xml.parsers.expat â Fast XML parsing using Expat¶
Note
If you need to parse untrusted or unauthenticated data, see XML security.
The xml.parsers.expat module is a Python interface to the Expat non-validating XML parser. The module provides a single extension type, xmlparser, that represents the current state of an XML parser. After an xmlparser object has been created, various attributes of the object can be set to handler functions. When an XML document is then fed to the parser, the handler functions are called for the character data and markup in the XML document.
This module uses the pyexpat module to provide access to the Expat parser. Direct use of the pyexpat module is deprecated.
This module provides one exception and one type object:
The exception raised when Expat reports an error. See section ExpatError Exceptions for more information on interpreting Expat errors.
Alias for ExpatError.
The type of the return values from the ParserCreate() function.
The xml.parsers.expat module contains two functions:
Returns an explanatory string for a given error number errno.
Creates and returns a new xmlparser object. encoding, if specified, must be a string naming the encoding used by the XML data. Expat doesnât support as many encodings as Python does, and its repertoire of encodings canât be extended; it supports UTF-8, UTF-16, ISO-8859-1 (Latin1), and ASCII. If encoding [1] is given it will override the implicit or explicit encoding of the document.
Expat can optionally do XML namespace processing for you, enabled by providing a value for namespace_separator. The value must be a one-character string; a ValueError will be raised if the string has an illegal length (None is considered the same as omission). When namespace processing is enabled, element type names and attribute names that belong to a namespace will be expanded. The element name passed to the element handlers StartElementHandler and EndElementHandler will be the concatenation of the namespace URI, the namespace separator character, and the local part of the name. If the namespace separator is a zero byte (chr(0)) then the namespace URI and the local part will be concatenated without any separator.
For example, if namespace_separator is set to a space character (' ') and the following document is parsed:
<?xml version="1.0"?>
<root xmlns = "http://default-namespace.org/"
xmlns:py = "http://www.python.org/ns/">
<py:elem1 />
<elem2 xmlns="" />
</root>
StartElementHandler will receive the following strings for each element:
http://default-namespace.org/ root http://www.python.org/ns/ elem1 elem2
Due to limitations in the Expat library used by pyexpat, the xmlparser instance returned can only be used to parse a single XML document. Call ParserCreate for each document to provide unique parser instances.
xmlparser objects have the following methods:
Parses the contents of the string data, calling the appropriate handler functions to process the parsed data. isfinal must be true on the final call to this method; it allows the parsing of a single file in fragments, not the submission of multiple files. data can be the empty string at any time.
Parse XML data reading from the object file. file only needs to provide the read(nbytes) method, returning the empty string when thereâs no more data.
Sets the base to be used for resolving relative URIs in system identifiers in declarations. Resolving relative identifiers is left to the application: this value will be passed through as the base argument to the ExternalEntityRefHandler(), NotationDeclHandler(), and UnparsedEntityDeclHandler() functions.
Returns a string containing the base set by a previous call to SetBase(), or None if SetBase() hasnât been called.
Returns the input data that generated the current event as a string. The data is in the encoding of the entity which contains the text. When called while an event handler is not active, the return value is None.
Create a âchildâ parser which can be used to parse an external parsed entity referred to by content parsed by the parent parser. The context parameter should be the string passed to the ExternalEntityRefHandler() handler function, described below. The child parser is created with the ordered_attributes and specified_attributes set to the values of this parser.
Control parsing of parameter entities (including the external DTD subset). Possible flag values are XML_PARAM_ENTITY_PARSING_NEVER, XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE and XML_PARAM_ENTITY_PARSING_ALWAYS. Return true if setting the flag was successful.
Calling this with a true value for flag (the default) will cause Expat to call the ExternalEntityRefHandler with None for all arguments to allow an alternate DTD to be loaded. If the document does not contain a document type declaration, the ExternalEntityRefHandler will still be called, but the StartDoctypeDeclHandler and EndDoctypeDeclHandler will not be called.
Passing a false value for flag will cancel a previous call that passed a true value, but otherwise has no effect.
This method can only be called before the Parse() or ParseFile() methods are called; calling it after either of those have been called causes ExpatError to be raised with the code attribute set to errors.codes[errors.XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING].
Warning
Calling SetReparseDeferralEnabled(False) has security implications, as detailed below; please make sure to understand these consequences prior to using the SetReparseDeferralEnabled method.
Expat 2.6.0 introduced a security mechanism called âreparse deferralâ where instead of causing denial of service through quadratic runtime from reparsing large tokens, reparsing of unfinished tokens is now delayed by default until a sufficient amount of input is reached. Due to this delay, registered handlers may â depending of the sizing of input chunks pushed to Expat â no longer be called right after pushing new input to the parser. Where immediate feedback and taking over responsibility of protecting against denial of service from large tokens are both wanted, calling SetReparseDeferralEnabled(False) disables reparse deferral for the current Expat parser instance, temporarily or altogether. Calling SetReparseDeferralEnabled(True) allows re-enabling reparse deferral.
Note that SetReparseDeferralEnabled() has been backported to some prior releases of CPython as a security fix. Check for availability of SetReparseDeferralEnabled() using hasattr() if used in code running across a variety of Python versions.
Added in version 3.13.
Returns whether reparse deferral is currently enabled for the given Expat parser instance.
Added in version 3.13.
xmlparser objects have the following attributes:
The size of the buffer used when buffer_text is true. A new buffer size can be set by assigning a new integer value to this attribute. When the size is changed, the buffer will be flushed.
Setting this to true causes the xmlparser object to buffer textual content returned by Expat to avoid multiple calls to the CharacterDataHandler() callback whenever possible. This can improve performance substantially since Expat normally breaks character data into chunks at every line ending. This attribute is false by default, and may be changed at any time. Note that when it is false, data that does not contain newlines may be chunked too.
If buffer_text is enabled, the number of bytes stored in the buffer. These bytes represent UTF-8 encoded text. This attribute has no meaningful interpretation when buffer_text is false.
Setting this attribute to a non-zero integer causes the attributes to be reported as a list rather than a dictionary. The attributes are presented in the order found in the document text. For each attribute, two list entries are presented: the attribute name and the attribute value. (Older versions of this module also used this format.) By default, this attribute is false; it may be changed at any time.
If set to a non-zero integer, the parser will report only those attributes which were specified in the document instance and not those which were derived from attribute declarations. Applications which set this need to be especially careful to use what additional information is available from the declarations as needed to comply with the standards for the behavior of XML processors. By default, this attribute is false; it may be changed at any time.
The following attributes contain values relating to the most recent error encountered by an xmlparser object, and will only have correct values once a call to Parse() or ParseFile() has raised an xml.parsers.expat.ExpatError exception.
Byte index at which an error occurred.
Numeric code specifying the problem. This value can be passed to the ErrorString() function, or compared to one of the constants defined in the errors object.
Column number at which an error occurred.
Line number at which an error occurred.
The following attributes contain values relating to the current parse location in an xmlparser object. During a callback reporting a parse event they indicate the location of the first of the sequence of characters that generated the event. When called outside of a callback, the position indicated will be just past the last parse event (regardless of whether there was an associated callback).
Current byte index in the parser input.
Current column number in the parser input.
Current line number in the parser input.
Here is the list of handlers that can be set. To set a handler on an xmlparser object o, use o.handlername = func. handlername must be taken from the following list, and func must be a callable object accepting the correct number of arguments. The arguments are all strings, unless otherwise stated.
Called when the XML declaration is parsed. The XML declaration is the (optional) declaration of the applicable version of the XML recommendation, the encoding of the document text, and an optional âstandaloneâ declaration. version and encoding will be strings, and standalone will be 1 if the document is declared standalone, 0 if it is declared not to be standalone, or -1 if the standalone clause was omitted. This is only available with Expat version 1.95.0 or newer.
Called when Expat begins parsing the document type declaration (<!DOCTYPE ...). The doctypeName is provided exactly as presented. The systemId and publicId parameters give the system and public identifiers if specified, or None if omitted. has_internal_subset will be true if the document contains and internal document declaration subset. This requires Expat version 1.2 or newer.
Called when Expat is done parsing the document type declaration. This requires Expat version 1.2 or newer.
Called once for each element type declaration. name is the name of the element type, and model is a representation of the content model.
Called for each declared attribute for an element type. If an attribute list declaration declares three attributes, this handler is called three times, once for each attribute. elname is the name of the element to which the declaration applies and attname is the name of the attribute declared. The attribute type is a string passed as type; the possible values are 'CDATA', 'ID', 'IDREF', ⦠default gives the default value for the attribute used when the attribute is not specified by the document instance, or None if there is no default value (#IMPLIED values). If the attribute is required to be given in the document instance, required will be true. This requires Expat version 1.95.0 or newer.
Called for the start of every element. name is a string containing the element name, and attributes is the element attributes. If ordered_attributes is true, this is a list (see ordered_attributes for a full description). Otherwise itâs a dictionary mapping names to values.
Called for the end of every element.
Called for every processing instruction.
Called for character data. This will be called for normal character data, CDATA marked content, and ignorable whitespace. Applications which must distinguish these cases can use the StartCdataSectionHandler, EndCdataSectionHandler, and ElementDeclHandler callbacks to collect the required information. Note that the character data may be chunked even if it is short and so you may receive more than one call to CharacterDataHandler(). Set the buffer_text instance attribute to True to avoid that.
Called for unparsed (NDATA) entity declarations. This is only present for version 1.2 of the Expat library; for more recent versions, use EntityDeclHandler instead. (The underlying function in the Expat library has been declared obsolete.)
Called for all entity declarations. For parameter and internal entities, value will be a string giving the declared contents of the entity; this will be None for external entities. The notationName parameter will be None for parsed entities, and the name of the notation for unparsed entities. is_parameter_entity will be true if the entity is a parameter entity or false for general entities (most applications only need to be concerned with general entities). This is only available starting with version 1.95.0 of the Expat library.
Called for notation declarations. notationName, base, and systemId, and publicId are strings if given. If the public identifier is omitted, publicId will be None.
Called when an element contains a namespace declaration. Namespace declarations are processed before the StartElementHandler is called for the element on which declarations are placed.
Called when the closing tag is reached for an element that contained a namespace declaration. This is called once for each namespace declaration on the element in the reverse of the order for which the StartNamespaceDeclHandler was called to indicate the start of each namespace declarationâs scope. Calls to this handler are made after the corresponding EndElementHandler for the end of the element.
Called for comments. data is the text of the comment, excluding the leading '<!--' and trailing '-->'.
Called at the start of a CDATA section. This and EndCdataSectionHandler are needed to be able to identify the syntactical start and end for CDATA sections.
Called at the end of a CDATA section.
Called for any characters in the XML document for which no applicable handler has been specified. This means characters that are part of a construct which could be reported, but for which no handler has been supplied.
This is the same as the DefaultHandler(), but doesnât inhibit expansion of internal entities. The entity reference will not be passed to the default handler.
Called if the XML document hasnât been declared as being a standalone document. This happens when there is an external subset or a reference to a parameter entity, but the XML declaration does not set standalone to yes in an XML declaration. If this handler returns 0, then the parser will raise an XML_ERROR_NOT_STANDALONE error. If this handler is not set, no exception is raised by the parser for this condition.
Called for references to external entities. base is the current base, as set by a previous call to SetBase(). The public and system identifiers, systemId and publicId, are strings if given; if the public identifier is not given, publicId will be None. The context value is opaque and should only be used as described below.
For external entities to be parsed, this handler must be implemented. It is responsible for creating the sub-parser using ExternalEntityParserCreate(context), initializing it with the appropriate callbacks, and parsing the entity. This handler should return an integer; if it returns 0, the parser will raise an XML_ERROR_EXTERNAL_ENTITY_HANDLING error, otherwise parsing will continue.
If this handler is not provided, external entities are reported by the DefaultHandler callback, if provided.
ExpatError exceptions have a number of interesting attributes:
Expatâs internal error number for the specific error. The errors.messages dictionary maps these error numbers to Expatâs error messages. For example:
from xml.parsers.expat import ParserCreate, ExpatError, errors
p = ParserCreate()
try:
p.Parse(some_xml_document)
except ExpatError as err:
print("Error:", errors.messages[err.code])
The errors module also provides error message constants and a dictionary codes mapping these messages back to the error codes, see below.
Line number on which the error was detected. The first line is numbered 1.
Character offset into the line where the error occurred. The first column is numbered 0.
The following program defines three handlers that just print out their arguments.
import xml.parsers.expat
# 3 handler functions
def start_element(name, attrs):
print('Start element:', name, attrs)
def end_element(name):
print('End element:', name)
def char_data(data):
print('Character data:', repr(data))
p = xml.parsers.expat.ParserCreate()
p.StartElementHandler = start_element
p.EndElementHandler = end_element
p.CharacterDataHandler = char_data
p.Parse("""<?xml version="1.0"?>
<parent id="top"><child1 name="paul">Text goes here</child1>
<child2 name="fred">More text</child2>
</parent>""", 1)
The output from this program is:
Start element: parent {'id': 'top'}
Start element: child1 {'name': 'paul'}
Character data: 'Text goes here'
End element: child1
Character data: '\n'
Start element: child2 {'name': 'fred'}
Character data: 'More text'
End element: child2
Character data: '\n'
End element: parent
Content Model Descriptions¶
Content models are described using nested tuples. Each tuple contains four values: the type, the quantifier, the name, and a tuple of children. Children are simply additional content model descriptions.
The values of the first two fields are constants defined in the xml.parsers.expat.model module. These constants can be collected in two groups: the model type group and the quantifier group.
The constants in the model type group are:
The element named by the model name was declared to have a content model of ANY.
The named element allows a choice from a number of options; this is used for content models such as (A | B | C).
Elements which are declared to be EMPTY have this model type.
Models which represent a series of models which follow one after the other are indicated with this model type. This is used for models such as (A, B, C).
The constants in the quantifier group are:
No modifier is given, so it can appear exactly once, as for A.
The model is optional: it can appear once or not at all, as for A?.
The model must occur one or more times (like A+).
The model must occur zero or more times, as for A*.
The following constants are provided in the xml.parsers.expat.errors module. These constants are useful in interpreting some of the attributes of the ExpatError exception objects raised when an error has occurred. Since for backwards compatibility reasons, the constantsâ value is the error message and not the numeric error code, you do this by comparing its code attribute with errors.codes[errors.XML_ERROR_CONSTANT_NAME].
The errors module has the following attributes:
A dictionary mapping string descriptions to their error codes.
Added in version 3.2.
A dictionary mapping numeric error codes to their string descriptions.
Added in version 3.2.
An entity reference in an attribute value referred to an external entity instead of an internal entity.
A character reference referred to a character which is illegal in XML (for example, character 0, or â�â).
An entity reference referred to an entity which was declared with a notation, so cannot be parsed.
An attribute was used more than once in a start tag.
Raised when an input byte could not properly be assigned to a character; for example, a NUL byte (value 0) in a UTF-8 input stream.
Something other than whitespace occurred after the document element.
An XML declaration was found somewhere other than the start of the input data.
The document contains no elements (XML requires all documents to contain exactly one top-level element)..
Expat was not able to allocate memory internally.
A parameter entity reference was found where it was not allowed.
An incomplete character was found in the input.
An entity reference contained another reference to the same entity; possibly via a different name, and possibly indirectly.
Some unspecified syntax error was encountered.
An end tag did not match the innermost open start tag.
Some token (such as a start tag) was not closed before the end of the stream or the next token was encountered.
A reference was made to an entity which was not defined.
The document encoding is not supported by Expat.
A CDATA marked section was not closed.
The parser determined that the document was not âstandaloneâ though it declared itself to be in the XML declaration, and the NotStandaloneHandler was set and returned 0.
An operation was requested that requires DTD support to be compiled in, but Expat was configured without DTD support. This should never be reported by a standard build of the xml.parsers.expat module.
A behavioral change was requested after parsing started that can only be changed before parsing has started. This is (currently) only raised by UseForeignDTD().
An undeclared prefix was found when namespace processing was enabled.
The document attempted to remove the namespace declaration associated with a prefix.
A parameter entity contained incomplete markup.
The document contained no document element at all.
There was an error parsing a text declaration in an external entity.
Characters were found in the public id that are not allowed.
The requested operation was made on a suspended parser, but isnât allowed. This includes attempts to provide additional input or to stop the parser.
An attempt to resume the parser was made when the parser had not been suspended.
This should not be reported to Python applications.
The requested operation was made on a parser which was finished parsing input, but isnât allowed. This includes attempts to provide additional input or to stop the parser.
An attempt was made to undeclare reserved namespace prefix xml or to bind it to another namespace URI.
An attempt was made to declare or undeclare reserved namespace prefix xmlns.
An attempt was made to bind the URI of one the reserved namespace prefixes xml and xmlns to another namespace prefix.
This should not be reported to Python applications.
This should not be reported to Python applications.
The limit on input amplification factor (from DTD and entities) has been breached.
Footnotes
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4