diff options
Diffstat (limited to 'Lib/xml/dom')
| -rw-r--r-- | Lib/xml/dom/NodeFilter.py | 27 | ||||
| -rw-r--r-- | Lib/xml/dom/__init__.py | 139 | ||||
| -rw-r--r-- | Lib/xml/dom/domreg.py | 99 | ||||
| -rw-r--r-- | Lib/xml/dom/expatbuilder.py | 983 | ||||
| -rw-r--r-- | Lib/xml/dom/minicompat.py | 110 | ||||
| -rw-r--r-- | Lib/xml/dom/minidom.py | 1936 | ||||
| -rw-r--r-- | Lib/xml/dom/pulldom.py | 351 | ||||
| -rw-r--r-- | Lib/xml/dom/xmlbuilder.py | 386 |
8 files changed, 4031 insertions, 0 deletions
diff --git a/Lib/xml/dom/NodeFilter.py b/Lib/xml/dom/NodeFilter.py new file mode 100644 index 0000000..fc05245 --- /dev/null +++ b/Lib/xml/dom/NodeFilter.py @@ -0,0 +1,27 @@ +# This is the Python mapping for interface NodeFilter from +# DOM2-Traversal-Range. It contains only constants. + +class NodeFilter: + """ + This is the DOM2 NodeFilter interface. It contains only constants. + """ + FILTER_ACCEPT = 1 + FILTER_REJECT = 2 + FILTER_SKIP = 3 + + SHOW_ALL = 0xFFFFFFFFL + SHOW_ELEMENT = 0x00000001 + SHOW_ATTRIBUTE = 0x00000002 + SHOW_TEXT = 0x00000004 + SHOW_CDATA_SECTION = 0x00000008 + SHOW_ENTITY_REFERENCE = 0x00000010 + SHOW_ENTITY = 0x00000020 + SHOW_PROCESSING_INSTRUCTION = 0x00000040 + SHOW_COMMENT = 0x00000080 + SHOW_DOCUMENT = 0x00000100 + SHOW_DOCUMENT_TYPE = 0x00000200 + SHOW_DOCUMENT_FRAGMENT = 0x00000400 + SHOW_NOTATION = 0x00000800 + + def acceptNode(self, node): + raise NotImplementedError diff --git a/Lib/xml/dom/__init__.py b/Lib/xml/dom/__init__.py new file mode 100644 index 0000000..6363d00 --- /dev/null +++ b/Lib/xml/dom/__init__.py @@ -0,0 +1,139 @@ +"""W3C Document Object Model implementation for Python. + +The Python mapping of the Document Object Model is documented in the +Python Library Reference in the section on the xml.dom package. + +This package contains the following modules: + +minidom -- A simple implementation of the Level 1 DOM with namespace + support added (based on the Level 2 specification) and other + minor Level 2 functionality. + +pulldom -- DOM builder supporting on-demand tree-building for selected + subtrees of the document. + +""" + + +class Node: + """Class giving the NodeType constants.""" + + # DOM implementations may use this as a base class for their own + # Node implementations. If they don't, the constants defined here + # should still be used as the canonical definitions as they match + # the values given in the W3C recommendation. Client code can + # safely refer to these values in all tests of Node.nodeType + # values. + + ELEMENT_NODE = 1 + ATTRIBUTE_NODE = 2 + TEXT_NODE = 3 + CDATA_SECTION_NODE = 4 + ENTITY_REFERENCE_NODE = 5 + ENTITY_NODE = 6 + PROCESSING_INSTRUCTION_NODE = 7 + COMMENT_NODE = 8 + DOCUMENT_NODE = 9 + DOCUMENT_TYPE_NODE = 10 + DOCUMENT_FRAGMENT_NODE = 11 + NOTATION_NODE = 12 + + +#ExceptionCode +INDEX_SIZE_ERR = 1 +DOMSTRING_SIZE_ERR = 2 +HIERARCHY_REQUEST_ERR = 3 +WRONG_DOCUMENT_ERR = 4 +INVALID_CHARACTER_ERR = 5 +NO_DATA_ALLOWED_ERR = 6 +NO_MODIFICATION_ALLOWED_ERR = 7 +NOT_FOUND_ERR = 8 +NOT_SUPPORTED_ERR = 9 +INUSE_ATTRIBUTE_ERR = 10 +INVALID_STATE_ERR = 11 +SYNTAX_ERR = 12 +INVALID_MODIFICATION_ERR = 13 +NAMESPACE_ERR = 14 +INVALID_ACCESS_ERR = 15 +VALIDATION_ERR = 16 + + +class DOMException(Exception): + """Abstract base class for DOM exceptions. + Exceptions with specific codes are specializations of this class.""" + + def __init__(self, *args, **kw): + if self.__class__ is DOMException: + raise RuntimeError( + "DOMException should not be instantiated directly") + Exception.__init__(self, *args, **kw) + + def _get_code(self): + return self.code + + +class IndexSizeErr(DOMException): + code = INDEX_SIZE_ERR + +class DomstringSizeErr(DOMException): + code = DOMSTRING_SIZE_ERR + +class HierarchyRequestErr(DOMException): + code = HIERARCHY_REQUEST_ERR + +class WrongDocumentErr(DOMException): + code = WRONG_DOCUMENT_ERR + +class InvalidCharacterErr(DOMException): + code = INVALID_CHARACTER_ERR + +class NoDataAllowedErr(DOMException): + code = NO_DATA_ALLOWED_ERR + +class NoModificationAllowedErr(DOMException): + code = NO_MODIFICATION_ALLOWED_ERR + +class NotFoundErr(DOMException): + code = NOT_FOUND_ERR + +class NotSupportedErr(DOMException): + code = NOT_SUPPORTED_ERR + +class InuseAttributeErr(DOMException): + code = INUSE_ATTRIBUTE_ERR + +class InvalidStateErr(DOMException): + code = INVALID_STATE_ERR + +class SyntaxErr(DOMException): + code = SYNTAX_ERR + +class InvalidModificationErr(DOMException): + code = INVALID_MODIFICATION_ERR + +class NamespaceErr(DOMException): + code = NAMESPACE_ERR + +class InvalidAccessErr(DOMException): + code = INVALID_ACCESS_ERR + +class ValidationErr(DOMException): + code = VALIDATION_ERR + +class UserDataHandler: + """Class giving the operation constants for UserDataHandler.handle().""" + + # Based on DOM Level 3 (WD 9 April 2002) + + NODE_CLONED = 1 + NODE_IMPORTED = 2 + NODE_DELETED = 3 + NODE_RENAMED = 4 + +XML_NAMESPACE = "http://www.w3.org/XML/1998/namespace" +XMLNS_NAMESPACE = "http://www.w3.org/2000/xmlns/" +XHTML_NAMESPACE = "http://www.w3.org/1999/xhtml" +EMPTY_NAMESPACE = None +EMPTY_PREFIX = None + +from domreg import getDOMImplementation,registerDOMImplementation diff --git a/Lib/xml/dom/domreg.py b/Lib/xml/dom/domreg.py new file mode 100644 index 0000000..684c436 --- /dev/null +++ b/Lib/xml/dom/domreg.py @@ -0,0 +1,99 @@ +"""Registration facilities for DOM. This module should not be used +directly. Instead, the functions getDOMImplementation and +registerDOMImplementation should be imported from xml.dom.""" + +from xml.dom.minicompat import * # isinstance, StringTypes + +# This is a list of well-known implementations. Well-known names +# should be published by posting to xml-sig@python.org, and are +# subsequently recorded in this file. + +well_known_implementations = { + 'minidom':'xml.dom.minidom', + '4DOM': 'xml.dom.DOMImplementation', + } + +# DOM implementations not officially registered should register +# themselves with their + +registered = {} + +def registerDOMImplementation(name, factory): + """registerDOMImplementation(name, factory) + + Register the factory function with the name. The factory function + should return an object which implements the DOMImplementation + interface. The factory function can either return the same object, + or a new one (e.g. if that implementation supports some + customization).""" + + registered[name] = factory + +def _good_enough(dom, features): + "_good_enough(dom, features) -> Return 1 if the dom offers the features" + for f,v in features: + if not dom.hasFeature(f,v): + return 0 + return 1 + +def getDOMImplementation(name = None, features = ()): + """getDOMImplementation(name = None, features = ()) -> DOM implementation. + + Return a suitable DOM implementation. The name is either + well-known, the module name of a DOM implementation, or None. If + it is not None, imports the corresponding module and returns + DOMImplementation object if the import succeeds. + + If name is not given, consider the available implementations to + find one with the required feature set. If no implementation can + be found, raise an ImportError. The features list must be a sequence + of (feature, version) pairs which are passed to hasFeature.""" + + import os + creator = None + mod = well_known_implementations.get(name) + if mod: + mod = __import__(mod, {}, {}, ['getDOMImplementation']) + return mod.getDOMImplementation() + elif name: + return registered[name]() + elif os.environ.has_key("PYTHON_DOM"): + return getDOMImplementation(name = os.environ["PYTHON_DOM"]) + + # User did not specify a name, try implementations in arbitrary + # order, returning the one that has the required features + if isinstance(features, StringTypes): + features = _parse_feature_string(features) + for creator in registered.values(): + dom = creator() + if _good_enough(dom, features): + return dom + + for creator in well_known_implementations.keys(): + try: + dom = getDOMImplementation(name = creator) + except StandardError: # typically ImportError, or AttributeError + continue + if _good_enough(dom, features): + return dom + + raise ImportError,"no suitable DOM implementation found" + +def _parse_feature_string(s): + features = [] + parts = s.split() + i = 0 + length = len(parts) + while i < length: + feature = parts[i] + if feature[0] in "0123456789": + raise ValueError, "bad feature name: %r" % (feature,) + i = i + 1 + version = None + if i < length: + v = parts[i] + if v[0] in "0123456789": + i = i + 1 + version = v + features.append((feature, version)) + return tuple(features) diff --git a/Lib/xml/dom/expatbuilder.py b/Lib/xml/dom/expatbuilder.py new file mode 100644 index 0000000..a2f8a33 --- /dev/null +++ b/Lib/xml/dom/expatbuilder.py @@ -0,0 +1,983 @@ +"""Facility to use the Expat parser to load a minidom instance +from a string or file. + +This avoids all the overhead of SAX and pulldom to gain performance. +""" + +# Warning! +# +# This module is tightly bound to the implementation details of the +# minidom DOM and can't be used with other DOM implementations. This +# is due, in part, to a lack of appropriate methods in the DOM (there is +# no way to create Entity and Notation nodes via the DOM Level 2 +# interface), and for performance. The later is the cause of some fairly +# cryptic code. +# +# Performance hacks: +# +# - .character_data_handler() has an extra case in which continuing +# data is appended to an existing Text node; this can be a +# speedup since pyexpat can break up character data into multiple +# callbacks even though we set the buffer_text attribute on the +# parser. This also gives us the advantage that we don't need a +# separate normalization pass. +# +# - Determining that a node exists is done using an identity comparison +# with None rather than a truth test; this avoids searching for and +# calling any methods on the node object if it exists. (A rather +# nice speedup is achieved this way as well!) + +from xml.dom import xmlbuilder, minidom, Node +from xml.dom import EMPTY_NAMESPACE, EMPTY_PREFIX, XMLNS_NAMESPACE +from xml.parsers import expat +from xml.dom.minidom import _append_child, _set_attribute_node +from xml.dom.NodeFilter import NodeFilter + +from xml.dom.minicompat import * + +TEXT_NODE = Node.TEXT_NODE +CDATA_SECTION_NODE = Node.CDATA_SECTION_NODE +DOCUMENT_NODE = Node.DOCUMENT_NODE + +FILTER_ACCEPT = xmlbuilder.DOMBuilderFilter.FILTER_ACCEPT +FILTER_REJECT = xmlbuilder.DOMBuilderFilter.FILTER_REJECT +FILTER_SKIP = xmlbuilder.DOMBuilderFilter.FILTER_SKIP +FILTER_INTERRUPT = xmlbuilder.DOMBuilderFilter.FILTER_INTERRUPT + +theDOMImplementation = minidom.getDOMImplementation() + +# Expat typename -> TypeInfo +_typeinfo_map = { + "CDATA": minidom.TypeInfo(None, "cdata"), + "ENUM": minidom.TypeInfo(None, "enumeration"), + "ENTITY": minidom.TypeInfo(None, "entity"), + "ENTITIES": minidom.TypeInfo(None, "entities"), + "ID": minidom.TypeInfo(None, "id"), + "IDREF": minidom.TypeInfo(None, "idref"), + "IDREFS": minidom.TypeInfo(None, "idrefs"), + "NMTOKEN": minidom.TypeInfo(None, "nmtoken"), + "NMTOKENS": minidom.TypeInfo(None, "nmtokens"), + } + +class ElementInfo(object): + __slots__ = '_attr_info', '_model', 'tagName' + + def __init__(self, tagName, model=None): + self.tagName = tagName + self._attr_info = [] + self._model = model + + def __getstate__(self): + return self._attr_info, self._model, self.tagName + + def __setstate__(self, state): + self._attr_info, self._model, self.tagName = state + + def getAttributeType(self, aname): + for info in self._attr_info: + if info[1] == aname: + t = info[-2] + if t[0] == "(": + return _typeinfo_map["ENUM"] + else: + return _typeinfo_map[info[-2]] + return minidom._no_type + + def getAttributeTypeNS(self, namespaceURI, localName): + return minidom._no_type + + def isElementContent(self): + if self._model: + type = self._model[0] + return type not in (expat.model.XML_CTYPE_ANY, + expat.model.XML_CTYPE_MIXED) + else: + return False + + def isEmpty(self): + if self._model: + return self._model[0] == expat.model.XML_CTYPE_EMPTY + else: + return False + + def isId(self, aname): + for info in self._attr_info: + if info[1] == aname: + return info[-2] == "ID" + return False + + def isIdNS(self, euri, ename, auri, aname): + # not sure this is meaningful + return self.isId((auri, aname)) + +def _intern(builder, s): + return builder._intern_setdefault(s, s) + +def _parse_ns_name(builder, name): + assert ' ' in name + parts = name.split(' ') + intern = builder._intern_setdefault + if len(parts) == 3: + uri, localname, prefix = parts + prefix = intern(prefix, prefix) + qname = "%s:%s" % (prefix, localname) + qname = intern(qname, qname) + localname = intern(localname, localname) + else: + uri, localname = parts + prefix = EMPTY_PREFIX + qname = localname = intern(localname, localname) + return intern(uri, uri), localname, prefix, qname + + +class ExpatBuilder: + """Document builder that uses Expat to build a ParsedXML.DOM document + instance.""" + + def __init__(self, options=None): + if options is None: + options = xmlbuilder.Options() + self._options = options + if self._options.filter is not None: + self._filter = FilterVisibilityController(self._options.filter) + else: + self._filter = None + # This *really* doesn't do anything in this case, so + # override it with something fast & minimal. + self._finish_start_element = id + self._parser = None + self.reset() + + def createParser(self): + """Create a new parser object.""" + return expat.ParserCreate() + + def getParser(self): + """Return the parser object, creating a new one if needed.""" + if not self._parser: + self._parser = self.createParser() + self._intern_setdefault = self._parser.intern.setdefault + self._parser.buffer_text = True + self._parser.ordered_attributes = True + self._parser.specified_attributes = True + self.install(self._parser) + return self._parser + + def reset(self): + """Free all data structures used during DOM construction.""" + self.document = theDOMImplementation.createDocument( + EMPTY_NAMESPACE, None, None) + self.curNode = self.document + self._elem_info = self.document._elem_info + self._cdata = False + + def install(self, parser): + """Install the callbacks needed to build the DOM into the parser.""" + # This creates circular references! + parser.StartDoctypeDeclHandler = self.start_doctype_decl_handler + parser.StartElementHandler = self.first_element_handler + parser.EndElementHandler = self.end_element_handler + parser.ProcessingInstructionHandler = self.pi_handler + if self._options.entities: + parser.EntityDeclHandler = self.entity_decl_handler + parser.NotationDeclHandler = self.notation_decl_handler + if self._options.comments: + parser.CommentHandler = self.comment_handler + if self._options.cdata_sections: + parser.StartCdataSectionHandler = self.start_cdata_section_handler + parser.EndCdataSectionHandler = self.end_cdata_section_handler + parser.CharacterDataHandler = self.character_data_handler_cdata + else: + parser.CharacterDataHandler = self.character_data_handler + parser.ExternalEntityRefHandler = self.external_entity_ref_handler + parser.XmlDeclHandler = self.xml_decl_handler + parser.ElementDeclHandler = self.element_decl_handler + parser.AttlistDeclHandler = self.attlist_decl_handler + + def parseFile(self, file): + """Parse a document from a file object, returning the document + node.""" + parser = self.getParser() + first_buffer = True + try: + while 1: + buffer = file.read(16*1024) + if not buffer: + break + parser.Parse(buffer, 0) + if first_buffer and self.document.documentElement: + self._setup_subset(buffer) + first_buffer = False + parser.Parse("", True) + except ParseEscape: + pass + doc = self.document + self.reset() + self._parser = None + return doc + + def parseString(self, string): + """Parse a document from a string, returning the document node.""" + parser = self.getParser() + try: + parser.Parse(string, True) + self._setup_subset(string) + except ParseEscape: + pass + doc = self.document + self.reset() + self._parser = None + return doc + + def _setup_subset(self, buffer): + """Load the internal subset if there might be one.""" + if self.document.doctype: + extractor = InternalSubsetExtractor() + extractor.parseString(buffer) + subset = extractor.getSubset() + self.document.doctype.internalSubset = subset + + def start_doctype_decl_handler(self, doctypeName, systemId, publicId, + has_internal_subset): + doctype = self.document.implementation.createDocumentType( + doctypeName, publicId, systemId) + doctype.ownerDocument = self.document + self.document.childNodes.append(doctype) + self.document.doctype = doctype + if self._filter and self._filter.acceptNode(doctype) == FILTER_REJECT: + self.document.doctype = None + del self.document.childNodes[-1] + doctype = None + self._parser.EntityDeclHandler = None + self._parser.NotationDeclHandler = None + if has_internal_subset: + if doctype is not None: + doctype.entities._seq = [] + doctype.notations._seq = [] + self._parser.CommentHandler = None + self._parser.ProcessingInstructionHandler = None + self._parser.EndDoctypeDeclHandler = self.end_doctype_decl_handler + + def end_doctype_decl_handler(self): + if self._options.comments: + self._parser.CommentHandler = self.comment_handler + self._parser.ProcessingInstructionHandler = self.pi_handler + if not (self._elem_info or self._filter): + self._finish_end_element = id + + def pi_handler(self, target, data): + node = self.document.createProcessingInstruction(target, data) + _append_child(self.curNode, node) + if self._filter and self._filter.acceptNode(node) == FILTER_REJECT: + self.curNode.removeChild(node) + + def character_data_handler_cdata(self, data): + childNodes = self.curNode.childNodes + if self._cdata: + if ( self._cdata_continue + and childNodes[-1].nodeType == CDATA_SECTION_NODE): + childNodes[-1].appendData(data) + return + node = self.document.createCDATASection(data) + self._cdata_continue = True + elif childNodes and childNodes[-1].nodeType == TEXT_NODE: + node = childNodes[-1] + value = node.data + data + d = node.__dict__ + d['data'] = d['nodeValue'] = value + return + else: + node = minidom.Text() + d = node.__dict__ + d['data'] = d['nodeValue'] = data + d['ownerDocument'] = self.document + _append_child(self.curNode, node) + + def character_data_handler(self, data): + childNodes = self.curNode.childNodes + if childNodes and childNodes[-1].nodeType == TEXT_NODE: + node = childNodes[-1] + d = node.__dict__ + d['data'] = d['nodeValue'] = node.data + data + return + node = minidom.Text() + d = node.__dict__ + d['data'] = d['nodeValue'] = node.data + data + d['ownerDocument'] = self.document + _append_child(self.curNode, node) + + def entity_decl_handler(self, entityName, is_parameter_entity, value, + base, systemId, publicId, notationName): + if is_parameter_entity: + # we don't care about parameter entities for the DOM + return + if not self._options.entities: + return + node = self.document._create_entity(entityName, publicId, + systemId, notationName) + if value is not None: + # internal entity + # node *should* be readonly, but we'll cheat + child = self.document.createTextNode(value) + node.childNodes.append(child) + self.document.doctype.entities._seq.append(node) + if self._filter and self._filter.acceptNode(node) == FILTER_REJECT: + del self.document.doctype.entities._seq[-1] + + def notation_decl_handler(self, notationName, base, systemId, publicId): + node = self.document._create_notation(notationName, publicId, systemId) + self.document.doctype.notations._seq.append(node) + if self._filter and self._filter.acceptNode(node) == FILTER_ACCEPT: + del self.document.doctype.notations._seq[-1] + + def comment_handler(self, data): + node = self.document.createComment(data) + _append_child(self.curNode, node) + if self._filter and self._filter.acceptNode(node) == FILTER_REJECT: + self.curNode.removeChild(node) + + def start_cdata_section_handler(self): + self._cdata = True + self._cdata_continue = False + + def end_cdata_section_handler(self): + self._cdata = False + self._cdata_continue = False + + def external_entity_ref_handler(self, context, base, systemId, publicId): + return 1 + + def first_element_handler(self, name, attributes): + if self._filter is None and not self._elem_info: + self._finish_end_element = id + self.getParser().StartElementHandler = self.start_element_handler + self.start_element_handler(name, attributes) + + def start_element_handler(self, name, attributes): + node = self.document.createElement(name) + _append_child(self.curNode, node) + self.curNode = node + + if attributes: + for i in range(0, len(attributes), 2): + a = minidom.Attr(attributes[i], EMPTY_NAMESPACE, + None, EMPTY_PREFIX) + value = attributes[i+1] + d = a.childNodes[0].__dict__ + d['data'] = d['nodeValue'] = value + d = a.__dict__ + d['value'] = d['nodeValue'] = value + d['ownerDocument'] = self.document + _set_attribute_node(node, a) + + if node is not self.document.documentElement: + self._finish_start_element(node) + + def _finish_start_element(self, node): + if self._filter: + # To be general, we'd have to call isSameNode(), but this + # is sufficient for minidom: + if node is self.document.documentElement: + return + filt = self._filter.startContainer(node) + if filt == FILTER_REJECT: + # ignore this node & all descendents + Rejecter(self) + elif filt == FILTER_SKIP: + # ignore this node, but make it's children become + # children of the parent node + Skipper(self) + else: + return + self.curNode = node.parentNode + node.parentNode.removeChild(node) + node.unlink() + + # If this ever changes, Namespaces.end_element_handler() needs to + # be changed to match. + # + def end_element_handler(self, name): + curNode = self.curNode + self.curNode = curNode.parentNode + self._finish_end_element(curNode) + + def _finish_end_element(self, curNode): + info = self._elem_info.get(curNode.tagName) + if info: + self._handle_white_text_nodes(curNode, info) + if self._filter: + if curNode is self.document.documentElement: + return + if self._filter.acceptNode(curNode) == FILTER_REJECT: + self.curNode.removeChild(curNode) + curNode.unlink() + + def _handle_white_text_nodes(self, node, info): + if (self._options.whitespace_in_element_content + or not info.isElementContent()): + return + + # We have element type information and should remove ignorable + # whitespace; identify for text nodes which contain only + # whitespace. + L = [] + for child in node.childNodes: + if child.nodeType == TEXT_NODE and not child.data.strip(): + L.append(child) + + # Remove ignorable whitespace from the tree. + for child in L: + node.removeChild(child) + + def element_decl_handler(self, name, model): + info = self._elem_info.get(name) + if info is None: + self._elem_info[name] = ElementInfo(name, model) + else: + assert info._model is None + info._model = model + + def attlist_decl_handler(self, elem, name, type, default, required): + info = self._elem_info.get(elem) + if info is None: + info = ElementInfo(elem) + self._elem_info[elem] = info + info._attr_info.append( + [None, name, None, None, default, 0, type, required]) + + def xml_decl_handler(self, version, encoding, standalone): + self.document.version = version + self.document.encoding = encoding + # This is still a little ugly, thanks to the pyexpat API. ;-( + if standalone >= 0: + if standalone: + self.document.standalone = True + else: + self.document.standalone = False + + +# Don't include FILTER_INTERRUPT, since that's checked separately +# where allowed. +_ALLOWED_FILTER_RETURNS = (FILTER_ACCEPT, FILTER_REJECT, FILTER_SKIP) + +class FilterVisibilityController(object): + """Wrapper around a DOMBuilderFilter which implements the checks + to make the whatToShow filter attribute work.""" + + __slots__ = 'filter', + + def __init__(self, filter): + self.filter = filter + + def startContainer(self, node): + mask = self._nodetype_mask[node.nodeType] + if self.filter.whatToShow & mask: + val = self.filter.startContainer(node) + if val == FILTER_INTERRUPT: + raise ParseEscape + if val not in _ALLOWED_FILTER_RETURNS: + raise ValueError, \ + "startContainer() returned illegal value: " + repr(val) + return val + else: + return FILTER_ACCEPT + + def acceptNode(self, node): + mask = self._nodetype_mask[node.nodeType] + if self.filter.whatToShow & mask: + val = self.filter.acceptNode(node) + if val == FILTER_INTERRUPT: + raise ParseEscape + if val == FILTER_SKIP: + # move all child nodes to the parent, and remove this node + parent = node.parentNode + for child in node.childNodes[:]: + parent.appendChild(child) + # node is handled by the caller + return FILTER_REJECT + if val not in _ALLOWED_FILTER_RETURNS: + raise ValueError, \ + "acceptNode() returned illegal value: " + repr(val) + return val + else: + return FILTER_ACCEPT + + _nodetype_mask = { + Node.ELEMENT_NODE: NodeFilter.SHOW_ELEMENT, + Node.ATTRIBUTE_NODE: NodeFilter.SHOW_ATTRIBUTE, + Node.TEXT_NODE: NodeFilter.SHOW_TEXT, + Node.CDATA_SECTION_NODE: NodeFilter.SHOW_CDATA_SECTION, + Node.ENTITY_REFERENCE_NODE: NodeFilter.SHOW_ENTITY_REFERENCE, + Node.ENTITY_NODE: NodeFilter.SHOW_ENTITY, + Node.PROCESSING_INSTRUCTION_NODE: NodeFilter.SHOW_PROCESSING_INSTRUCTION, + Node.COMMENT_NODE: NodeFilter.SHOW_COMMENT, + Node.DOCUMENT_NODE: NodeFilter.SHOW_DOCUMENT, + Node.DOCUMENT_TYPE_NODE: NodeFilter.SHOW_DOCUMENT_TYPE, + Node.DOCUMENT_FRAGMENT_NODE: NodeFilter.SHOW_DOCUMENT_FRAGMENT, + Node.NOTATION_NODE: NodeFilter.SHOW_NOTATION, + } + + +class FilterCrutch(object): + __slots__ = '_builder', '_level', '_old_start', '_old_end' + + def __init__(self, builder): + self._level = 0 + self._builder = builder + parser = builder._parser + self._old_start = parser.StartElementHandler + self._old_end = parser.EndElementHandler + parser.StartElementHandler = self.start_element_handler + parser.EndElementHandler = self.end_element_handler + +class Rejecter(FilterCrutch): + __slots__ = () + + def __init__(self, builder): + FilterCrutch.__init__(self, builder) + parser = builder._parser + for name in ("ProcessingInstructionHandler", + "CommentHandler", + "CharacterDataHandler", + "StartCdataSectionHandler", + "EndCdataSectionHandler", + "ExternalEntityRefHandler", + ): + setattr(parser, name, None) + + def start_element_handler(self, *args): + self._level = self._level + 1 + + def end_element_handler(self, *args): + if self._level == 0: + # restore the old handlers + parser = self._builder._parser + self._builder.install(parser) + parser.StartElementHandler = self._old_start + parser.EndElementHandler = self._old_end + else: + self._level = self._level - 1 + +class Skipper(FilterCrutch): + __slots__ = () + + def start_element_handler(self, *args): + node = self._builder.curNode + self._old_start(*args) + if self._builder.curNode is not node: + self._level = self._level + 1 + + def end_element_handler(self, *args): + if self._level == 0: + # We're popping back out of the node we're skipping, so we + # shouldn't need to do anything but reset the handlers. + self._builder._parser.StartElementHandler = self._old_start + self._builder._parser.EndElementHandler = self._old_end + self._builder = None + else: + self._level = self._level - 1 + self._old_end(*args) + + +# framework document used by the fragment builder. +# Takes a string for the doctype, subset string, and namespace attrs string. + +_FRAGMENT_BUILDER_INTERNAL_SYSTEM_ID = \ + "http://xml.python.org/entities/fragment-builder/internal" + +_FRAGMENT_BUILDER_TEMPLATE = ( + '''\ +<!DOCTYPE wrapper + %%s [ + <!ENTITY fragment-builder-internal + SYSTEM "%s"> +%%s +]> +<wrapper %%s +>&fragment-builder-internal;</wrapper>''' + % _FRAGMENT_BUILDER_INTERNAL_SYSTEM_ID) + + +class FragmentBuilder(ExpatBuilder): + """Builder which constructs document fragments given XML source + text and a context node. + + The context node is expected to provide information about the + namespace declarations which are in scope at the start of the + fragment. + """ + + def __init__(self, context, options=None): + if context.nodeType == DOCUMENT_NODE: + self.originalDocument = context + self.context = context + else: + self.originalDocument = context.ownerDocument + self.context = context + ExpatBuilder.__init__(self, options) + + def reset(self): + ExpatBuilder.reset(self) + self.fragment = None + + def parseFile(self, file): + """Parse a document fragment from a file object, returning the + fragment node.""" + return self.parseString(file.read()) + + def parseString(self, string): + """Parse a document fragment from a string, returning the + fragment node.""" + self._source = string + parser = self.getParser() + doctype = self.originalDocument.doctype + ident = "" + if doctype: + subset = doctype.internalSubset or self._getDeclarations() + if doctype.publicId: + ident = ('PUBLIC "%s" "%s"' + % (doctype.publicId, doctype.systemId)) + elif doctype.systemId: + ident = 'SYSTEM "%s"' % doctype.systemId + else: + subset = "" + nsattrs = self._getNSattrs() # get ns decls from node's ancestors + document = _FRAGMENT_BUILDER_TEMPLATE % (ident, subset, nsattrs) + try: + parser.Parse(document, 1) + except: + self.reset() + raise + fragment = self.fragment + self.reset() +## self._parser = None + return fragment + + def _getDeclarations(self): + """Re-create the internal subset from the DocumentType node. + + This is only needed if we don't already have the + internalSubset as a string. + """ + doctype = self.context.ownerDocument.doctype + s = "" + if doctype: + for i in range(doctype.notations.length): + notation = doctype.notations.item(i) + if s: + s = s + "\n " + s = "%s<!NOTATION %s" % (s, notation.nodeName) + if notation.publicId: + s = '%s PUBLIC "%s"\n "%s">' \ + % (s, notation.publicId, notation.systemId) + else: + s = '%s SYSTEM "%s">' % (s, notation.systemId) + for i in range(doctype.entities.length): + entity = doctype.entities.item(i) + if s: + s = s + "\n " + s = "%s<!ENTITY %s" % (s, entity.nodeName) + if entity.publicId: + s = '%s PUBLIC "%s"\n "%s"' \ + % (s, entity.publicId, entity.systemId) + elif entity.systemId: + s = '%s SYSTEM "%s"' % (s, entity.systemId) + else: + s = '%s "%s"' % (s, entity.firstChild.data) + if entity.notationName: + s = "%s NOTATION %s" % (s, entity.notationName) + s = s + ">" + return s + + def _getNSattrs(self): + return "" + + def external_entity_ref_handler(self, context, base, systemId, publicId): + if systemId == _FRAGMENT_BUILDER_INTERNAL_SYSTEM_ID: + # this entref is the one that we made to put the subtree + # in; all of our given input is parsed in here. + old_document = self.document + old_cur_node = self.curNode + parser = self._parser.ExternalEntityParserCreate(context) + # put the real document back, parse into the fragment to return + self.document = self.originalDocument + self.fragment = self.document.createDocumentFragment() + self.curNode = self.fragment + try: + parser.Parse(self._source, 1) + finally: + self.curNode = old_cur_node + self.document = old_document + self._source = None + return -1 + else: + return ExpatBuilder.external_entity_ref_handler( + self, context, base, systemId, publicId) + + +class Namespaces: + """Mix-in class for builders; adds support for namespaces.""" + + def _initNamespaces(self): + # list of (prefix, uri) ns declarations. Namespace attrs are + # constructed from this and added to the element's attrs. + self._ns_ordered_prefixes = [] + + def createParser(self): + """Create a new namespace-handling parser.""" + parser = expat.ParserCreate(namespace_separator=" ") + parser.namespace_prefixes = True + return parser + + def install(self, parser): + """Insert the namespace-handlers onto the parser.""" + ExpatBuilder.install(self, parser) + if self._options.namespace_declarations: + parser.StartNamespaceDeclHandler = ( + self.start_namespace_decl_handler) + + def start_namespace_decl_handler(self, prefix, uri): + """Push this namespace declaration on our storage.""" + self._ns_ordered_prefixes.append((prefix, uri)) + + def start_element_handler(self, name, attributes): + if ' ' in name: + uri, localname, prefix, qname = _parse_ns_name(self, name) + else: + uri = EMPTY_NAMESPACE + qname = name + localname = None + prefix = EMPTY_PREFIX + node = minidom.Element(qname, uri, prefix, localname) + node.ownerDocument = self.document + _append_child(self.curNode, node) + self.curNode = node + + if self._ns_ordered_prefixes: + for prefix, uri in self._ns_ordered_prefixes: + if prefix: + a = minidom.Attr(_intern(self, 'xmlns:' + prefix), + XMLNS_NAMESPACE, prefix, "xmlns") + else: + a = minidom.Attr("xmlns", XMLNS_NAMESPACE, + "xmlns", EMPTY_PREFIX) + d = a.childNodes[0].__dict__ + d['data'] = d['nodeValue'] = uri + d = a.__dict__ + d['value'] = d['nodeValue'] = uri + d['ownerDocument'] = self.document + _set_attribute_node(node, a) + del self._ns_ordered_prefixes[:] + + if attributes: + _attrs = node._attrs + _attrsNS = node._attrsNS + for i in range(0, len(attributes), 2): + aname = attributes[i] + value = attributes[i+1] + if ' ' in aname: + uri, localname, prefix, qname = _parse_ns_name(self, aname) + a = minidom.Attr(qname, uri, localname, prefix) + _attrs[qname] = a + _attrsNS[(uri, localname)] = a + else: + a = minidom.Attr(aname, EMPTY_NAMESPACE, + aname, EMPTY_PREFIX) + _attrs[aname] = a + _attrsNS[(EMPTY_NAMESPACE, aname)] = a + d = a.childNodes[0].__dict__ + d['data'] = d['nodeValue'] = value + d = a.__dict__ + d['ownerDocument'] = self.document + d['value'] = d['nodeValue'] = value + d['ownerElement'] = node + + if __debug__: + # This only adds some asserts to the original + # end_element_handler(), so we only define this when -O is not + # used. If changing one, be sure to check the other to see if + # it needs to be changed as well. + # + def end_element_handler(self, name): + curNode = self.curNode + if ' ' in name: + uri, localname, prefix, qname = _parse_ns_name(self, name) + assert (curNode.namespaceURI == uri + and curNode.localName == localname + and curNode.prefix == prefix), \ + "element stack messed up! (namespace)" + else: + assert curNode.nodeName == name, \ + "element stack messed up - bad nodeName" + assert curNode.namespaceURI == EMPTY_NAMESPACE, \ + "element stack messed up - bad namespaceURI" + self.curNode = curNode.parentNode + self._finish_end_element(curNode) + + +class ExpatBuilderNS(Namespaces, ExpatBuilder): + """Document builder that supports namespaces.""" + + def reset(self): + ExpatBuilder.reset(self) + self._initNamespaces() + + +class FragmentBuilderNS(Namespaces, FragmentBuilder): + """Fragment builder that supports namespaces.""" + + def reset(self): + FragmentBuilder.reset(self) + self._initNamespaces() + + def _getNSattrs(self): + """Return string of namespace attributes from this element and + ancestors.""" + # XXX This needs to be re-written to walk the ancestors of the + # context to build up the namespace information from + # declarations, elements, and attributes found in context. + # Otherwise we have to store a bunch more data on the DOM + # (though that *might* be more reliable -- not clear). + attrs = "" + context = self.context + L = [] + while context: + if hasattr(context, '_ns_prefix_uri'): + for prefix, uri in context._ns_prefix_uri.items(): + # add every new NS decl from context to L and attrs string + if prefix in L: + continue + L.append(prefix) + if prefix: + declname = "xmlns:" + prefix + else: + declname = "xmlns" + if attrs: + attrs = "%s\n %s='%s'" % (attrs, declname, uri) + else: + attrs = " %s='%s'" % (declname, uri) + context = context.parentNode + return attrs + + +class ParseEscape(Exception): + """Exception raised to short-circuit parsing in InternalSubsetExtractor.""" + pass + +class InternalSubsetExtractor(ExpatBuilder): + """XML processor which can rip out the internal document type subset.""" + + subset = None + + def getSubset(self): + """Return the internal subset as a string.""" + return self.subset + + def parseFile(self, file): + try: + ExpatBuilder.parseFile(self, file) + except ParseEscape: + pass + + def parseString(self, string): + try: + ExpatBuilder.parseString(self, string) + except ParseEscape: + pass + + def install(self, parser): + parser.StartDoctypeDeclHandler = self.start_doctype_decl_handler + parser.StartElementHandler = self.start_element_handler + + def start_doctype_decl_handler(self, name, publicId, systemId, + has_internal_subset): + if has_internal_subset: + parser = self.getParser() + self.subset = [] + parser.DefaultHandler = self.subset.append + parser.EndDoctypeDeclHandler = self.end_doctype_decl_handler + else: + raise ParseEscape() + + def end_doctype_decl_handler(self): + s = ''.join(self.subset).replace('\r\n', '\n').replace('\r', '\n') + self.subset = s + raise ParseEscape() + + def start_element_handler(self, name, attrs): + raise ParseEscape() + + +def parse(file, namespaces=True): + """Parse a document, returning the resulting Document node. + + 'file' may be either a file name or an open file object. + """ + if namespaces: + builder = ExpatBuilderNS() + else: + builder = ExpatBuilder() + + if isinstance(file, StringTypes): + fp = open(file, 'rb') + try: + result = builder.parseFile(fp) + finally: + fp.close() + else: + result = builder.parseFile(file) + return result + + +def parseString(string, namespaces=True): + """Parse a document from a string, returning the resulting + Document node. + """ + if namespaces: + builder = ExpatBuilderNS() + else: + builder = ExpatBuilder() + return builder.parseString(string) + + +def parseFragment(file, context, namespaces=True): + """Parse a fragment of a document, given the context from which it + was originally extracted. context should be the parent of the + node(s) which are in the fragment. + + 'file' may be either a file name or an open file object. + """ + if namespaces: + builder = FragmentBuilderNS(context) + else: + builder = FragmentBuilder(context) + + if isinstance(file, StringTypes): + fp = open(file, 'rb') + try: + result = builder.parseFile(fp) + finally: + fp.close() + else: + result = builder.parseFile(file) + return result + + +def parseFragmentString(string, context, namespaces=True): + """Parse a fragment of a document from a string, given the context + from which it was originally extracted. context should be the + parent of the node(s) which are in the fragment. + """ + if namespaces: + builder = FragmentBuilderNS(context) + else: + builder = FragmentBuilder(context) + return builder.parseString(string) + + +def makeBuilder(options): + """Create a builder based on an Options object.""" + if options.namespaces: + return ExpatBuilderNS(options) + else: + return ExpatBuilder(options) diff --git a/Lib/xml/dom/minicompat.py b/Lib/xml/dom/minicompat.py new file mode 100644 index 0000000..d491fb6 --- /dev/null +++ b/Lib/xml/dom/minicompat.py @@ -0,0 +1,110 @@ +"""Python version compatibility support for minidom.""" + +# This module should only be imported using "import *". +# +# The following names are defined: +# +# NodeList -- lightest possible NodeList implementation +# +# EmptyNodeList -- lightest possible NodeList that is guarateed to +# remain empty (immutable) +# +# StringTypes -- tuple of defined string types +# +# defproperty -- function used in conjunction with GetattrMagic; +# using these together is needed to make them work +# as efficiently as possible in both Python 2.2+ +# and older versions. For example: +# +# class MyClass(GetattrMagic): +# def _get_myattr(self): +# return something +# +# defproperty(MyClass, "myattr", +# "return some value") +# +# For Python 2.2 and newer, this will construct a +# property object on the class, which avoids +# needing to override __getattr__(). It will only +# work for read-only attributes. +# +# For older versions of Python, inheriting from +# GetattrMagic will use the traditional +# __getattr__() hackery to achieve the same effect, +# but less efficiently. +# +# defproperty() should be used for each version of +# the relevant _get_<property>() function. + +__all__ = ["NodeList", "EmptyNodeList", "StringTypes", "defproperty"] + +import xml.dom + +try: + unicode +except NameError: + StringTypes = type(''), +else: + StringTypes = type(''), type(unicode('')) + + +class NodeList(list): + __slots__ = () + + def item(self, index): + if 0 <= index < len(self): + return self[index] + + def _get_length(self): + return len(self) + + def _set_length(self, value): + raise xml.dom.NoModificationAllowedErr( + "attempt to modify read-only attribute 'length'") + + length = property(_get_length, _set_length, + doc="The number of nodes in the NodeList.") + + def __getstate__(self): + return list(self) + + def __setstate__(self, state): + self[:] = state + + +class EmptyNodeList(tuple): + __slots__ = () + + def __add__(self, other): + NL = NodeList() + NL.extend(other) + return NL + + def __radd__(self, other): + NL = NodeList() + NL.extend(other) + return NL + + def item(self, index): + return None + + def _get_length(self): + return 0 + + def _set_length(self, value): + raise xml.dom.NoModificationAllowedErr( + "attempt to modify read-only attribute 'length'") + + length = property(_get_length, _set_length, + doc="The number of nodes in the NodeList.") + + +def defproperty(klass, name, doc): + get = getattr(klass, ("_get_" + name)).im_func + def set(self, value, name=name): + raise xml.dom.NoModificationAllowedErr( + "attempt to modify read-only attribute " + repr(name)) + assert not hasattr(klass, "_set_" + name), \ + "expected not to find _set_" + name + prop = property(get, set, doc=doc) + setattr(klass, name, prop) diff --git a/Lib/xml/dom/minidom.py b/Lib/xml/dom/minidom.py new file mode 100644 index 0000000..3a35781 --- /dev/null +++ b/Lib/xml/dom/minidom.py @@ -0,0 +1,1936 @@ +"""\ +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("&", "&").replace("<", "<") + 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 + def __setitem__(self, attname, value): + if isinstance(value, StringTypes): + try: + node = self._attrs[attname] + except KeyError: + node = Attr(attname) + node.ownerDocument = self._ownerElement.ownerDocument + self.setNamedItem(node) + node.value = value + else: + if not isinstance(value, Attr): + raise TypeError, "value must be a string or Attr object" + node = value + self.setNamedItem(node) + + def getNamedItem(self, name): + try: + return self._attrs[name] + except KeyError: + return None + + def getNamedItemNS(self, namespaceURI, localName): + try: + return self._attrsNS[(namespaceURI, localName)] + except KeyError: + return None + + def removeNamedItem(self, name): + n = self.getNamedItem(name) + if n is not None: + _clear_id_cache(self._ownerElement) + del self._attrs[n.nodeName] + del self._attrsNS[(n.namespaceURI, n.localName)] + if n.__dict__.has_key('ownerElement'): + n.__dict__['ownerElement'] = None + return n + else: + raise xml.dom.NotFoundErr() + + def removeNamedItemNS(self, namespaceURI, localName): + n = self.getNamedItemNS(namespaceURI, localName) + if n is not None: + _clear_id_cache(self._ownerElement) + del self._attrsNS[(n.namespaceURI, n.localName)] + del self._attrs[n.nodeName] + if n.__dict__.has_key('ownerElement'): + n.__dict__['ownerElement'] = None + return n + else: + raise xml.dom.NotFoundErr() + + def setNamedItem(self, node): + if not isinstance(node, Attr): + raise xml.dom.HierarchyRequestErr( + "%s cannot be child of %s" % (repr(node), repr(self))) + old = self._attrs.get(node.name) + if old: + old.unlink() + self._attrs[node.name] = node + self._attrsNS[(node.namespaceURI, node.localName)] = node + node.ownerElement = self._ownerElement + _clear_id_cache(node.ownerElement) + return old + + def setNamedItemNS(self, node): + return self.setNamedItem(node) + + def __delitem__(self, attname_or_tuple): + node = self[attname_or_tuple] + _clear_id_cache(node.ownerElement) + node.unlink() + + def __getstate__(self): + return self._attrs, self._attrsNS, self._ownerElement + + def __setstate__(self, state): + self._attrs, self._attrsNS, self._ownerElement = state + +defproperty(NamedNodeMap, "length", + doc="Number of nodes in the NamedNodeMap.") + +AttributeList = NamedNodeMap + + +class TypeInfo(object): + __slots__ = 'namespace', 'name' + + def __init__(self, namespace, name): + self.namespace = namespace + self.name = name + + def __repr__(self): + if self.namespace: + return "<TypeInfo %r (from %r)>" % (self.name, self.namespace) + else: + return "<TypeInfo %r>" % self.name + + def _get_name(self): + return self.name + + def _get_namespace(self): + return self.namespace + +_no_type = TypeInfo(None, None) + +class Element(Node): + nodeType = Node.ELEMENT_NODE + nodeValue = None + schemaType = _no_type + + _magic_id_nodes = 0 + + _child_node_types = (Node.ELEMENT_NODE, + Node.PROCESSING_INSTRUCTION_NODE, + Node.COMMENT_NODE, + Node.TEXT_NODE, + Node.CDATA_SECTION_NODE, + Node.ENTITY_REFERENCE_NODE) + + def __init__(self, tagName, namespaceURI=EMPTY_NAMESPACE, prefix=None, + localName=None): + self.tagName = self.nodeName = tagName + self.prefix = prefix + self.namespaceURI = namespaceURI + self.childNodes = NodeList() + + self._attrs = {} # attributes are double-indexed: + self._attrsNS = {} # tagName -> Attribute + # URI,localName -> Attribute + # in the future: consider lazy generation + # of attribute objects this is too tricky + # for now because of headaches with + # namespaces. + + def _get_localName(self): + return self.tagName.split(":", 1)[-1] + + def _get_tagName(self): + return self.tagName + + def unlink(self): + for attr in self._attrs.values(): + attr.unlink() + self._attrs = None + self._attrsNS = None + Node.unlink(self) + + def getAttribute(self, attname): + try: + return self._attrs[attname].value + except KeyError: + return "" + + def getAttributeNS(self, namespaceURI, localName): + try: + return self._attrsNS[(namespaceURI, localName)].value + except KeyError: + return "" + + def setAttribute(self, attname, value): + attr = self.getAttributeNode(attname) + if attr is None: + attr = Attr(attname) + # for performance + d = attr.__dict__ + d["value"] = d["nodeValue"] = value + d["ownerDocument"] = self.ownerDocument + self.setAttributeNode(attr) + elif value != attr.value: + d = attr.__dict__ + d["value"] = d["nodeValue"] = value + if attr.isId: + _clear_id_cache(self) + + def setAttributeNS(self, namespaceURI, qualifiedName, value): + prefix, localname = _nssplit(qualifiedName) + attr = self.getAttributeNodeNS(namespaceURI, localname) + if attr is None: + # for performance + attr = Attr(qualifiedName, namespaceURI, localname, prefix) + d = attr.__dict__ + d["prefix"] = prefix + d["nodeName"] = qualifiedName + d["value"] = d["nodeValue"] = value + d["ownerDocument"] = self.ownerDocument + self.setAttributeNode(attr) + else: + d = attr.__dict__ + if value != attr.value: + d["value"] = d["nodeValue"] = value + if attr.isId: + _clear_id_cache(self) + if attr.prefix != prefix: + d["prefix"] = prefix + d["nodeName"] = qualifiedName + + def getAttributeNode(self, attrname): + return self._attrs.get(attrname) + + def getAttributeNodeNS(self, namespaceURI, localName): + return self._attrsNS.get((namespaceURI, localName)) + + def setAttributeNode(self, attr): + if attr.ownerElement not in (None, self): + raise xml.dom.InuseAttributeErr("attribute node already owned") + old1 = self._attrs.get(attr.name, None) + if old1 is not None: + self.removeAttributeNode(old1) + old2 = self._attrsNS.get((attr.namespaceURI, attr.localName), None) + if old2 is not None and old2 is not old1: + self.removeAttributeNode(old2) + _set_attribute_node(self, attr) + + if old1 is not attr: + # It might have already been part of this node, in which case + # it doesn't represent a change, and should not be returned. + return old1 + if old2 is not attr: + return old2 + + setAttributeNodeNS = setAttributeNode + + def removeAttribute(self, name): + try: + attr = self._attrs[name] + except KeyError: + raise xml.dom.NotFoundErr() + self.removeAttributeNode(attr) + + def removeAttributeNS(self, namespaceURI, localName): + try: + attr = self._attrsNS[(namespaceURI, localName)] + except KeyError: + raise xml.dom.NotFoundErr() + self.removeAttributeNode(attr) + + def removeAttributeNode(self, node): + if node is None: + raise xml.dom.NotFoundErr() + try: + self._attrs[node.name] + except KeyError: + raise xml.dom.NotFoundErr() + _clear_id_cache(self) + node.unlink() + # Restore this since the node is still useful and otherwise + # unlinked + node.ownerDocument = self.ownerDocument + + removeAttributeNodeNS = removeAttributeNode + + def hasAttribute(self, name): + return self._attrs.has_key(name) + + def hasAttributeNS(self, namespaceURI, localName): + return self._attrsNS.has_key((namespaceURI, localName)) + + def getElementsByTagName(self, name): + return _get_elements_by_tagName_helper(self, name, NodeList()) + + def getElementsByTagNameNS(self, namespaceURI, localName): + return _get_elements_by_tagName_ns_helper( + self, namespaceURI, localName, NodeList()) + + def __repr__(self): + return "<DOM Element: %s at %#x>" % (self.tagName, id(self)) + + def writexml(self, writer, indent="", addindent="", newl=""): + # 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() + a_names.sort() + + 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(">%s"%(newl)) + for node in self.childNodes: + node.writexml(writer,indent+addindent,addindent,newl) + writer.write("%s</%s>%s" % (indent,self.tagName,newl)) + else: + writer.write("/>%s"%(newl)) + + def _get_attributes(self): + return NamedNodeMap(self._attrs, self._attrsNS, self) + + def hasAttributes(self): + if self._attrs: + return True + else: + return False + + # DOM Level 3 attributes, based on the 22 Oct 2002 draft + + def setIdAttribute(self, name): + idAttr = self.getAttributeNode(name) + self.setIdAttributeNode(idAttr) + + def setIdAttributeNS(self, namespaceURI, localName): + idAttr = self.getAttributeNodeNS(namespaceURI, localName) + self.setIdAttributeNode(idAttr) + + def setIdAttributeNode(self, idAttr): + if idAttr is None or not self.isSameNode(idAttr.ownerElement): + raise xml.dom.NotFoundErr() + if _get_containing_entref(self) is not None: + raise xml.dom.NoModificationAllowedErr() + if not idAttr._is_id: + idAttr.__dict__['_is_id'] = True + self._magic_id_nodes += 1 + self.ownerDocument._magic_id_count += 1 + _clear_id_cache(self) + +defproperty(Element, "attributes", + doc="NamedNodeMap of attributes on the element.") +defproperty(Element, "localName", + doc="Namespace-local name of this element.") + + +def _set_attribute_node(element, attr): + _clear_id_cache(element) + element._attrs[attr.name] = attr + element._attrsNS[(attr.namespaceURI, attr.localName)] = attr + + # This creates a circular reference, but Element.unlink() + # breaks the cycle since the references to the attribute + # dictionaries are tossed. + attr.__dict__['ownerElement'] = element + + +class Childless: + """Mixin that makes childless-ness easy to implement and avoids + the complexity of the Node methods that deal with children. + """ + + attributes = None + childNodes = EmptyNodeList() + firstChild = None + lastChild = None + + def _get_firstChild(self): + return None + + def _get_lastChild(self): + return None + + def appendChild(self, node): + raise xml.dom.HierarchyRequestErr( + self.nodeName + " nodes cannot have children") + + def hasChildNodes(self): + return False + + def insertBefore(self, newChild, refChild): + raise xml.dom.HierarchyRequestErr( + self.nodeName + " nodes do not have children") + + def removeChild(self, oldChild): + raise xml.dom.NotFoundErr( + self.nodeName + " nodes do not have children") + + def replaceChild(self, newChild, oldChild): + raise xml.dom.HierarchyRequestErr( + self.nodeName + " nodes do not have children") + + +class ProcessingInstruction(Childless, Node): + nodeType = Node.PROCESSING_INSTRUCTION_NODE + + def __init__(self, target, data): + self.target = self.nodeName = target + self.data = self.nodeValue = data + + def _get_data(self): + return self.data + def _set_data(self, value): + d = self.__dict__ + d['data'] = d['nodeValue'] = value + + def _get_target(self): + return self.target + def _set_target(self, value): + d = self.__dict__ + d['target'] = d['nodeName'] = value + + def __setattr__(self, name, value): + if name == "data" or name == "nodeValue": + self.__dict__['data'] = self.__dict__['nodeValue'] = value + elif name == "target" or name == "nodeName": + self.__dict__['target'] = self.__dict__['nodeName'] = value + else: + self.__dict__[name] = value + + def writexml(self, writer, indent="", addindent="", newl=""): + writer.write("%s<?%s %s?>%s" % (indent,self.target, self.data, newl)) + + +class CharacterData(Childless, Node): + def _get_length(self): + return len(self.data) + __len__ = _get_length + + def _get_data(self): + return self.__dict__['data'] + def _set_data(self, data): + d = self.__dict__ + d['data'] = d['nodeValue'] = data + + _get_nodeValue = _get_data + _set_nodeValue = _set_data + + def __setattr__(self, name, value): + if name == "data" or name == "nodeValue": + self.__dict__['data'] = self.__dict__['nodeValue'] = value + else: + self.__dict__[name] = value + + def __repr__(self): + data = self.data + if len(data) > 10: + dotdotdot = "..." + else: + dotdotdot = "" + return "<DOM %s node \"%s%s\">" % ( + self.__class__.__name__, data[0:10], dotdotdot) + + def substringData(self, offset, count): + if offset < 0: + raise xml.dom.IndexSizeErr("offset cannot be negative") + if offset >= len(self.data): + raise xml.dom.IndexSizeErr("offset cannot be beyond end of data") + if count < 0: + raise xml.dom.IndexSizeErr("count cannot be negative") + return self.data[offset:offset+count] + + def appendData(self, arg): + self.data = self.data + arg + + def insertData(self, offset, arg): + if offset < 0: + raise xml.dom.IndexSizeErr("offset cannot be negative") + if offset >= len(self.data): + raise xml.dom.IndexSizeErr("offset cannot be beyond end of data") + if arg: + self.data = "%s%s%s" % ( + self.data[:offset], arg, self.data[offset:]) + + def deleteData(self, offset, count): + if offset < 0: + raise xml.dom.IndexSizeErr("offset cannot be negative") + if offset >= len(self.data): + raise xml.dom.IndexSizeErr("offset cannot be beyond end of data") + if count < 0: + raise xml.dom.IndexSizeErr("count cannot be negative") + if count: + self.data = self.data[:offset] + self.data[offset+count:] + + def replaceData(self, offset, count, arg): + if offset < 0: + raise xml.dom.IndexSizeErr("offset cannot be negative") + if offset >= len(self.data): + raise xml.dom.IndexSizeErr("offset cannot be beyond end of data") + if count < 0: + raise xml.dom.IndexSizeErr("count cannot be negative") + if count: + self.data = "%s%s%s" % ( + self.data[:offset], arg, self.data[offset+count:]) + +defproperty(CharacterData, "length", doc="Length of the string data.") + + +class Text(CharacterData): + # Make sure we don't add an instance __dict__ if we don't already + # have one, at least when that's possible: + # XXX this does not work, CharacterData is an old-style class + # __slots__ = () + + nodeType = Node.TEXT_NODE + nodeName = "#text" + attributes = None + + def splitText(self, offset): + if offset < 0 or offset > len(self.data): + raise xml.dom.IndexSizeErr("illegal offset value") + newText = self.__class__() + newText.data = self.data[offset:] + newText.ownerDocument = self.ownerDocument + next = self.nextSibling + if self.parentNode and self in self.parentNode.childNodes: + if next is None: + self.parentNode.appendChild(newText) + else: + self.parentNode.insertBefore(newText, next) + self.data = self.data[:offset] + return newText + + def writexml(self, writer, indent="", addindent="", newl=""): + _write_data(writer, "%s%s%s"%(indent, self.data, newl)) + + # DOM Level 3 (WD 9 April 2002) + + def _get_wholeText(self): + L = [self.data] + n = self.previousSibling + while n is not None: + if n.nodeType in (Node.TEXT_NODE, Node.CDATA_SECTION_NODE): + L.insert(0, n.data) + n = n.previousSibling + else: + break + n = self.nextSibling + while n is not None: + if n.nodeType in (Node.TEXT_NODE, Node.CDATA_SECTION_NODE): + L.append(n.data) + n = n.nextSibling + else: + break + return ''.join(L) + + def replaceWholeText(self, content): + # XXX This needs to be seriously changed if minidom ever + # supports EntityReference nodes. + parent = self.parentNode + n = self.previousSibling + while n is not None: + if n.nodeType in (Node.TEXT_NODE, Node.CDATA_SECTION_NODE): + next = n.previousSibling + parent.removeChild(n) + n = next + else: + break + n = self.nextSibling + if not content: + parent.removeChild(self) + while n is not None: + if n.nodeType in (Node.TEXT_NODE, Node.CDATA_SECTION_NODE): + next = n.nextSibling + parent.removeChild(n) + n = next + else: + break + if content: + d = self.__dict__ + d['data'] = content + d['nodeValue'] = content + return self + else: + return None + + def _get_isWhitespaceInElementContent(self): + if self.data.strip(): + return False + elem = _get_containing_element(self) + if elem is None: + return False + info = self.ownerDocument._get_elem_info(elem) + if info is None: + return False + else: + return info.isElementContent() + +defproperty(Text, "isWhitespaceInElementContent", + doc="True iff this text node contains only whitespace" + " and is in element content.") +defproperty(Text, "wholeText", + doc="The text of all logically-adjacent text nodes.") + + +def _get_containing_element(node): + c = node.parentNode + while c is not None: + if c.nodeType == Node.ELEMENT_NODE: + return c + c = c.parentNode + return None + +def _get_containing_entref(node): + c = node.parentNode + while c is not None: + if c.nodeType == Node.ENTITY_REFERENCE_NODE: + return c + c = c.parentNode + return None + + +class Comment(Childless, CharacterData): + nodeType = Node.COMMENT_NODE + nodeName = "#comment" + + def __init__(self, data): + self.data = self.nodeValue = data + + def writexml(self, writer, indent="", addindent="", newl=""): + writer.write("%s<!--%s-->%s" % (indent, self.data, newl)) + + +class CDATASection(Text): + # Make sure we don't add an instance __dict__ if we don't already + # have one, at least when that's possible: + # XXX this does not work, Text is an old-style class + # __slots__ = () + + nodeType = Node.CDATA_SECTION_NODE + nodeName = "#cdata-section" + + def writexml(self, writer, indent="", addindent="", newl=""): + if self.data.find("]]>") >= 0: + raise ValueError("']]>' not allowed in a CDATA section") + writer.write("<![CDATA[%s]]>" % self.data) + + +class ReadOnlySequentialNamedNodeMap(object): + __slots__ = '_seq', + + def __init__(self, seq=()): + # seq should be a list or tuple + self._seq = seq + + def __len__(self): + return len(self._seq) + + def _get_length(self): + return len(self._seq) + + def getNamedItem(self, name): + for n in self._seq: + if n.nodeName == name: + return n + + def getNamedItemNS(self, namespaceURI, localName): + for n in self._seq: + if n.namespaceURI == namespaceURI and n.localName == localName: + return n + + def __getitem__(self, name_or_tuple): + if isinstance(name_or_tuple, tuple): + node = self.getNamedItemNS(*name_or_tuple) + else: + node = self.getNamedItem(name_or_tuple) + if node is None: + raise KeyError, name_or_tuple + return node + + def item(self, index): + if index < 0: + return None + try: + return self._seq[index] + except IndexError: + return None + + def removeNamedItem(self, name): + raise xml.dom.NoModificationAllowedErr( + "NamedNodeMap instance is read-only") + + def removeNamedItemNS(self, namespaceURI, localName): + raise xml.dom.NoModificationAllowedErr( + "NamedNodeMap instance is read-only") + + def setNamedItem(self, node): + raise xml.dom.NoModificationAllowedErr( + "NamedNodeMap instance is read-only") + + def setNamedItemNS(self, node): + raise xml.dom.NoModificationAllowedErr( + "NamedNodeMap instance is read-only") + + def __getstate__(self): + return [self._seq] + + def __setstate__(self, state): + self._seq = state[0] + +defproperty(ReadOnlySequentialNamedNodeMap, "length", + doc="Number of entries in the NamedNodeMap.") + + +class Identified: + """Mix-in class that supports the publicId and systemId attributes.""" + + # XXX this does not work, this is an old-style class + # __slots__ = 'publicId', 'systemId' + + def _identified_mixin_init(self, publicId, systemId): + self.publicId = publicId + self.systemId = systemId + + def _get_publicId(self): + return self.publicId + + def _get_systemId(self): + return self.systemId + +class DocumentType(Identified, Childless, Node): + nodeType = Node.DOCUMENT_TYPE_NODE + nodeValue = None + name = None + publicId = None + systemId = None + internalSubset = None + + def __init__(self, qualifiedName): + self.entities = ReadOnlySequentialNamedNodeMap() + self.notations = ReadOnlySequentialNamedNodeMap() + if qualifiedName: + prefix, localname = _nssplit(qualifiedName) + self.name = localname + self.nodeName = self.name + + def _get_internalSubset(self): + return self.internalSubset + + def cloneNode(self, deep): + if self.ownerDocument is None: + # it's ok + clone = DocumentType(None) + clone.name = self.name + clone.nodeName = self.name + operation = xml.dom.UserDataHandler.NODE_CLONED + if deep: + clone.entities._seq = [] + clone.notations._seq = [] + for n in self.notations._seq: + notation = Notation(n.nodeName, n.publicId, n.systemId) + clone.notations._seq.append(notation) + n._call_user_data_handler(operation, n, notation) + for e in self.entities._seq: + entity = Entity(e.nodeName, e.publicId, e.systemId, + e.notationName) + entity.actualEncoding = e.actualEncoding + entity.encoding = e.encoding + entity.version = e.version + clone.entities._seq.append(entity) + e._call_user_data_handler(operation, n, entity) + self._call_user_data_handler(operation, self, clone) + return clone + else: + return None + + def writexml(self, writer, indent="", addindent="", newl=""): + writer.write("<!DOCTYPE ") + writer.write(self.name) + if self.publicId: + writer.write("%s PUBLIC '%s'%s '%s'" + % (newl, self.publicId, newl, self.systemId)) + elif self.systemId: + writer.write("%s SYSTEM '%s'" % (newl, self.systemId)) + if self.internalSubset is not None: + writer.write(" [") + writer.write(self.internalSubset) + writer.write("]") + writer.write(">"+newl) + +class Entity(Identified, Node): + attributes = None + nodeType = Node.ENTITY_NODE + nodeValue = None + + actualEncoding = None + encoding = None + version = None + + def __init__(self, name, publicId, systemId, notation): + self.nodeName = name + self.notationName = notation + self.childNodes = NodeList() + self._identified_mixin_init(publicId, systemId) + + def _get_actualEncoding(self): + return self.actualEncoding + + def _get_encoding(self): + return self.encoding + + def _get_version(self): + return self.version + + def appendChild(self, newChild): + raise xml.dom.HierarchyRequestErr( + "cannot append children to an entity node") + + def insertBefore(self, newChild, refChild): + raise xml.dom.HierarchyRequestErr( + "cannot insert children below an entity node") + + def removeChild(self, oldChild): + raise xml.dom.HierarchyRequestErr( + "cannot remove children from an entity node") + + def replaceChild(self, newChild, oldChild): + raise xml.dom.HierarchyRequestErr( + "cannot replace children of an entity node") + +class Notation(Identified, Childless, Node): + nodeType = Node.NOTATION_NODE + nodeValue = None + + def __init__(self, name, publicId, systemId): + self.nodeName = name + self._identified_mixin_init(publicId, systemId) + + +class DOMImplementation(DOMImplementationLS): + _features = [("core", "1.0"), + ("core", "2.0"), + ("core", "3.0"), + ("core", None), + ("xml", "1.0"), + ("xml", "2.0"), + ("xml", "3.0"), + ("xml", None), + ("ls-load", "3.0"), + ("ls-load", None), + ] + + def hasFeature(self, feature, version): + if version == "": + version = None + return (feature.lower(), version) in self._features + + def createDocument(self, namespaceURI, qualifiedName, doctype): + if doctype and doctype.parentNode is not None: + raise xml.dom.WrongDocumentErr( + "doctype object owned by another DOM tree") + doc = self._create_document() + + add_root_element = not (namespaceURI is None + and qualifiedName is None + and doctype is None) + + if not qualifiedName and add_root_element: + # The spec is unclear what to raise here; SyntaxErr + # would be the other obvious candidate. Since Xerces raises + # InvalidCharacterErr, and since SyntaxErr is not listed + # for createDocument, that seems to be the better choice. + # XXX: need to check for illegal characters here and in + # createElement. + + # DOM Level III clears this up when talking about the return value + # of this function. If namespaceURI, qName and DocType are + # Null the document is returned without a document element + # Otherwise if doctype or namespaceURI are not None + # Then we go back to the above problem + raise xml.dom.InvalidCharacterErr("Element with no name") + + if add_root_element: + prefix, localname = _nssplit(qualifiedName) + if prefix == "xml" \ + and namespaceURI != "http://www.w3.org/XML/1998/namespace": + raise xml.dom.NamespaceErr("illegal use of 'xml' prefix") + if prefix and not namespaceURI: + raise xml.dom.NamespaceErr( + "illegal use of prefix without namespaces") + element = doc.createElementNS(namespaceURI, qualifiedName) + if doctype: + doc.appendChild(doctype) + doc.appendChild(element) + + if doctype: + doctype.parentNode = doctype.ownerDocument = doc + + doc.doctype = doctype + doc.implementation = self + return doc + + def createDocumentType(self, qualifiedName, publicId, systemId): + doctype = DocumentType(qualifiedName) + doctype.publicId = publicId + doctype.systemId = systemId + return doctype + + # DOM Level 3 (WD 9 April 2002) + + def getInterface(self, feature): + if self.hasFeature(feature, None): + return self + else: + return None + + # internal + def _create_document(self): + return Document() + +class ElementInfo(object): + """Object that represents content-model information for an element. + + This implementation is not expected to be used in practice; DOM + builders should provide implementations which do the right thing + using information available to it. + + """ + + __slots__ = 'tagName', + + def __init__(self, name): + self.tagName = name + + def getAttributeType(self, aname): + return _no_type + + def getAttributeTypeNS(self, namespaceURI, localName): + return _no_type + + def isElementContent(self): + return False + + def isEmpty(self): + """Returns true iff this element is declared to have an EMPTY + content model.""" + return False + + def isId(self, aname): + """Returns true iff the named attribte is a DTD-style ID.""" + return False + + def isIdNS(self, namespaceURI, localName): + """Returns true iff the identified attribute is a DTD-style ID.""" + return False + + def __getstate__(self): + return self.tagName + + def __setstate__(self, state): + self.tagName = state + +def _clear_id_cache(node): + if node.nodeType == Node.DOCUMENT_NODE: + node._id_cache.clear() + node._id_search_stack = None + elif _in_document(node): + node.ownerDocument._id_cache.clear() + node.ownerDocument._id_search_stack= None + +class Document(Node, DocumentLS): + _child_node_types = (Node.ELEMENT_NODE, Node.PROCESSING_INSTRUCTION_NODE, + Node.COMMENT_NODE, Node.DOCUMENT_TYPE_NODE) + + nodeType = Node.DOCUMENT_NODE + nodeName = "#document" + nodeValue = None + attributes = None + doctype = None + parentNode = None + previousSibling = nextSibling = None + + implementation = DOMImplementation() + + # Document attributes from Level 3 (WD 9 April 2002) + + actualEncoding = None + encoding = None + standalone = None + version = None + strictErrorChecking = False + errorHandler = None + documentURI = None + + _magic_id_count = 0 + + def __init__(self): + self.childNodes = NodeList() + # mapping of (namespaceURI, localName) -> ElementInfo + # and tagName -> ElementInfo + self._elem_info = {} + self._id_cache = {} + self._id_search_stack = None + + def _get_elem_info(self, element): + if element.namespaceURI: + key = element.namespaceURI, element.localName + else: + key = element.tagName + return self._elem_info.get(key) + + def _get_actualEncoding(self): + return self.actualEncoding + + def _get_doctype(self): + return self.doctype + + def _get_documentURI(self): + return self.documentURI + + def _get_encoding(self): + return self.encoding + + def _get_errorHandler(self): + return self.errorHandler + + def _get_standalone(self): + return self.standalone + + def _get_strictErrorChecking(self): + return self.strictErrorChecking + + def _get_version(self): + return self.version + + def appendChild(self, node): + if node.nodeType not in self._child_node_types: + raise xml.dom.HierarchyRequestErr( + "%s cannot be child of %s" % (repr(node), repr(self))) + if node.parentNode is not None: + # This needs to be done before the next test since this + # may *be* the document element, in which case it should + # end up re-ordered to the end. + node.parentNode.removeChild(node) + + if node.nodeType == Node.ELEMENT_NODE \ + and self._get_documentElement(): + raise xml.dom.HierarchyRequestErr( + "two document elements disallowed") + return Node.appendChild(self, node) + + def removeChild(self, oldChild): + try: + self.childNodes.remove(oldChild) + except ValueError: + raise xml.dom.NotFoundErr() + oldChild.nextSibling = oldChild.previousSibling = None + oldChild.parentNode = None + if self.documentElement is oldChild: + self.documentElement = None + + return oldChild + + def _get_documentElement(self): + for node in self.childNodes: + if node.nodeType == Node.ELEMENT_NODE: + return node + + def unlink(self): + if self.doctype is not None: + self.doctype.unlink() + self.doctype = None + Node.unlink(self) + + def cloneNode(self, deep): + if not deep: + return None + clone = self.implementation.createDocument(None, None, None) + clone.encoding = self.encoding + clone.standalone = self.standalone + clone.version = self.version + for n in self.childNodes: + childclone = _clone_node(n, deep, clone) + assert childclone.ownerDocument.isSameNode(clone) + clone.childNodes.append(childclone) + if childclone.nodeType == Node.DOCUMENT_NODE: + assert clone.documentElement is None + elif childclone.nodeType == Node.DOCUMENT_TYPE_NODE: + assert clone.doctype is None + clone.doctype = childclone + childclone.parentNode = clone + self._call_user_data_handler(xml.dom.UserDataHandler.NODE_CLONED, + self, clone) + return clone + + def createDocumentFragment(self): + d = DocumentFragment() + d.ownerDocument = self + return d + + def createElement(self, tagName): + e = Element(tagName) + e.ownerDocument = self + return e + + def createTextNode(self, data): + if not isinstance(data, StringTypes): + raise TypeError, "node contents must be a string" + t = Text() + t.data = data + t.ownerDocument = self + return t + + def createCDATASection(self, data): + if not isinstance(data, StringTypes): + raise TypeError, "node contents must be a string" + c = CDATASection() + c.data = data + c.ownerDocument = self + return c + + def createComment(self, data): + c = Comment(data) + c.ownerDocument = self + return c + + def createProcessingInstruction(self, target, data): + p = ProcessingInstruction(target, data) + p.ownerDocument = self + return p + + def createAttribute(self, qName): + a = Attr(qName) + a.ownerDocument = self + a.value = "" + return a + + def createElementNS(self, namespaceURI, qualifiedName): + prefix, localName = _nssplit(qualifiedName) + e = Element(qualifiedName, namespaceURI, prefix) + e.ownerDocument = self + return e + + def createAttributeNS(self, namespaceURI, qualifiedName): + prefix, localName = _nssplit(qualifiedName) + a = Attr(qualifiedName, namespaceURI, localName, prefix) + a.ownerDocument = self + a.value = "" + return a + + # A couple of implementation-specific helpers to create node types + # not supported by the W3C DOM specs: + + def _create_entity(self, name, publicId, systemId, notationName): + e = Entity(name, publicId, systemId, notationName) + e.ownerDocument = self + return e + + def _create_notation(self, name, publicId, systemId): + n = Notation(name, publicId, systemId) + n.ownerDocument = self + return n + + def getElementById(self, id): + if self._id_cache.has_key(id): + return self._id_cache[id] + if not (self._elem_info or self._magic_id_count): + return None + + stack = self._id_search_stack + if stack is None: + # we never searched before, or the cache has been cleared + stack = [self.documentElement] + self._id_search_stack = stack + elif not stack: + # Previous search was completed and cache is still valid; + # no matching node. + return None + + result = None + while stack: + node = stack.pop() + # add child elements to stack for continued searching + stack.extend([child for child in node.childNodes + if child.nodeType in _nodeTypes_with_children]) + # check this node + info = self._get_elem_info(node) + if info: + # We have to process all ID attributes before + # returning in order to get all the attributes set to + # be IDs using Element.setIdAttribute*(). + for attr in node.attributes.values(): + if attr.namespaceURI: + if info.isIdNS(attr.namespaceURI, attr.localName): + self._id_cache[attr.value] = node + if attr.value == id: + result = node + elif not node._magic_id_nodes: + break + elif info.isId(attr.name): + self._id_cache[attr.value] = node + if attr.value == id: + result = node + elif not node._magic_id_nodes: + break + elif attr._is_id: + self._id_cache[attr.value] = node + if attr.value == id: + result = node + elif node._magic_id_nodes == 1: + break + elif node._magic_id_nodes: + for attr in node.attributes.values(): + if attr._is_id: + self._id_cache[attr.value] = node + if attr.value == id: + result = node + if result is not None: + break + return result + + def getElementsByTagName(self, name): + return _get_elements_by_tagName_helper(self, name, NodeList()) + + def getElementsByTagNameNS(self, namespaceURI, localName): + return _get_elements_by_tagName_ns_helper( + self, namespaceURI, localName, NodeList()) + + def isSupported(self, feature, version): + return self.implementation.hasFeature(feature, version) + + def importNode(self, node, deep): + if node.nodeType == Node.DOCUMENT_NODE: + raise xml.dom.NotSupportedErr("cannot import document nodes") + elif node.nodeType == Node.DOCUMENT_TYPE_NODE: + raise xml.dom.NotSupportedErr("cannot import document type nodes") + return _clone_node(node, deep, self) + + def writexml(self, writer, indent="", addindent="", newl="", + encoding = None): + 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) + + # DOM Level 3 (WD 9 April 2002) + + def renameNode(self, n, namespaceURI, name): + if n.ownerDocument is not self: + raise xml.dom.WrongDocumentErr( + "cannot rename nodes from other documents;\n" + "expected %s,\nfound %s" % (self, n.ownerDocument)) + if n.nodeType not in (Node.ELEMENT_NODE, Node.ATTRIBUTE_NODE): + raise xml.dom.NotSupportedErr( + "renameNode() only applies to element and attribute nodes") + if namespaceURI != EMPTY_NAMESPACE: + if ':' in name: + prefix, localName = name.split(':', 1) + if ( prefix == "xmlns" + and namespaceURI != xml.dom.XMLNS_NAMESPACE): + raise xml.dom.NamespaceErr( + "illegal use of 'xmlns' prefix") + else: + if ( name == "xmlns" + and namespaceURI != xml.dom.XMLNS_NAMESPACE + and n.nodeType == Node.ATTRIBUTE_NODE): + raise xml.dom.NamespaceErr( + "illegal use of the 'xmlns' attribute") + prefix = None + localName = name + else: + prefix = None + localName = None + if n.nodeType == Node.ATTRIBUTE_NODE: + element = n.ownerElement + if element is not None: + is_id = n._is_id + element.removeAttributeNode(n) + else: + element = None + # avoid __setattr__ + d = n.__dict__ + d['prefix'] = prefix + d['localName'] = localName + d['namespaceURI'] = namespaceURI + d['nodeName'] = name + if n.nodeType == Node.ELEMENT_NODE: + d['tagName'] = name + else: + # attribute node + d['name'] = name + if element is not None: + element.setAttributeNode(n) + if is_id: + element.setIdAttributeNode(n) + # It's not clear from a semantic perspective whether we should + # call the user data handlers for the NODE_RENAMED event since + # we're re-using the existing node. The draft spec has been + # interpreted as meaning "no, don't call the handler unless a + # new node is created." + return n + +defproperty(Document, "documentElement", + doc="Top-level element of this document.") + + +def _clone_node(node, deep, newOwnerDocument): + """ + Clone a node and give it the new owner document. + Called by Node.cloneNode and Document.importNode + """ + if node.ownerDocument.isSameNode(newOwnerDocument): + operation = xml.dom.UserDataHandler.NODE_CLONED + else: + operation = xml.dom.UserDataHandler.NODE_IMPORTED + if node.nodeType == Node.ELEMENT_NODE: + clone = newOwnerDocument.createElementNS(node.namespaceURI, + node.nodeName) + for attr in node.attributes.values(): + clone.setAttributeNS(attr.namespaceURI, attr.nodeName, attr.value) + a = clone.getAttributeNodeNS(attr.namespaceURI, attr.localName) + a.specified = attr.specified + + if deep: + for child in node.childNodes: + c = _clone_node(child, deep, newOwnerDocument) + clone.appendChild(c) + + elif node.nodeType == Node.DOCUMENT_FRAGMENT_NODE: + clone = newOwnerDocument.createDocumentFragment() + if deep: + for child in node.childNodes: + c = _clone_node(child, deep, newOwnerDocument) + clone.appendChild(c) + + elif node.nodeType == Node.TEXT_NODE: + clone = newOwnerDocument.createTextNode(node.data) + elif node.nodeType == Node.CDATA_SECTION_NODE: + clone = newOwnerDocument.createCDATASection(node.data) + elif node.nodeType == Node.PROCESSING_INSTRUCTION_NODE: + clone = newOwnerDocument.createProcessingInstruction(node.target, + node.data) + elif node.nodeType == Node.COMMENT_NODE: + clone = newOwnerDocument.createComment(node.data) + elif node.nodeType == Node.ATTRIBUTE_NODE: + clone = newOwnerDocument.createAttributeNS(node.namespaceURI, + node.nodeName) + clone.specified = True + clone.value = node.value + elif node.nodeType == Node.DOCUMENT_TYPE_NODE: + assert node.ownerDocument is not newOwnerDocument + operation = xml.dom.UserDataHandler.NODE_IMPORTED + clone = newOwnerDocument.implementation.createDocumentType( + node.name, node.publicId, node.systemId) + clone.ownerDocument = newOwnerDocument + if deep: + clone.entities._seq = [] + clone.notations._seq = [] + for n in node.notations._seq: + notation = Notation(n.nodeName, n.publicId, n.systemId) + notation.ownerDocument = newOwnerDocument + clone.notations._seq.append(notation) + if hasattr(n, '_call_user_data_handler'): + n._call_user_data_handler(operation, n, notation) + for e in node.entities._seq: + entity = Entity(e.nodeName, e.publicId, e.systemId, + e.notationName) + entity.actualEncoding = e.actualEncoding + entity.encoding = e.encoding + entity.version = e.version + entity.ownerDocument = newOwnerDocument + clone.entities._seq.append(entity) + if hasattr(e, '_call_user_data_handler'): + e._call_user_data_handler(operation, n, entity) + else: + # Note the cloning of Document and DocumentType nodes is + # implemenetation specific. minidom handles those cases + # directly in the cloneNode() methods. + raise xml.dom.NotSupportedErr("Cannot clone node %s" % repr(node)) + + # Check for _call_user_data_handler() since this could conceivably + # used with other DOM implementations (one of the FourThought + # DOMs, perhaps?). + if hasattr(node, '_call_user_data_handler'): + node._call_user_data_handler(operation, node, clone) + return clone + + +def _nssplit(qualifiedName): + fields = qualifiedName.split(':', 1) + if len(fields) == 2: + return fields + else: + return (None, fields[0]) + + +def _get_StringIO(): + # we can't use cStringIO since it doesn't support Unicode strings + from StringIO import StringIO + return StringIO() + +def _do_pulldom_parse(func, args, kwargs): + events = func(*args, **kwargs) + toktype, rootNode = events.getEvent() + events.expandNode(rootNode) + events.clear() + return rootNode + +def parse(file, parser=None, bufsize=None): + """Parse a file into a DOM by filename or file object.""" + if parser is None and not bufsize: + from xml.dom import expatbuilder + return expatbuilder.parse(file) + else: + from xml.dom import pulldom + return _do_pulldom_parse(pulldom.parse, (file,), + {'parser': parser, 'bufsize': bufsize}) + +def parseString(string, parser=None): + """Parse a file into a DOM from a string.""" + if parser is None: + from xml.dom import expatbuilder + return expatbuilder.parseString(string) + else: + from xml.dom import pulldom + return _do_pulldom_parse(pulldom.parseString, (string,), + {'parser': parser}) + +def getDOMImplementation(features=None): + if features: + if isinstance(features, StringTypes): + features = domreg._parse_feature_string(features) + for f, v in features: + if not Document.implementation.hasFeature(f, v): + return None + return Document.implementation diff --git a/Lib/xml/dom/pulldom.py b/Lib/xml/dom/pulldom.py new file mode 100644 index 0000000..18f49b5 --- /dev/null +++ b/Lib/xml/dom/pulldom.py @@ -0,0 +1,351 @@ +import xml.sax +import xml.sax.handler +import types + +try: + _StringTypes = [types.StringType, types.UnicodeType] +except AttributeError: + _StringTypes = [types.StringType] + +START_ELEMENT = "START_ELEMENT" +END_ELEMENT = "END_ELEMENT" +COMMENT = "COMMENT" +START_DOCUMENT = "START_DOCUMENT" +END_DOCUMENT = "END_DOCUMENT" +PROCESSING_INSTRUCTION = "PROCESSING_INSTRUCTION" +IGNORABLE_WHITESPACE = "IGNORABLE_WHITESPACE" +CHARACTERS = "CHARACTERS" + +class PullDOM(xml.sax.ContentHandler): + _locator = None + document = None + + def __init__(self, documentFactory=None): + from xml.dom import XML_NAMESPACE + self.documentFactory = documentFactory + self.firstEvent = [None, None] + self.lastEvent = self.firstEvent + self.elementStack = [] + self.push = self.elementStack.append + try: + self.pop = self.elementStack.pop + except AttributeError: + # use class' pop instead + pass + self._ns_contexts = [{XML_NAMESPACE:'xml'}] # contains uri -> prefix dicts + self._current_context = self._ns_contexts[-1] + self.pending_events = [] + + def pop(self): + result = self.elementStack[-1] + del self.elementStack[-1] + return result + + def setDocumentLocator(self, locator): + self._locator = locator + + def startPrefixMapping(self, prefix, uri): + if not hasattr(self, '_xmlns_attrs'): + self._xmlns_attrs = [] + self._xmlns_attrs.append((prefix or 'xmlns', uri)) + self._ns_contexts.append(self._current_context.copy()) + self._current_context[uri] = prefix or None + + def endPrefixMapping(self, prefix): + self._current_context = self._ns_contexts.pop() + + def startElementNS(self, name, tagName , attrs): + # Retrieve xml namespace declaration attributes. + xmlns_uri = 'http://www.w3.org/2000/xmlns/' + xmlns_attrs = getattr(self, '_xmlns_attrs', None) + if xmlns_attrs is not None: + for aname, value in xmlns_attrs: + attrs._attrs[(xmlns_uri, aname)] = value + self._xmlns_attrs = [] + uri, localname = name + if uri: + # When using namespaces, the reader may or may not + # provide us with the original name. If not, create + # *a* valid tagName from the current context. + if tagName is None: + prefix = self._current_context[uri] + if prefix: + tagName = prefix + ":" + localname + else: + tagName = localname + if self.document: + node = self.document.createElementNS(uri, tagName) + else: + node = self.buildDocument(uri, tagName) + else: + # When the tagname is not prefixed, it just appears as + # localname + if self.document: + node = self.document.createElement(localname) + else: + node = self.buildDocument(None, localname) + + for aname,value in attrs.items(): + a_uri, a_localname = aname + if a_uri == xmlns_uri: + if a_localname == 'xmlns': + qname = a_localname + else: + qname = 'xmlns:' + a_localname + attr = self.document.createAttributeNS(a_uri, qname) + node.setAttributeNodeNS(attr) + elif a_uri: + prefix = self._current_context[a_uri] + if prefix: + qname = prefix + ":" + a_localname + else: + qname = a_localname + attr = self.document.createAttributeNS(a_uri, qname) + node.setAttributeNodeNS(attr) + else: + attr = self.document.createAttribute(a_localname) + node.setAttributeNode(attr) + attr.value = value + + self.lastEvent[1] = [(START_ELEMENT, node), None] + self.lastEvent = self.lastEvent[1] + self.push(node) + + def endElementNS(self, name, tagName): + self.lastEvent[1] = [(END_ELEMENT, self.pop()), None] + self.lastEvent = self.lastEvent[1] + + def startElement(self, name, attrs): + if self.document: + node = self.document.createElement(name) + else: + node = self.buildDocument(None, name) + + for aname,value in attrs.items(): + attr = self.document.createAttribute(aname) + attr.value = value + node.setAttributeNode(attr) + + self.lastEvent[1] = [(START_ELEMENT, node), None] + self.lastEvent = self.lastEvent[1] + self.push(node) + + def endElement(self, name): + self.lastEvent[1] = [(END_ELEMENT, self.pop()), None] + self.lastEvent = self.lastEvent[1] + + def comment(self, s): + if self.document: + node = self.document.createComment(s) + self.lastEvent[1] = [(COMMENT, node), None] + self.lastEvent = self.lastEvent[1] + else: + event = [(COMMENT, s), None] + self.pending_events.append(event) + + def processingInstruction(self, target, data): + if self.document: + node = self.document.createProcessingInstruction(target, data) + self.lastEvent[1] = [(PROCESSING_INSTRUCTION, node), None] + self.lastEvent = self.lastEvent[1] + else: + event = [(PROCESSING_INSTRUCTION, target, data), None] + self.pending_events.append(event) + + def ignorableWhitespace(self, chars): + node = self.document.createTextNode(chars) + self.lastEvent[1] = [(IGNORABLE_WHITESPACE, node), None] + self.lastEvent = self.lastEvent[1] + + def characters(self, chars): + node = self.document.createTextNode(chars) + self.lastEvent[1] = [(CHARACTERS, node), None] + self.lastEvent = self.lastEvent[1] + + def startDocument(self): + if self.documentFactory is None: + import xml.dom.minidom + self.documentFactory = xml.dom.minidom.Document.implementation + + def buildDocument(self, uri, tagname): + # Can't do that in startDocument, since we need the tagname + # XXX: obtain DocumentType + node = self.documentFactory.createDocument(uri, tagname, None) + self.document = node + self.lastEvent[1] = [(START_DOCUMENT, node), None] + self.lastEvent = self.lastEvent[1] + self.push(node) + # Put everything we have seen so far into the document + for e in self.pending_events: + if e[0][0] == PROCESSING_INSTRUCTION: + _,target,data = e[0] + n = self.document.createProcessingInstruction(target, data) + e[0] = (PROCESSING_INSTRUCTION, n) + elif e[0][0] == COMMENT: + n = self.document.createComment(e[0][1]) + e[0] = (COMMENT, n) + else: + raise AssertionError("Unknown pending event ",e[0][0]) + self.lastEvent[1] = e + self.lastEvent = e + self.pending_events = None + return node.firstChild + + def endDocument(self): + self.lastEvent[1] = [(END_DOCUMENT, self.document), None] + self.pop() + + def clear(self): + "clear(): Explicitly release parsing structures" + self.document = None + +class ErrorHandler: + def warning(self, exception): + print exception + def error(self, exception): + raise exception + def fatalError(self, exception): + raise exception + +class DOMEventStream: + def __init__(self, stream, parser, bufsize): + self.stream = stream + self.parser = parser + self.bufsize = bufsize + if not hasattr(self.parser, 'feed'): + self.getEvent = self._slurp + self.reset() + + def reset(self): + self.pulldom = PullDOM() + # This content handler relies on namespace support + self.parser.setFeature(xml.sax.handler.feature_namespaces, 1) + self.parser.setContentHandler(self.pulldom) + + def __getitem__(self, pos): + rc = self.getEvent() + if rc: + return rc + raise IndexError + + def next(self): + rc = self.getEvent() + if rc: + return rc + raise StopIteration + + def __iter__(self): + return self + + def expandNode(self, node): + event = self.getEvent() + parents = [node] + while event: + token, cur_node = event + if cur_node is node: + return + if token != END_ELEMENT: + parents[-1].appendChild(cur_node) + if token == START_ELEMENT: + parents.append(cur_node) + elif token == END_ELEMENT: + del parents[-1] + event = self.getEvent() + + def getEvent(self): + # use IncrementalParser interface, so we get the desired + # pull effect + if not self.pulldom.firstEvent[1]: + self.pulldom.lastEvent = self.pulldom.firstEvent + while not self.pulldom.firstEvent[1]: + buf = self.stream.read(self.bufsize) + if not buf: + self.parser.close() + return None + self.parser.feed(buf) + rc = self.pulldom.firstEvent[1][0] + self.pulldom.firstEvent[1] = self.pulldom.firstEvent[1][1] + return rc + + def _slurp(self): + """ Fallback replacement for getEvent() using the + standard SAX2 interface, which means we slurp the + SAX events into memory (no performance gain, but + we are compatible to all SAX parsers). + """ + self.parser.parse(self.stream) + self.getEvent = self._emit + return self._emit() + + def _emit(self): + """ Fallback replacement for getEvent() that emits + the events that _slurp() read previously. + """ + rc = self.pulldom.firstEvent[1][0] + self.pulldom.firstEvent[1] = self.pulldom.firstEvent[1][1] + return rc + + def clear(self): + """clear(): Explicitly release parsing objects""" + self.pulldom.clear() + del self.pulldom + self.parser = None + self.stream = None + +class SAX2DOM(PullDOM): + + def startElementNS(self, name, tagName , attrs): + PullDOM.startElementNS(self, name, tagName, attrs) + curNode = self.elementStack[-1] + parentNode = self.elementStack[-2] + parentNode.appendChild(curNode) + + def startElement(self, name, attrs): + PullDOM.startElement(self, name, attrs) + curNode = self.elementStack[-1] + parentNode = self.elementStack[-2] + parentNode.appendChild(curNode) + + def processingInstruction(self, target, data): + PullDOM.processingInstruction(self, target, data) + node = self.lastEvent[0][1] + parentNode = self.elementStack[-1] + parentNode.appendChild(node) + + def ignorableWhitespace(self, chars): + PullDOM.ignorableWhitespace(self, chars) + node = self.lastEvent[0][1] + parentNode = self.elementStack[-1] + parentNode.appendChild(node) + + def characters(self, chars): + PullDOM.characters(self, chars) + node = self.lastEvent[0][1] + parentNode = self.elementStack[-1] + parentNode.appendChild(node) + + +default_bufsize = (2 ** 14) - 20 + +def parse(stream_or_string, parser=None, bufsize=None): + if bufsize is None: + bufsize = default_bufsize + if type(stream_or_string) in _StringTypes: + stream = open(stream_or_string) + else: + stream = stream_or_string + if not parser: + parser = xml.sax.make_parser() + return DOMEventStream(stream, parser, bufsize) + +def parseString(string, parser=None): + try: + from cStringIO import StringIO + except ImportError: + from StringIO import StringIO + + bufsize = len(string) + buf = StringIO(string) + if not parser: + parser = xml.sax.make_parser() + return DOMEventStream(buf, parser, bufsize) diff --git a/Lib/xml/dom/xmlbuilder.py b/Lib/xml/dom/xmlbuilder.py new file mode 100644 index 0000000..ac1d448 --- /dev/null +++ b/Lib/xml/dom/xmlbuilder.py @@ -0,0 +1,386 @@ +"""Implementation of the DOM Level 3 'LS-Load' feature.""" + +import copy +import xml.dom + +from xml.dom.NodeFilter import NodeFilter + + +__all__ = ["DOMBuilder", "DOMEntityResolver", "DOMInputSource"] + + +class Options: + """Features object that has variables set for each DOMBuilder feature. + + The DOMBuilder class uses an instance of this class to pass settings to + the ExpatBuilder class. + """ + + # Note that the DOMBuilder class in LoadSave constrains which of these + # values can be set using the DOM Level 3 LoadSave feature. + + namespaces = 1 + namespace_declarations = True + validation = False + external_parameter_entities = True + external_general_entities = True + external_dtd_subset = True + validate_if_schema = False + validate = False + datatype_normalization = False + create_entity_ref_nodes = True + entities = True + whitespace_in_element_content = True + cdata_sections = True + comments = True + charset_overrides_xml_encoding = True + infoset = False + supported_mediatypes_only = False + + errorHandler = None + filter = None + + +class DOMBuilder: + entityResolver = None + errorHandler = None + filter = None + + ACTION_REPLACE = 1 + ACTION_APPEND_AS_CHILDREN = 2 + ACTION_INSERT_AFTER = 3 + ACTION_INSERT_BEFORE = 4 + + _legal_actions = (ACTION_REPLACE, ACTION_APPEND_AS_CHILDREN, + ACTION_INSERT_AFTER, ACTION_INSERT_BEFORE) + + def __init__(self): + self._options = Options() + + def _get_entityResolver(self): + return self.entityResolver + def _set_entityResolver(self, entityResolver): + self.entityResolver = entityResolver + + def _get_errorHandler(self): + return self.errorHandler + def _set_errorHandler(self, errorHandler): + self.errorHandler = errorHandler + + def _get_filter(self): + return self.filter + def _set_filter(self, filter): + self.filter = filter + + def setFeature(self, name, state): + if self.supportsFeature(name): + state = state and 1 or 0 + try: + settings = self._settings[(_name_xform(name), state)] + except KeyError: + raise xml.dom.NotSupportedErr( + "unsupported feature: %r" % (name,)) + else: + for name, value in settings: + setattr(self._options, name, value) + else: + raise xml.dom.NotFoundErr("unknown feature: " + repr(name)) + + def supportsFeature(self, name): + return hasattr(self._options, _name_xform(name)) + + def canSetFeature(self, name, state): + key = (_name_xform(name), state and 1 or 0) + return self._settings.has_key(key) + + # This dictionary maps from (feature,value) to a list of + # (option,value) pairs that should be set on the Options object. + # If a (feature,value) setting is not in this dictionary, it is + # not supported by the DOMBuilder. + # + _settings = { + ("namespace_declarations", 0): [ + ("namespace_declarations", 0)], + ("namespace_declarations", 1): [ + ("namespace_declarations", 1)], + ("validation", 0): [ + ("validation", 0)], + ("external_general_entities", 0): [ + ("external_general_entities", 0)], + ("external_general_entities", 1): [ + ("external_general_entities", 1)], + ("external_parameter_entities", 0): [ + ("external_parameter_entities", 0)], + ("external_parameter_entities", 1): [ + ("external_parameter_entities", 1)], + ("validate_if_schema", 0): [ + ("validate_if_schema", 0)], + ("create_entity_ref_nodes", 0): [ + ("create_entity_ref_nodes", 0)], + ("create_entity_ref_nodes", 1): [ + ("create_entity_ref_nodes", 1)], + ("entities", 0): [ + ("create_entity_ref_nodes", 0), + ("entities", 0)], + ("entities", 1): [ + ("entities", 1)], + ("whitespace_in_element_content", 0): [ + ("whitespace_in_element_content", 0)], + ("whitespace_in_element_content", 1): [ + ("whitespace_in_element_content", 1)], + ("cdata_sections", 0): [ + ("cdata_sections", 0)], + ("cdata_sections", 1): [ + ("cdata_sections", 1)], + ("comments", 0): [ + ("comments", 0)], + ("comments", 1): [ + ("comments", 1)], + ("charset_overrides_xml_encoding", 0): [ + ("charset_overrides_xml_encoding", 0)], + ("charset_overrides_xml_encoding", 1): [ + ("charset_overrides_xml_encoding", 1)], + ("infoset", 0): [], + ("infoset", 1): [ + ("namespace_declarations", 0), + ("validate_if_schema", 0), + ("create_entity_ref_nodes", 0), + ("entities", 0), + ("cdata_sections", 0), + ("datatype_normalization", 1), + ("whitespace_in_element_content", 1), + ("comments", 1), + ("charset_overrides_xml_encoding", 1)], + ("supported_mediatypes_only", 0): [ + ("supported_mediatypes_only", 0)], + ("namespaces", 0): [ + ("namespaces", 0)], + ("namespaces", 1): [ + ("namespaces", 1)], + } + + def getFeature(self, name): + xname = _name_xform(name) + try: + return getattr(self._options, xname) + except AttributeError: + if name == "infoset": + options = self._options + return (options.datatype_normalization + and options.whitespace_in_element_content + and options.comments + and options.charset_overrides_xml_encoding + and not (options.namespace_declarations + or options.validate_if_schema + or options.create_entity_ref_nodes + or options.entities + or options.cdata_sections)) + raise xml.dom.NotFoundErr("feature %s not known" % repr(name)) + + def parseURI(self, uri): + if self.entityResolver: + input = self.entityResolver.resolveEntity(None, uri) + else: + input = DOMEntityResolver().resolveEntity(None, uri) + return self.parse(input) + + def parse(self, input): + options = copy.copy(self._options) + options.filter = self.filter + options.errorHandler = self.errorHandler + fp = input.byteStream + if fp is None and options.systemId: + import urllib2 + fp = urllib2.urlopen(input.systemId) + return self._parse_bytestream(fp, options) + + def parseWithContext(self, input, cnode, action): + if action not in self._legal_actions: + raise ValueError("not a legal action") + raise NotImplementedError("Haven't written this yet...") + + def _parse_bytestream(self, stream, options): + import xml.dom.expatbuilder + builder = xml.dom.expatbuilder.makeBuilder(options) + return builder.parseFile(stream) + + +def _name_xform(name): + return name.lower().replace('-', '_') + + +class DOMEntityResolver(object): + __slots__ = '_opener', + + def resolveEntity(self, publicId, systemId): + assert systemId is not None + source = DOMInputSource() + source.publicId = publicId + source.systemId = systemId + source.byteStream = self._get_opener().open(systemId) + + # determine the encoding if the transport provided it + source.encoding = self._guess_media_encoding(source) + + # determine the base URI is we can + import posixpath, urlparse + parts = urlparse.urlparse(systemId) + scheme, netloc, path, params, query, fragment = parts + # XXX should we check the scheme here as well? + if path and not path.endswith("/"): + path = posixpath.dirname(path) + "/" + parts = scheme, netloc, path, params, query, fragment + source.baseURI = urlparse.urlunparse(parts) + + return source + + def _get_opener(self): + try: + return self._opener + except AttributeError: + self._opener = self._create_opener() + return self._opener + + def _create_opener(self): + import urllib2 + return urllib2.build_opener() + + def _guess_media_encoding(self, source): + info = source.byteStream.info() + if info.has_key("Content-Type"): + for param in info.getplist(): + if param.startswith("charset="): + return param.split("=", 1)[1].lower() + + +class DOMInputSource(object): + __slots__ = ('byteStream', 'characterStream', 'stringData', + 'encoding', 'publicId', 'systemId', 'baseURI') + + def __init__(self): + self.byteStream = None + self.characterStream = None + self.stringData = None + self.encoding = None + self.publicId = None + self.systemId = None + self.baseURI = None + + def _get_byteStream(self): + return self.byteStream + def _set_byteStream(self, byteStream): + self.byteStream = byteStream + + def _get_characterStream(self): + return self.characterStream + def _set_characterStream(self, characterStream): + self.characterStream = characterStream + + def _get_stringData(self): + return self.stringData + def _set_stringData(self, data): + self.stringData = data + + def _get_encoding(self): + return self.encoding + def _set_encoding(self, encoding): + self.encoding = encoding + + def _get_publicId(self): + return self.publicId + def _set_publicId(self, publicId): + self.publicId = publicId + + def _get_systemId(self): + return self.systemId + def _set_systemId(self, systemId): + self.systemId = systemId + + def _get_baseURI(self): + return self.baseURI + def _set_baseURI(self, uri): + self.baseURI = uri + + +class DOMBuilderFilter: + """Element filter which can be used to tailor construction of + a DOM instance. + """ + + # There's really no need for this class; concrete implementations + # should just implement the endElement() and startElement() + # methods as appropriate. Using this makes it easy to only + # implement one of them. + + FILTER_ACCEPT = 1 + FILTER_REJECT = 2 + FILTER_SKIP = 3 + FILTER_INTERRUPT = 4 + + whatToShow = NodeFilter.SHOW_ALL + + def _get_whatToShow(self): + return self.whatToShow + + def acceptNode(self, element): + return self.FILTER_ACCEPT + + def startContainer(self, element): + return self.FILTER_ACCEPT + +del NodeFilter + + +class DocumentLS: + """Mixin to create documents that conform to the load/save spec.""" + + async = False + + def _get_async(self): + return False + def _set_async(self, async): + if async: + raise xml.dom.NotSupportedErr( + "asynchronous document loading is not supported") + + def abort(self): + # What does it mean to "clear" a document? Does the + # documentElement disappear? + raise NotImplementedError( + "haven't figured out what this means yet") + + def load(self, uri): + raise NotImplementedError("haven't written this yet") + + def loadXML(self, source): + raise NotImplementedError("haven't written this yet") + + def saveXML(self, snode): + if snode is None: + snode = self + elif snode.ownerDocument is not self: + raise xml.dom.WrongDocumentErr() + return snode.toxml() + + +class DOMImplementationLS: + MODE_SYNCHRONOUS = 1 + MODE_ASYNCHRONOUS = 2 + + def createDOMBuilder(self, mode, schemaType): + if schemaType is not None: + raise xml.dom.NotSupportedErr( + "schemaType not yet supported") + if mode == self.MODE_SYNCHRONOUS: + return DOMBuilder() + if mode == self.MODE_ASYNCHRONOUS: + raise xml.dom.NotSupportedErr( + "asynchronous builders are not supported") + raise ValueError("unknown value for mode") + + def createDOMWriter(self): + raise NotImplementedError( + "the writer interface hasn't been written yet!") + + def createDOMInputSource(self): + return DOMInputSource() |