2
2
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
4
4
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
5
<meta name="generator" content="Docutils 0.4.1: http://docutils.sourceforge.net/" />
5
<meta name="generator" content="Docutils 0.5: http://docutils.sourceforge.net/" />
6
6
<title>Using custom Element classes in lxml</title>
7
7
<link rel="stylesheet" href="style.css" type="text/css" />
10
10
<div class="document" id="using-custom-element-classes-in-lxml">
11
<div class="sidemenu"><ul id="lxml-section"><li><span class="section title">lxml</span><ul class="menu foreign" id="index-menu"><li class="menu title"><a href="index.html">lxml</a><ul class="submenu"><li class="menu item"><a href="index.html#introduction">Introduction</a></li><li class="menu item"><a href="index.html#documentation">Documentation</a></li><li class="menu item"><a href="index.html#download">Download</a></li><li class="menu item"><a href="index.html#mailing-list">Mailing list</a></li><li class="menu item"><a href="index.html#bug-tracker">Bug tracker</a></li><li class="menu item"><a href="index.html#license">License</a></li><li class="menu item"><a href="index.html#old-versions">Old Versions</a></li></ul></li></ul><ul class="menu foreign" id="intro-menu"><li class="menu title"><a href="intro.html">Why lxml?</a><ul class="submenu"><li class="menu item"><a href="intro.html#motto">Motto</a></li><li class="menu item"><a href="intro.html#aims">Aims</a></li></ul></li></ul><ul class="menu foreign" id="installation-menu"><li class="menu title"><a href="installation.html">Installing lxml</a><ul class="submenu"><li class="menu item"><a href="installation.html#requirements">Requirements</a></li><li class="menu item"><a href="installation.html#installation">Installation</a></li><li class="menu item"><a href="installation.html#building-lxml-from-sources">Building lxml from sources</a></li><li class="menu item"><a href="installation.html#ms-windows">MS Windows</a></li><li class="menu item"><a href="installation.html#macos-x">MacOS-X</a></li></ul></li></ul><ul class="menu foreign" id="lxml2-menu"><li class="menu title"><a href="lxml2.html">What's new in lxml 2.0?</a><ul class="submenu"><li class="menu item"><a href="lxml2.html#changes-in-etree-and-objectify">Changes in etree and objectify</a></li><li class="menu item"><a href="lxml2.html#new-modules">New modules</a></li></ul></li></ul><ul class="menu foreign" id="performance-menu"><li class="menu title"><a href="performance.html">Benchmarks and Speed</a><ul class="submenu"><li class="menu item"><a href="performance.html#general-notes">General notes</a></li><li class="menu item"><a href="performance.html#how-to-read-the-timings">How to read the timings</a></li><li class="menu item"><a href="performance.html#parsing-and-serialising">Parsing and Serialising</a></li><li class="menu item"><a href="performance.html#the-elementtree-api">The ElementTree API</a></li><li class="menu item"><a href="performance.html#xpath">XPath</a></li><li class="menu item"><a href="performance.html#a-longer-example">A longer example</a></li><li class="menu item"><a href="performance.html#lxml-objectify">lxml.objectify</a></li></ul></li></ul><ul class="menu foreign" id="compatibility-menu"><li class="menu title"><a href="compatibility.html">ElementTree compatibility of lxml.etree</a></li></ul><ul class="menu foreign" id="FAQ-menu"><li class="menu title"><a href="FAQ.html">lxml FAQ - Frequently Asked Questions</a><ul class="submenu"><li class="menu item"><a href="FAQ.html#general-questions">General Questions</a></li><li class="menu item"><a href="FAQ.html#installation">Installation</a></li><li class="menu item"><a href="FAQ.html#contributing">Contributing</a></li><li class="menu item"><a href="FAQ.html#bugs">Bugs</a></li><li class="menu item"><a href="FAQ.html#threading">Threading</a></li><li class="menu item"><a href="FAQ.html#parsing-and-serialisation">Parsing and Serialisation</a></li><li class="menu item"><a href="FAQ.html#xpath-and-document-traversal">XPath and Document Traversal</a></li></ul></li></ul></li></ul><ul id="Developing with lxml-section"><li><span class="section title">Developing with lxml</span><ul class="menu foreign" id="tutorial-menu"><li class="menu title"><a href="tutorial.html">The lxml.etree Tutorial</a><ul class="submenu"><li class="menu item"><a href="tutorial.html#the-element-class">The Element class</a></li><li class="menu item"><a href="tutorial.html#the-elementtree-class">The ElementTree class</a></li><li class="menu item"><a href="tutorial.html#parsing-from-strings-and-files">Parsing from strings and files</a></li><li class="menu item"><a href="tutorial.html#namespaces">Namespaces</a></li><li class="menu item"><a href="tutorial.html#the-e-factory">The E-factory</a></li><li class="menu item"><a href="tutorial.html#elementpath">ElementPath</a></li></ul></li></ul><ul class="menu foreign" id="api index-menu"><li class="menu title"><a href="api/index.html">API reference</a></li></ul><ul class="menu foreign" id="api-menu"><li class="menu title"><a href="api.html">APIs specific to lxml.etree</a><ul class="submenu"><li class="menu item"><a href="api.html#lxml-etree">lxml.etree</a></li><li class="menu item"><a href="api.html#other-element-apis">Other Element APIs</a></li><li class="menu item"><a href="api.html#trees-and-documents">Trees and Documents</a></li><li class="menu item"><a href="api.html#iteration">Iteration</a></li><li class="menu item"><a href="api.html#error-handling-on-exceptions">Error handling on exceptions</a></li><li class="menu item"><a href="api.html#error-logging">Error logging</a></li><li class="menu item"><a href="api.html#serialisation">Serialisation</a></li><li class="menu item"><a href="api.html#cdata">CDATA</a></li><li class="menu item"><a href="api.html#xinclude-and-elementinclude">XInclude and ElementInclude</a></li><li class="menu item"><a href="api.html#write-c14n-on-elementtree">write_c14n on ElementTree</a></li></ul></li></ul><ul class="menu foreign" id="parsing-menu"><li class="menu title"><a href="parsing.html">Parsing XML and HTML with lxml</a><ul class="submenu"><li class="menu item"><a href="parsing.html#parsers">Parsers</a></li><li class="menu item"><a href="parsing.html#the-target-parser-interface">The target parser interface</a></li><li class="menu item"><a href="parsing.html#the-feed-parser-interface">The feed parser interface</a></li><li class="menu item"><a href="parsing.html#iterparse-and-iterwalk">iterparse and iterwalk</a></li><li class="menu item"><a href="parsing.html#python-unicode-strings">Python unicode strings</a></li></ul></li></ul><ul class="menu foreign" id="validation-menu"><li class="menu title"><a href="validation.html">Validation with lxml</a><ul class="submenu"><li class="menu item"><a href="validation.html#validation-at-parse-time">Validation at parse time</a></li><li class="menu item"><a href="validation.html#dtd">DTD</a></li><li class="menu item"><a href="validation.html#relaxng">RelaxNG</a></li><li class="menu item"><a href="validation.html#xmlschema">XMLSchema</a></li><li class="menu item"><a href="validation.html#schematron">Schematron</a></li></ul></li></ul><ul class="menu foreign" id="xpathxslt-menu"><li class="menu title"><a href="xpathxslt.html">XPath and XSLT with lxml</a><ul class="submenu"><li class="menu item"><a href="xpathxslt.html#xpath">XPath</a></li><li class="menu item"><a href="xpathxslt.html#xslt">XSLT</a></li></ul></li></ul><ul class="menu foreign" id="objectify-menu"><li class="menu title"><a href="objectify.html">lxml.objectify</a><ul class="submenu"><li class="menu item"><a href="objectify.html#the-lxml-objectify-api">The lxml.objectify API</a></li><li class="menu item"><a href="objectify.html#asserting-a-schema">Asserting a Schema</a></li><li class="menu item"><a href="objectify.html#objectpath">ObjectPath</a></li><li class="menu item"><a href="objectify.html#python-data-types">Python data types</a></li><li class="menu item"><a href="objectify.html#how-data-types-are-matched">How data types are matched</a></li><li class="menu item"><a href="objectify.html#what-is-different-from-lxml-etree">What is different from lxml.etree?</a></li></ul></li></ul><ul class="menu foreign" id="lxmlhtml-menu"><li class="menu title"><a href="lxmlhtml.html">lxml.html</a><ul class="submenu"><li class="menu item"><a href="lxmlhtml.html#parsing-html">Parsing HTML</a></li><li class="menu item"><a href="lxmlhtml.html#html-element-methods">HTML Element Methods</a></li><li class="menu item"><a href="lxmlhtml.html#running-html-doctests">Running HTML doctests</a></li><li class="menu item"><a href="lxmlhtml.html#creating-html-with-the-e-factory">Creating HTML with the E-factory</a></li><li class="menu item"><a href="lxmlhtml.html#working-with-links">Working with links</a></li><li class="menu item"><a href="lxmlhtml.html#forms">Forms</a></li><li class="menu item"><a href="lxmlhtml.html#cleaning-up-html">Cleaning up HTML</a></li><li class="menu item"><a href="lxmlhtml.html#html-diff">HTML Diff</a></li><li class="menu item"><a href="lxmlhtml.html#examples">Examples</a></li></ul></li></ul><ul class="menu foreign" id="cssselect-menu"><li class="menu title"><a href="cssselect.html">lxml.cssselect</a><ul class="submenu"><li class="menu item"><a href="cssselect.html#the-cssselector-class">The CSSSelector class</a></li><li class="menu item"><a href="cssselect.html#css-selectors">CSS Selectors</a></li><li class="menu item"><a href="cssselect.html#namespaces">Namespaces</a></li><li class="menu item"><a href="cssselect.html#limitations">Limitations</a></li></ul></li></ul><ul class="menu foreign" id="elementsoup-menu"><li class="menu title"><a href="elementsoup.html">BeautifulSoup Parser</a><ul class="submenu"><li class="menu item"><a href="elementsoup.html#parsing-with-the-soupparser">Parsing with the soupparser</a></li><li class="menu item"><a href="elementsoup.html#entity-handling">Entity handling</a></li><li class="menu item"><a href="elementsoup.html#using-soupparser-as-a-fallback">Using soupparser as a fallback</a></li></ul></li></ul></li></ul><ul id="Extending lxml-section"><li><span class="section title">Extending lxml</span><ul class="menu foreign" id="resolvers-menu"><li class="menu title"><a href="resolvers.html">Document loading and URL resolving</a><ul class="submenu"><li class="menu item"><a href="resolvers.html#uri-resolvers">URI Resolvers</a></li><li class="menu item"><a href="resolvers.html#document-loading-in-context">Document loading in context</a></li><li class="menu item"><a href="resolvers.html#i-o-access-control-in-xslt">I/O access control in XSLT</a></li></ul></li></ul><ul class="menu foreign" id="extensions-menu"><li class="menu title"><a href="extensions.html">Python extensions for XPath and XSLT</a><ul class="submenu"><li class="menu item"><a href="extensions.html#xpath-extension-functions">XPath Extension functions</a></li><li class="menu item"><a href="extensions.html#xslt-extension-elements">XSLT extension elements</a></li></ul></li></ul><ul class="menu foreign" id="element classes-menu"><li class="menu title"><a href="element_classes.html">Using custom Element classes in lxml</a><ul class="submenu"><li class="menu item"><a href="element_classes.html#element-initialization">Element initialization</a></li><li class="menu item"><a href="element_classes.html#setting-up-a-class-lookup-scheme">Setting up a class lookup scheme</a></li><li class="menu item"><a href="element_classes.html#implementing-namespaces">Implementing namespaces</a></li></ul></li></ul><ul class="menu foreign" id="sax-menu"><li class="menu title"><a href="sax.html">Sax support</a><ul class="submenu"><li class="menu item"><a href="sax.html#building-a-tree-from-sax-events">Building a tree from SAX events</a></li><li class="menu item"><a href="sax.html#producing-sax-events-from-an-elementtree-or-element">Producing SAX events from an ElementTree or Element</a></li><li class="menu item"><a href="sax.html#interfacing-with-pulldom-minidom">Interfacing with pulldom/minidom</a></li></ul></li></ul><ul class="menu foreign" id="capi-menu"><li class="menu title"><a href="capi.html">The public C-API of lxml.etree</a><ul class="submenu"><li class="menu item"><a href="capi.html#writing-external-modules-in-cython">Writing external modules in Cython</a></li><li class="menu item"><a href="capi.html#writing-external-modules-in-c">Writing external modules in C</a></li></ul></li></ul></li></ul><ul id="Developing lxml-section"><li><span class="section title">Developing lxml</span><ul class="menu foreign" id="build-menu"><li class="menu title"><a href="build.html">How to build lxml from source</a><ul class="submenu"><li class="menu item"><a href="build.html#cython">Cython</a></li><li class="menu item"><a href="build.html#subversion">Subversion</a></li><li class="menu item"><a href="build.html#setuptools">Setuptools</a></li><li class="menu item"><a href="build.html#running-the-tests-and-reporting-errors">Running the tests and reporting errors</a></li><li class="menu item"><a href="build.html#contributing-an-egg">Contributing an egg</a></li><li class="menu item"><a href="build.html#providing-newer-library-versions-on-mac-os-x">Providing newer library versions on Mac-OS X</a></li><li class="menu item"><a href="build.html#static-linking-on-windows">Static linking on Windows</a></li><li class="menu item"><a href="build.html#building-debian-packages-from-svn-sources">Building Debian packages from SVN sources</a></li></ul></li></ul><ul class="menu foreign" id="lxml source howto-menu"><li class="menu title"><a href="lxml-source-howto.html">How to read the source of lxml</a><ul class="submenu"><li class="menu item"><a href="lxml-source-howto.html#what-is-cython">What is Cython?</a></li><li class="menu item"><a href="lxml-source-howto.html#where-to-start">Where to start?</a></li><li class="menu item"><a href="lxml-source-howto.html#lxml-etree">lxml.etree</a></li><li class="menu item"><a href="lxml-source-howto.html#python-modules">Python modules</a></li><li class="menu item"><a href="lxml-source-howto.html#lxml-objectify">lxml.objectify</a></li><li class="menu item"><a href="lxml-source-howto.html#lxml-html">lxml.html</a></li></ul></li></ul><ul class="menu foreign" id="changes 2 1 5-menu"><li class="menu title"><a href="changes-2.1.5.html">Release Changelog</a></li></ul><ul class="menu foreign" id="credits-menu"><li class="menu title"><a href="credits.html">Credits</a><ul class="submenu"><li class="menu item"><a href="credits.html#main-contributors">Main contributors</a></li><li class="menu item"><a href="credits.html#special-thanks-goes-to">Special thanks goes to:</a></li></ul></li></ul></li></ul></div><h1 class="title">Using custom Element classes in lxml</h1>
12
<p>lxml has very sophisticated support for custom Element classes. You can
13
provide your own classes for Elements and have lxml use them by default, for
14
all elements generated by a specific parser, for a specific tag name in a
15
specific namespace or for an exact element at a specific position in the tree.</p>
11
<div class="sidemenu"><ul id="lxml-section"><li><span class="section title">lxml</span><ul class="menu foreign" id="index-menu"><li class="menu title"><a href="index.html">lxml</a><ul class="submenu"><li class="menu item"><a href="index.html#introduction">Introduction</a></li><li class="menu item"><a href="index.html#documentation">Documentation</a></li><li class="menu item"><a href="index.html#download">Download</a></li><li class="menu item"><a href="index.html#mailing-list">Mailing list</a></li><li class="menu item"><a href="index.html#bug-tracker">Bug tracker</a></li><li class="menu item"><a href="index.html#license">License</a></li><li class="menu item"><a href="index.html#old-versions">Old Versions</a></li></ul></li></ul><ul class="menu foreign" id="intro-menu"><li class="menu title"><a href="intro.html">Why lxml?</a><ul class="submenu"><li class="menu item"><a href="intro.html#motto">Motto</a></li><li class="menu item"><a href="intro.html#aims">Aims</a></li></ul></li></ul><ul class="menu foreign" id="installation-menu"><li class="menu title"><a href="installation.html">Installing lxml</a><ul class="submenu"><li class="menu item"><a href="installation.html#requirements">Requirements</a></li><li class="menu item"><a href="installation.html#installation">Installation</a></li><li class="menu item"><a href="installation.html#building-lxml-from-sources">Building lxml from sources</a></li><li class="menu item"><a href="installation.html#ms-windows">MS Windows</a></li><li class="menu item"><a href="installation.html#macos-x">MacOS-X</a></li></ul></li></ul><ul class="menu foreign" id="lxml2-menu"><li class="menu title"><a href="lxml2.html">What's new in lxml 2.0?</a><ul class="submenu"><li class="menu item"><a href="lxml2.html#changes-in-etree-and-objectify">Changes in etree and objectify</a></li><li class="menu item"><a href="lxml2.html#new-modules">New modules</a></li></ul></li></ul><ul class="menu foreign" id="performance-menu"><li class="menu title"><a href="performance.html">Benchmarks and Speed</a><ul class="submenu"><li class="menu item"><a href="performance.html#general-notes">General notes</a></li><li class="menu item"><a href="performance.html#how-to-read-the-timings">How to read the timings</a></li><li class="menu item"><a href="performance.html#parsing-and-serialising">Parsing and Serialising</a></li><li class="menu item"><a href="performance.html#the-elementtree-api">The ElementTree API</a></li><li class="menu item"><a href="performance.html#xpath">XPath</a></li><li class="menu item"><a href="performance.html#a-longer-example">A longer example</a></li><li class="menu item"><a href="performance.html#lxml-objectify">lxml.objectify</a></li></ul></li></ul><ul class="menu foreign" id="compatibility-menu"><li class="menu title"><a href="compatibility.html">ElementTree compatibility of lxml.etree</a></li></ul><ul class="menu foreign" id="FAQ-menu"><li class="menu title"><a href="FAQ.html">lxml FAQ - Frequently Asked Questions</a><ul class="submenu"><li class="menu item"><a href="FAQ.html#general-questions">General Questions</a></li><li class="menu item"><a href="FAQ.html#installation">Installation</a></li><li class="menu item"><a href="FAQ.html#contributing">Contributing</a></li><li class="menu item"><a href="FAQ.html#bugs">Bugs</a></li><li class="menu item"><a href="FAQ.html#threading">Threading</a></li><li class="menu item"><a href="FAQ.html#parsing-and-serialisation">Parsing and Serialisation</a></li><li class="menu item"><a href="FAQ.html#xpath-and-document-traversal">XPath and Document Traversal</a></li></ul></li></ul></li></ul><ul id="Developing with lxml-section"><li><span class="section title">Developing with lxml</span><ul class="menu foreign" id="tutorial-menu"><li class="menu title"><a href="tutorial.html">The lxml.etree Tutorial</a><ul class="submenu"><li class="menu item"><a href="tutorial.html#the-element-class">The Element class</a></li><li class="menu item"><a href="tutorial.html#the-elementtree-class">The ElementTree class</a></li><li class="menu item"><a href="tutorial.html#parsing-from-strings-and-files">Parsing from strings and files</a></li><li class="menu item"><a href="tutorial.html#namespaces">Namespaces</a></li><li class="menu item"><a href="tutorial.html#the-e-factory">The E-factory</a></li><li class="menu item"><a href="tutorial.html#elementpath">ElementPath</a></li></ul></li></ul><ul class="menu foreign" id="api index-menu"><li class="menu title"><a href="api/index.html">API reference</a></li></ul><ul class="menu foreign" id="api-menu"><li class="menu title"><a href="api.html">APIs specific to lxml.etree</a><ul class="submenu"><li class="menu item"><a href="api.html#lxml-etree">lxml.etree</a></li><li class="menu item"><a href="api.html#other-element-apis">Other Element APIs</a></li><li class="menu item"><a href="api.html#trees-and-documents">Trees and Documents</a></li><li class="menu item"><a href="api.html#iteration">Iteration</a></li><li class="menu item"><a href="api.html#error-handling-on-exceptions">Error handling on exceptions</a></li><li class="menu item"><a href="api.html#error-logging">Error logging</a></li><li class="menu item"><a href="api.html#serialisation">Serialisation</a></li><li class="menu item"><a href="api.html#cdata">CDATA</a></li><li class="menu item"><a href="api.html#xinclude-and-elementinclude">XInclude and ElementInclude</a></li><li class="menu item"><a href="api.html#write-c14n-on-elementtree">write_c14n on ElementTree</a></li></ul></li></ul><ul class="menu foreign" id="parsing-menu"><li class="menu title"><a href="parsing.html">Parsing XML and HTML with lxml</a><ul class="submenu"><li class="menu item"><a href="parsing.html#parsers">Parsers</a></li><li class="menu item"><a href="parsing.html#the-target-parser-interface">The target parser interface</a></li><li class="menu item"><a href="parsing.html#the-feed-parser-interface">The feed parser interface</a></li><li class="menu item"><a href="parsing.html#iterparse-and-iterwalk">iterparse and iterwalk</a></li><li class="menu item"><a href="parsing.html#python-unicode-strings">Python unicode strings</a></li></ul></li></ul><ul class="menu foreign" id="validation-menu"><li class="menu title"><a href="validation.html">Validation with lxml</a><ul class="submenu"><li class="menu item"><a href="validation.html#validation-at-parse-time">Validation at parse time</a></li><li class="menu item"><a href="validation.html#dtd">DTD</a></li><li class="menu item"><a href="validation.html#relaxng">RelaxNG</a></li><li class="menu item"><a href="validation.html#xmlschema">XMLSchema</a></li><li class="menu item"><a href="validation.html#schematron">Schematron</a></li></ul></li></ul><ul class="menu foreign" id="xpathxslt-menu"><li class="menu title"><a href="xpathxslt.html">XPath and XSLT with lxml</a><ul class="submenu"><li class="menu item"><a href="xpathxslt.html#xpath">XPath</a></li><li class="menu item"><a href="xpathxslt.html#xslt">XSLT</a></li></ul></li></ul><ul class="menu foreign" id="objectify-menu"><li class="menu title"><a href="objectify.html">lxml.objectify</a><ul class="submenu"><li class="menu item"><a href="objectify.html#the-lxml-objectify-api">The lxml.objectify API</a></li><li class="menu item"><a href="objectify.html#asserting-a-schema">Asserting a Schema</a></li><li class="menu item"><a href="objectify.html#objectpath">ObjectPath</a></li><li class="menu item"><a href="objectify.html#python-data-types">Python data types</a></li><li class="menu item"><a href="objectify.html#how-data-types-are-matched">How data types are matched</a></li><li class="menu item"><a href="objectify.html#what-is-different-from-lxml-etree">What is different from lxml.etree?</a></li></ul></li></ul><ul class="menu foreign" id="lxmlhtml-menu"><li class="menu title"><a href="lxmlhtml.html">lxml.html</a><ul class="submenu"><li class="menu item"><a href="lxmlhtml.html#parsing-html">Parsing HTML</a></li><li class="menu item"><a href="lxmlhtml.html#html-element-methods">HTML Element Methods</a></li><li class="menu item"><a href="lxmlhtml.html#running-html-doctests">Running HTML doctests</a></li><li class="menu item"><a href="lxmlhtml.html#creating-html-with-the-e-factory">Creating HTML with the E-factory</a></li><li class="menu item"><a href="lxmlhtml.html#working-with-links">Working with links</a></li><li class="menu item"><a href="lxmlhtml.html#forms">Forms</a></li><li class="menu item"><a href="lxmlhtml.html#cleaning-up-html">Cleaning up HTML</a></li><li class="menu item"><a href="lxmlhtml.html#html-diff">HTML Diff</a></li><li class="menu item"><a href="lxmlhtml.html#examples">Examples</a></li></ul></li></ul><ul class="menu foreign" id="cssselect-menu"><li class="menu title"><a href="cssselect.html">lxml.cssselect</a><ul class="submenu"><li class="menu item"><a href="cssselect.html#the-cssselector-class">The CSSSelector class</a></li><li class="menu item"><a href="cssselect.html#css-selectors">CSS Selectors</a></li><li class="menu item"><a href="cssselect.html#namespaces">Namespaces</a></li><li class="menu item"><a href="cssselect.html#limitations">Limitations</a></li></ul></li></ul><ul class="menu foreign" id="elementsoup-menu"><li class="menu title"><a href="elementsoup.html">BeautifulSoup Parser</a><ul class="submenu"><li class="menu item"><a href="elementsoup.html#parsing-with-the-soupparser">Parsing with the soupparser</a></li><li class="menu item"><a href="elementsoup.html#entity-handling">Entity handling</a></li><li class="menu item"><a href="elementsoup.html#using-soupparser-as-a-fallback">Using soupparser as a fallback</a></li></ul></li></ul><ul class="menu foreign" id="html5parser-menu"><li class="menu title"><a href="html5parser.html">html5lib Parser</a><ul class="submenu"><li class="menu item"><a href="html5parser.html#differences-to-regular-html-parsing">Differences to regular HTML parsing</a></li><li class="menu item"><a href="html5parser.html#function-reference">Function Reference</a></li></ul></li></ul></li></ul><ul id="Extending lxml-section"><li><span class="section title">Extending lxml</span><ul class="menu foreign" id="resolvers-menu"><li class="menu title"><a href="resolvers.html">Document loading and URL resolving</a><ul class="submenu"><li class="menu item"><a href="resolvers.html#uri-resolvers">URI Resolvers</a></li><li class="menu item"><a href="resolvers.html#document-loading-in-context">Document loading in context</a></li><li class="menu item"><a href="resolvers.html#i-o-access-control-in-xslt">I/O access control in XSLT</a></li></ul></li></ul><ul class="menu foreign" id="extensions-menu"><li class="menu title"><a href="extensions.html">Python extensions for XPath and XSLT</a><ul class="submenu"><li class="menu item"><a href="extensions.html#xpath-extension-functions">XPath Extension functions</a></li><li class="menu item"><a href="extensions.html#xslt-extension-elements">XSLT extension elements</a></li></ul></li></ul><ul class="menu foreign" id="element classes-menu"><li class="menu title"><a href="element_classes.html">Using custom Element classes in lxml</a><ul class="submenu"><li class="menu item"><a href="element_classes.html#background-on-element-proxies">Background on Element proxies</a></li><li class="menu item"><a href="element_classes.html#element-initialization">Element initialization</a></li><li class="menu item"><a href="element_classes.html#setting-up-a-class-lookup-scheme">Setting up a class lookup scheme</a></li><li class="menu item"><a href="element_classes.html#generating-xml-with-custom-classes">Generating XML with custom classes</a></li><li class="menu item"><a href="element_classes.html#implementing-namespaces">Implementing namespaces</a></li></ul></li></ul><ul class="menu foreign" id="sax-menu"><li class="menu title"><a href="sax.html">Sax support</a><ul class="submenu"><li class="menu item"><a href="sax.html#building-a-tree-from-sax-events">Building a tree from SAX events</a></li><li class="menu item"><a href="sax.html#producing-sax-events-from-an-elementtree-or-element">Producing SAX events from an ElementTree or Element</a></li><li class="menu item"><a href="sax.html#interfacing-with-pulldom-minidom">Interfacing with pulldom/minidom</a></li></ul></li></ul><ul class="menu foreign" id="capi-menu"><li class="menu title"><a href="capi.html">The public C-API of lxml.etree</a><ul class="submenu"><li class="menu item"><a href="capi.html#writing-external-modules-in-cython">Writing external modules in Cython</a></li><li class="menu item"><a href="capi.html#writing-external-modules-in-c">Writing external modules in C</a></li></ul></li></ul></li></ul><ul id="Developing lxml-section"><li><span class="section title">Developing lxml</span><ul class="menu foreign" id="build-menu"><li class="menu title"><a href="build.html">How to build lxml from source</a><ul class="submenu"><li class="menu item"><a href="build.html#cython">Cython</a></li><li class="menu item"><a href="build.html#subversion">Subversion</a></li><li class="menu item"><a href="build.html#setuptools">Setuptools</a></li><li class="menu item"><a href="build.html#running-the-tests-and-reporting-errors">Running the tests and reporting errors</a></li><li class="menu item"><a href="build.html#building-an-egg">Building an egg</a></li><li class="menu item"><a href="build.html#building-lxml-on-macos-x">Building lxml on MacOS-X</a></li><li class="menu item"><a href="build.html#static-linking-on-windows">Static linking on Windows</a></li><li class="menu item"><a href="build.html#building-debian-packages-from-svn-sources">Building Debian packages from SVN sources</a></li></ul></li></ul><ul class="menu foreign" id="lxml source howto-menu"><li class="menu title"><a href="lxml-source-howto.html">How to read the source of lxml</a><ul class="submenu"><li class="menu item"><a href="lxml-source-howto.html#what-is-cython">What is Cython?</a></li><li class="menu item"><a href="lxml-source-howto.html#where-to-start">Where to start?</a></li><li class="menu item"><a href="lxml-source-howto.html#lxml-etree">lxml.etree</a></li><li class="menu item"><a href="lxml-source-howto.html#python-modules">Python modules</a></li><li class="menu item"><a href="lxml-source-howto.html#lxml-objectify">lxml.objectify</a></li><li class="menu item"><a href="lxml-source-howto.html#lxml-html">lxml.html</a></li></ul></li></ul><ul class="menu foreign" id="changes 2 2 2-menu"><li class="menu title"><a href="changes-2.2.2.html">Release Changelog</a></li></ul><ul class="menu foreign" id="credits-menu"><li class="menu title"><a href="credits.html">Credits</a><ul class="submenu"><li class="menu item"><a href="credits.html#main-contributors">Main contributors</a></li><li class="menu item"><a href="credits.html#special-thanks-goes-to">Special thanks goes to:</a></li></ul></li></ul></li></ul></div><h1 class="title">Using custom Element classes in lxml</h1>
13
<p>lxml has very sophisticated support for custom Element classes. You
14
can provide your own classes for Elements and have lxml use them by
15
default for all elements generated by a specific parser, only for a
16
specific tag name in a specific namespace or even for an exact element
17
at a specific position in the tree.</p>
16
18
<p>Custom Elements must inherit from the <tt class="docutils literal"><span class="pre">lxml.etree.ElementBase</span></tt> class, which
17
19
provides the Element interface for subclasses:</p>
18
20
<div class="syntax"><pre><span class="gp">>>> </span><span class="k">from</span> <span class="nn">lxml</span> <span class="k">import</span> <span class="n">etree</span>
19
<span class="gp">>>> </span><span class="k">class</span> <span class="nc">HonkElement</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">ElementBase</span><span class="p">):</span>
22
<span class="gp">>>> </span><span class="k">class</span> <span class="nc">honk</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">ElementBase</span><span class="p">):</span>
20
23
<span class="gp">... </span> <span class="k">def</span> <span class="nf">honking</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
21
24
<span class="gp">... </span> <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s">'honking'</span><span class="p">)</span> <span class="o">==</span> <span class="s">'true'</span>
22
25
<span class="gp">... </span> <span class="n">honking</span> <span class="o">=</span> <span class="nb">property</span><span class="p">(</span><span class="n">honking</span><span class="p">)</span>
24
<p>This defines a new Element class <tt class="docutils literal"><span class="pre">HonkElement</span></tt> with a property <tt class="docutils literal"><span class="pre">honking</span></tt>.</p>
25
<p>Note that you cannot (or rather <em>must not</em>) instantiate this class
26
yourself. lxml.etree will do that for you through its normal
27
ElementTree API. All you have to do is tell lxml which class to use
28
for which kind of Element. This is done through a class lookup
29
scheme, as described below.</p>
30
<div class="contents topic">
31
<p class="topic-title first"><a id="contents" name="contents">Contents</a></p>
27
<p>This defines a new Element class <tt class="docutils literal"><span class="pre">honk</span></tt> with a property <tt class="docutils literal"><span class="pre">honking</span></tt>.</p>
28
<p>The following document describes how you can make lxml.etree use these
29
custom Element classes.</p>
30
<div class="contents topic" id="contents">
31
<p class="topic-title first">Contents</p>
32
32
<ul class="simple">
33
<li><a class="reference" href="#element-initialization" id="id2" name="id2">Element initialization</a></li>
34
<li><a class="reference" href="#setting-up-a-class-lookup-scheme" id="id3" name="id3">Setting up a class lookup scheme</a><ul>
35
<li><a class="reference" href="#default-class-lookup" id="id4" name="id4">Default class lookup</a></li>
36
<li><a class="reference" href="#namespace-class-lookup" id="id5" name="id5">Namespace class lookup</a></li>
37
<li><a class="reference" href="#attribute-based-lookup" id="id6" name="id6">Attribute based lookup</a></li>
38
<li><a class="reference" href="#custom-element-class-lookup" id="id7" name="id7">Custom element class lookup</a></li>
39
<li><a class="reference" href="#tree-based-element-class-lookup-in-python" id="id8" name="id8">Tree based element class lookup in Python</a></li>
33
<li><a class="reference internal" href="#background-on-element-proxies" id="id2">Background on Element proxies</a></li>
34
<li><a class="reference internal" href="#element-initialization" id="id3">Element initialization</a></li>
35
<li><a class="reference internal" href="#setting-up-a-class-lookup-scheme" id="id4">Setting up a class lookup scheme</a><ul>
36
<li><a class="reference internal" href="#default-class-lookup" id="id5">Default class lookup</a></li>
37
<li><a class="reference internal" href="#namespace-class-lookup" id="id6">Namespace class lookup</a></li>
38
<li><a class="reference internal" href="#attribute-based-lookup" id="id7">Attribute based lookup</a></li>
39
<li><a class="reference internal" href="#custom-element-class-lookup" id="id8">Custom element class lookup</a></li>
40
<li><a class="reference internal" href="#tree-based-element-class-lookup-in-python" id="id9">Tree based element class lookup in Python</a></li>
42
<li><a class="reference" href="#id1" id="id9" name="id9">Implementing namespaces</a></li>
43
<li><a class="reference internal" href="#generating-xml-with-custom-classes" id="id10">Generating XML with custom classes</a></li>
44
<li><a class="reference internal" href="#id1" id="id11">Implementing namespaces</a></li>
46
<h1><a id="element-initialization" name="element-initialization">Element initialization</a></h1>
47
<p>There is one thing to know up front. Element classes <em>must not</em> have a
48
constructor, neither must there be any internal state (except for the data
49
stored in the underlying XML tree). Element instances are created and garbage
50
collected at need, so there is no way to predict when and how often a
51
constructor would be called. Even worse, when the <tt class="docutils literal"><span class="pre">__init__</span></tt> method is
52
called, the object may not even be initialized yet to represent the XML tag,
53
so there is not much use in providing an <tt class="docutils literal"><span class="pre">__init__</span></tt> method in subclasses.</p>
47
<div class="section" id="background-on-element-proxies">
48
<h1>Background on Element proxies</h1>
49
<p>Being based on libxml2, lxml.etree holds the entire XML tree in a C
50
structure. To communicate with Python code, it creates Python proxy
51
objects for the XML elements on demand.</p>
53
<img alt="proxies.png" src="proxies.png" />
55
<p>The mapping between C elements and Python Element classes is
56
completely configurable. When you ask lxml.etree for an Element by
57
using its API, it will instantiate your classes for you. All you have
58
to do is tell lxml which class to use for which kind of Element. This
59
is done through a class lookup scheme, as described in the sections
62
<div class="section" id="element-initialization">
63
<h1>Element initialization</h1>
64
<p>There is one thing to know up front. Element classes <em>must not</em> have
65
an <tt class="docutils literal"><span class="pre">__init___</span></tt> or <tt class="docutils literal"><span class="pre">__new__</span></tt> method. There should not be any
66
internal state either, except for the data stored in the underlying
67
XML tree. Element instances are created and garbage collected at
68
need, so there is no way to predict when and how often a proxy is
69
created for them. Even worse, when the <tt class="docutils literal"><span class="pre">__init__</span></tt> method is called,
70
the object is not even initialized yet to represent the XML tag, so
71
there is not much use in providing an <tt class="docutils literal"><span class="pre">__init__</span></tt> method in
54
73
<p>Most use cases will not require any class initialisation, so you can content
55
74
yourself with skipping to the next section for now. However, if you really
56
75
need to set up your element class on instantiation, there is one possible way
59
78
children or verify and update attributes.</p>
60
79
<p>The semantics of <tt class="docutils literal"><span class="pre">_init()</span></tt> are as follows:</p>
61
80
<ul class="simple">
62
<li>It is called at least once on element instantiation time. That is, when a
63
Python representation of the element is created by lxml. At that time, the
64
element object is completely initialized to represent a specific XML element
81
<li>It is called once on Element class instantiation time. That is,
82
when a Python representation of the element is created by lxml. At
83
that time, the element object is completely initialized to represent
84
a specific XML element within the tree.</li>
66
85
<li>The method has complete access to the XML tree. Modifications can be done
67
86
in exactly the same way as anywhere else in the program.</li>
68
87
<li>Python representations of elements may be created multiple times during the
69
lifetime of an XML element in the underlying tree. The <tt class="docutils literal"><span class="pre">_init()</span></tt> code
88
lifetime of an XML element in the underlying C tree. The <tt class="docutils literal"><span class="pre">_init()</span></tt> code
70
89
provided by subclasses must take special care by itself that multiple
71
90
executions either are harmless or that they are prevented by some kind of
72
91
flag in the XML tree. The latter can be achieved by modifying an attribute
77
96
you write here as its exceptions may turn up in various unexpected places.</li>
81
<h1><a id="setting-up-a-class-lookup-scheme" name="setting-up-a-class-lookup-scheme">Setting up a class lookup scheme</a></h1>
99
<div class="section" id="setting-up-a-class-lookup-scheme">
100
<h1>Setting up a class lookup scheme</h1>
82
101
<p>The first thing to do when deploying custom element classes is to register a
83
102
class lookup scheme on a parser. lxml.etree provides quite a number of
84
different schemes, that also support class lookup based on namespaces or
103
different schemes that also support class lookup based on namespaces or
85
104
attribute values. Most lookups support fallback chaining, which allows the
86
105
next lookup mechanism to take over when the previous one fails to find a
88
<p>For example, setting a different default element class for a parser works as
90
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">parser_lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementDefaultClassLookup</span><span class="p">(</span><span class="n">element</span><span class="o">=</span><span class="n">HonkElement</span><span class="p">)</span>
107
<p>For example, setting the <tt class="docutils literal"><span class="pre">honk</span></tt> Element as a default element class
108
for a parser works as follows:</p>
109
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">parser_lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementDefaultClassLookup</span><span class="p">(</span><span class="n">element</span><span class="o">=</span><span class="n">honk</span><span class="p">)</span>
91
110
<span class="gp">>>> </span><span class="n">parser</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XMLParser</span><span class="p">()</span>
92
111
<span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">set_element_class_lookup</span><span class="p">(</span><span class="n">parser_lookup</span><span class="p">)</span>
95
114
does not know about your specialised parser and creates a new document that
96
115
deploys the default parser:</p>
97
116
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">el</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">Element</span><span class="p">(</span><span class="s">"root"</span><span class="p">)</span>
98
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
117
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
99
118
<span class="go">False</span>
101
<p>You should therefore avoid using this function in code that uses custom
102
classes. The <tt class="docutils literal"><span class="pre">makeelement()</span></tt> method of parsers provides a simple
120
<p>You should therefore avoid using this factory function in code that
121
uses custom classes. The <tt class="docutils literal"><span class="pre">makeelement()</span></tt> method of parsers provides
122
a simple replacement:</p>
104
123
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">el</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">makeelement</span><span class="p">(</span><span class="s">"root"</span><span class="p">)</span>
105
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
124
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
106
125
<span class="go">True</span>
108
127
<p>If you use a parser at the module level, you can easily redirect a module
109
128
level <tt class="docutils literal"><span class="pre">Element()</span></tt> factory to the parser method by adding code like this:</p>
110
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">MODULE_PARSER</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XMLParser</span><span class="p">()</span>
111
<span class="gp">>>> </span><span class="n">Element</span> <span class="o">=</span> <span class="n">MODULE_PARSER</span><span class="o">.</span><span class="n">makeelement</span>
129
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">module_level_parser</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XMLParser</span><span class="p">()</span>
130
<span class="gp">>>> </span><span class="n">Element</span> <span class="o">=</span> <span class="n">module_level_parser</span><span class="o">.</span><span class="n">makeelement</span>
113
132
<p>While the <tt class="docutils literal"><span class="pre">XML()</span></tt> and <tt class="docutils literal"><span class="pre">HTML()</span></tt> factories also depend on the default
114
133
parser, you can pass them a different parser as second argument:</p>
115
134
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">element</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XML</span><span class="p">(</span><span class="s">"<test/>"</span><span class="p">)</span>
116
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
135
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
117
136
<span class="go">False</span>
119
138
<span class="gp">>>> </span><span class="n">element</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XML</span><span class="p">(</span><span class="s">"<test/>"</span><span class="p">,</span> <span class="n">parser</span><span class="p">)</span>
120
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
139
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
121
140
<span class="go">True</span>
123
142
<p>Whenever you create a document with a parser, it will inherit the lookup
124
143
scheme and all subsequent element instantiations for this document will use
126
145
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">element</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">fromstring</span><span class="p">(</span><span class="s">"<test/>"</span><span class="p">,</span> <span class="n">parser</span><span class="p">)</span>
127
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
146
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
128
147
<span class="go">True</span>
129
148
<span class="gp">>>> </span><span class="n">el</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">SubElement</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="s">"subel"</span><span class="p">)</span>
130
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
149
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
131
150
<span class="go">True</span>
133
<p>For small projects, you may also consider setting a lookup scheme on the
134
default parser. To avoid interfering with other modules, however, it is
135
usually a better idea to use a dedicated parser for each module (or a parser
136
pool when using threads) and then register the required lookup scheme only for
152
<p>For testing code in the Python interpreter and for small projects, you
153
may also consider setting a lookup scheme on the default parser. To
154
avoid interfering with other modules, however, it is usually a better
155
idea to use a dedicated parser for each module (or a parser pool when
156
using threads) and then register the required lookup scheme only for
138
<div class="section">
139
<h2><a id="default-class-lookup" name="default-class-lookup">Default class lookup</a></h2>
158
<div class="section" id="default-class-lookup">
159
<h2>Default class lookup</h2>
140
160
<p>This is the most simple lookup mechanism. It always returns the default
141
161
element class. Consequently, no further fallbacks are supported, but this
142
scheme is a good fallback for other custom lookup mechanisms.</p>
162
scheme is a nice fallback for other custom lookup mechanisms.</p>
144
164
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementDefaultClassLookup</span><span class="p">()</span>
145
165
<span class="gp">>>> </span><span class="n">parser</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XMLParser</span><span class="p">()</span>
151
171
the constructor. While it accepts classes for <tt class="docutils literal"><span class="pre">element</span></tt>, <tt class="docutils literal"><span class="pre">comment</span></tt> and
152
172
<tt class="docutils literal"><span class="pre">pi</span></tt> nodes, most use cases will only override the element class:</p>
153
173
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">el</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">makeelement</span><span class="p">(</span><span class="s">"myelement"</span><span class="p">)</span>
154
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
174
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
155
175
<span class="go">False</span>
157
<span class="gp">>>> </span><span class="n">lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementDefaultClassLookup</span><span class="p">(</span><span class="n">element</span><span class="o">=</span><span class="n">HonkElement</span><span class="p">)</span>
177
<span class="gp">>>> </span><span class="n">lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementDefaultClassLookup</span><span class="p">(</span><span class="n">element</span><span class="o">=</span><span class="n">honk</span><span class="p">)</span>
158
178
<span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">set_element_class_lookup</span><span class="p">(</span><span class="n">lookup</span><span class="p">)</span>
160
180
<span class="gp">>>> </span><span class="n">el</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">makeelement</span><span class="p">(</span><span class="s">"myelement"</span><span class="p">)</span>
161
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">HonkElement</span><span class="p">))</span>
181
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">honk</span><span class="p">))</span>
162
182
<span class="go">True</span>
163
183
<span class="gp">>>> </span><span class="n">el</span><span class="o">.</span><span class="n">honking</span>
164
184
<span class="go">False</span>
169
189
<span class="go">True</span>
172
<div class="section">
173
<h2><a id="namespace-class-lookup" name="namespace-class-lookup">Namespace class lookup</a></h2>
192
<div class="section" id="namespace-class-lookup">
193
<h2>Namespace class lookup</h2>
174
194
<p>This is an advanced lookup mechanism that supports namespace/tag-name specific
175
195
element classes. You can select it by calling:</p>
176
196
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementNamespaceClassLookup</span><span class="p">()</span>
177
197
<span class="gp">>>> </span><span class="n">parser</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XMLParser</span><span class="p">()</span>
178
198
<span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">set_element_class_lookup</span><span class="p">(</span><span class="n">lookup</span><span class="p">)</span>
180
<p>See the separate section on <a class="reference" href="#implementing-namespaces">implementing namespaces</a> below to learn how to
200
<p>See the separate section on <a class="reference external" href="#implementing-namespaces">implementing namespaces</a> below to learn how to
181
201
make use of it.</p>
182
202
<p>This scheme supports a fallback mechanism that is used in the case where the
183
203
namespace is not found or no class was registered for the element name.
184
204
Normally, the default class lookup is used here. To change it, pass the
185
205
desired fallback lookup scheme to the constructor:</p>
186
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">fallback</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementDefaultClassLookup</span><span class="p">(</span><span class="n">element</span><span class="o">=</span><span class="n">HonkElement</span><span class="p">)</span>
206
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">fallback</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementDefaultClassLookup</span><span class="p">(</span><span class="n">element</span><span class="o">=</span><span class="n">honk</span><span class="p">)</span>
187
207
<span class="gp">>>> </span><span class="n">lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementNamespaceClassLookup</span><span class="p">(</span><span class="n">fallback</span><span class="p">)</span>
188
208
<span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">set_element_class_lookup</span><span class="p">(</span><span class="n">lookup</span><span class="p">)</span>
191
<div class="section">
192
<h2><a id="attribute-based-lookup" name="attribute-based-lookup">Attribute based lookup</a></h2>
211
<div class="section" id="attribute-based-lookup">
212
<h2>Attribute based lookup</h2>
193
213
<p>This scheme uses a mapping from attribute values to classes. An attribute
194
214
name is set at initialisation time and is then used to find the corresponding
195
value. It is set up as follows:</p>
196
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">id_class_mapping</span> <span class="o">=</span> <span class="p">{}</span> <span class="c"># maps attribute values to element classes</span>
215
value in a dictionary. It is set up as follows:</p>
216
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">id_class_mapping</span> <span class="o">=</span> <span class="p">{</span><span class="s">'1234'</span> <span class="p">:</span> <span class="n">honk</span><span class="p">}</span> <span class="c"># maps attribute values to classes</span>
197
218
<span class="gp">>>> </span><span class="n">lookup</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">AttributeBasedElementClassLookup</span><span class="p">(</span>
198
219
<span class="gp">... </span> <span class="s">'id'</span><span class="p">,</span> <span class="n">id_class_mapping</span><span class="p">)</span>
199
220
<span class="gp">>>> </span><span class="n">parser</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XMLParser</span><span class="p">()</span>
200
221
<span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">set_element_class_lookup</span><span class="p">(</span><span class="n">lookup</span><span class="p">)</span>
202
<p>Instead of a global setup of this scheme, you should consider using a
203
per-parser setup.</p>
204
223
<p>This class uses its fallback if the attribute is not found or its value is not
205
224
in the mapping. Normally, the default class lookup is used here. If you want
206
225
to use the namespace lookup, for example, you can use this code:</p>
211
230
<span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">set_element_class_lookup</span><span class="p">(</span><span class="n">lookup</span><span class="p">)</span>
214
<div class="section">
215
<h2><a id="custom-element-class-lookup" name="custom-element-class-lookup">Custom element class lookup</a></h2>
233
<div class="section" id="custom-element-class-lookup">
234
<h2>Custom element class lookup</h2>
216
235
<p>This is the most customisable way of finding element classes on a per-element
217
236
basis. It allows you to implement a custom lookup scheme in a subclass:</p>
218
237
<div class="syntax"><pre><span class="gp">>>> </span><span class="k">class</span> <span class="nc">MyLookup</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">CustomElementClassLookup</span><span class="p">):</span>
219
238
<span class="gp">... </span> <span class="k">def</span> <span class="nf">lookup</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">node_type</span><span class="p">,</span> <span class="n">document</span><span class="p">,</span> <span class="n">namespace</span><span class="p">,</span> <span class="n">name</span><span class="p">):</span>
220
<span class="gp">... </span> <span class="k">return</span> <span class="n">MyElementClass</span> <span class="c"># defined elsewhere</span>
239
<span class="gp">... </span> <span class="k">return</span> <span class="n">honk</span> <span class="c"># be a bit more selective here ...</span>
222
241
<span class="gp">>>> </span><span class="n">parser</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XMLParser</span><span class="p">()</span>
223
242
<span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">set_element_class_lookup</span><span class="p">(</span><span class="n">MyLookup</span><span class="p">())</span>
225
<p>The <tt class="docutils literal"><span class="pre">lookup()</span></tt> method must either return None (which triggers the fallback
226
mechanism) or a subclass of <tt class="docutils literal"><span class="pre">lxml.etree.ElementBase</span></tt>. It can take any
227
decision it wants based on the node type (one of "element", "comment", "PI"),
228
the XML document of the element, or its namespace or tag name.</p>
229
<p>Instead of a global setup of this scheme, you should consider using a
230
per-parser setup.</p>
244
<p>The <tt class="docutils literal"><span class="pre">.lookup()</span></tt> method must return either None (which triggers the
245
fallback mechanism) or a subclass of <tt class="docutils literal"><span class="pre">lxml.etree.ElementBase</span></tt>. It
246
can take any decision it wants based on the node type (one of
247
"element", "comment", "PI", "entity"), the XML document of the
248
element, or its namespace or tag name.</p>
232
<div class="section">
233
<h2><a id="tree-based-element-class-lookup-in-python" name="tree-based-element-class-lookup-in-python">Tree based element class lookup in Python</a></h2>
234
<p>Taking more elaborate decisions than allowed by the custom scheme is difficult
235
to achieve in pure Python. It would require access to the tree - before the
236
elements in the tree have been instantiated as Python Element objects.</p>
250
<div class="section" id="tree-based-element-class-lookup-in-python">
251
<h2>Tree based element class lookup in Python</h2>
252
<p>Taking more elaborate decisions than allowed by the custom scheme is
253
difficult to achieve in pure Python, as it results in a
254
chicken-and-egg problem. It would require access to the tree - before
255
the elements in the tree have been instantiated as Python Element
237
257
<p>Luckily, there is a way to do this. The <tt class="docutils literal"><span class="pre">PythonElementClassLookup</span></tt>
238
258
works similar to the custom lookup scheme:</p>
239
259
<div class="syntax"><pre><span class="gp">>>> </span><span class="k">class</span> <span class="nc">MyLookup</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">PythonElementClassLookup</span><span class="p">):</span>
260
280
provide. Once you have taken the decision which class to use for this
261
281
element, you can simply return it and have lxml take care of cleaning up the
262
282
instantiated proxy classes.</p>
263
<p>Note: this lookup scheme originally lived in a separate module called
283
<p>Sidenote: this lookup scheme originally lived in a separate module called
264
284
<tt class="docutils literal"><span class="pre">lxml.pyclasslookup</span></tt>.</p>
267
<div class="section">
268
<h1><a id="id1" name="id1">Implementing namespaces</a></h1>
287
<div class="section" id="generating-xml-with-custom-classes">
288
<h1>Generating XML with custom classes</h1>
289
<p>Up to lxml 2.1, you could not instantiate proxy classes yourself.
290
Only lxml.etree could do that when creating an object representation
291
of an existing XML element. Since lxml 2.2, however, instantiating
292
this class will simply create a new Element:</p>
293
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">el</span> <span class="o">=</span> <span class="n">honk</span><span class="p">(</span><span class="n">honking</span> <span class="o">=</span> <span class="s">'true'</span><span class="p">)</span>
294
<span class="gp">>>> </span><span class="n">el</span><span class="o">.</span><span class="n">tag</span>
295
<span class="go">'honk'</span>
296
<span class="gp">>>> </span><span class="n">el</span><span class="o">.</span><span class="n">honking</span>
297
<span class="go">True</span>
299
<p>Note, however, that the proxy you create here will be garbage
300
collected just like any other proxy. You can therefore not count on
301
lxml.etree using the same class that you instantiated when you access
302
this Element a second time after letting its reference go. You should
303
therefore always use a corresponding class lookup scheme that returns
304
your Element proxy classes for the elements that they create. The
305
<tt class="docutils literal"><span class="pre">ElementNamespaceClassLookup</span></tt> is generally a good match.</p>
306
<p>You can use custom Element classes to quickly create XML fragments:</p>
307
<div class="syntax"><pre><span class="gp">>>> </span><span class="k">class</span> <span class="nc">hale</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">ElementBase</span><span class="p">):</span> <span class="k">pass</span>
308
<span class="gp">>>> </span><span class="k">class</span> <span class="nc">bopp</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">ElementBase</span><span class="p">):</span> <span class="k">pass</span>
310
<span class="gp">>>> </span><span class="n">el</span> <span class="o">=</span> <span class="n">hale</span><span class="p">(</span> <span class="s">"some "</span><span class="p">,</span> <span class="n">honk</span><span class="p">(</span><span class="n">honking</span> <span class="o">=</span> <span class="s">'true'</span><span class="p">),</span> <span class="n">bopp</span><span class="p">,</span> <span class="s">" text"</span> <span class="p">)</span>
312
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">el</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="nb">unicode</span><span class="p">))</span>
313
<span class="go"><hale>some <honk honking="true"/><bopp/> text</hale></span>
316
<div class="section" id="id1">
317
<h1>Implementing namespaces</h1>
269
318
<p>lxml allows you to implement namespaces, in a rather literal sense. After
270
319
setting up the namespace class lookup mechanism as described above, you can
271
320
build a new element namespace (or retrieve an existing one) by calling the
293
347
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">honk_element</span><span class="o">.</span><span class="n">honking</span><span class="p">)</span>
294
348
<span class="go">True</span>
296
<p>Essentially, what this allows you to do, is to give elements a custom API
350
<p>Essentially, what this allows you to do, is to give Elements a custom API
297
351
based on their namespace and tag name.</p>
298
<p>A somewhat related topic are <a class="reference" href="extensions.html">extension functions</a> which use a similar
352
<p>A somewhat related topic are <a class="reference external" href="extensions.html">extension functions</a> which use a similar
299
353
mechanism for registering extension functions in XPath and XSLT.</p>
300
<p>In the setup example above, we associated the HonkElement class only with the
301
'honk' element. If an XML tree contains different elements in the same
302
namespace, they do not pick up the same implementation:</p>
354
<p>In the setup example above, we associated the <tt class="docutils literal"><span class="pre">honk</span></tt> Element class
355
only with the 'honk' element. If an XML tree contains different
356
elements in the same namespace, they do not pick up the same
303
358
<div class="syntax"><pre><span class="gp">>>> </span><span class="n">xml</span> <span class="o">=</span> <span class="s">'<honk xmlns="http://hui.de/honk" honking="true"><bla/></honk>'</span>
304
359
<span class="gp">>>> </span><span class="n">honk_element</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">XML</span><span class="p">(</span><span class="n">xml</span><span class="p">,</span> <span class="n">parser</span><span class="p">)</span>
305
360
<span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">honk_element</span><span class="o">.</span><span class="n">honking</span><span class="p">)</span>
321
376
<div class="syntax"><pre><span class="gp">>>> </span><span class="k">class</span> <span class="nc">HonkNSElement</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">ElementBase</span><span class="p">):</span>
322
377
<span class="gp">... </span> <span class="k">def</span> <span class="nf">honk</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
323
378
<span class="gp">... </span> <span class="k">return</span> <span class="s">"HONK"</span>
324
<span class="gp">>>> </span><span class="n">namespace</span><span class="p">[</span><span class="bp">None</span><span class="p">]</span> <span class="o">=</span> <span class="n">HonkNSElement</span>
379
<span class="gp">>>> </span><span class="n">namespace</span><span class="p">[</span><span class="bp">None</span><span class="p">]</span> <span class="o">=</span> <span class="n">HonkNSElement</span> <span class="c"># default Element for namespace</span>
326
381
<span class="gp">>>> </span><span class="k">class</span> <span class="nc">HonkElement</span><span class="p">(</span><span class="n">HonkNSElement</span><span class="p">):</span>
327
382
<span class="gp">... </span> <span class="k">def</span> <span class="nf">honking</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
328
383
<span class="gp">... </span> <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s">'honking'</span><span class="p">)</span> <span class="o">==</span> <span class="s">'true'</span>
329
384
<span class="gp">... </span> <span class="n">honking</span> <span class="o">=</span> <span class="nb">property</span><span class="p">(</span><span class="n">honking</span><span class="p">)</span>
330
<span class="gp">>>> </span><span class="n">namespace</span><span class="p">[</span><span class="s">'honk'</span><span class="p">]</span> <span class="o">=</span> <span class="n">HonkElement</span>
385
<span class="gp">>>> </span><span class="n">namespace</span><span class="p">[</span><span class="s">'honk'</span><span class="p">]</span> <span class="o">=</span> <span class="n">HonkElement</span> <span class="c"># Element for specific tag</span>
332
387
<p>Now you can rely on lxml to always return objects of type HonkNSElement or its
333
388
subclasses for elements of this namespace:</p>