xml.dom —- The Document Object Model API

Source code:Lib/xml/dom/init.py

The Document Object Model, or "DOM," is a cross-language API from the World WideWeb Consortium (W3C) for accessing and modifying XML documents. A DOMimplementation presents an XML document as a tree structure, or allows clientcode to build such a structure from scratch. It then gives access to thestructure through a set of objects which provided well-known interfaces.

The DOM is extremely useful for random-access applications. SAX only allows youa view of one bit of the document at a time. If you are looking at one SAXelement, you have no access to another. If you are looking at a text node, youhave no access to a containing element. When you write a SAX application, youneed to keep track of your program's position in the document somewhere in yourown code. SAX does not do it for you. Also, if you need to look ahead in theXML document, you are just out of luck.

Some applications are simply impossible in an event driven model with no accessto a tree. Of course you could build some sort of tree yourself in SAX events,but the DOM allows you to avoid writing that code. The DOM is a standard treerepresentation for XML data.

The Document Object Model is being defined by the W3C in stages, or "levels" intheir terminology. The Python mapping of the API is substantially based on theDOM Level 2 recommendation.

DOM applications typically start by parsing some XML into a DOM. How this isaccomplished is not covered at all by DOM Level 1, and Level 2 provides onlylimited improvements: There is a DOMImplementation object class whichprovides access to Document creation methods, but no way to access anXML reader/parser/Document builder in an implementation-independent way. Thereis also no well-defined way to access these methods without an existingDocument object. In Python, each DOM implementation will provide afunction getDOMImplementation(). DOM Level 3 adds a Load/Storespecification, which defines an interface to the reader, but this is not yetavailable in the Python standard library.

Once you have a DOM document object, you can access the parts of your XMLdocument through its properties and methods. These properties are defined inthe DOM specification; this portion of the reference manual describes theinterpretation of the specification in Python.

The specification provided by the W3C defines the DOM API for Java, ECMAScript,and OMG IDL. The Python mapping defined here is based in large part on the IDLversion of the specification, but strict compliance is not required (thoughimplementations are free to support the strict mapping from IDL). See section一致性 for a detailed discussion of mapping requirements.

参见

• Document Object Model (DOM) Level 2 Specification

• The W3C recommendation upon which the Python DOM API is based.

• Document Object Model (DOM) Level 1 Specification

• The W3C recommendation for the DOM supported by xml.dom.minidom.

• Python Language Mapping Specification

• This specifies the mapping from OMG IDL to Python.

模块内容

The xml.dom contains the following functions:

• xml.dom.registerDOMImplementation(name, factory)

• Register the factory function with the name name. The factory functionshould return an object which implements the DOMImplementationinterface. The factory function can return the same object every time, or a newone for each call, as appropriate for the specific implementation (e.g. if thatimplementation supports some customization).

• xml.dom.getDOMImplementation(name=None, features=())

• Return a suitable DOM implementation. The name is either well-known, themodule name of a DOM implementation, or None. If it is not None, importsthe corresponding module and returns a DOMImplementation object if theimport succeeds. If no name is given, and if the environment variablePYTHON_DOM is set, this variable is used to find the implementation.

If name is not given, this examines the available implementations to find onewith the required feature set. If no implementation can be found, raise anImportError. The features list must be a sequence of (feature,version) pairs which are passed to the hasFeature() method on availableDOMImplementation objects.

Some convenience constants are also provided:

• xml.dom.EMPTY_NAMESPACE

• The value used to indicate that no namespace is associated with a node in theDOM. This is typically found as the namespaceURI of a node, or used asthe namespaceURI parameter to a namespaces-specific method.

• xml.dom.XML_NAMESPACE

• The namespace URI associated with the reserved prefix xml, as defined byNamespaces in XML (section 4).

• xml.dom.XMLNS_NAMESPACE

• The namespace URI for namespace declarations, as defined by Document ObjectModel (DOM) Level 2 Core Specification (section 1.1.8).

• xml.dom.XHTML_NAMESPACE

• The URI of the XHTML namespace as defined by XHTML 1.0: The ExtensibleHyperText Markup Language (section 3.1.1).

In addition, xml.dom contains a base Node class and the DOMexception classes. The Node class provided by this module does notimplement any of the methods or attributes defined by the DOM specification;concrete DOM implementations must provide those. The Node classprovided as part of this module does provide the constants used for thenodeType attribute on concrete Node objects; they are locatedwithin the class rather than at the module level to conform with the DOMspecifications.

Objects in the DOM

DOM的权威文档是来自W3C的DOM规范。

请注意，DOM属性也可以作为节点而不是简单的字符串进行操作。然而，必须这样做的情况相当少见，所以这种用法还没有记录下来。

另一节描述了在Python中使用DOM定义的异常。

DOMImplementation Objects

The DOMImplementation interface provides a way for applications todetermine the availability of particular features in the DOM they are using.DOM Level 2 added the ability to create new Document andDocumentType objects using the DOMImplementation as well.

• DOMImplementation.hasFeature(feature, version)

• Return True if the feature identified by the pair of strings feature andversion is implemented.

• DOMImplementation.createDocument(namespaceUri, qualifiedName, doctype)

• Return a new Document object (the root of the DOM), with a childElement object having the given namespaceUri and qualifiedName. Thedoctype must be a DocumentType object created bycreateDocumentType(), or None. In the Python DOM API, the first twoarguments can also be None in order to indicate that no Elementchild is to be created.

• DOMImplementation.createDocumentType(qualifiedName, publicId, systemId)

• Return a new DocumentType object that encapsulates the givenqualifiedName, publicId, and systemId strings, representing theinformation contained in an XML document type declaration.

节点对象

All of the components of an XML document are subclasses of Node.

• Node.nodeType

• An integer representing the node type. Symbolic constants for the types are onthe Node object: ELEMENT_NODE, ATTRIBUTE_NODE,TEXT_NODE, CDATA_SECTION_NODE, ENTITY_NODE,PROCESSING_INSTRUCTION_NODE, COMMENT_NODE,DOCUMENT_NODE, DOCUMENT_TYPE_NODE, NOTATION_NODE.This is a read-only attribute.

• Node.parentNode

• The parent of the current node, or None for the document node. The value isalways a Node object or None. For Element nodes, thiswill be the parent element, except for the root element, in which case it willbe the Document object. For Attr nodes, this is alwaysNone. This is a read-only attribute.

• Node.attributes

• A NamedNodeMap of attribute objects. Only elements have actual valuesfor this; others provide None for this attribute. This is a read-onlyattribute.

• Node.previousSibling

• The node that immediately precedes this one with the same parent. Forinstance the element with an end-tag that comes just before the _self_element's start-tag. Of course, XML documents are made up of more than justelements so the previous sibling could be text, a comment, or something else.If this node is the first child of the parent, this attribute will beNone. This is a read-only attribute.

• Node.nextSibling

• The node that immediately follows this one with the same parent. See alsopreviousSibling. If this is the last child of the parent, thisattribute will be None. This is a read-only attribute.

• Node.childNodes

• A list of nodes contained within this node. This is a read-only attribute.

• Node.firstChild

• The first child of the node, if there are any, or None. This is a read-onlyattribute.

• Node.lastChild

• The last child of the node, if there are any, or None. This is a read-onlyattribute.

• Node.localName

• The part of the tagName following the colon if there is one, else theentire tagName. The value is a string.

• Node.prefix

• The part of the tagName preceding the colon if there is one, else theempty string. The value is a string, or None.

• Node.namespaceURI

• The namespace associated with the element name. This will be a string orNone. This is a read-only attribute.

• Node.nodeName

• This has a different meaning for each node type; see the DOM specification fordetails. You can always get the information you would get here from anotherproperty such as the tagName property for elements or the nameproperty for attributes. For all node types, the value of this attribute will beeither a string or None. This is a read-only attribute.

• Node.nodeValue

• This has a different meaning for each node type; see the DOM specification fordetails. The situation is similar to that with nodeName. The value isa string or None.

• Node.hasAttributes()

• Return True if the node has any attributes.

• Node.hasChildNodes()

• Return True if the node has any child nodes.

• Node.isSameNode(other)

• Return True if other refers to the same node as this node. This is especiallyuseful for DOM implementations which use any sort of proxy architecture (becausemore than one object can refer to the same node).

注解

This is based on a proposed DOM Level 3 API which is still in the "workingdraft" stage, but this particular interface appears uncontroversial. Changesfrom the W3C will not necessarily affect this method in the Python DOM interface(though any new W3C API for this would also be supported).

• Node.appendChild(newChild)

• Add a new child node to this node at the end of the list ofchildren, returning newChild. If the node was already inthe tree, it is removed first.

• Node.insertBefore(newChild, refChild)

• Insert a new child node before an existing child. It must be the case thatrefChild is a child of this node; if not, ValueError is raised.newChild is returned. If refChild is None, it inserts newChild at theend of the children's list.

• Node.removeChild(oldChild)

• Remove a child node. oldChild must be a child of this node; if not,ValueError is raised. oldChild is returned on success. If _oldChild_will not be used further, its unlink() method should be called.

• Node.replaceChild(newChild, oldChild)

• Replace an existing node with a new node. It must be the case that _oldChild_is a child of this node; if not, ValueError is raised.

• Node.normalize()

• Join adjacent text nodes so that all stretches of text are stored as singleText instances. This simplifies processing text from a DOM tree formany applications.

• Node.cloneNode(deep)

• Clone this node. Setting deep means to clone all child nodes as well. Thisreturns the clone.

节点列表对象

A NodeList represents a sequence of nodes. These objects are used intwo ways in the DOM Core recommendation: an Element object providesone as its list of child nodes, and the getElementsByTagName() andgetElementsByTagNameNS() methods of Node return objects with thisinterface to represent query results.

The DOM Level 2 recommendation defines one method and one attribute for theseobjects:

• NodeList.item(i)

• Return the i'th item from the sequence, if there is one, or None. Theindex i is not allowed to be less than zero or greater than or equal to thelength of the sequence.

• NodeList.length

• The number of nodes in the sequence.

In addition, the Python DOM interface requires that some additional support isprovided to allow NodeList objects to be used as Python sequences. AllNodeList implementations must include support forlen() andgetitem(); this allows iteration over the NodeList infor statements and proper support for the len() built-infunction.

If a DOM implementation supports modification of the document, theNodeList implementation must also support thesetitem() and delitem() methods.

文档类型对象

Information about the notations and entities declared by a document (includingthe external subset if the parser uses it and can provide the information) isavailable from a DocumentType object. The DocumentType for adocument is available from the Document object's doctypeattribute; if there is no DOCTYPE declaration for the document, thedocument's doctype attribute will be set to None instead of aninstance of this interface.

DocumentType is a specialization of Node, and adds thefollowing attributes:

• DocumentType.publicId

• The public identifier for the external subset of the document type definition.This will be a string or None.

• DocumentType.systemId

• The system identifier for the external subset of the document type definition.This will be a URI as a string, or None.

• DocumentType.internalSubset

• A string giving the complete internal subset from the document. This does notinclude the brackets which enclose the subset. If the document has no internalsubset, this should be None.

• DocumentType.name

• The name of the root element as given in the DOCTYPE declaration, ifpresent.

• DocumentType.entities

• This is a NamedNodeMap giving the definitions of external entities.For entity names defined more than once, only the first definition is provided(others are ignored as required by the XML recommendation). This may beNone if the information is not provided by the parser, or if no entities aredefined.

• DocumentType.notations

• This is a NamedNodeMap giving the definitions of notations. Fornotation names defined more than once, only the first definition is provided(others are ignored as required by the XML recommendation). This may beNone if the information is not provided by the parser, or if no notationsare defined.

文档对象

A Document represents an entire XML document, including its constituentelements, attributes, processing instructions, comments etc. Remember that itinherits properties from Node.

• Document.documentElement

• The one and only root element of the document.

• Document.createElement(tagName)

• Create and return a new element node. The element is not inserted into thedocument when it is created. You need to explicitly insert it with one of theother methods such as insertBefore() or appendChild().

• Document.createElementNS(namespaceURI, tagName)

• Create and return a new element with a namespace. The tagName may have aprefix. The element is not inserted into the document when it is created. Youneed to explicitly insert it with one of the other methods such asinsertBefore() or appendChild().

• Document.createTextNode(data)

• Create and return a text node containing the data passed as a parameter. Aswith the other creation methods, this one does not insert the node into thetree.

• Document.createComment(data)

• Create and return a comment node containing the data passed as a parameter. Aswith the other creation methods, this one does not insert the node into thetree.

• Document.createProcessingInstruction(target, data)

• Create and return a processing instruction node containing the target anddata passed as parameters. As with the other creation methods, this one doesnot insert the node into the tree.

• Document.createAttribute(name)

• Create and return an attribute node. This method does not associate theattribute node with any particular element. You must usesetAttributeNode() on the appropriate Element object to use thenewly created attribute instance.

• Document.createAttributeNS(namespaceURI, qualifiedName)

• Create and return an attribute node with a namespace. The tagName may have aprefix. This method does not associate the attribute node with any particularelement. You must use setAttributeNode() on the appropriateElement object to use the newly created attribute instance.

• Document.getElementsByTagName(tagName)

• Search for all descendants (direct children, children's children, etc.) with aparticular element type name.

• Document.getElementsByTagNameNS(namespaceURI, localName)

• Search for all descendants (direct children, children's children, etc.) with aparticular namespace URI and localname. The localname is the part of thenamespace after the prefix.

元素对象

Element is a subclass of Node, so inherits all the attributesof that class.

• Element.tagName

• The element type name. In a namespace-using document it may have colons in it.The value is a string.

• Element.getElementsByTagName(tagName)

• Same as equivalent method in the Document class.

• Element.getElementsByTagNameNS(namespaceURI, localName)

• Same as equivalent method in the Document class.

• Element.hasAttribute(name)

• Return True if the element has an attribute named by name.

• Element.hasAttributeNS(namespaceURI, localName)

• Return True if the element has an attribute named by namespaceURI andlocalName.

• Element.getAttribute(name)

• Return the value of the attribute named by name as a string. If no suchattribute exists, an empty string is returned, as if the attribute had no value.

• Element.getAttributeNode(attrname)

• Return the Attr node for the attribute named by attrname.

• Element.getAttributeNS(namespaceURI, localName)

• Return the value of the attribute named by namespaceURI and localName as astring. If no such attribute exists, an empty string is returned, as if theattribute had no value.

• Element.getAttributeNodeNS(namespaceURI, localName)

• Return an attribute value as a node, given a namespaceURI and localName.

• Element.removeAttribute(name)

• Remove an attribute by name. If there is no matching attribute, aNotFoundErr is raised.

• Element.removeAttributeNode(oldAttr)

• Remove and return oldAttr from the attribute list, if present. If oldAttr isnot present, NotFoundErr is raised.

• Element.removeAttributeNS(namespaceURI, localName)

• Remove an attribute by name. Note that it uses a localName, not a qname. Noexception is raised if there is no matching attribute.

• Element.setAttribute(name, value)

• Set an attribute value from a string.

• Element.setAttributeNode(newAttr)

• Add a new attribute node to the element, replacing an existing attribute ifnecessary if the name attribute matches. If a replacement occurs, theold attribute node will be returned. If newAttr is already in use,InuseAttributeErr will be raised.

• Element.setAttributeNodeNS(newAttr)

• Add a new attribute node to the element, replacing an existing attribute ifnecessary if the namespaceURI and localName attributes match.If a replacement occurs, the old attribute node will be returned. If _newAttr_is already in use, InuseAttributeErr will be raised.

• Element.setAttributeNS(namespaceURI, qname, value)

• Set an attribute value from a string, given a namespaceURI and a qname.Note that a qname is the whole attribute name. This is different than above.

Attr 对象

Attr inherits from Node, so inherits all its attributes.

• Attr.name

• The attribute name.In a namespace-using document it may include a colon.

• Attr.localName

• The part of the name following the colon if there is one, else theentire name.This is a read-only attribute.

• Attr.prefix

• The part of the name preceding the colon if there is one, else theempty string.

• Attr.value

• The text value of the attribute. This is a synonym for thenodeValue attribute.

NamedNodeMap 对象

NamedNodeMap does not inherit from Node.

• NamedNodeMap.length

• The length of the attribute list.

• NamedNodeMap.item(index)

• Return an attribute with a particular index. The order you get the attributesin is arbitrary but will be consistent for the life of a DOM. Each item is anattribute node. Get its value with the value attribute.

There are also experimental methods that give this class more mapping behavior.You can use them or you can use the standardized getAttribute*() familyof methods on the Element objects.

注释对象

Comment represents a comment in the XML document. It is a subclass ofNode, but cannot have child nodes.

• Comment.data
• The content of the comment as a string. The attribute contains all charactersbetween the leading <!-- and trailing -->, but does notinclude them.

Text 和 CDATASection 对象

The Text interface represents text in the XML document. If the parserand DOM implementation support the DOM's XML extension, portions of the textenclosed in CDATA marked sections are stored in CDATASection objects.These two interfaces are identical, but provide different values for thenodeType attribute.

These interfaces extend the Node interface. They cannot have childnodes.

• Text.data
• The content of the text node as a string.

注解

The use of a CDATASection node does not indicate that the noderepresents a complete CDATA marked section, only that the content of the nodewas part of a CDATA section. A single CDATA section may be represented by morethan one node in the document tree. There is no way to determine whether twoadjacent CDATASection nodes represent different CDATA marked sections.

ProcessingInstruction 对象

Represents a processing instruction in the XML document; this inherits from theNode interface and cannot have child nodes.

• ProcessingInstruction.target

• The content of the processing instruction up to the first whitespace character.This is a read-only attribute.

• ProcessingInstruction.data

• The content of the processing instruction following the first whitespacecharacter.

异常

The DOM Level 2 recommendation defines a single exception, DOMException,and a number of constants that allow applications to determine what sort oferror occurred. DOMException instances carry a code attributethat provides the appropriate value for the specific exception.

The Python DOM interface provides the constants, but also expands the set ofexceptions so that a specific exception exists for each of the exception codesdefined by the DOM. The implementations must raise the appropriate specificexception, each of which carries the appropriate value for the codeattribute.

• exception xml.dom.DOMException

• Base exception class used for all specific DOM exceptions. This exception classcannot be directly instantiated.

• exception xml.dom.DomstringSizeErr

• Raised when a specified range of text does not fit into a string. This is notknown to be used in the Python DOM implementations, but may be received from DOMimplementations not written in Python.

• exception xml.dom.HierarchyRequestErr

• Raised when an attempt is made to insert a node where the node type is notallowed.

• exception xml.dom.IndexSizeErr

• Raised when an index or size parameter to a method is negative or exceeds theallowed values.

• exception xml.dom.InuseAttributeErr

• Raised when an attempt is made to insert an Attr node that is alreadypresent elsewhere in the document.

• exception xml.dom.InvalidAccessErr

• Raised if a parameter or an operation is not supported on the underlying object.

• exception xml.dom.InvalidCharacterErr

• This exception is raised when a string parameter contains a character that isnot permitted in the context it's being used in by the XML 1.0 recommendation.For example, attempting to create an Element node with a space in theelement type name will cause this error to be raised.

• exception xml.dom.InvalidModificationErr

• Raised when an attempt is made to modify the type of a node.

• exception xml.dom.InvalidStateErr

• Raised when an attempt is made to use an object that is not defined or is nolonger usable.

• exception xml.dom.NamespaceErr

• If an attempt is made to change any object in a way that is not permitted withregard to the Namespaces in XMLrecommendation, this exception is raised.

• exception xml.dom.NotFoundErr

• Exception when a node does not exist in the referenced context. For example,NamedNodeMap.removeNamedItem() will raise this if the node passed in doesnot exist in the map.

• exception xml.dom.NotSupportedErr

• Raised when the implementation does not support the requested type of object oroperation.

• exception xml.dom.NoDataAllowedErr

• This is raised if data is specified for a node which does not support data.

• exception xml.dom.NoModificationAllowedErr

• Raised on attempts to modify an object where modifications are not allowed (suchas for read-only nodes).

• exception xml.dom.SyntaxErr

• Raised when an invalid or illegal string is specified.

• exception xml.dom.WrongDocumentErr

• Raised when a node is inserted in a different document than it currently belongsto, and the implementation does not support migrating the node from one documentto the other.

The exception codes defined in the DOM recommendation map to the exceptionsdescribed above according to this table:

一致性

This section describes the conformance requirements and relationships betweenthe Python DOM API, the W3C DOM recommendations, and the OMG IDL mapping forPython.

类型映射

将根据下表，将DOM规范中使用的IDL类型映射为Python类型。

Accessor Methods

The mapping from OMG IDL to Python defines accessor functions for IDLattribute declarations in much the way the Java mapping does.Mapping the IDL declarations

readonly attribute string someValue;
         attribute string anotherValue;

yields three accessor functions: a "get" method for someValue(get_someValue()), and "get" and "set" methods for anotherValue(_get_anotherValue() and _set_anotherValue()). The mapping, inparticular, does not require that the IDL attributes are accessible as normalPython attributes: object.someValue is _not required to work, and mayraise an AttributeError.

The Python DOM API, however, does require that normal attribute access work.This means that the typical surrogates generated by Python IDL compilers are notlikely to work, and wrapper objects may be needed on the client if the DOMobjects are accessed via CORBA. While this does require some additionalconsideration for CORBA DOM clients, the implementers with experience using DOMover CORBA from Python do not consider this a problem. Attributes that aredeclared readonly may not restrict write access in all DOMimplementations.

In the Python DOM API, accessor functions are not required. If provided, theyshould take the form defined by the Python IDL mapping, but these methods areconsidered unnecessary since the attributes are accessible directly from Python."Set" accessors should never be provided for readonly attributes.

The IDL definitions do not fully embody the requirements of the W3C DOM API,such as the notion of certain objects, such as the return value ofgetElementsByTagName(), being "live". The Python DOM API does not requireimplementations to enforce such requirements.