"""\

minidom.py -- a lightweight DOM implementation.

parse("foo.xml")

parseString("<foo><bar/></foo>")

Todo:

=====

 * convenience methods for getting elements and text.

 * more testing

 * bring some of the writer and linearizer code into conformance with this

        interface

 * SAX 2 namespaces

"""

import xml.dom

from xml.dom import EMPTY_NAMESPACE, EMPTY_PREFIX, XMLNS_NAMESPACE, domreg

from xml.dom.minicompat import *

from xml.dom.xmlbuilder import DOMImplementationLS, DocumentLS

# This is used by the ID-cache invalidation checks; the list isn't

# actually complete, since the nodes being checked will never be the

# DOCUMENT_NODE or DOCUMENT_FRAGMENT_NODE.  (The node being checked is

# the node being added or removed, not the node being modified.)

#

_nodeTypes_with_children = (xml.dom.Node.ELEMENT_NODE,

                            xml.dom.Node.ENTITY_REFERENCE_NODE)

class Node(xml.dom.Node):

    namespaceURI = None # this is non-null only for elements and attributes

    parentNode = None

    ownerDocument = None

    nextSibling = None

    previousSibling = None

    prefix = EMPTY_PREFIX # non-null only for NS elements and attributes

    def __nonzero__(self):

        return True

    def toxml(self, encoding = None):

        return self.toprettyxml("", "", encoding)

    def toprettyxml(self, indent="\t", newl="\n", encoding = None):

        # indent = the indentation string to prepend, per level

        # newl = the newline string to append

        writer = _get_StringIO()

        if encoding is not None:

            import codecs

            # Can't use codecs.getwriter to preserve 2.0 compatibility

            writer = codecs.lookup(encoding)[3](writer)

        if self.nodeType == Node.DOCUMENT_NODE:

            # Can pass encoding only to document, to put it into XML header

            self.writexml(writer, "", indent, newl, encoding)

        else:

            self.writexml(writer, "", indent, newl)

        return writer.getvalue()

    def hasChildNodes(self):

        if self.childNodes:

            return True

        else:

            return False

    def _get_childNodes(self):

        return self.childNodes

    def _get_firstChild(self):

        if self.childNodes:

            return self.childNodes[0]

    def _get_lastChild(self):

        if self.childNodes:

            return self.childNodes[-1]

    def insertBefore(self, newChild, refChild):

        if newChild.nodeType == self.DOCUMENT_FRAGMENT_NODE:

            for c in tuple(newChild.childNodes):

                self.insertBefore(c, refChild)

            ### The DOM does not clearly specify what to return in this case

            return newChild

        if newChild.nodeType not in self._child_node_types:

            raise xml.dom.HierarchyRequestErr(

                "%s cannot be child of %s" % (repr(newChild), repr(self)))

        if newChild.parentNode is not None:

            newChild.parentNode.removeChild(newChild)

        if refChild is None:

            self.appendChild(newChild)

        else:

            try:

                index = self.childNodes.index(refChild)

            except ValueError:

                raise xml.dom.NotFoundErr()

            if newChild.nodeType in _nodeTypes_with_children:

                _clear_id_cache(self)

            self.childNodes.insert(index, newChild)

            newChild.nextSibling = refChild

            refChild.previousSibling = newChild

            if index:

                node = self.childNodes[index-1]

                node.nextSibling = newChild

                newChild.previousSibling = node

            else:

                newChild.previousSibling = None

            newChild.parentNode = self

        return newChild

    def appendChild(self, node):

        if node.nodeType == self.DOCUMENT_FRAGMENT_NODE:

            for c in tuple(node.childNodes):

                self.appendChild(c)

            ### The DOM does not clearly specify what to return in this case

            return node

        if node.nodeType not in self._child_node_types:

            raise xml.dom.HierarchyRequestErr(

                "%s cannot be child of %s" % (repr(node), repr(self)))

        elif node.nodeType in _nodeTypes_with_children:

            _clear_id_cache(self)

        if node.parentNode is not None:

            node.parentNode.removeChild(node)

        _append_child(self, node)

        node.nextSibling = None

        return node

    def replaceChild(self, newChild, oldChild):

        if newChild.nodeType == self.DOCUMENT_FRAGMENT_NODE:

            refChild = oldChild.nextSibling

            self.removeChild(oldChild)

            return self.insertBefore(newChild, refChild)

        if newChild.nodeType not in self._child_node_types:

            raise xml.dom.HierarchyRequestErr(

                "%s cannot be child of %s" % (repr(newChild), repr(self)))

        if newChild is oldChild:

            return

        if newChild.parentNode is not None:

            newChild.parentNode.removeChild(newChild)

        try:

            index = self.childNodes.index(oldChild)

        except ValueError:

            raise xml.dom.NotFoundErr()

        self.childNodes[index] = newChild

        newChild.parentNode = self

        oldChild.parentNode = None

        if (newChild.nodeType in _nodeTypes_with_children

            or oldChild.nodeType in _nodeTypes_with_children):

            _clear_id_cache(self)

        newChild.nextSibling = oldChild.nextSibling

        newChild.previousSibling = oldChild.previousSibling

        oldChild.nextSibling = None

        oldChild.previousSibling = None

        if newChild.previousSibling:

            newChild.previousSibling.nextSibling = newChild

        if newChild.nextSibling:

            newChild.nextSibling.previousSibling = newChild

        return oldChild

    def removeChild(self, oldChild):

        try:

            self.childNodes.remove(oldChild)

        except ValueError:

            raise xml.dom.NotFoundErr()

        if oldChild.nextSibling is not None:

            oldChild.nextSibling.previousSibling = oldChild.previousSibling

        if oldChild.previousSibling is not None:

            oldChild.previousSibling.nextSibling = oldChild.nextSibling

        oldChild.nextSibling = oldChild.previousSibling = None

        if oldChild.nodeType in _nodeTypes_with_children:

            _clear_id_cache(self)

        oldChild.parentNode = None

        return oldChild

    def normalize(self):

        L = []

        for child in self.childNodes:

            if child.nodeType == Node.TEXT_NODE:

                data = child.data

                if data and L and L[-1].nodeType == child.nodeType:

                    # collapse text node

                    node = L[-1]

                    node.data = node.data + child.data

                    node.nextSibling = child.nextSibling

                    child.unlink()

                elif data:

                    if L:

                        L[-1].nextSibling = child

                        child.previousSibling = L[-1]

                    else:

                        child.previousSibling = None

                    L.append(child)

                else:

                    # empty text node; discard

                    child.unlink()

            else:

                if L:

                    L[-1].nextSibling = child

                    child.previousSibling = L[-1]

                else:

                    child.previousSibling = None

                L.append(child)

                if child.nodeType == Node.ELEMENT_NODE:

                    child.normalize()

        self.childNodes[:] = L

    def cloneNode(self, deep):

        return _clone_node(self, deep, self.ownerDocument or self)

    def isSupported(self, feature, version):

        return self.ownerDocument.implementation.hasFeature(feature, version)

    def _get_localName(self):

        # Overridden in Element and Attr where localName can be Non-Null

        return None

    # Node interfaces from Level 3 (WD 9 April 2002)

    def isSameNode(self, other):

        return self is other

    def getInterface(self, feature):

        if self.isSupported(feature, None):

            return self

        else:

            return None

    # The "user data" functions use a dictionary that is only present

    # if some user data has been set, so be careful not to assume it

    # exists.

    def getUserData(self, key):

        try:

            return self._user_data[key][0]

        except (AttributeError, KeyError):

            return None

    def setUserData(self, key, data, handler):

        old = None

        try:

            d = self._user_data

        except AttributeError:

            d = {}

            self._user_data = d

        if d.has_key(key):

            old = d[key][0]

        if data is None:

            # ignore handlers passed for None

            handler = None

            if old is not None:

                del d[key]

        else:

            d[key] = (data, handler)

        return old

    def _call_user_data_handler(self, operation, src, dst):

        if hasattr(self, "_user_data"):

            for key, (data, handler) in self._user_data.items():

                if handler is not None:

                    handler.handle(operation, key, data, src, dst)

    # minidom-specific API:

    def unlink(self):

        self.parentNode = self.ownerDocument = None

        if self.childNodes:

            for child in self.childNodes:

                child.unlink()

            self.childNodes = NodeList()

        self.previousSibling = None

        self.nextSibling = None

defproperty(Node, "firstChild", doc="First child node, or None.")

defproperty(Node, "lastChild",  doc="Last child node, or None.")

defproperty(Node, "localName",  doc="Namespace-local name of this node.")

def _append_child(self, node):

    # fast path with less checks; usable by DOM builders if careful

    childNodes = self.childNodes

    if childNodes:

        last = childNodes[-1]

        node.__dict__["previousSibling"] = last

        last.__dict__["nextSibling"] = node

    childNodes.append(node)

    node.__dict__["parentNode"] = self

def _in_document(node):

    # return True iff node is part of a document tree

    while node is not None:

        if node.nodeType == Node.DOCUMENT_NODE:

            return True

        node = node.parentNode

    return False

def _write_data(writer, data):

    "Writes datachars to writer."

    data = data.replace("&", "&amp;").replace("<", "&lt;")

    data = data.replace("\"", """).replace(">", ">")

    writer.write(data)

def _get_elements_by_tagName_helper(parent, name, rc):

    for node in parent.childNodes:

        if node.nodeType == Node.ELEMENT_NODE and \

            (name == "*" or node.tagName == name):

            rc.append(node)

        _get_elements_by_tagName_helper(node, name, rc)

    return rc

def _get_elements_by_tagName_ns_helper(parent, nsURI, localName, rc):

    for node in parent.childNodes:

        if node.nodeType == Node.ELEMENT_NODE:

            if ((localName == "*" or node.localName == localName) and

                (nsURI == "*" or node.namespaceURI == nsURI)):

                rc.append(node)

            _get_elements_by_tagName_ns_helper(node, nsURI, localName, rc)

    return rc

class DocumentFragment(Node):

    nodeType = Node.DOCUMENT_FRAGMENT_NODE

    nodeName = "#document-fragment"

    nodeValue = None

    attributes = None

    parentNode = None

    _child_node_types = (Node.ELEMENT_NODE,

                         Node.TEXT_NODE,

                         Node.CDATA_SECTION_NODE,

                         Node.ENTITY_REFERENCE_NODE,

                         Node.PROCESSING_INSTRUCTION_NODE,

                         Node.COMMENT_NODE,

                         Node.NOTATION_NODE)

    def __init__(self):

        self.childNodes = NodeList()

class Attr(Node):

    nodeType = Node.ATTRIBUTE_NODE

    attributes = None

    ownerElement = None

    specified = False

    _is_id = False

    _child_node_types = (Node.TEXT_NODE, Node.ENTITY_REFERENCE_NODE)

    def __init__(self, qName, namespaceURI=EMPTY_NAMESPACE, localName=None,

                 prefix=None):

        # skip setattr for performance

        d = self.__dict__

        d["nodeName"] = d["name"] = qName

        d["namespaceURI"] = namespaceURI

        d["prefix"] = prefix

        d['childNodes'] = NodeList()

        # Add the single child node that represents the value of the attr

        self.childNodes.append(Text())

        # nodeValue and value are set elsewhere

    def _get_localName(self):

        return self.nodeName.split(":", 1)[-1]

    def _get_name(self):

        return self.name

    def _get_specified(self):

        return self.specified

    def __setattr__(self, name, value):

        d = self.__dict__

        if name in ("value", "nodeValue"):

            d["value"] = d["nodeValue"] = value

            d2 = self.childNodes[0].__dict__

            d2["data"] = d2["nodeValue"] = value

            if self.ownerElement is not None:

                _clear_id_cache(self.ownerElement)

        elif name in ("name", "nodeName"):

            d["name"] = d["nodeName"] = value

            if self.ownerElement is not None:

                _clear_id_cache(self.ownerElement)

        else:

            d[name] = value

    def _set_prefix(self, prefix):

        nsuri = self.namespaceURI

        if prefix == "xmlns":

            if nsuri and nsuri != XMLNS_NAMESPACE:

                raise xml.dom.NamespaceErr(

                    "illegal use of 'xmlns' prefix for the wrong namespace")

        d = self.__dict__

        d['prefix'] = prefix

        if prefix is None:

            newName = self.localName

        else:

            newName = "%s:%s" % (prefix, self.localName)

        if self.ownerElement:

            _clear_id_cache(self.ownerElement)

        d['nodeName'] = d['name'] = newName

    def _set_value(self, value):

        d = self.__dict__

        d['value'] = d['nodeValue'] = value

        if self.ownerElement:

            _clear_id_cache(self.ownerElement)

        self.childNodes[0].data = value

    def unlink(self):

        # This implementation does not call the base implementation

        # since most of that is not needed, and the expense of the

        # method call is not warranted.  We duplicate the removal of

        # children, but that's all we needed from the base class.

        elem = self.ownerElement

        if elem is not None:

            del elem._attrs[self.nodeName]

            del elem._attrsNS[(self.namespaceURI, self.localName)]

            if self._is_id:

                self._is_id = False

                elem._magic_id_nodes -= 1

                self.ownerDocument._magic_id_count -= 1

        for child in self.childNodes:

            child.unlink()

        del self.childNodes[:]

    def _get_isId(self):

        if self._is_id:

            return True

        doc = self.ownerDocument

        elem = self.ownerElement

        if doc is None or elem is None:

            return False

        info = doc._get_elem_info(elem)

        if info is None:

            return False

        if self.namespaceURI:

            return info.isIdNS(self.namespaceURI, self.localName)

        else:

            return info.isId(self.nodeName)

    def _get_schemaType(self):

        doc = self.ownerDocument

        elem = self.ownerElement

        if doc is None or elem is None:

            return _no_type

        info = doc._get_elem_info(elem)

        if info is None:

            return _no_type

        if self.namespaceURI:

            return info.getAttributeTypeNS(self.namespaceURI, self.localName)

        else:

            return info.getAttributeType(self.nodeName)

defproperty(Attr, "isId",       doc="True if this attribute is an ID.")

defproperty(Attr, "localName",  doc="Namespace-local name of this attribute.")

defproperty(Attr, "schemaType", doc="Schema type for this attribute.")

class NamedNodeMap(object):

    """The attribute list is a transient interface to the underlying

    dictionaries.  Mutations here will change the underlying element's

    dictionary.

    Ordering is imposed artificially and does not reflect the order of

    attributes as found in an input document.

    """

    __slots__ = ('_attrs', '_attrsNS', '_ownerElement')

    def __init__(self, attrs, attrsNS, ownerElement):

        self._attrs = attrs

        self._attrsNS = attrsNS

        self._ownerElement = ownerElement

    def _get_length(self):

        return len(self._attrs)

    def item(self, index):

        try:

            return self[self._attrs.keys()[index]]

        except IndexError:

            return None

    def items(self):

        L = []

        for node in self._attrs.values():

            L.append((node.nodeName, node.value))

        return L

    def itemsNS(self):

        L = []

        for node in self._attrs.values():

            L.append(((node.namespaceURI, node.localName), node.value))

        return L

    def has_key(self, key):

        if isinstance(key, StringTypes):

            return self._attrs.has_key(key)

        else:

            return self._attrsNS.has_key(key)

    def keys(self):

        return self._attrs.keys()

    def keysNS(self):

        return self._attrsNS.keys()

    def values(self):

        return self._attrs.values()

    def get(self, name, value=None):

        return self._attrs.get(name, value)

    __len__ = _get_length

    def __cmp__(self, other):

        if self._attrs is getattr(other, "_attrs", None):

            return 0

        else:

            return cmp(id(self), id(other))

    def __getitem__(self, attname_or_tuple):

        if isinstance(attname_or_tuple, tuple):

            return self._attrsNS[attname_or_tuple]

        else:

            return self._attrs[attname_or_tuple]

    # same as set