag_xmlns = xml.tag.split('}', 1)[0][1:]
+    else:
+        tag_xmlns = ''
+
+    # Output the tag name and derived namespace of the element.
+    namespace = ''
+    if tag_xmlns not in ['', xmlns, stanza_ns]:
+        namespace = ' xmlns="%s"' % tag_xmlns
+        if stream and tag_xmlns in stream.namespace_map:
+            mapped_namespace = stream.namespace_map[tag_xmlns]
+            if mapped_namespace:
+                tag = "%s:%s" % (mapped_namespace, tag_name)
+    output.append("<%s" % tag_name)
+    output.append(namespace)
+
+    # Output escaped attribute values.
+    for attrib, value in xml.attrib.items():
+        if '{' not in attrib:
+            value = xml_escape(value)
+            output.append(' %s="%s"' % (attrib, value))
+
+    if len(xml) or xml.text:
+        # If there are additional child elements to serialize.
+        output.append(">")
+        if xml.text:
+            output.append(xml_escape(xml.text))
+        if len(xml):
+            for child in xml.getchildren():
+                output.append(tostring(child, tag_xmlns, stanza_ns, stream))
+        output.append("</%s>" % tag_name)
+    elif xml.text:
+        # If we only have text content.
+        output.append(">%s</%s>" % (xml_escape(xml.text), tag_name))
+    else:
+        # Empty element.
+        output.append(" />")
+    if xml.tail:
+        # If there is additional text after the element.
+        output.append(xml_escape(xml.tail))
+    return ''.join(output)
+
+
+def xml_escape(text):
+    """
+    Convert special characters in XML to escape sequences.
+
+    Arguments:
+        text -- The XML text to convert.
+    """
+    text = list(text)
+    escapes = {'&': '&amp;',
+               '<': '&lt;',
+               '>': '&gt;',
+               "'": '&apos;',
+               '"': '&quot;'}
+    for i, c in enumerate(text):
+        text[i] = escapes.get(c, c)
+    return ''.join(text)

sleekxmpp/xmlstream/tostring/tostring26.py

@@ -1,65 +1,100 @@
+"""
+    SleekXMPP: The Sleek XMPP Library
+    Copyright (C) 2010  Nathanael C. Fritz
+    This file is part of SleekXMPP.
+
+    See the file LICENSE for copying permission.
+"""
+
+from __future__ import unicode_literals
 import types
 
-class ToString(object):
-	def __str__(self, xml=None, xmlns='', stringbuffer=''):
-		if xml is None:
-			xml = self.xml
-		newoutput = [stringbuffer]
-		#TODO respect ET mapped namespaces
-		itag = xml.tag.split('}', 1)[-1]
-		if '}' in xml.tag:
-			ixmlns = xml.tag.split('}', 1)[0][1:]
-		else:
-			ixmlns = ''
-		nsbuffer = ''
-		if xmlns != ixmlns and ixmlns != u'' and ixmlns != self.namespace:
-			if self.stream is not None and ixmlns in self.stream.namespace_map:
-				if self.stream.namespace_map[ixmlns] != u'':
-					itag = "%s:%s" % (self.stream.namespace_map[ixmlns], itag)
-			else:
-				nsbuffer = """ xmlns="%s\"""" % ixmlns
-		if ixmlns not in ('', xmlns, self.namespace):
-			nsbuffer = """ xmlns="%s\"""" % ixmlns
-		newoutput.append("<%s" % itag)
-		newoutput.append(nsbuffer)
-		for attrib in xml.attrib:
-			if '{' not in attrib:
-				newoutput.append(""" %s="%s\"""" % (attrib, self.xmlesc(xml.attrib[attrib])))
-		if len(xml) or xml.text or xml.tail:
-			newoutput.append(u">")
-			if xml.text:
-				newoutput.append(self.xmlesc(xml.text))
-			if len(xml):
-				for child in xml.getchildren():
-					newoutput.append(self.__str__(child, ixmlns))
-			newoutput.append(u"</%s>" % (itag, ))
-			if xml.tail:
-				newoutput.append(self.xmlesc(xml.tail))
-		elif xml.text:
-			newoutput.append(">%s</%s>" % (self.xmlesc(xml.text), itag))
-		else:
-			newoutput.append(" />")
-		return u''.join(newoutput)
 
-	def xmlesc(self, text):
+def tostring(xml=None, xmlns='', stanza_ns='', stream=None, outbuffer=''):
-		if type(text) != types.UnicodeType:
+    """
-			text = list(unicode(text, 'utf-8', 'ignore'))
+    Serialize an XML object to a Unicode string.
-		else:
-			text = list(text)
 
-		cc = 0
+    Arguments:
-		matches = (u'&', u'<', u'"', u'>', u"'")
+        xml       -- The XML object to serialize. If the value is None,
-		for c in text:
+                     then the XML object contained in this stanza
-			if c in matches:
+                     object will be used.
-				if c == u'&':
+        xmlns     -- Optional namespace of an element wrapping the XML
-					text[cc] = u'&amp;'
+                     object.
-				elif c == u'<':
+        stanza_ns -- The namespace of the stanza object that contains
-					text[cc] = u'&lt;'
+                     the XML object.
-				elif c == u'>':
+        stream    -- The XML stream that generated the XML object.
-					text[cc] = u'&gt;'
+        outbuffer -- Optional buffer for storing serializations during
-				elif c == u"'":
+                     recursive calls.
-					text[cc] = u'&apos;'
+    """
-				else:
+    # Add previous results to the start of the output.
-					text[cc] = u'&quot;'
+    output = [outbuffer]
-			cc += 1
+
-		return ''.join(text)
+    # Extract the element's tag name.
+    tag_name = xml.tag.split('}', 1)[-1]
+
+    # Extract the element's namespace if it is defined.
+    if '}' in xml.tag:
+        tag_xmlns = xml.tag.split('}', 1)[0][1:]
+    else:
+        tag_xmlns = u''
+
+    # Output the tag name and derived namespace of the element.
+    namespace = u''
+    if tag_xmlns not in ['', xmlns, stanza_ns]:
+        namespace = u' xmlns="%s"' % tag_xmlns
+        if stream and tag_xmlns in stream.namespace_map:
+            mapped_namespace = stream.namespace_map[tag_xmlns]
+            if mapped_namespace:
+                tag = u"%s:%s" % (mapped_namespace, tag_name)
+    output.append(u"<%s" % tag_name)
+    output.append(namespace)
+
+    # Output escaped attribute values.
+    for attrib, value in xml.attrib.items():
+        if '{' not in attrib:
+            value = xml_escape(value)
+            output.append(u' %s="%s"' % (attrib, value))
+
+    if len(xml) or xml.text:
+        # If there are additional child elements to serialize.
+        output.append(u">")
+        if xml.text:
+            output.append(xml_escape(xml.text))
+        if len(xml):
+            for child in xml.getchildren():
+                output.append(tostring(child, tag_xmlns, stanza_ns, stream))
+        output.append(u"</%s>" % tag_name)
+        if xml.tail:
+            # If there is additional text after the element.
+            output.append(xml_escape(xml.tail))
+    elif xml.text:
+        # If we only have text content.
+        output.append(u">%s</%s>" % (xml_escape(xml.text), tag_name))
+    else:
+        # Empty element.
+        output.append(u" />")
+    if xml.tail:
+        # If there is additional text after the element.
+        output.append(xml_escape(xml.tail))
+    return u''.join(output)
+
+
+def xml_escape(text):
+    """
+    Convert special characters in XML to escape sequences.
+
+    Arguments:
+        text -- The XML text to convert.
+    """
+    if type(text) != types.UnicodeType:
+        text = list(unicode(text, 'utf-8', 'ignore'))
+    else:
+        text = list(text)
+    escapes = {u'&': u'&amp;',
+               u'<': u'&lt;',
+               u'>': u'&gt;',
+               u"'": u'&apos;',
+               u'"': u'&quot;'}
+    for i, c in enumerate(text):
+        text[i] = escapes.get(c, c)
+    return u''.join(text)

sleekxmpp/xmlstream/tostring26/__init__.py

@@ -1,65 +0,0 @@
-import types
-
-class ToString(object):
-	def __str__(self, xml=None, xmlns='', stringbuffer=''):
-		if xml is None:
-			xml = self.xml
-		newoutput = [stringbuffer]
-		#TODO respect ET mapped namespaces
-		itag = xml.tag.split('}', 1)[-1]
-		if '}' in xml.tag:
-			ixmlns = xml.tag.split('}', 1)[0][1:]
-		else:
-			ixmlns = ''
-		nsbuffer = ''
-		if xmlns != ixmlns and ixmlns != u'' and ixmlns != self.namespace:
-			if self.stream is not None and ixmlns in self.stream.namespace_map:
-				if self.stream.namespace_map[ixmlns] != u'':
-					itag = "%s:%s" % (self.stream.namespace_map[ixmlns], itag)
-			else:
-				nsbuffer = """ xmlns="%s\"""" % ixmlns
-		if ixmlns not in ('', xmlns, self.namespace):
-			nsbuffer = """ xmlns="%s\"""" % ixmlns
-		newoutput.append("<%s" % itag)
-		newoutput.append(nsbuffer)
-		for attrib in xml.attrib:
-			if '{' not in attrib:
-				newoutput.append(""" %s="%s\"""" % (attrib, self.xmlesc(xml.attrib[attrib])))
-		if len(xml) or xml.text or xml.tail:
-			newoutput.append(u">")
-			if xml.text:
-				newoutput.append(self.xmlesc(xml.text))
-			if len(xml):
-				for child in xml.getchildren():
-					newoutput.append(self.__str__(child, ixmlns))
-			newoutput.append(u"</%s>" % (itag, ))
-			if xml.tail:
-				newoutput.append(self.xmlesc(xml.tail))
-		elif xml.text:
-			newoutput.append(">%s</%s>" % (self.xmlesc(xml.text), itag))
-		else:
-			newoutput.append(" />")
-		return u''.join(newoutput)
-
-	def xmlesc(self, text):
-		if type(text) != types.UnicodeType:
-			text = list(unicode(text, 'utf-8', 'ignore'))
-		else:
-			text = list(text)
-
-		cc = 0
-		matches = (u'&', u'<', u'"', u'>', u"'")
-		for c in text:
-			if c in matches:
-				if c == u'&':
-					text[cc] = u'&amp;'
-				elif c == u'<':
-					text[cc] = u'&lt;'
-				elif c == u'>':
-					text[cc] = u'&gt;'
-				elif c == u"'":
-					text[cc] = u'&apos;'
-				else:
-					text[cc] = u'&quot;'
-			cc += 1
-		return ''.join(text)

sleekxmpp/xmlstream/xmlstream.py

@@ -23,6 +23,7 @@ import types
 import copy
 import xml.sax.saxutils
 from . import scheduler
+from sleekxmpp.xmlstream.tostring import tostring
 
 RESPONSE_TIMEOUT = 10
 HANDLER_THREADS = 1
@@ -303,7 +304,7 @@ class XMLStream(object):
 	def __spawnEvent(self, xmlobj):
 		"watching xmlOut and processes handlers"
 		#convert XML into Stanza
-		logging.debug("RECV: %s" % cElementTree.tostring(xmlobj))
+		logging.debug("RECV: %s" % tostring(xmlobj))
 		xmlobj = self.incoming_filter(xmlobj)
 		stanza_type = StanzaBase
 		for stanza_class in self.__root_stanza:
@@ -388,60 +389,6 @@ class XMLStream(object):
 	def removeStanzaExtension(self, stanza_class, stanza_extension):
 		stanza_extension[stanza_class].pop(stanza_extension)
 
-	def tostring(self, xml, xmlns='', stringbuffer=''):
-		newoutput = [stringbuffer]
-		#TODO respect ET mapped namespaces
-		itag = xml.tag.split('}', 1)[-1]
-		if '}' in xml.tag:
-			ixmlns = xml.tag.split('}', 1)[0][1:]
-		else:
-			ixmlns = ''
-		nsbuffer = ''
-		if xmlns != ixmlns and ixmlns != '':
-			if ixmlns in self.namespace_map:
-				if self.namespace_map[ixmlns] != '':
-					itag = "%s:%s" % (self.namespace_map[ixmlns], itag)
-			else:
-				nsbuffer = """ xmlns="%s\"""" % ixmlns
-		newoutput.append("<%s" % itag)
-		newoutput.append(nsbuffer)
-		for attrib in xml.attrib:
-			newoutput.append(""" %s="%s\"""" % (attrib, self.xmlesc(xml.attrib[attrib])))
-		if len(xml) or xml.text or xml.tail:
-			newoutput.append(">")
-			if xml.text:
-				newoutput.append(self.xmlesc(xml.text))
-			if len(xml):
-				for child in xml.getchildren():
-					newoutput.append(self.tostring(child, ixmlns))
-			newoutput.append("</%s>" % (itag, ))
-			if xml.tail:
-				newoutput.append(self.xmlesc(xml.tail))
-		elif xml.text:
-			newoutput.append(">%s</%s>" % (self.xmlesc(xml.text), itag))
-		else:
-			newoutput.append(" />")
-		return ''.join(newoutput)
-
-	def xmlesc(self, text):
-		text = list(text)
-		cc = 0
-		matches = ('&', '<', '"', '>', "'")
-		for c in text:
-			if c in matches:
-				if c == '&':
-					text[cc] = '&amp;'
-				elif c == '<':
-					text[cc] = '&lt;'
-				elif c == '>':
-					text[cc] = '&gt;'
-				elif c == "'":
-					text[cc] = '&apos;'
-				elif self.escape_quotes:
-					text[cc] = '&quot;'
-			cc += 1
-		return ''.join(text)
-	
 	def start_stream_handler(self, xml):
 		"""Meant to be overridden"""
 		pass