bpo-34160: Make sorting attributes in XML serialization optional.

Doc/library/xml.dom.minidom.rst

@@ -132,7 +132,7 @@ module documentation.  This section lists the differences between the API and
           ... # Work with dom.
 
 
-.. method:: Node.writexml(writer, indent="", addindent="", newl="")
+.. method:: Node.writexml(writer, indent="", addindent="", newl="", *, sort_attrs=True)
 
    Write XML to the writer object.  The writer should have a :meth:`write` method
    which matches that of the file object interface.  The *indent* parameter is the
@@ -143,11 +143,13 @@ module documentation.  This section lists the differences between the API and
    For the :class:`Document` node, an additional keyword argument *encoding* can
    be used to specify the encoding field of the XML header.
 
+   If *sort_attrs* is true (by default), then element attributes will be
+   written in lexical order.
+
    .. versionchanged:: 3.8
-      The :meth:`writexml` method now preserves the attribute order specified
-      by the user.
+      Added support for the *sort_attrs* argument.
 
-.. method:: Node.toxml(encoding=None)
+.. method:: Node.toxml(encoding=None, *, sort_attrs=True)
 
    Return a string or byte string containing the XML represented by
    the DOM node.
@@ -159,22 +161,23 @@ module documentation.  This section lists the differences between the API and
    encoding. Encoding this string in an encoding other than UTF-8 is
    likely incorrect, since UTF-8 is the default encoding of XML.
 
+   The *sort_attrs* argument behaves like the corresponding argument of
+   :meth:`writexml`.
+
    .. versionchanged:: 3.8
-      The :meth:`toxml` method now preserves the attribute order specified
-      by the user.
+      Added support for the *sort_attrs* argument.
 
-.. method:: Node.toprettyxml(indent="", newl="", encoding="")
+.. method:: Node.toprettyxml(indent="", newl="", encoding="", *, sort_attrs=True)
 
    Return a pretty-printed version of the document. *indent* specifies the
    indentation string and defaults to a tabulator; *newl* specifies the string
    emitted at the end of each line and defaults to ``\n``.
 
-   The *encoding* argument behaves like the corresponding argument of
-   :meth:`toxml`.
+   The *encoding* and *sort_attrs* arguments behave like the corresponding
+   arguments of :meth:`toxml`.
 
    .. versionchanged:: 3.8
-      The :meth:`toprettyxml` method now preserves the attribute order specified
-      by the user.
+      Added support for the *sort_attrs* argument.
 
 
 .. _dom-example:

Doc/library/xml.etree.elementtree.rst

@@ -592,29 +592,34 @@ Functions
 
 
 .. function:: tostring(element, encoding="us-ascii", method="xml", *, \
-                       short_empty_elements=True)
+                       short_empty_elements=True, sort_attrs=True)
 
    Generates a string representation of an XML element, including all
    subelements.  *element* is an :class:`Element` instance.  *encoding* [1]_ is
    the output encoding (default is US-ASCII).  Use ``encoding="unicode"`` to
    generate a Unicode string (otherwise, a bytestring is generated).  *method*
    is either ``"xml"``, ``"html"`` or ``"text"`` (default is ``"xml"``).
-   *short_empty_elements* has the same meaning as in :meth:`ElementTree.write`.
+   *short_empty_elements* and *sort_attrs* have the same meaning as in
+   :meth:`ElementTree.write`.
    Returns an (optionally) encoded string containing the XML data.
 
    .. versionadded:: 3.4
       The *short_empty_elements* parameter.
 
+   .. versionadded:: 3.8
+      The *sort_attrs* parameter.
+
 
 .. function:: tostringlist(element, encoding="us-ascii", method="xml", *, \
-                           short_empty_elements=True)
+                           short_empty_elements=True, sort_attrs=True)
 
    Generates a string representation of an XML element, including all
    subelements.  *element* is an :class:`Element` instance.  *encoding* [1]_ is
    the output encoding (default is US-ASCII).  Use ``encoding="unicode"`` to
    generate a Unicode string (otherwise, a bytestring is generated).  *method*
    is either ``"xml"``, ``"html"`` or ``"text"`` (default is ``"xml"``).
-   *short_empty_elements* has the same meaning as in :meth:`ElementTree.write`.
+   *short_empty_elements* and *sort_attrs* have the same meaning as in
+   :meth:`ElementTree.write`.
    Returns a list of (optionally) encoded strings containing the XML data.
    It does not guarantee any specific sequence, except that
    ``b"".join(tostringlist(element)) == tostring(element)``.
@@ -624,6 +629,9 @@ Functions
    .. versionadded:: 3.4
       The *short_empty_elements* parameter.
 
+   .. versionadded:: 3.8
+      The *sort_attrs* parameter.
+
 
 .. function:: XML(text, parser=None)
 
@@ -925,7 +933,7 @@ ElementTree Objects
 
    .. method:: write(file, encoding="us-ascii", xml_declaration=None, \
                      default_namespace=None, method="xml", *, \
-                     short_empty_elements=True)
+                     short_empty_elements=True, sort_attrs=True)
 
       Writes the element tree to a file, as XML.  *file* is a file name, or a
       :term:`file object` opened for writing.  *encoding* [1]_ is the output
@@ -940,6 +948,8 @@ ElementTree Objects
       of elements that contain no content.  If ``True`` (the default), they are
       emitted as a single self-closed tag, otherwise they are emitted as a pair
       of start/end tags.
+      If *sort_attrs* is true (by default), then element attributes will be
+      written in lexical order.
 
       The output is either a string (:class:`str`) or binary (:class:`bytes`).
       This is controlled by the *encoding* argument.  If *encoding* is
@@ -951,9 +961,8 @@ ElementTree Objects
       .. versionadded:: 3.4
          The *short_empty_elements* parameter.
 
-      .. versionchanged:: 3.8
-         The :meth:`write` method now preserves the attribute order specified
-         by the user.
+      .. versionadded:: 3.8
+         The *sort_attrs* parameter.
 
 
 This is the XML file that is going to be manipulated::

Lib/test/test_minidom.py

@@ -1567,17 +1567,26 @@ def test_minidom_attribute_order(self):
         doc = parseString(xml_str)
         output = io.StringIO()
         doc.writexml(output)
+        self.assertEqual(output.getvalue(),
+            '<?xml version="1.0" ?><curriculum company="example" status="public"/>')
+        output = io.StringIO()
+        doc.writexml(output, sort_attrs=False)
         self.assertEqual(output.getvalue(), xml_str)
 
     def test_toxml_with_attributes_ordered(self):
         xml_str = '<?xml version="1.0" ?><curriculum status="public" company="example"/>'
         doc = parseString(xml_str)
-        self.assertEqual(doc.toxml(), xml_str)
+        self.assertEqual(doc.toxml(),
+            '<?xml version="1.0" ?><curriculum company="example" status="public"/>')
+        self.assertEqual(doc.toxml(sort_attrs=False), xml_str)
 
     def test_toprettyxml_with_attributes_ordered(self):
         xml_str = '<?xml version="1.0" ?><curriculum status="public" company="example"/>'
         doc = parseString(xml_str)
         self.assertEqual(doc.toprettyxml(),
+                         '<?xml version="1.0" ?>\n'
+                         '<curriculum company="example" status="public"/>\n')
+        self.assertEqual(doc.toprettyxml(sort_attrs=False),
                          '<?xml version="1.0" ?>\n'
                          '<curriculum status="public" company="example"/>\n')
 

Lib/test/test_xml_etree.py

@@ -1054,10 +1054,15 @@ def test_dump_attribute_order(self):
 
     def test_tree_write_attribute_order(self):
         # See BPO 34160
-        root = ET.Element('cirriculum', status='public', company='example')
+        root = ET.XML('<cirriculum status="public" company="example"></cirriculum>')
+        #root = ET.Element('cirriculum', status='public', company='example')
         self.assertEqual(serialize(root),
+                         '<cirriculum company="example" status="public" />')
+        self.assertEqual(serialize(root, sort_attrs=False),
                          '<cirriculum status="public" company="example" />')
         self.assertEqual(serialize(root, method='html'),
+                '<cirriculum company="example" status="public"></cirriculum>')
+        self.assertEqual(serialize(root, method='html', sort_attrs=False),
                 '<cirriculum status="public" company="example"></cirriculum>')
 
 

Lib/xml/dom/minidom.py

@@ -43,10 +43,10 @@ class Node(xml.dom.Node):
     def __bool__(self):
         return True
 
-    def toxml(self, encoding=None):
-        return self.toprettyxml("", "", encoding)
+    def toxml(self, encoding=None, *, sort_attrs=True):
+        return self.toprettyxml("", "", encoding, sort_attrs=sort_attrs)
 
-    def toprettyxml(self, indent="\t", newl="\n", encoding=None):
+    def toprettyxml(self, indent="\t", newl="\n", encoding=None, *, sort_attrs=True):
         if encoding is None:
             writer = io.StringIO()
         else:
@@ -56,9 +56,11 @@ def toprettyxml(self, indent="\t", newl="\n", encoding=None):
                                       newline='\n')
         if self.nodeType == Node.DOCUMENT_NODE:
             # Can pass encoding only to document, to put it into XML header
-            self.writexml(writer, "", indent, newl, encoding)
+            self.writexml(writer, "", indent, newl, encoding,
+                          sort_attrs=sort_attrs)
         else:
-            self.writexml(writer, "", indent, newl)
+            self.writexml(writer, "", indent, newl,
+                          sort_attrs=sort_attrs)
         if encoding is None:
             return writer.getvalue()
         else:
@@ -847,27 +849,33 @@ def getElementsByTagNameNS(self, namespaceURI, localName):
     def __repr__(self):
         return "<DOM Element: %s at %#x>" % (self.tagName, id(self))
 
-    def writexml(self, writer, indent="", addindent="", newl=""):
+    def writexml(self, writer, indent="", addindent="", newl="", *,
+                 sort_attrs=True):
         # indent = current indentation
         # addindent = indentation to add to higher levels
         # newl = newline string
         writer.write(indent+"<" + self.tagName)
 
         attrs = self._get_attributes()
+        a_names = attrs.keys()
+        if sort_attrs:
+            a_names = sorted(a_names)
 
-        for a_name in attrs.keys():
+        for a_name in a_names:
             writer.write(" %s=\"" % a_name)
             _write_data(writer, attrs[a_name].value)
             writer.write("\"")
         if self.childNodes:
             writer.write(">")
             if (len(self.childNodes) == 1 and
                 self.childNodes[0].nodeType == Node.TEXT_NODE):
-                self.childNodes[0].writexml(writer, '', '', '')
+                self.childNodes[0].writexml(writer, '', '', '',
+                                            sort_attrs=sort_attrs)
             else:
                 writer.write(newl)
                 for node in self.childNodes:
-                    node.writexml(writer, indent+addindent, addindent, newl)
+                    node.writexml(writer, indent+addindent, addindent, newl,
+                                  sort_attrs=sort_attrs)
                 writer.write(indent)
             writer.write("</%s>%s" % (self.tagName, newl))
         else:
@@ -984,7 +992,8 @@ def _set_nodeName(self, value):
         self.target = value
     nodeName = property(_get_nodeName, _set_nodeName)
 
-    def writexml(self, writer, indent="", addindent="", newl=""):
+    def writexml(self, writer, indent="", addindent="", newl="", *,
+                 sort_attrs=True):
         writer.write("%s<?%s %s?>%s" % (indent,self.target, self.data, newl))
 
 
@@ -1084,7 +1093,8 @@ def splitText(self, offset):
         self.data = self.data[:offset]
         return newText
 
-    def writexml(self, writer, indent="", addindent="", newl=""):
+    def writexml(self, writer, indent="", addindent="", newl="", *,
+                 sort_attrs=True):
         _write_data(writer, "%s%s%s" % (indent, self.data, newl))
 
     # DOM Level 3 (WD 9 April 2002)
@@ -1179,7 +1189,8 @@ def __init__(self, data):
         CharacterData.__init__(self)
         self._data = data
 
-    def writexml(self, writer, indent="", addindent="", newl=""):
+    def writexml(self, writer, indent="", addindent="", newl="", *,
+                 sort_attrs=True):
         if "--" in self.data:
             raise ValueError("'--' is not allowed in a comment node")
         writer.write("%s<!--%s-->%s" % (indent, self.data, newl))
@@ -1191,7 +1202,8 @@ class CDATASection(Text):
     nodeType = Node.CDATA_SECTION_NODE
     nodeName = "#cdata-section"
 
-    def writexml(self, writer, indent="", addindent="", newl=""):
+    def writexml(self, writer, indent="", addindent="", newl="", *,
+                 sort_attrs=True):
         if self.data.find("]]>") >= 0:
             raise ValueError("']]>' not allowed in a CDATA section")
         writer.write("<![CDATA[%s]]>" % self.data)
@@ -1324,7 +1336,8 @@ def cloneNode(self, deep):
         else:
             return None
 
-    def writexml(self, writer, indent="", addindent="", newl=""):
+    def writexml(self, writer, indent="", addindent="", newl="", *,
+                 sort_attrs=True):
         writer.write("<!DOCTYPE ")
         writer.write(self.name)
         if self.publicId:
@@ -1786,14 +1799,15 @@ def importNode(self, node, deep):
             raise xml.dom.NotSupportedErr("cannot import document type nodes")
         return _clone_node(node, deep, self)
 
-    def writexml(self, writer, indent="", addindent="", newl="", encoding=None):
+    def writexml(self, writer, indent="", addindent="", newl="", encoding=None,
+                 *, sort_attrs=True):
         if encoding is None:
             writer.write('<?xml version="1.0" ?>'+newl)
         else:
             writer.write('<?xml version="1.0" encoding="%s"?>%s' % (
                 encoding, newl))
         for node in self.childNodes:
-            node.writexml(writer, indent, addindent, newl)
+            node.writexml(writer, indent, addindent, newl, sort_attrs=sort_attrs)
 
     # DOM Level 3 (WD 9 April 2002)