--- /dev/null
+#\r
+# ElementTree\r
+# $Id: ElementTree.py 2326 2005-03-17 07:45:21Z fredrik $\r
+#\r
+# light-weight XML support for Python 1.5.2 and later.\r
+#\r
+# history:\r
+# 2001-10-20 fl created (from various sources)\r
+# 2001-11-01 fl return root from parse method\r
+# 2002-02-16 fl sort attributes in lexical order\r
+# 2002-04-06 fl TreeBuilder refactoring, added PythonDoc markup\r
+# 2002-05-01 fl finished TreeBuilder refactoring\r
+# 2002-07-14 fl added basic namespace support to ElementTree.write\r
+# 2002-07-25 fl added QName attribute support\r
+# 2002-10-20 fl fixed encoding in write\r
+# 2002-11-24 fl changed default encoding to ascii; fixed attribute encoding\r
+# 2002-11-27 fl accept file objects or file names for parse/write\r
+# 2002-12-04 fl moved XMLTreeBuilder back to this module\r
+# 2003-01-11 fl fixed entity encoding glitch for us-ascii\r
+# 2003-02-13 fl added XML literal factory\r
+# 2003-02-21 fl added ProcessingInstruction/PI factory\r
+# 2003-05-11 fl added tostring/fromstring helpers\r
+# 2003-05-26 fl added ElementPath support\r
+# 2003-07-05 fl added makeelement factory method\r
+# 2003-07-28 fl added more well-known namespace prefixes\r
+# 2003-08-15 fl fixed typo in ElementTree.findtext (Thomas Dartsch)\r
+# 2003-09-04 fl fall back on emulator if ElementPath is not installed\r
+# 2003-10-31 fl markup updates\r
+# 2003-11-15 fl fixed nested namespace bug\r
+# 2004-03-28 fl added XMLID helper\r
+# 2004-06-02 fl added default support to findtext\r
+# 2004-06-08 fl fixed encoding of non-ascii element/attribute names\r
+# 2004-08-23 fl take advantage of post-2.1 expat features\r
+# 2005-02-01 fl added iterparse implementation\r
+# 2005-03-02 fl fixed iterparse support for pre-2.2 versions\r
+#\r
+# Copyright (c) 1999-2005 by Fredrik Lundh. All rights reserved.\r
+#\r
+# fredrik@pythonware.com\r
+# http://www.pythonware.com\r
+#\r
+# --------------------------------------------------------------------\r
+# The ElementTree toolkit is\r
+#\r
+# Copyright (c) 1999-2005 by Fredrik Lundh\r
+#\r
+# By obtaining, using, and/or copying this software and/or its\r
+# associated documentation, you agree that you have read, understood,\r
+# and will comply with the following terms and conditions:\r
+#\r
+# Permission to use, copy, modify, and distribute this software and\r
+# its associated documentation for any purpose and without fee is\r
+# hereby granted, provided that the above copyright notice appears in\r
+# all copies, and that both that copyright notice and this permission\r
+# notice appear in supporting documentation, and that the name of\r
+# Secret Labs AB or the author not be used in advertising or publicity\r
+# pertaining to distribution of the software without specific, written\r
+# prior permission.\r
+#\r
+# SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD\r
+# TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANT-\r
+# ABILITY AND FITNESS. IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR\r
+# BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY\r
+# DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,\r
+# WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS\r
+# ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE\r
+# OF THIS SOFTWARE.\r
+# --------------------------------------------------------------------\r
+\r
+__all__ = [\r
+ # public symbols\r
+ "Comment",\r
+ "dump",\r
+ "Element", "ElementTree",\r
+ "fromstring",\r
+ "iselement", "iterparse",\r
+ "parse",\r
+ "PI", "ProcessingInstruction",\r
+ "QName",\r
+ "SubElement",\r
+ "tostring",\r
+ "TreeBuilder",\r
+ "VERSION", "XML",\r
+ "XMLTreeBuilder",\r
+ ]\r
+\r
+##\r
+# The <b>Element</b> type is a flexible container object, designed to\r
+# store hierarchical data structures in memory. The type can be\r
+# described as a cross between a list and a dictionary.\r
+# <p>\r
+# Each element has a number of properties associated with it:\r
+# <ul>\r
+# <li>a <i>tag</i>. This is a string identifying what kind of data\r
+# this element represents (the element type, in other words).</li>\r
+# <li>a number of <i>attributes</i>, stored in a Python dictionary.</li>\r
+# <li>a <i>text</i> string.</li>\r
+# <li>an optional <i>tail</i> string.</li>\r
+# <li>a number of <i>child elements</i>, stored in a Python sequence</li>\r
+# </ul>\r
+#\r
+# To create an element instance, use the {@link #Element} or {@link\r
+# #SubElement} factory functions.\r
+# <p>\r
+# The {@link #ElementTree} class can be used to wrap an element\r
+# structure, and convert it from and to XML.\r
+##\r
+\r
+import string, sys, re\r
+\r
+class _SimpleElementPath:\r
+ # emulate pre-1.2 find/findtext/findall behaviour\r
+ def find(self, element, tag):\r
+ for elem in element:\r
+ if elem.tag == tag:\r
+ return elem\r
+ return None\r
+ def findtext(self, element, tag, default=None):\r
+ for elem in element:\r
+ if elem.tag == tag:\r
+ return elem.text or ""\r
+ return default\r
+ def findall(self, element, tag):\r
+ if tag[:3] == ".//":\r
+ return element.getiterator(tag[3:])\r
+ result = []\r
+ for elem in element:\r
+ if elem.tag == tag:\r
+ result.append(elem)\r
+ return result\r
+\r
+try:\r
+ import ElementPath\r
+except ImportError:\r
+ # FIXME: issue warning in this case?\r
+ ElementPath = _SimpleElementPath()\r
+\r
+# TODO: add support for custom namespace resolvers/default namespaces\r
+# TODO: add improved support for incremental parsing\r
+\r
+VERSION = "1.2.6"\r
+\r
+##\r
+# Internal element class. This class defines the Element interface,\r
+# and provides a reference implementation of this interface.\r
+# <p>\r
+# You should not create instances of this class directly. Use the\r
+# appropriate factory functions instead, such as {@link #Element}\r
+# and {@link #SubElement}.\r
+#\r
+# @see Element\r
+# @see SubElement\r
+# @see Comment\r
+# @see ProcessingInstruction\r
+\r
+class _ElementInterface:\r
+ # <tag attrib>text<child/>...</tag>tail\r
+\r
+ ##\r
+ # (Attribute) Element tag.\r
+\r
+ tag = None\r
+\r
+ ##\r
+ # (Attribute) Element attribute dictionary. Where possible, use\r
+ # {@link #_ElementInterface.get},\r
+ # {@link #_ElementInterface.set},\r
+ # {@link #_ElementInterface.keys}, and\r
+ # {@link #_ElementInterface.items} to access\r
+ # element attributes.\r
+\r
+ attrib = None\r
+\r
+ ##\r
+ # (Attribute) Text before first subelement. This is either a\r
+ # string or the value None, if there was no text.\r
+\r
+ text = None\r
+\r
+ ##\r
+ # (Attribute) Text after this element's end tag, but before the\r
+ # next sibling element's start tag. This is either a string or\r
+ # the value None, if there was no text.\r
+\r
+ tail = None # text after end tag, if any\r
+\r
+ def __init__(self, tag, attrib):\r
+ self.tag = tag\r
+ self.attrib = attrib\r
+ self._children = []\r
+\r
+ def __repr__(self):\r
+ return "<Element %s at %x>" % (self.tag, id(self))\r
+\r
+ ##\r
+ # Creates a new element object of the same type as this element.\r
+ #\r
+ # @param tag Element tag.\r
+ # @param attrib Element attributes, given as a dictionary.\r
+ # @return A new element instance.\r
+\r
+ def makeelement(self, tag, attrib):\r
+ return Element(tag, attrib)\r
+\r
+ ##\r
+ # Returns the number of subelements.\r
+ #\r
+ # @return The number of subelements.\r
+\r
+ def __len__(self):\r
+ return len(self._children)\r
+\r
+ ##\r
+ # Returns the given subelement.\r
+ #\r
+ # @param index What subelement to return.\r
+ # @return The given subelement.\r
+ # @exception IndexError If the given element does not exist.\r
+\r
+ def __getitem__(self, index):\r
+ return self._children[index]\r
+\r
+ ##\r
+ # Replaces the given subelement.\r
+ #\r
+ # @param index What subelement to replace.\r
+ # @param element The new element value.\r
+ # @exception IndexError If the given element does not exist.\r
+ # @exception AssertionError If element is not a valid object.\r
+\r
+ def __setitem__(self, index, element):\r
+ assert iselement(element)\r
+ self._children[index] = element\r
+\r
+ ##\r
+ # Deletes the given subelement.\r
+ #\r
+ # @param index What subelement to delete.\r
+ # @exception IndexError If the given element does not exist.\r
+\r
+ def __delitem__(self, index):\r
+ del self._children[index]\r
+\r
+ ##\r
+ # Returns a list containing subelements in the given range.\r
+ #\r
+ # @param start The first subelement to return.\r
+ # @param stop The first subelement that shouldn't be returned.\r
+ # @return A sequence object containing subelements.\r
+\r
+ def __getslice__(self, start, stop):\r
+ return self._children[start:stop]\r
+\r
+ ##\r
+ # Replaces a number of subelements with elements from a sequence.\r
+ #\r
+ # @param start The first subelement to replace.\r
+ # @param stop The first subelement that shouldn't be replaced.\r
+ # @param elements A sequence object with zero or more elements.\r
+ # @exception AssertionError If a sequence member is not a valid object.\r
+\r
+ def __setslice__(self, start, stop, elements):\r
+ for element in elements:\r
+ assert iselement(element)\r
+ self._children[start:stop] = list(elements)\r
+\r
+ ##\r
+ # Deletes a number of subelements.\r
+ #\r
+ # @param start The first subelement to delete.\r
+ # @param stop The first subelement to leave in there.\r
+\r
+ def __delslice__(self, start, stop):\r
+ del self._children[start:stop]\r
+\r
+ ##\r
+ # Adds a subelement to the end of this element.\r
+ #\r
+ # @param element The element to add.\r
+ # @exception AssertionError If a sequence member is not a valid object.\r
+\r
+ def append(self, element):\r
+ assert iselement(element)\r
+ self._children.append(element)\r
+\r
+ ##\r
+ # Inserts a subelement at the given position in this element.\r
+ #\r
+ # @param index Where to insert the new subelement.\r
+ # @exception AssertionError If the element is not a valid object.\r
+\r
+ def insert(self, index, element):\r
+ assert iselement(element)\r
+ self._children.insert(index, element)\r
+\r
+ ##\r
+ # Removes a matching subelement. Unlike the <b>find</b> methods,\r
+ # this method compares elements based on identity, not on tag\r
+ # value or contents.\r
+ #\r
+ # @param element What element to remove.\r
+ # @exception ValueError If a matching element could not be found.\r
+ # @exception AssertionError If the element is not a valid object.\r
+\r
+ def remove(self, element):\r
+ assert iselement(element)\r
+ self._children.remove(element)\r
+\r
+ ##\r
+ # Returns all subelements. The elements are returned in document\r
+ # order.\r
+ #\r
+ # @return A list of subelements.\r
+ # @defreturn list of Element instances\r
+\r
+ def getchildren(self):\r
+ return self._children\r
+\r
+ ##\r
+ # Finds the first matching subelement, by tag name or path.\r
+ #\r
+ # @param path What element to look for.\r
+ # @return The first matching element, or None if no element was found.\r
+ # @defreturn Element or None\r
+\r
+ def find(self, path):\r
+ return ElementPath.find(self, path)\r
+\r
+ ##\r
+ # Finds text for the first matching subelement, by tag name or path.\r
+ #\r
+ # @param path What element to look for.\r
+ # @param default What to return if the element was not found.\r
+ # @return The text content of the first matching element, or the\r
+ # default value no element was found. Note that if the element\r
+ # has is found, but has no text content, this method returns an\r
+ # empty string.\r
+ # @defreturn string\r
+\r
+ def findtext(self, path, default=None):\r
+ return ElementPath.findtext(self, path, default)\r
+\r
+ ##\r
+ # Finds all matching subelements, by tag name or path.\r
+ #\r
+ # @param path What element to look for.\r
+ # @return A list or iterator containing all matching elements,\r
+ # in document order.\r
+ # @defreturn list of Element instances\r
+\r
+ def findall(self, path):\r
+ return ElementPath.findall(self, path)\r
+\r
+ ##\r
+ # Resets an element. This function removes all subelements, clears\r
+ # all attributes, and sets the text and tail attributes to None.\r
+\r
+ def clear(self):\r
+ self.attrib.clear()\r
+ self._children = []\r
+ self.text = self.tail = None\r
+\r
+ ##\r
+ # Gets an element attribute.\r
+ #\r
+ # @param key What attribute to look for.\r
+ # @param default What to return if the attribute was not found.\r
+ # @return The attribute value, or the default value, if the\r
+ # attribute was not found.\r
+ # @defreturn string or None\r
+\r
+ def get(self, key, default=None):\r
+ return self.attrib.get(key, default)\r
+\r
+ ##\r
+ # Sets an element attribute.\r
+ #\r
+ # @param key What attribute to set.\r
+ # @param value The attribute value.\r
+\r
+ def set(self, key, value):\r
+ self.attrib[key] = value\r
+\r
+ ##\r
+ # Gets a list of attribute names. The names are returned in an\r
+ # arbitrary order (just like for an ordinary Python dictionary).\r
+ #\r
+ # @return A list of element attribute names.\r
+ # @defreturn list of strings\r
+\r
+ def keys(self):\r
+ return self.attrib.keys()\r
+\r
+ ##\r
+ # Gets element attributes, as a sequence. The attributes are\r
+ # returned in an arbitrary order.\r
+ #\r
+ # @return A list of (name, value) tuples for all attributes.\r
+ # @defreturn list of (string, string) tuples\r
+\r
+ def items(self):\r
+ return self.attrib.items()\r
+\r
+ ##\r
+ # Creates a tree iterator. The iterator loops over this element\r
+ # and all subelements, in document order, and returns all elements\r
+ # with a matching tag.\r
+ # <p>\r
+ # If the tree structure is modified during iteration, the result\r
+ # is undefined.\r
+ #\r
+ # @param tag What tags to look for (default is to return all elements).\r
+ # @return A list or iterator containing all the matching elements.\r
+ # @defreturn list or iterator\r
+\r
+ def getiterator(self, tag=None):\r
+ nodes = []\r
+ if tag == "*":\r
+ tag = None\r
+ if tag is None or self.tag == tag:\r
+ nodes.append(self)\r
+ for node in self._children:\r
+ nodes.extend(node.getiterator(tag))\r
+ return nodes\r
+\r
+# compatibility\r
+_Element = _ElementInterface\r
+\r
+##\r
+# Element factory. This function returns an object implementing the\r
+# standard Element interface. The exact class or type of that object\r
+# is implementation dependent, but it will always be compatible with\r
+# the {@link #_ElementInterface} class in this module.\r
+# <p>\r
+# The element name, attribute names, and attribute values can be\r
+# either 8-bit ASCII strings or Unicode strings.\r
+#\r
+# @param tag The element name.\r
+# @param attrib An optional dictionary, containing element attributes.\r
+# @param **extra Additional attributes, given as keyword arguments.\r
+# @return An element instance.\r
+# @defreturn Element\r
+\r
+def Element(tag, attrib={}, **extra):\r
+ attrib = attrib.copy()\r
+ attrib.update(extra)\r
+ return _ElementInterface(tag, attrib)\r
+\r
+##\r
+# Subelement factory. This function creates an element instance, and\r
+# appends it to an existing element.\r
+# <p>\r
+# The element name, attribute names, and attribute values can be\r
+# either 8-bit ASCII strings or Unicode strings.\r
+#\r
+# @param parent The parent element.\r
+# @param tag The subelement name.\r
+# @param attrib An optional dictionary, containing element attributes.\r
+# @param **extra Additional attributes, given as keyword arguments.\r
+# @return An element instance.\r
+# @defreturn Element\r
+\r
+def SubElement(parent, tag, attrib={}, **extra):\r
+ attrib = attrib.copy()\r
+ attrib.update(extra)\r
+ element = parent.makeelement(tag, attrib)\r
+ parent.append(element)\r
+ return element\r
+\r
+##\r
+# Comment element factory. This factory function creates a special\r
+# element that will be serialized as an XML comment.\r
+# <p>\r
+# The comment string can be either an 8-bit ASCII string or a Unicode\r
+# string.\r
+#\r
+# @param text A string containing the comment string.\r
+# @return An element instance, representing a comment.\r
+# @defreturn Element\r
+\r
+def Comment(text=None):\r
+ element = Element(Comment)\r
+ element.text = text\r
+ return element\r
+\r
+##\r
+# PI element factory. This factory function creates a special element\r
+# that will be serialized as an XML processing instruction.\r
+#\r
+# @param target A string containing the PI target.\r
+# @param text A string containing the PI contents, if any.\r
+# @return An element instance, representing a PI.\r
+# @defreturn Element\r
+\r
+def ProcessingInstruction(target, text=None):\r
+ element = Element(ProcessingInstruction)\r
+ element.text = target\r
+ if text:\r
+ element.text = element.text + " " + text\r
+ return element\r
+\r
+PI = ProcessingInstruction\r
+\r
+##\r
+# QName wrapper. This can be used to wrap a QName attribute value, in\r
+# order to get proper namespace handling on output.\r
+#\r
+# @param text A string containing the QName value, in the form {uri}local,\r
+# or, if the tag argument is given, the URI part of a QName.\r
+# @param tag Optional tag. If given, the first argument is interpreted as\r
+# an URI, and this argument is interpreted as a local name.\r
+# @return An opaque object, representing the QName.\r
+\r
+class QName:\r
+ def __init__(self, text_or_uri, tag=None):\r
+ if tag:\r
+ text_or_uri = "{%s}%s" % (text_or_uri, tag)\r
+ self.text = text_or_uri\r
+ def __str__(self):\r
+ return self.text\r
+ def __hash__(self):\r
+ return hash(self.text)\r
+ def __cmp__(self, other):\r
+ if isinstance(other, QName):\r
+ return cmp(self.text, other.text)\r
+ return cmp(self.text, other)\r
+\r
+##\r
+# ElementTree wrapper class. This class represents an entire element\r
+# hierarchy, and adds some extra support for serialization to and from\r
+# standard XML.\r
+#\r
+# @param element Optional root element.\r
+# @keyparam file Optional file handle or name. If given, the\r
+# tree is initialized with the contents of this XML file.\r
+\r
+class ElementTree:\r
+\r
+ def __init__(self, element=None, file=None):\r
+ assert element is None or iselement(element)\r
+ self._root = element # first node\r
+ if file:\r
+ self.parse(file)\r
+\r
+ ##\r
+ # Gets the root element for this tree.\r
+ #\r
+ # @return An element instance.\r
+ # @defreturn Element\r
+\r
+ def getroot(self):\r
+ return self._root\r
+\r
+ ##\r
+ # Replaces the root element for this tree. This discards the\r
+ # current contents of the tree, and replaces it with the given\r
+ # element. Use with care.\r
+ #\r
+ # @param element An element instance.\r
+\r
+ def _setroot(self, element):\r
+ assert iselement(element)\r
+ self._root = element\r
+\r
+ ##\r
+ # Loads an external XML document into this element tree.\r
+ #\r
+ # @param source A file name or file object.\r
+ # @param parser An optional parser instance. If not given, the\r
+ # standard {@link XMLTreeBuilder} parser is used.\r
+ # @return The document root element.\r
+ # @defreturn Element\r
+\r
+ def parse(self, source, parser=None):\r
+ if not hasattr(source, "read"):\r
+ source = open(source, "rb")\r
+ if not parser:\r
+ parser = XMLTreeBuilder()\r
+ while 1:\r
+ data = source.read(32768)\r
+ if not data:\r
+ break\r
+ parser.feed(data)\r
+ self._root = parser.close()\r
+ return self._root\r
+\r
+ ##\r
+ # Creates a tree iterator for the root element. The iterator loops\r
+ # over all elements in this tree, in document order.\r
+ #\r
+ # @param tag What tags to look for (default is to return all elements)\r
+ # @return An iterator.\r
+ # @defreturn iterator\r
+\r
+ def getiterator(self, tag=None):\r
+ assert self._root is not None\r
+ return self._root.getiterator(tag)\r
+\r
+ ##\r
+ # Finds the first toplevel element with given tag.\r
+ # Same as getroot().find(path).\r
+ #\r
+ # @param path What element to look for.\r
+ # @return The first matching element, or None if no element was found.\r
+ # @defreturn Element or None\r
+\r
+ def find(self, path):\r
+ assert self._root is not None\r
+ if path[:1] == "/":\r
+ path = "." + path\r
+ return self._root.find(path)\r
+\r
+ ##\r
+ # Finds the element text for the first toplevel element with given\r
+ # tag. Same as getroot().findtext(path).\r
+ #\r
+ # @param path What toplevel element to look for.\r
+ # @param default What to return if the element was not found.\r
+ # @return The text content of the first matching element, or the\r
+ # default value no element was found. Note that if the element\r
+ # has is found, but has no text content, this method returns an\r
+ # empty string.\r
+ # @defreturn string\r
+\r
+ def findtext(self, path, default=None):\r
+ assert self._root is not None\r
+ if path[:1] == "/":\r
+ path = "." + path\r
+ return self._root.findtext(path, default)\r
+\r
+ ##\r
+ # Finds all toplevel elements with the given tag.\r
+ # Same as getroot().findall(path).\r
+ #\r
+ # @param path What element to look for.\r
+ # @return A list or iterator containing all matching elements,\r
+ # in document order.\r
+ # @defreturn list of Element instances\r
+\r
+ def findall(self, path):\r
+ assert self._root is not None\r
+ if path[:1] == "/":\r
+ path = "." + path\r
+ return self._root.findall(path)\r
+\r
+ ##\r
+ # Writes the element tree to a file, as XML.\r
+ #\r
+ # @param file A file name, or a file object opened for writing.\r
+ # @param encoding Optional output encoding (default is US-ASCII).\r
+\r
+ def write(self, file, encoding="us-ascii"):\r
+ assert self._root is not None\r
+ if not hasattr(file, "write"):\r
+ file = open(file, "wb")\r
+ if not encoding:\r
+ encoding = "us-ascii"\r
+ elif encoding != "utf-8" and encoding != "us-ascii":\r
+ file.write("<?xml version='1.0' encoding='%s'?>\n" % encoding)\r
+ self._write(file, self._root, encoding, {})\r
+\r
+ def _write(self, file, node, encoding, namespaces):\r
+ # write XML to file\r
+ tag = node.tag\r
+ if tag is Comment:\r
+ file.write("<!-- %s -->" % _escape_cdata(node.text, encoding))\r
+ elif tag is ProcessingInstruction:\r
+ file.write("<?%s?>" % _escape_cdata(node.text, encoding))\r
+ else:\r
+ items = node.items()\r
+ xmlns_items = [] # new namespaces in this scope\r
+ try:\r
+ if isinstance(tag, QName) or tag[:1] == "{":\r
+ tag, xmlns = fixtag(tag, namespaces)\r
+ if xmlns: xmlns_items.append(xmlns)\r
+ except TypeError:\r
+ _raise_serialization_error(tag)\r
+ file.write("<" + _encode(tag, encoding))\r
+ if items or xmlns_items:\r
+ items.sort() # lexical order\r
+ for k, v in items:\r
+ try:\r
+ if isinstance(k, QName) or k[:1] == "{":\r
+ k, xmlns = fixtag(k, namespaces)\r
+ if xmlns: xmlns_items.append(xmlns)\r
+ except TypeError:\r
+ _raise_serialization_error(k)\r
+ try:\r
+ if isinstance(v, QName):\r
+ v, xmlns = fixtag(v, namespaces)\r
+ if xmlns: xmlns_items.append(xmlns)\r
+ except TypeError:\r
+ _raise_serialization_error(v)\r
+ file.write(" %s=\"%s\"" % (_encode(k, encoding),\r
+ _escape_attrib(v, encoding)))\r
+ for k, v in xmlns_items:\r
+ file.write(" %s=\"%s\"" % (_encode(k, encoding),\r
+ _escape_attrib(v, encoding)))\r
+ if node.text or len(node):\r
+ file.write(">")\r
+ if node.text:\r
+ file.write(_escape_cdata(node.text, encoding))\r
+ for n in node:\r
+ self._write(file, n, encoding, namespaces)\r
+ file.write("</" + _encode(tag, encoding) + ">")\r
+ else:\r
+ file.write(" />")\r
+ for k, v in xmlns_items:\r
+ del namespaces[v]\r
+ if node.tail:\r
+ file.write(_escape_cdata(node.tail, encoding))\r
+\r
+# --------------------------------------------------------------------\r
+# helpers\r
+\r
+##\r
+# Checks if an object appears to be a valid element object.\r
+#\r
+# @param An element instance.\r
+# @return A true value if this is an element object.\r
+# @defreturn flag\r
+\r
+def iselement(element):\r
+ # FIXME: not sure about this; might be a better idea to look\r
+ # for tag/attrib/text attributes\r
+ return isinstance(element, _ElementInterface) or hasattr(element, "tag")\r
+\r
+##\r
+# Writes an element tree or element structure to sys.stdout. This\r
+# function should be used for debugging only.\r
+# <p>\r
+# The exact output format is implementation dependent. In this\r
+# version, it's written as an ordinary XML file.\r
+#\r
+# @param elem An element tree or an individual element.\r
+\r
+def dump(elem):\r
+ # debugging\r
+ if not isinstance(elem, ElementTree):\r
+ elem = ElementTree(elem)\r
+ elem.write(sys.stdout)\r
+ tail = elem.getroot().tail\r
+ if not tail or tail[-1] != "\n":\r
+ sys.stdout.write("\n")\r
+\r
+def _encode(s, encoding):\r
+ try:\r
+ return s.encode(encoding)\r
+ except AttributeError:\r
+ return s # 1.5.2: assume the string uses the right encoding\r
+\r
+if sys.version[:3] == "1.5":\r
+ _escape = re.compile(r"[&<>\"\x80-\xff]+") # 1.5.2\r
+else:\r
+ _escape = re.compile(eval(r'u"[&<>\"\u0080-\uffff]+"'))\r
+\r
+_escape_map = {\r
+ "&": "&",\r
+ "<": "<",\r
+ ">": ">",\r
+ '"': """,\r
+}\r
+\r
+_namespace_map = {\r
+ # "well-known" namespace prefixes\r
+ "http://www.w3.org/XML/1998/namespace": "xml",\r
+ "http://www.w3.org/1999/xhtml": "html",\r
+ "http://www.w3.org/1999/02/22-rdf-syntax-ns#": "rdf",\r
+ "http://schemas.xmlsoap.org/wsdl/": "wsdl",\r
+}\r
+\r
+def _raise_serialization_error(text):\r
+ raise TypeError(\r
+ "cannot serialize %r (type %s)" % (text, type(text).__name__)\r
+ )\r
+\r
+def _encode_entity(text, pattern=_escape):\r
+ # map reserved and non-ascii characters to numerical entities\r
+ def escape_entities(m, map=_escape_map):\r
+ out = []\r
+ append = out.append\r
+ for char in m.group():\r
+ text = map.get(char)\r
+ if text is None:\r
+ text = "&#%d;" % ord(char)\r
+ append(text)\r
+ return string.join(out, "")\r
+ try:\r
+ return _encode(pattern.sub(escape_entities, text), "ascii")\r
+ except TypeError:\r
+ _raise_serialization_error(text)\r
+\r
+#\r
+# the following functions assume an ascii-compatible encoding\r
+# (or "utf-16")\r
+\r
+def _escape_cdata(text, encoding=None, replace=string.replace):\r
+ # escape character data\r
+ try:\r
+ if encoding:\r
+ try:\r
+ text = _encode(text, encoding)\r
+ except UnicodeError:\r
+ return _encode_entity(text)\r
+ text = replace(text, "&", "&")\r
+ text = replace(text, "<", "<")\r
+ text = replace(text, ">", ">")\r
+ return text\r
+ except (TypeError, AttributeError):\r
+ _raise_serialization_error(text)\r
+\r
+def _escape_attrib(text, encoding=None, replace=string.replace):\r
+ # escape attribute value\r
+ try:\r
+ if encoding:\r
+ try:\r
+ text = _encode(text, encoding)\r
+ except UnicodeError:\r
+ return _encode_entity(text)\r
+ text = replace(text, "&", "&")\r
+ text = replace(text, "'", "'") # FIXME: overkill\r
+ text = replace(text, "\"", """)\r
+ text = replace(text, "<", "<")\r
+ text = replace(text, ">", ">")\r
+ return text\r
+ except (TypeError, AttributeError):\r
+ _raise_serialization_error(text)\r
+\r
+def fixtag(tag, namespaces):\r
+ # given a decorated tag (of the form {uri}tag), return prefixed\r
+ # tag and namespace declaration, if any\r
+ if isinstance(tag, QName):\r
+ tag = tag.text\r
+ namespace_uri, tag = string.split(tag[1:], "}", 1)\r
+ prefix = namespaces.get(namespace_uri)\r
+ if prefix is None:\r
+ prefix = _namespace_map.get(namespace_uri)\r
+ if prefix is None:\r
+ prefix = "ns%d" % len(namespaces)\r
+ namespaces[namespace_uri] = prefix\r
+ if prefix == "xml":\r
+ xmlns = None\r
+ else:\r
+ xmlns = ("xmlns:%s" % prefix, namespace_uri)\r
+ else:\r
+ xmlns = None\r
+ return "%s:%s" % (prefix, tag), xmlns\r
+\r
+##\r
+# Parses an XML document into an element tree.\r
+#\r
+# @param source A filename or file object containing XML data.\r
+# @param parser An optional parser instance. If not given, the\r
+# standard {@link XMLTreeBuilder} parser is used.\r
+# @return An ElementTree instance\r
+\r
+def parse(source, parser=None):\r
+ tree = ElementTree()\r
+ tree.parse(source, parser)\r
+ return tree\r
+\r
+##\r
+# Parses an XML document into an element tree incrementally, and reports\r
+# what's going on to the user.\r
+#\r
+# @param source A filename or file object containing XML data.\r
+# @param events A list of events to report back. If omitted, only "end"\r
+# events are reported.\r
+# @return A (event, elem) iterator.\r
+\r
+class iterparse:\r
+\r
+ def __init__(self, source, events=None):\r
+ if not hasattr(source, "read"):\r
+ source = open(source, "rb")\r
+ self._file = source\r
+ self._events = []\r
+ self._index = 0\r
+ self.root = self._root = None\r
+ self._parser = XMLTreeBuilder()\r
+ # wire up the parser for event reporting\r
+ parser = self._parser._parser\r
+ append = self._events.append\r
+ if events is None:\r
+ events = ["end"]\r
+ for event in events:\r
+ if event == "start":\r
+ try:\r
+ parser.ordered_attributes = 1\r
+ parser.specified_attributes = 1\r
+ def handler(tag, attrib_in, event=event, append=append,\r
+ start=self._parser._start_list):\r
+ append((event, start(tag, attrib_in)))\r
+ parser.StartElementHandler = handler\r
+ except AttributeError:\r
+ def handler(tag, attrib_in, event=event, append=append,\r
+ start=self._parser._start):\r
+ append((event, start(tag, attrib_in)))\r
+ parser.StartElementHandler = handler\r
+ elif event == "end":\r
+ def handler(tag, event=event, append=append,\r
+ end=self._parser._end):\r
+ append((event, end(tag)))\r
+ parser.EndElementHandler = handler\r
+ elif event == "start-ns":\r
+ def handler(prefix, uri, event=event, append=append):\r
+ try:\r
+ uri = _encode(uri, "ascii")\r
+ except UnicodeError:\r
+ pass\r
+ append((event, (prefix or "", uri)))\r
+ parser.StartNamespaceDeclHandler = handler\r
+ elif event == "end-ns":\r
+ def handler(prefix, event=event, append=append):\r
+ append((event, None))\r
+ parser.EndNamespaceDeclHandler = handler\r
+\r
+ def next(self):\r
+ while 1:\r
+ try:\r
+ item = self._events[self._index]\r
+ except IndexError:\r
+ if self._parser is None:\r
+ self.root = self._root\r
+ try:\r
+ raise StopIteration\r
+ except NameError:\r
+ raise IndexError\r
+ # load event buffer\r
+ del self._events[:]\r
+ self._index = 0\r
+ data = self._file.read(16384)\r
+ if data:\r
+ self._parser.feed(data)\r
+ else:\r
+ self._root = self._parser.close()\r
+ self._parser = None\r
+ else:\r
+ self._index = self._index + 1\r
+ return item\r
+\r
+ try:\r
+ iter\r
+ def __iter__(self):\r
+ return self\r
+ except NameError:\r
+ def __getitem__(self, index):\r
+ return self.next()\r
+\r
+##\r
+# Parses an XML document from a string constant. This function can\r
+# be used to embed "XML literals" in Python code.\r
+#\r
+# @param source A string containing XML data.\r
+# @return An Element instance.\r
+# @defreturn Element\r
+\r
+def XML(text):\r
+ parser = XMLTreeBuilder()\r
+ parser.feed(text)\r
+ return parser.close()\r
+\r
+##\r
+# Parses an XML document from a string constant, and also returns\r
+# a dictionary which maps from element id:s to elements.\r
+#\r
+# @param source A string containing XML data.\r
+# @return A tuple containing an Element instance and a dictionary.\r
+# @defreturn (Element, dictionary)\r
+\r
+def XMLID(text):\r
+ parser = XMLTreeBuilder()\r
+ parser.feed(text)\r
+ tree = parser.close()\r
+ ids = {}\r
+ for elem in tree.getiterator():\r
+ id = elem.get("id")\r
+ if id:\r
+ ids[id] = elem\r
+ return tree, ids\r
+\r
+##\r
+# Parses an XML document from a string constant. Same as {@link #XML}.\r
+#\r
+# @def fromstring(text)\r
+# @param source A string containing XML data.\r
+# @return An Element instance.\r
+# @defreturn Element\r
+\r
+fromstring = XML\r
+\r
+##\r
+# Generates a string representation of an XML element, including all\r
+# subelements.\r
+#\r
+# @param element An Element instance.\r
+# @return An encoded string containing the XML data.\r
+# @defreturn string\r
+\r
+def tostring(element, encoding=None):\r
+ class dummy:\r
+ pass\r
+ data = []\r
+ file = dummy()\r
+ file.write = data.append\r
+ ElementTree(element).write(file, encoding)\r
+ return string.join(data, "")\r
+\r
+##\r
+# Generic element structure builder. This builder converts a sequence\r
+# of {@link #TreeBuilder.start}, {@link #TreeBuilder.data}, and {@link\r
+# #TreeBuilder.end} method calls to a well-formed element structure.\r
+# <p>\r
+# You can use this class to build an element structure using a custom XML\r
+# parser, or a parser for some other XML-like format.\r
+#\r
+# @param element_factory Optional element factory. This factory\r
+# is called to create new Element instances, as necessary.\r
+\r
+class TreeBuilder:\r
+\r
+ def __init__(self, element_factory=None):\r
+ self._data = [] # data collector\r
+ self._elem = [] # element stack\r
+ self._last = None # last element\r
+ self._tail = None # true if we're after an end tag\r
+ if element_factory is None:\r
+ element_factory = _ElementInterface\r
+ self._factory = element_factory\r
+\r
+ ##\r
+ # Flushes the parser buffers, and returns the toplevel documen\r
+ # element.\r
+ #\r
+ # @return An Element instance.\r
+ # @defreturn Element\r
+\r
+ def close(self):\r
+ assert len(self._elem) == 0, "missing end tags"\r
+ assert self._last != None, "missing toplevel element"\r
+ return self._last\r
+\r
+ def _flush(self):\r
+ if self._data:\r
+ if self._last is not None:\r
+ text = string.join(self._data, "")\r
+ if self._tail:\r
+ assert self._last.tail is None, "internal error (tail)"\r
+ self._last.tail = text\r
+ else:\r
+ assert self._last.text is None, "internal error (text)"\r
+ self._last.text = text\r
+ self._data = []\r
+\r
+ ##\r
+ # Adds text to the current element.\r
+ #\r
+ # @param data A string. This should be either an 8-bit string\r
+ # containing ASCII text, or a Unicode string.\r
+\r
+ def data(self, data):\r
+ self._data.append(data)\r
+\r
+ ##\r
+ # Opens a new element.\r
+ #\r
+ # @param tag The element name.\r
+ # @param attrib A dictionary containing element attributes.\r
+ # @return The opened element.\r
+ # @defreturn Element\r
+\r
+ def start(self, tag, attrs):\r
+ self._flush()\r
+ self._last = elem = self._factory(tag, attrs)\r
+ if self._elem:\r
+ self._elem[-1].append(elem)\r
+ self._elem.append(elem)\r
+ self._tail = 0\r
+ return elem\r
+\r
+ ##\r
+ # Closes the current element.\r
+ #\r
+ # @param tag The element name.\r
+ # @return The closed element.\r
+ # @defreturn Element\r
+\r
+ def end(self, tag):\r
+ self._flush()\r
+ self._last = self._elem.pop()\r
+ assert self._last.tag == tag,\\r
+ "end tag mismatch (expected %s, got %s)" % (\r
+ self._last.tag, tag)\r
+ self._tail = 1\r
+ return self._last\r
+\r
+##\r
+# Element structure builder for XML source data, based on the\r
+# <b>expat</b> parser.\r
+#\r
+# @keyparam target Target object. If omitted, the builder uses an\r
+# instance of the standard {@link #TreeBuilder} class.\r
+# @keyparam html Predefine HTML entities. This flag is not supported\r
+# by the current implementation.\r
+# @see #ElementTree\r
+# @see #TreeBuilder\r
+\r
+class XMLTreeBuilder:\r
+\r
+ def __init__(self, html=0, target=None):\r
+ try:\r
+ from xml.parsers import expat\r
+ except ImportError:\r
+ raise ImportError(\r
+ "No module named expat; use SimpleXMLTreeBuilder instead"\r
+ )\r
+ self._parser = parser = expat.ParserCreate(None, "}")\r
+ if target is None:\r
+ target = TreeBuilder()\r
+ self._target = target\r
+ self._names = {} # name memo cache\r
+ # callbacks\r
+ parser.DefaultHandlerExpand = self._default\r
+ parser.StartElementHandler = self._start\r
+ parser.EndElementHandler = self._end\r
+ parser.CharacterDataHandler = self._data\r
+ # let expat do the buffering, if supported\r
+ try:\r
+ self._parser.buffer_text = 1\r
+ except AttributeError:\r
+ pass\r
+ # use new-style attribute handling, if supported\r
+ try:\r
+ self._parser.ordered_attributes = 1\r
+ self._parser.specified_attributes = 1\r
+ parser.StartElementHandler = self._start_list\r
+ except AttributeError:\r
+ pass\r
+ encoding = None\r
+ if not parser.returns_unicode:\r
+ encoding = "utf-8"\r
+ # target.xml(encoding, None)\r
+ self._doctype = None\r
+ self.entity = {}\r
+\r
+ def _fixtext(self, text):\r
+ # convert text string to ascii, if possible\r
+ try:\r
+ return _encode(text, "ascii")\r
+ except UnicodeError:\r
+ return text\r
+\r
+ def _fixname(self, key):\r
+ # expand qname, and convert name string to ascii, if possible\r
+ try:\r
+ name = self._names[key]\r
+ except KeyError:\r
+ name = key\r
+ if "}" in name:\r
+ name = "{" + name\r
+ self._names[key] = name = self._fixtext(name)\r
+ return name\r
+\r
+ def _start(self, tag, attrib_in):\r
+ fixname = self._fixname\r
+ tag = fixname(tag)\r
+ attrib = {}\r
+ for key, value in attrib_in.items():\r
+ attrib[fixname(key)] = self._fixtext(value)\r
+ return self._target.start(tag, attrib)\r
+\r
+ def _start_list(self, tag, attrib_in):\r
+ fixname = self._fixname\r
+ tag = fixname(tag)\r
+ attrib = {}\r
+ if attrib_in:\r
+ for i in range(0, len(attrib_in), 2):\r
+ attrib[fixname(attrib_in[i])] = self._fixtext(attrib_in[i+1])\r
+ return self._target.start(tag, attrib)\r
+\r
+ def _data(self, text):\r
+ return self._target.data(self._fixtext(text))\r
+\r
+ def _end(self, tag):\r
+ return self._target.end(self._fixname(tag))\r
+\r
+ def _default(self, text):\r
+ prefix = text[:1]\r
+ if prefix == "&":\r
+ # deal with undefined entities\r
+ try:\r
+ self._target.data(self.entity[text[1:-1]])\r
+ except KeyError:\r
+ from xml.parsers import expat\r
+ raise expat.error(\r
+ "undefined entity %s: line %d, column %d" %\r
+ (text, self._parser.ErrorLineNumber,\r
+ self._parser.ErrorColumnNumber)\r
+ )\r
+ elif prefix == "<" and text[:9] == "<!DOCTYPE":\r
+ self._doctype = [] # inside a doctype declaration\r
+ elif self._doctype is not None:\r
+ # parse doctype contents\r
+ if prefix == ">":\r
+ self._doctype = None\r
+ return\r
+ text = string.strip(text)\r
+ if not text:\r
+ return\r
+ self._doctype.append(text)\r
+ n = len(self._doctype)\r
+ if n > 2:\r
+ type = self._doctype[1]\r
+ if type == "PUBLIC" and n == 4:\r
+ name, type, pubid, system = self._doctype\r
+ elif type == "SYSTEM" and n == 3:\r
+ name, type, system = self._doctype\r
+ pubid = None\r
+ else:\r
+ return\r
+ if pubid:\r
+ pubid = pubid[1:-1]\r
+ self.doctype(name, pubid, system[1:-1])\r
+ self._doctype = None\r
+\r
+ ##\r
+ # Handles a doctype declaration.\r
+ #\r
+ # @param name Doctype name.\r
+ # @param pubid Public identifier.\r
+ # @param system System identifier.\r
+\r
+ def doctype(self, name, pubid, system):\r
+ pass\r
+\r
+ ##\r
+ # Feeds data to the parser.\r
+ #\r
+ # @param data Encoded data.\r
+\r
+ def feed(self, data):\r
+ self._parser.Parse(data, 0)\r
+\r
+ ##\r
+ # Finishes feeding data to the parser.\r
+ #\r
+ # @return An element structure.\r
+ # @defreturn Element\r
+\r
+ def close(self):\r
+ self._parser.Parse("", 1) # end of data\r
+ tree = self._target.close()\r
+ del self._target, self._parser # get rid of circular references\r
+ return tree\r
projects / python / commitdiff