1 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
2 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
4 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
5 <meta name="generator" content="Docutils 0.14: http://docutils.sourceforge.net/" />
6 <title>Parsing XML and HTML with lxml</title>
7 <link rel="stylesheet" href="style.css" type="text/css" />
8 <script type="text/javascript">
9 function trigger_menu(event) {
10 var sidemenu = document.getElementById("sidemenu");
11 var classes = sidemenu.getAttribute("class");
12 classes = (classes.indexOf(" visible") === -1) ? classes + " visible" : classes.replace(" visible", "");
13 sidemenu.setAttribute("class", classes);
14 event.preventDefault();
15 event.stopPropagation();
17 function hide_menu() {
18 var sidemenu = document.getElementById("sidemenu");
19 var classes = sidemenu.getAttribute("class");
20 if (classes.indexOf(" visible") !== -1) {
21 sidemenu.setAttribute("class", classes.replace(" visible", ""));
24 </script><meta content="width=device-width, initial-scale=1" name="viewport" /></head>
25 <body onclick="hide_menu()">
26 <div class="document" id="parsing-xml-and-html-with-lxml">
27 <div class="sidemenu" id="sidemenu"><div class="menutrigger" onclick="trigger_menu(event)">Menu</div><div class="menu"><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#support-the-project">Support the project</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><li class="menu item"><a href="index.html#legal-notice-for-donations">Legal Notice for Donations</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#where-to-get-it">Where to get it</a></li><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-dev-sources">Building lxml from dev sources</a></li><li class="menu item"><a href="installation.html#using-lxml-with-python-libxml2">Using lxml with python-libxml2</a></li><li class="menu item"><a href="installation.html#source-builds-on-ms-windows">Source builds on MS Windows</a></li><li class="menu item"><a href="installation.html#source-builds-on-macos-x">Source builds on MacOS-X</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#id1">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#incremental-xml-generation">Incremental XML generation</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></ul></li></ul><ul class="menu current" 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#incremental-event-parsing">Incremental event parsing</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#id1">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#id2">Schematron</a></li><li class="menu item"><a href="validation.html#id3">(Pre-ISO-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#the-cssselect-method">The cssselect method</a></li><li class="menu item"><a href="cssselect.html#supported-selectors">Supported Selectors</a></li><li class="menu item"><a href="cssselect.html#namespaces">Namespaces</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><li class="menu item"><a href="elementsoup.html#using-only-the-encoding-detection">Using only the encoding detection</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#xml-catalogs">XML Catalogs</a></li><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#id1">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#passing-generated-trees-through-python">Passing generated trees through Python</a></li><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#github-git-and-hg">Github, git and hg</a></li><li class="menu item"><a href="build.html#building-the-sources">Building the sources</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-or-wheel">Building an egg or wheel</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 4 4 3-menu"><li class="menu title"><a href="changes-4.4.3.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><li><a href="/sitemap.html">Sitemap</a></li></ul></div></div><h1 class="title">Parsing XML and HTML with lxml</h1>
29 <p>lxml provides a very simple and powerful API for parsing XML and HTML. It
30 supports one-step parsing as well as step-by-step parsing using an
31 event-driven API (currently only for XML).</p>
32 <div class="contents topic" id="contents">
33 <p class="topic-title first">Contents</p>
35 <li><a class="reference internal" href="#parsers" id="id1">Parsers</a><ul>
36 <li><a class="reference internal" href="#parser-options" id="id2">Parser options</a></li>
37 <li><a class="reference internal" href="#error-log" id="id3">Error log</a></li>
38 <li><a class="reference internal" href="#parsing-html" id="id4">Parsing HTML</a></li>
39 <li><a class="reference internal" href="#doctype-information" id="id5">Doctype information</a></li>
42 <li><a class="reference internal" href="#the-target-parser-interface" id="id6">The target parser interface</a></li>
43 <li><a class="reference internal" href="#the-feed-parser-interface" id="id7">The feed parser interface</a></li>
44 <li><a class="reference internal" href="#incremental-event-parsing" id="id8">Incremental event parsing</a><ul>
45 <li><a class="reference internal" href="#event-types" id="id9">Event types</a></li>
46 <li><a class="reference internal" href="#modifying-the-tree" id="id10">Modifying the tree</a></li>
47 <li><a class="reference internal" href="#selective-tag-events" id="id11">Selective tag events</a></li>
48 <li><a class="reference internal" href="#comments-and-pis" id="id12">Comments and PIs</a></li>
49 <li><a class="reference internal" href="#events-with-custom-targets" id="id13">Events with custom targets</a></li>
52 <li><a class="reference internal" href="#iterparse-and-iterwalk" id="id14">iterparse and iterwalk</a><ul>
53 <li><a class="reference internal" href="#iterwalk" id="id15">iterwalk</a></li>
56 <li><a class="reference internal" href="#python-unicode-strings" id="id16">Python unicode strings</a><ul>
57 <li><a class="reference internal" href="#serialising-to-unicode-strings" id="id17">Serialising to Unicode strings</a></li>
62 <p>The usual setup procedure:</p>
63 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">lxml</span> <span class="kn">import</span> <span class="n">etree</span>
65 <p>The following examples also use StringIO or BytesIO to show how to parse
66 from files and file-like objects. Both are available in the <tt class="docutils literal">io</tt> module:</p>
67 <div class="syntax"><pre><span></span><span class="kn">from</span> <span class="nn">io</span> <span class="kn">import</span> <span class="n">StringIO</span><span class="p">,</span> <span class="n">BytesIO</span>
69 <div class="section" id="parsers">
71 <p>Parsers are represented by parser objects. There is support for parsing both
72 XML and (broken) HTML. Note that XHTML is best parsed as XML, parsing it with
73 the HTML parser can lead to unexpected results. Here is a simple example for
74 parsing XML from an in-memory string:</p>
75 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">xml</span> <span class="o">=</span> <span class="s1">'<a xmlns="test"><b xmlns="test"/></a>'</span>
77 <span class="gp">>>> </span><span class="n">root</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="n">xml</span><span class="p">)</span>
78 <span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
79 <span class="go">b'<a xmlns="test"><b xmlns="test"/></a>'</span>
81 <p>To read from a file or file-like object, you can use the <tt class="docutils literal">parse()</tt> function,
82 which returns an <tt class="docutils literal">ElementTree</tt> object:</p>
83 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">StringIO</span><span class="p">(</span><span class="n">xml</span><span class="p">))</span>
84 <span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">tree</span><span class="o">.</span><span class="n">getroot</span><span class="p">())</span>
85 <span class="go">b'<a xmlns="test"><b xmlns="test"/></a>'</span>
87 <p>Note how the <tt class="docutils literal">parse()</tt> function reads from a file-like object here. If
88 parsing is done from a real file, it is more common (and also somewhat more
89 efficient) to pass a filename:</p>
90 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s2">"doc/test.xml"</span><span class="p">)</span>
92 <p>lxml can parse from a local file, an HTTP URL or an FTP URL. It also
93 auto-detects and reads gzip-compressed XML files (.gz).</p>
94 <p>If you want to parse from memory and still provide a base URL for the document
95 (e.g. to support relative paths in an XInclude), you can pass the <tt class="docutils literal">base_url</tt>
97 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">root</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="n">xml</span><span class="p">,</span> <span class="n">base_url</span><span class="o">=</span><span class="s2">"http://where.it/is/from.xml"</span><span class="p">)</span>
99 <div class="section" id="parser-options">
100 <h2>Parser options</h2>
101 <p>The parsers accept a number of setup options as keyword arguments. The above
102 example is easily extended to clean up namespaces during parsing:</p>
103 <div class="syntax"><pre><span></span><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><span class="n">ns_clean</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
104 <span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">StringIO</span><span class="p">(</span><span class="n">xml</span><span class="p">),</span> <span class="n">parser</span><span class="p">)</span>
105 <span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">tree</span><span class="o">.</span><span class="n">getroot</span><span class="p">())</span>
106 <span class="go">b'<a xmlns="test"><b/></a>'</span>
108 <p>The keyword arguments in the constructor are mainly based on the libxml2
109 parser configuration. A DTD will also be loaded if validation or attribute
110 default values are requested.</p>
111 <p>Available boolean keyword arguments:</p>
113 <li>attribute_defaults - read the DTD (if referenced by the document) and add
114 the default attributes from it</li>
115 <li>dtd_validation - validate while parsing (if a DTD was referenced)</li>
116 <li>load_dtd - load and parse the DTD while parsing (no validation is performed)</li>
117 <li>no_network - prevent network access when looking up external
118 documents (on by default)</li>
119 <li>ns_clean - try to clean up redundant namespace declarations</li>
120 <li>recover - try hard to parse through broken XML</li>
121 <li>remove_blank_text - discard blank text nodes between tags, also known as
122 ignorable whitespace. This is best used together with a DTD or schema
123 (which tells data and noise apart), otherwise a heuristic will be applied.</li>
124 <li>remove_comments - discard comments</li>
125 <li>remove_pis - discard processing instructions</li>
126 <li>strip_cdata - replace CDATA sections by normal text content (on by
128 <li>resolve_entities - replace entities by their text value (on by
130 <li>huge_tree - disable security restrictions and support very deep trees
131 and very long text content (only affects libxml2 2.7+)</li>
132 <li>compact - use compact storage for short text content (on by default)</li>
133 <li>collect_ids - collect XML IDs in a hash table while parsing (on by default).
134 Disabling this can substantially speed up parsing of documents with many
135 different IDs if the hash lookup is not used afterwards.</li>
137 <p>Other keyword arguments:</p>
139 <li>encoding - override the document encoding</li>
140 <li>target - a parser target object that will receive the parse events
141 (see <a class="reference internal" href="#the-target-parser-interface">The target parser interface</a>)</li>
142 <li>schema - an XMLSchema to validate against (see <a class="reference external" href="validation.html#xmlschema">validation</a>)</li>
145 <div class="section" id="error-log">
147 <p>Parsers have an <tt class="docutils literal">error_log</tt> property that lists the errors and
148 warnings of the last parser run:</p>
149 <div class="syntax"><pre><span></span><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>
150 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">parser</span><span class="o">.</span><span class="n">error_log</span><span class="p">))</span>
151 <span class="go">0</span>
153 <span class="gp">>>> </span><span class="n">tree</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="s2">"<root></span><span class="se">\n</span><span class="s2"></b>"</span><span class="p">,</span> <span class="n">parser</span><span class="p">)</span> <span class="c1"># doctest: +ELLIPSIS</span>
154 <span class="gt">Traceback (most recent call last):</span>
155 <span class="c">...</span>
156 <span class="gr">lxml.etree.XMLSyntaxError</span>: <span class="n">Opening and ending tag mismatch: root line 1 and b, line 2, column 5...</span>
158 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">parser</span><span class="o">.</span><span class="n">error_log</span><span class="p">))</span>
159 <span class="go">1</span>
161 <span class="gp">>>> </span><span class="n">error</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">error_log</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
162 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">error</span><span class="o">.</span><span class="n">message</span><span class="p">)</span>
163 <span class="go">Opening and ending tag mismatch: root line 1 and b</span>
164 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">error</span><span class="o">.</span><span class="n">line</span><span class="p">)</span>
165 <span class="go">2</span>
166 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">error</span><span class="o">.</span><span class="n">column</span><span class="p">)</span>
167 <span class="go">5</span>
169 <p>Each entry in the log has the following properties:</p>
171 <li><tt class="docutils literal">message</tt>: the message text</li>
172 <li><tt class="docutils literal">domain</tt>: the domain ID (see the lxml.etree.ErrorDomains class)</li>
173 <li><tt class="docutils literal">type</tt>: the message type ID (see the lxml.etree.ErrorTypes class)</li>
174 <li><tt class="docutils literal">level</tt>: the log level ID (see the lxml.etree.ErrorLevels class)</li>
175 <li><tt class="docutils literal">line</tt>: the line at which the message originated (if applicable)</li>
176 <li><tt class="docutils literal">column</tt>: the character column at which the message originated (if applicable)</li>
177 <li><tt class="docutils literal">filename</tt>: the name of the file in which the message originated (if applicable)</li>
179 <p>For convenience, there are also three properties that provide readable
180 names for the ID values:</p>
182 <li><tt class="docutils literal">domain_name</tt></li>
183 <li><tt class="docutils literal">type_name</tt></li>
184 <li><tt class="docutils literal">level_name</tt></li>
186 <p>To filter for a specific kind of message, use the different
187 <tt class="docutils literal"><span class="pre">filter_*()</span></tt> methods on the error log (see the
188 lxml.etree._ListErrorLog class).</p>
190 <div class="section" id="parsing-html">
191 <h2>Parsing HTML</h2>
192 <p>HTML parsing is similarly simple. The parsers have a <tt class="docutils literal">recover</tt>
193 keyword argument that the HTMLParser sets by default. It lets libxml2
194 try its best to return a valid HTML tree with all content it can
195 manage to parse. It will not raise an exception on parser errors.
196 You should use libxml2 version 2.6.21 or newer to take advantage of
198 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">broken_html</span> <span class="o">=</span> <span class="s2">"<html><head><title>test<body><h1>page title</h3>"</span>
200 <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">HTMLParser</span><span class="p">()</span>
201 <span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">StringIO</span><span class="p">(</span><span class="n">broken_html</span><span class="p">),</span> <span class="n">parser</span><span class="p">)</span>
203 <span class="gp">>>> </span><span class="n">result</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">tree</span><span class="o">.</span><span class="n">getroot</span><span class="p">(),</span>
204 <span class="gp">... </span> <span class="n">pretty_print</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">method</span><span class="o">=</span><span class="s2">"html"</span><span class="p">)</span>
205 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
206 <span class="go"><html></span>
207 <span class="go"> <head></span>
208 <span class="go"> <title>test</title></span>
209 <span class="go"> </head></span>
210 <span class="go"> <body></span>
211 <span class="go"> <h1>page title</h1></span>
212 <span class="go"> </body></span>
213 <span class="go"></html></span>
215 <p>Lxml has an HTML function, similar to the XML shortcut known from
217 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">html</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">HTML</span><span class="p">(</span><span class="n">broken_html</span><span class="p">)</span>
218 <span class="gp">>>> </span><span class="n">result</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">html</span><span class="p">,</span> <span class="n">pretty_print</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">method</span><span class="o">=</span><span class="s2">"html"</span><span class="p">)</span>
219 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
220 <span class="go"><html></span>
221 <span class="go"> <head></span>
222 <span class="go"> <title>test</title></span>
223 <span class="go"> </head></span>
224 <span class="go"> <body></span>
225 <span class="go"> <h1>page title</h1></span>
226 <span class="go"> </body></span>
227 <span class="go"></html></span>
229 <p>The support for parsing broken HTML depends entirely on libxml2's recovery
230 algorithm. It is <em>not</em> the fault of lxml if you find documents that are so
231 heavily broken that the parser cannot handle them. There is also no guarantee
232 that the resulting tree will contain all data from the original document. The
233 parser may have to drop seriously broken parts when struggling to keep
234 parsing. Especially misplaced meta tags can suffer from this, which may lead
235 to encoding problems.</p>
236 <p>Note that the result is a valid HTML tree, but it may not be a
237 well-formed XML tree. For example, XML forbids double hyphens in
238 comments, which the HTML parser will happily accept in recovery mode.
239 Therefore, if your goal is to serialise an HTML document as an
240 XML/XHTML document after parsing, you may have to apply some manual
241 preprocessing first.</p>
242 <p>Also note that the HTML parser is meant to parse HTML documents. For
243 XHTML documents, use the XML parser, which is namespace aware.</p>
245 <div class="section" id="doctype-information">
246 <h2>Doctype information</h2>
247 <p>The use of the libxml2 parsers makes some additional information available at
248 the API level. Currently, ElementTree objects can access the DOCTYPE
249 information provided by a parsed document, as well as the XML version and the
250 original encoding. Since lxml 3.5, the doctype references are mutable.</p>
251 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">pub_id</span> <span class="o">=</span> <span class="s2">"-//W3C//DTD XHTML 1.0 Transitional//EN"</span>
252 <span class="gp">>>> </span><span class="n">sys_url</span> <span class="o">=</span> <span class="s2">"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"</span>
253 <span class="gp">>>> </span><span class="n">doctype_string</span> <span class="o">=</span> <span class="s1">'<!DOCTYPE html PUBLIC "</span><span class="si">%s</span><span class="s1">" "</span><span class="si">%s</span><span class="s1">">'</span> <span class="o">%</span> <span class="p">(</span><span class="n">pub_id</span><span class="p">,</span> <span class="n">sys_url</span><span class="p">)</span>
254 <span class="gp">>>> </span><span class="n">xml_header</span> <span class="o">=</span> <span class="s1">'<?xml version="1.0" encoding="ascii"?>'</span>
255 <span class="gp">>>> </span><span class="n">xhtml</span> <span class="o">=</span> <span class="n">xml_header</span> <span class="o">+</span> <span class="n">doctype_string</span> <span class="o">+</span> <span class="s1">'<html><body></body></html>'</span>
257 <span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">StringIO</span><span class="p">(</span><span class="n">xhtml</span><span class="p">))</span>
258 <span class="gp">>>> </span><span class="n">docinfo</span> <span class="o">=</span> <span class="n">tree</span><span class="o">.</span><span class="n">docinfo</span>
259 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">docinfo</span><span class="o">.</span><span class="n">public_id</span><span class="p">)</span>
260 <span class="go">-//W3C//DTD XHTML 1.0 Transitional//EN</span>
261 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">docinfo</span><span class="o">.</span><span class="n">system_url</span><span class="p">)</span>
262 <span class="go">http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd</span>
263 <span class="gp">>>> </span><span class="n">docinfo</span><span class="o">.</span><span class="n">doctype</span> <span class="o">==</span> <span class="n">doctype_string</span>
264 <span class="go">True</span>
266 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">docinfo</span><span class="o">.</span><span class="n">xml_version</span><span class="p">)</span>
267 <span class="go">1.0</span>
268 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">docinfo</span><span class="o">.</span><span class="n">encoding</span><span class="p">)</span>
269 <span class="go">ascii</span>
271 <span class="gp">>>> </span><span class="n">docinfo</span><span class="o">.</span><span class="n">system_url</span> <span class="o">=</span> <span class="bp">None</span>
272 <span class="gp">>>> </span><span class="n">docinfo</span><span class="o">.</span><span class="n">public_id</span> <span class="o">=</span> <span class="bp">None</span>
273 <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">tree</span><span class="p">))</span>
274 <span class="go"><!DOCTYPE html></span>
275 <span class="go"><html><body/></html></span>
279 <div class="section" id="the-target-parser-interface">
280 <h1>The target parser interface</h1>
281 <p><a class="reference external" href="http://effbot.org/elementtree/elementtree-xmlparser.htm">As in ElementTree</a>, and similar to a SAX event handler, you can pass
282 a target object to the parser:</p>
283 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="k">class</span> <span class="nc">EchoTarget</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
284 <span class="gp">... </span> <span class="k">def</span> <span class="nf">start</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">,</span> <span class="n">attrib</span><span class="p">):</span>
285 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"start </span><span class="si">%s</span><span class="s2"> </span><span class="si">%r</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">tag</span><span class="p">,</span> <span class="nb">dict</span><span class="p">(</span><span class="n">attrib</span><span class="p">)))</span>
286 <span class="gp">... </span> <span class="k">def</span> <span class="nf">end</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">):</span>
287 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"end </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">tag</span><span class="p">)</span>
288 <span class="gp">... </span> <span class="k">def</span> <span class="nf">data</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data</span><span class="p">):</span>
289 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"data </span><span class="si">%r</span><span class="s2">"</span> <span class="o">%</span> <span class="n">data</span><span class="p">)</span>
290 <span class="gp">... </span> <span class="k">def</span> <span class="nf">comment</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">text</span><span class="p">):</span>
291 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"comment </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">text</span><span class="p">)</span>
292 <span class="gp">... </span> <span class="k">def</span> <span class="nf">close</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
293 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"close"</span><span class="p">)</span>
294 <span class="gp">... </span> <span class="k">return</span> <span class="s2">"closed!"</span>
296 <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><span class="n">target</span> <span class="o">=</span> <span class="n">EchoTarget</span><span class="p">())</span>
298 <span class="gp">>>> </span><span class="n">result</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="s2">"<element>some<!--comment-->text</element>"</span><span class="p">,</span>
299 <span class="gp">... </span> <span class="n">parser</span><span class="p">)</span>
300 <span class="go">start element {}</span>
301 <span class="go">data u'some'</span>
302 <span class="go">comment comment</span>
303 <span class="go">data u'text'</span>
304 <span class="go">end element</span>
305 <span class="go">close</span>
307 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
308 <span class="go">closed!</span>
310 <p>It is important for the <tt class="docutils literal">.close()</tt> method to reset the parser target
311 to a usable state, so that you can reuse the parser as often as you
313 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">result</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="s2">"<element>some<!--comment-->text</element>"</span><span class="p">,</span>
314 <span class="gp">... </span> <span class="n">parser</span><span class="p">)</span>
315 <span class="go">start element {}</span>
316 <span class="go">data u'some'</span>
317 <span class="go">comment comment</span>
318 <span class="go">data u'text'</span>
319 <span class="go">end element</span>
320 <span class="go">close</span>
322 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
323 <span class="go">closed!</span>
325 <p>Starting with lxml 2.3, the <tt class="docutils literal">.close()</tt> method will also be called in
326 the error case. This diverges from the behaviour of ElementTree, but
327 allows target objects to clean up their state in all situations, so
328 that the parser can reuse them afterwards.</p>
329 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="k">class</span> <span class="nc">CollectorTarget</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
330 <span class="gp">... </span> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
331 <span class="gp">... </span> <span class="bp">self</span><span class="o">.</span><span class="n">events</span> <span class="o">=</span> <span class="p">[]</span>
332 <span class="gp">... </span> <span class="k">def</span> <span class="nf">start</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">,</span> <span class="n">attrib</span><span class="p">):</span>
333 <span class="gp">... </span> <span class="bp">self</span><span class="o">.</span><span class="n">events</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"start </span><span class="si">%s</span><span class="s2"> </span><span class="si">%r</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">tag</span><span class="p">,</span> <span class="nb">dict</span><span class="p">(</span><span class="n">attrib</span><span class="p">)))</span>
334 <span class="gp">... </span> <span class="k">def</span> <span class="nf">end</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">):</span>
335 <span class="gp">... </span> <span class="bp">self</span><span class="o">.</span><span class="n">events</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"end </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">tag</span><span class="p">)</span>
336 <span class="gp">... </span> <span class="k">def</span> <span class="nf">data</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data</span><span class="p">):</span>
337 <span class="gp">... </span> <span class="bp">self</span><span class="o">.</span><span class="n">events</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"data </span><span class="si">%r</span><span class="s2">"</span> <span class="o">%</span> <span class="n">data</span><span class="p">)</span>
338 <span class="gp">... </span> <span class="k">def</span> <span class="nf">comment</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">text</span><span class="p">):</span>
339 <span class="gp">... </span> <span class="bp">self</span><span class="o">.</span><span class="n">events</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"comment </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">text</span><span class="p">)</span>
340 <span class="gp">... </span> <span class="k">def</span> <span class="nf">close</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
341 <span class="gp">... </span> <span class="bp">self</span><span class="o">.</span><span class="n">events</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">"close"</span><span class="p">)</span>
342 <span class="gp">... </span> <span class="k">return</span> <span class="s2">"closed!"</span>
344 <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><span class="n">target</span> <span class="o">=</span> <span class="n">CollectorTarget</span><span class="p">())</span>
346 <span class="gp">>>> </span><span class="n">result</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="s2">"<element>some</error>"</span><span class="p">,</span>
347 <span class="gp">... </span> <span class="n">parser</span><span class="p">)</span> <span class="c1"># doctest: +ELLIPSIS</span>
348 <span class="gt">Traceback (most recent call last):</span>
349 <span class="c">...</span>
350 <span class="gr">lxml.etree.XMLSyntaxError</span>: <span class="n">Opening and ending tag mismatch...</span>
352 <span class="gp">>>> </span><span class="k">for</span> <span class="n">event</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">target</span><span class="o">.</span><span class="n">events</span><span class="p">:</span>
353 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="n">event</span><span class="p">)</span>
354 <span class="go">start element {}</span>
355 <span class="go">data u'some'</span>
356 <span class="go">close</span>
358 <p>Note that the parser does <em>not</em> build a tree when using a parser
359 target. The result of the parser run is whatever the target object
360 returns from its <tt class="docutils literal">.close()</tt> method. If you want to return an XML
361 tree here, you have to create it programmatically in the target
362 object. An example for a parser target that builds a tree is the
363 <tt class="docutils literal">TreeBuilder</tt>:</p>
364 <div class="syntax"><pre><span></span><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><span class="n">target</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">TreeBuilder</span><span class="p">())</span>
366 <span class="gp">>>> </span><span class="n">result</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="s2">"<element>some<!--comment-->text</element>"</span><span class="p">,</span>
367 <span class="gp">... </span> <span class="n">parser</span><span class="p">)</span>
369 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">result</span><span class="o">.</span><span class="n">tag</span><span class="p">)</span>
370 <span class="go">element</span>
371 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">result</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">text</span><span class="p">)</span>
372 <span class="go">comment</span>
375 <div class="section" id="the-feed-parser-interface">
376 <h1>The feed parser interface</h1>
377 <p>Since lxml 2.0, the parsers have a feed parser interface that is
378 compatible to the <a class="reference external" href="http://effbot.org/elementtree/elementtree-xmlparser.htm">ElementTree parsers</a>. You can use it to feed data
379 into the parser in a controlled step-by-step way.</p>
380 <p>In lxml.etree, you can use both interfaces to a parser at the same
381 time: the <tt class="docutils literal">parse()</tt> or <tt class="docutils literal">XML()</tt> functions, and the feed parser
382 interface. Both are independent and will not conflict (except if used
383 in conjunction with a parser target object as described above).</p>
384 <p>To start parsing with a feed parser, just call its <tt class="docutils literal">feed()</tt> method
385 to feed it some data.</p>
386 <div class="syntax"><pre><span></span><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>
388 <span class="gp">>>> </span><span class="k">for</span> <span class="n">data</span> <span class="ow">in</span> <span class="p">(</span><span class="s1">'<?xml versio'</span><span class="p">,</span> <span class="s1">'n="1.0"?'</span><span class="p">,</span> <span class="s1">'><roo'</span><span class="p">,</span> <span class="s1">'t><a'</span><span class="p">,</span> <span class="s1">'/></root>'</span><span class="p">):</span>
389 <span class="gp">... </span> <span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="n">data</span><span class="p">)</span>
391 <p>When you are done parsing, you <strong>must</strong> call the <tt class="docutils literal">close()</tt> method to
392 retrieve the root Element of the parse result document, and to unlock the
394 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">root</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
396 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">root</span><span class="o">.</span><span class="n">tag</span><span class="p">)</span>
397 <span class="go">root</span>
398 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">root</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">tag</span><span class="p">)</span>
399 <span class="go">a</span>
401 <p>If you do not call <tt class="docutils literal">close()</tt>, the parser will stay locked and
402 subsequent feeds will keep appending data, usually resulting in a non
403 well-formed document and an unexpected parser error. So make sure you
404 always close the parser after use, also in the exception case.</p>
405 <p>Another way of achieving the same step-by-step parsing is by writing your own
406 file-like object that returns a chunk of data on each <tt class="docutils literal">read()</tt> call. Where
407 the feed parser interface allows you to actively pass data chunks into the
408 parser, a file-like object passively responds to <tt class="docutils literal">read()</tt> requests of the
409 parser itself. Depending on the data source, either way may be more natural.</p>
410 <p>Note that the feed parser has its own error log called
411 <tt class="docutils literal">feed_error_log</tt>. Errors in the feed parser do not show up in the
412 normal <tt class="docutils literal">error_log</tt> and vice versa.</p>
413 <p>You can also combine the feed parser interface with the target parser:</p>
414 <div class="syntax"><pre><span></span><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><span class="n">target</span> <span class="o">=</span> <span class="n">EchoTarget</span><span class="p">())</span>
416 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s2">"<eleme"</span><span class="p">)</span>
417 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s2">"nt>some text</elem"</span><span class="p">)</span>
418 <span class="go">start element {}</span>
419 <span class="go">data u'some text'</span>
420 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s2">"ent>"</span><span class="p">)</span>
421 <span class="go">end element</span>
423 <span class="gp">>>> </span><span class="n">result</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
424 <span class="go">close</span>
425 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
426 <span class="go">closed!</span>
428 <p>Again, this prevents the automatic creation of an XML tree and leaves
429 all the event handling to the target object. The <tt class="docutils literal">close()</tt> method
430 of the parser forwards the return value of the target's <tt class="docutils literal">close()</tt>
433 <div class="section" id="incremental-event-parsing">
434 <h1>Incremental event parsing</h1>
435 <p>In Python 3.4, the <tt class="docutils literal">xml.etree.ElementTree</tt> package gained an extension
436 to the feed parser interface that is implemented by the <tt class="docutils literal">XMLPullParser</tt>
437 class. It additionally allows processing parse events after each
438 incremental parsing step, by calling the <tt class="docutils literal">.read_events()</tt> method and
439 iterating over the result. This is most useful for non-blocking execution
440 environments where data chunks arrive one after the other and should be
441 processed as far as possible in each step.</p>
442 <p>The same feature is available in lxml 3.3. The basic usage is as follows:</p>
443 <div class="syntax"><pre><span></span><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">XMLPullParser</span><span class="p">(</span><span class="n">events</span><span class="o">=</span><span class="p">(</span><span class="s1">'start'</span><span class="p">,</span> <span class="s1">'end'</span><span class="p">))</span>
445 <span class="gp">>>> </span><span class="k">def</span> <span class="nf">print_events</span><span class="p">(</span><span class="n">parser</span><span class="p">):</span>
446 <span class="gp">... </span> <span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">element</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">():</span>
447 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">: </span><span class="si">%s</span><span class="s1">'</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">element</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
449 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<root>some text'</span><span class="p">)</span>
450 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">parser</span><span class="p">)</span>
451 <span class="go">start: root</span>
452 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">parser</span><span class="p">)</span> <span class="c1"># well, no more events, as before ...</span>
454 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<child><a />'</span><span class="p">)</span>
455 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">parser</span><span class="p">)</span>
456 <span class="go">start: child</span>
457 <span class="go">start: a</span>
458 <span class="go">end: a</span>
460 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'</child></roo'</span><span class="p">)</span>
461 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">parser</span><span class="p">)</span>
462 <span class="go">end: child</span>
463 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'t>'</span><span class="p">)</span>
464 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">parser</span><span class="p">)</span>
465 <span class="go">end: root</span>
467 <p>Just like the normal feed parser, the <tt class="docutils literal">XMLPullParser</tt> builds a tree in
468 memory (and you should always call the <tt class="docutils literal">.close()</tt> method when done with
470 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">root</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
471 <span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
472 <span class="go">b'<root>some text<child><a/></child></root>'</span>
474 <p>However, since the parser provides incremental access to that tree,
475 you can explicitly delete content that you no longer need once you
476 have processed it. Read the section on <a class="reference internal" href="#modifying-the-tree">Modifying the tree</a> below
477 to see what you can do here and what kind of modifications you should
479 <p>In lxml, it is enough to call the <tt class="docutils literal">.read_events()</tt> method once as
480 the iterator it returns can be reused when new events are available.</p>
481 <p>Also, as known from other iterators in lxml, you can pass a <tt class="docutils literal">tag</tt>
482 argument that selects which parse events are returned by the
483 <tt class="docutils literal">.read_events()</tt> iterator.</p>
484 <div class="section" id="event-types">
486 <p>The parse events are tuples <tt class="docutils literal"><span class="pre">(event-type,</span> object)</tt>. The event types
487 supported by ElementTree and lxml.etree are the strings 'start', 'end',
488 'start-ns' and 'end-ns'. The 'start' and 'end' events represent opening
489 and closing elements. They are accompanied by the respective Element
490 instance. By default, only 'end' events are generated, whereas the
491 example above requested the generation of both 'start' and 'end' events.</p>
492 <p>The 'start-ns' and 'end-ns' events notify about namespace declarations.
493 They do not come with Elements. Instead, the value of the 'start-ns'
494 event is a tuple <tt class="docutils literal">(prefix, namespaceURI)</tt> that designates the beginning
495 of a prefix-namespace mapping. The corresponding <tt class="docutils literal"><span class="pre">end-ns</span></tt> event does
496 not have a value (None). It is common practice to use a list as namespace
497 stack and pop the last entry on the 'end-ns' event.</p>
498 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="k">def</span> <span class="nf">print_events</span><span class="p">(</span><span class="n">events</span><span class="p">):</span>
499 <span class="gp">... </span> <span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">events</span><span class="p">:</span>
500 <span class="gp">... </span> <span class="k">if</span> <span class="n">action</span> <span class="ow">in</span> <span class="p">(</span><span class="s1">'start'</span><span class="p">,</span> <span class="s1">'end'</span><span class="p">):</span>
501 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">obj</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
502 <span class="gp">... </span> <span class="k">elif</span> <span class="n">action</span> <span class="o">==</span> <span class="s1">'start-ns'</span><span class="p">:</span>
503 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">obj</span><span class="p">))</span>
504 <span class="gp">... </span> <span class="k">else</span><span class="p">:</span>
505 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="n">action</span><span class="p">)</span>
507 <span class="gp">>>> </span><span class="n">event_types</span> <span class="o">=</span> <span class="p">(</span><span class="s2">"start"</span><span class="p">,</span> <span class="s2">"end"</span><span class="p">,</span> <span class="s2">"start-ns"</span><span class="p">,</span> <span class="s2">"end-ns"</span><span class="p">)</span>
508 <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">XMLPullParser</span><span class="p">(</span><span class="n">event_types</span><span class="p">)</span>
509 <span class="gp">>>> </span><span class="n">events</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">()</span>
511 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<root><element>'</span><span class="p">)</span>
512 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">events</span><span class="p">)</span>
513 <span class="go">start: root</span>
514 <span class="go">start: element</span>
515 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'text</element><element>text</element>'</span><span class="p">)</span>
516 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">events</span><span class="p">)</span>
517 <span class="go">end: element</span>
518 <span class="go">start: element</span>
519 <span class="go">end: element</span>
520 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<empty-element xmlns="http://testns/" />'</span><span class="p">)</span>
521 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">events</span><span class="p">)</span>
522 <span class="go">start-ns: ('', 'http://testns/')</span>
523 <span class="go">start: {http://testns/}empty-element</span>
524 <span class="go">end: {http://testns/}empty-element</span>
525 <span class="go">end-ns</span>
526 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'</root>'</span><span class="p">)</span>
527 <span class="gp">>>> </span><span class="n">print_events</span><span class="p">(</span><span class="n">events</span><span class="p">)</span>
528 <span class="go">end: root</span>
531 <div class="section" id="modifying-the-tree">
532 <h2>Modifying the tree</h2>
533 <p>You can modify the element and its descendants when handling the
534 'end' event. To save memory, for example, you can remove subtrees
535 that are no longer needed:</p>
536 <div class="syntax"><pre><span></span><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">XMLPullParser</span><span class="p">()</span>
537 <span class="gp">>>> </span><span class="n">events</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">()</span>
539 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<root><element key="value">text</element>'</span><span class="p">)</span>
540 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<element><child /></element>'</span><span class="p">)</span>
541 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">events</span><span class="p">:</span>
542 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">: </span><span class="si">%d</span><span class="s1">'</span> <span class="o">%</span> <span class="p">(</span><span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">elem</span><span class="p">)))</span> <span class="c1"># processing</span>
543 <span class="gp">... </span> <span class="n">elem</span><span class="o">.</span><span class="n">clear</span><span class="p">(</span><span class="n">keep_tail</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span> <span class="c1"># delete children</span>
544 <span class="go">element: 0</span>
545 <span class="go">child: 0</span>
546 <span class="go">element: 1</span>
547 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<empty-element xmlns="http://testns/" /></root>'</span><span class="p">)</span>
548 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">events</span><span class="p">:</span>
549 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">: </span><span class="si">%d</span><span class="s1">'</span> <span class="o">%</span> <span class="p">(</span><span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">elem</span><span class="p">)))</span> <span class="c1"># processing</span>
550 <span class="gp">... </span> <span class="n">elem</span><span class="o">.</span><span class="n">clear</span><span class="p">(</span><span class="n">keep_tail</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span> <span class="c1"># delete children</span>
551 <span class="go">{http://testns/}empty-element: 0</span>
552 <span class="go">root: 3</span>
554 <span class="gp">>>> </span><span class="n">root</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
555 <span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
556 <span class="go">b'<root/>'</span>
558 <p><strong>WARNING</strong>: During the 'start' event, any content of the element,
559 such as the descendants, following siblings or text, is not yet
560 available and should not be accessed. Only attributes are guaranteed
561 to be set. During the 'end' event, the element and its descendants
562 can be freely modified, but its following siblings should not be
563 accessed. During either of the two events, you <strong>must not</strong> modify or
564 move the ancestors (parents) of the current element. You should also
565 avoid moving or discarding the element itself. The golden rule is: do
566 not touch anything that will have to be touched again by the parser
568 <p>If you have elements with a long list of children in your XML file and want
569 to save more memory during parsing, you can clean up the preceding siblings
570 of the current element:</p>
571 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="k">for</span> <span class="n">event</span><span class="p">,</span> <span class="n">element</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">():</span>
572 <span class="gp">... </span> <span class="c1"># ... do something with the element</span>
573 <span class="gp">... </span> <span class="n">element</span><span class="o">.</span><span class="n">clear</span><span class="p">(</span><span class="n">keep_tail</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span> <span class="c1"># clean up children</span>
574 <span class="gp">... </span> <span class="k">while</span> <span class="n">element</span><span class="o">.</span><span class="n">getprevious</span><span class="p">()</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span><span class="p">:</span>
575 <span class="gp">... </span> <span class="k">del</span> <span class="n">element</span><span class="o">.</span><span class="n">getparent</span><span class="p">()[</span><span class="mi">0</span><span class="p">]</span> <span class="c1"># clean up preceding siblings</span>
577 <p>The <tt class="docutils literal">while</tt> loop deletes multiple siblings in a row. This is only necessary
578 if you skipped over some of them using the <tt class="docutils literal">tag</tt> keyword argument.
579 Otherwise, a simple <tt class="docutils literal">if</tt> should do. The more selective your tag is,
580 however, the more thought you will have to put into finding the right way to
581 clean up the elements that were skipped. Therefore, it is sometimes easier to
582 traverse all elements and do the tag selection by hand in the event handler
585 <div class="section" id="selective-tag-events">
586 <h2>Selective tag events</h2>
587 <p>As an extension over ElementTree, lxml.etree accepts a <tt class="docutils literal">tag</tt> keyword
588 argument just like <tt class="docutils literal">element.iter(tag)</tt>. This restricts events to a
589 specific tag or namespace:</p>
590 <div class="syntax"><pre><span></span><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">XMLPullParser</span><span class="p">(</span><span class="n">tag</span><span class="o">=</span><span class="s2">"element"</span><span class="p">)</span>
592 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<root><element key="value">text</element>'</span><span class="p">)</span>
593 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<element><child /></element>'</span><span class="p">)</span>
594 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<empty-element xmlns="http://testns/" /></root>'</span><span class="p">)</span>
596 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">():</span>
597 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
598 <span class="go">end: element</span>
599 <span class="go">end: element</span>
601 <span class="gp">>>> </span><span class="n">event_types</span> <span class="o">=</span> <span class="p">(</span><span class="s2">"start"</span><span class="p">,</span> <span class="s2">"end"</span><span class="p">)</span>
602 <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">XMLPullParser</span><span class="p">(</span><span class="n">event_types</span><span class="p">,</span> <span class="n">tag</span><span class="o">=</span><span class="s2">"{http://testns/}*"</span><span class="p">)</span>
604 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<root><element key="value">text</element>'</span><span class="p">)</span>
605 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<element><child /></element>'</span><span class="p">)</span>
606 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<empty-element xmlns="http://testns/" /></root>'</span><span class="p">)</span>
608 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">():</span>
609 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
610 <span class="go">start: {http://testns/}empty-element</span>
611 <span class="go">end: {http://testns/}empty-element</span>
614 <div class="section" id="comments-and-pis">
615 <h2>Comments and PIs</h2>
616 <p>As an extension over ElementTree, the <tt class="docutils literal">XMLPullParser</tt> in lxml.etree
617 also supports the event types 'comment' and 'pi' for the respective
619 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">event_types</span> <span class="o">=</span> <span class="p">(</span><span class="s2">"start"</span><span class="p">,</span> <span class="s2">"end"</span><span class="p">,</span> <span class="s2">"comment"</span><span class="p">,</span> <span class="s2">"pi"</span><span class="p">)</span>
620 <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">XMLPullParser</span><span class="p">(</span><span class="n">event_types</span><span class="p">)</span>
622 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<?some pi ?><!-- a comment --><root>'</span><span class="p">)</span>
623 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<element key="value">text</element>'</span><span class="p">)</span>
624 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<!-- another comment -->'</span><span class="p">)</span>
625 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<element>text</element>tail'</span><span class="p">)</span>
626 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<empty-element xmlns="http://testns/" />'</span><span class="p">)</span>
627 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'</root>'</span><span class="p">)</span>
629 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">():</span>
630 <span class="gp">... </span> <span class="k">if</span> <span class="n">action</span> <span class="ow">in</span> <span class="p">(</span><span class="s1">'start'</span><span class="p">,</span> <span class="s1">'end'</span><span class="p">):</span>
631 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
632 <span class="gp">... </span> <span class="k">elif</span> <span class="n">action</span> <span class="o">==</span> <span class="s1">'pi'</span><span class="p">:</span>
633 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: -</span><span class="si">%s</span><span class="s2">=</span><span class="si">%s</span><span class="s2">-"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">target</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">text</span><span class="p">))</span>
634 <span class="gp">... </span> <span class="k">else</span><span class="p">:</span> <span class="c1"># 'comment'</span>
635 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: -</span><span class="si">%s</span><span class="s2">-"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">text</span><span class="p">))</span>
636 <span class="go">pi: -some=pi -</span>
637 <span class="go">comment: - a comment -</span>
638 <span class="go">start: root</span>
639 <span class="go">start: element</span>
640 <span class="go">end: element</span>
641 <span class="go">comment: - another comment -</span>
642 <span class="go">start: element</span>
643 <span class="go">end: element</span>
644 <span class="go">start: {http://testns/}empty-element</span>
645 <span class="go">end: {http://testns/}empty-element</span>
646 <span class="go">end: root</span>
648 <span class="gp">>>> </span><span class="n">root</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
649 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">root</span><span class="o">.</span><span class="n">tag</span><span class="p">)</span>
650 <span class="go">root</span>
653 <div class="section" id="events-with-custom-targets">
654 <h2>Events with custom targets</h2>
655 <p>You can combine the pull parser with a parser target. In that case,
656 it is the target's responsibility to generate event values. Whatever
657 it returns from its <tt class="docutils literal">.start()</tt> and <tt class="docutils literal">.end()</tt> methods will be returned
658 by the pull parser as the second item of the parse events tuple.</p>
659 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="k">class</span> <span class="nc">Target</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
660 <span class="gp">... </span> <span class="k">def</span> <span class="nf">start</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">,</span> <span class="n">attrib</span><span class="p">):</span>
661 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'-> start(</span><span class="si">%s</span><span class="s1">)'</span> <span class="o">%</span> <span class="n">tag</span><span class="p">)</span>
662 <span class="gp">... </span> <span class="k">return</span> <span class="s1">'>>START: </span><span class="si">%s</span><span class="s1"><<'</span> <span class="o">%</span> <span class="n">tag</span>
663 <span class="gp">... </span> <span class="k">def</span> <span class="nf">end</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">):</span>
664 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'-> end(</span><span class="si">%s</span><span class="s1">)'</span> <span class="o">%</span> <span class="n">tag</span><span class="p">)</span>
665 <span class="gp">... </span> <span class="k">return</span> <span class="s1">'>>END: </span><span class="si">%s</span><span class="s1"><<'</span> <span class="o">%</span> <span class="n">tag</span>
666 <span class="gp">... </span> <span class="k">def</span> <span class="nf">close</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
667 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'-> close()'</span><span class="p">)</span>
668 <span class="gp">... </span> <span class="k">return</span> <span class="s2">"CLOSED!"</span>
670 <span class="gp">>>> </span><span class="n">event_types</span> <span class="o">=</span> <span class="p">(</span><span class="s1">'start'</span><span class="p">,</span> <span class="s1">'end'</span><span class="p">)</span>
671 <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">XMLPullParser</span><span class="p">(</span><span class="n">event_types</span><span class="p">,</span> <span class="n">target</span><span class="o">=</span><span class="n">Target</span><span class="p">())</span>
673 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<root><child1 /><child2 /></root>'</span><span class="p">)</span>
674 <span class="go">-> start(root)</span>
675 <span class="go">-> start(child1)</span>
676 <span class="go">-> end(child1)</span>
677 <span class="go">-> start(child2)</span>
678 <span class="go">-> end(child2)</span>
679 <span class="go">-> end(root)</span>
681 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">():</span>
682 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">: </span><span class="si">%s</span><span class="s1">'</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">value</span><span class="p">))</span>
683 <span class="go">start: >>START: root<<</span>
684 <span class="go">start: >>START: child1<<</span>
685 <span class="go">end: >>END: child1<<</span>
686 <span class="go">start: >>START: child2<<</span>
687 <span class="go">end: >>END: child2<<</span>
688 <span class="go">end: >>END: root<<</span>
690 <span class="gp">>>> </span><span class="k">print</span><span class="p">(</span><span class="n">parser</span><span class="o">.</span><span class="n">close</span><span class="p">())</span>
691 <span class="go">-> close()</span>
692 <span class="go">CLOSED!</span>
694 <p>As you can see, the event values do not even have to be Element objects.
695 The target is generally free to decide how it wants to create an XML tree
696 or whatever else it wants to make of the parser callbacks. In many cases,
697 however, you will want to make your custom target inherit from the
698 <tt class="docutils literal">TreeBuilder</tt> class in order to have it build a tree that you can process
699 normally. The <tt class="docutils literal">start()</tt> and <tt class="docutils literal">.end()</tt> methods of <tt class="docutils literal">TreeBuilder</tt> return
700 the Element object that was created, so you can override them and modify
701 the input or output according to your needs. Here is an example that
702 filters attributes before they are being added to the tree:</p>
703 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="k">class</span> <span class="nc">AttributeFilter</span><span class="p">(</span><span class="n">etree</span><span class="o">.</span><span class="n">TreeBuilder</span><span class="p">):</span>
704 <span class="gp">... </span> <span class="k">def</span> <span class="nf">start</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">,</span> <span class="n">attrib</span><span class="p">):</span>
705 <span class="gp">... </span> <span class="n">attrib</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">attrib</span><span class="p">)</span>
706 <span class="gp">... </span> <span class="k">if</span> <span class="s1">'evil'</span> <span class="ow">in</span> <span class="n">attrib</span><span class="p">:</span>
707 <span class="gp">... </span> <span class="k">del</span> <span class="n">attrib</span><span class="p">[</span><span class="s1">'evil'</span><span class="p">]</span>
708 <span class="gp">... </span> <span class="k">return</span> <span class="nb">super</span><span class="p">(</span><span class="n">AttributeFilter</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">start</span><span class="p">(</span><span class="n">tag</span><span class="p">,</span> <span class="n">attrib</span><span class="p">)</span>
710 <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">XMLPullParser</span><span class="p">(</span><span class="n">target</span><span class="o">=</span><span class="n">AttributeFilter</span><span class="p">())</span>
711 <span class="gp">>>> </span><span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="s1">'<root><child1 test="123" /><child2 evil="YES" /></root>'</span><span class="p">)</span>
713 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">element</span> <span class="ow">in</span> <span class="n">parser</span><span class="o">.</span><span class="n">read_events</span><span class="p">():</span>
714 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s1">'</span><span class="si">%s</span><span class="s1">: </span><span class="si">%s</span><span class="s1">(</span><span class="si">%r</span><span class="s1">)'</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">element</span><span class="o">.</span><span class="n">tag</span><span class="p">,</span> <span class="n">element</span><span class="o">.</span><span class="n">attrib</span><span class="p">))</span>
715 <span class="go">end: child1({'test': '123'})</span>
716 <span class="go">end: child2({})</span>
717 <span class="go">end: root({})</span>
719 <span class="gp">>>> </span><span class="n">root</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
723 <div class="section" id="iterparse-and-iterwalk">
724 <h1>iterparse and iterwalk</h1>
725 <p>As known from ElementTree, the <tt class="docutils literal">iterparse()</tt> utility function
726 returns an iterator that generates parser events for an XML file (or
727 file-like object), while building the tree. You can think of it as
728 a blocking wrapper around the <tt class="docutils literal">XMLPullParser</tt> that automatically and
729 incrementally reads data from the input file for you and provides a
730 single iterator for them:</p>
731 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">xml</span> <span class="o">=</span> <span class="s1">'''</span>
732 <span class="gp">... </span><span class="s1"><root></span>
733 <span class="gp">... </span><span class="s1"> <element key='value'>text</element></span>
734 <span class="gp">... </span><span class="s1"> <element>text</element>tail</span>
735 <span class="gp">... </span><span class="s1"> <empty-element xmlns="http://testns/" /></span>
736 <span class="gp">... </span><span class="s1"></root></span>
737 <span class="gp">... </span><span class="s1">'''</span>
739 <span class="gp">>>> </span><span class="n">context</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">iterparse</span><span class="p">(</span><span class="n">StringIO</span><span class="p">(</span><span class="n">xml</span><span class="p">))</span>
740 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">context</span><span class="p">:</span>
741 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
742 <span class="go">end: element</span>
743 <span class="go">end: element</span>
744 <span class="go">end: {http://testns/}empty-element</span>
745 <span class="go">end: root</span>
747 <p>After parsing, the resulting tree is available through the <tt class="docutils literal">root</tt> property
749 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">context</span><span class="o">.</span><span class="n">root</span><span class="o">.</span><span class="n">tag</span>
750 <span class="go">'root'</span>
752 <p>The other event types can be activated with the <tt class="docutils literal">events</tt> keyword argument:</p>
753 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">events</span> <span class="o">=</span> <span class="p">(</span><span class="s2">"start"</span><span class="p">,</span> <span class="s2">"end"</span><span class="p">)</span>
754 <span class="gp">>>> </span><span class="n">context</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">iterparse</span><span class="p">(</span><span class="n">StringIO</span><span class="p">(</span><span class="n">xml</span><span class="p">),</span> <span class="n">events</span><span class="o">=</span><span class="n">events</span><span class="p">)</span>
755 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">context</span><span class="p">:</span>
756 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
757 <span class="go">start: root</span>
758 <span class="go">start: element</span>
759 <span class="go">end: element</span>
760 <span class="go">start: element</span>
761 <span class="go">end: element</span>
762 <span class="go">start: {http://testns/}empty-element</span>
763 <span class="go">end: {http://testns/}empty-element</span>
764 <span class="go">end: root</span>
766 <p><tt class="docutils literal">iterparse()</tt> also supports the <tt class="docutils literal">tag</tt> argument for selective event
767 iteration and several other parameters that control the parser setup.
768 The <tt class="docutils literal">tag</tt> argument can be a single tag or a sequence of tags.
769 You can also use it to parse HTML input by passing <tt class="docutils literal">html=True</tt>.</p>
770 <div class="section" id="iterwalk">
772 <p>For convenience, lxml also provides an <tt class="docutils literal">iterwalk()</tt> function.
773 It behaves exactly like <tt class="docutils literal">iterparse()</tt>, but works on Elements and
774 ElementTrees. Here is an example for a tree parsed by <tt class="docutils literal">iterparse()</tt>:</p>
775 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">f</span> <span class="o">=</span> <span class="n">StringIO</span><span class="p">(</span><span class="n">xml</span><span class="p">)</span>
776 <span class="gp">>>> </span><span class="n">context</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">iterparse</span><span class="p">(</span>
777 <span class="gp">... </span> <span class="n">f</span><span class="p">,</span> <span class="n">events</span><span class="o">=</span><span class="p">(</span><span class="s2">"start"</span><span class="p">,</span> <span class="s2">"end"</span><span class="p">),</span> <span class="n">tag</span><span class="o">=</span><span class="s2">"element"</span><span class="p">)</span>
779 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">context</span><span class="p">:</span>
780 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
781 <span class="go">start: element</span>
782 <span class="go">end: element</span>
783 <span class="go">start: element</span>
784 <span class="go">end: element</span>
786 <span class="gp">>>> </span><span class="n">root</span> <span class="o">=</span> <span class="n">context</span><span class="o">.</span><span class="n">root</span>
788 <p>And now we can take the resulting in-memory tree and iterate over it
789 using <tt class="docutils literal">iterwalk()</tt> to get the exact same events without parsing the
791 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">context</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">iterwalk</span><span class="p">(</span>
792 <span class="gp">... </span> <span class="n">root</span><span class="p">,</span> <span class="n">events</span><span class="o">=</span><span class="p">(</span><span class="s2">"start"</span><span class="p">,</span> <span class="s2">"end"</span><span class="p">),</span> <span class="n">tag</span><span class="o">=</span><span class="s2">"element"</span><span class="p">)</span>
794 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">context</span><span class="p">:</span>
795 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
796 <span class="go">start: element</span>
797 <span class="go">end: element</span>
798 <span class="go">start: element</span>
799 <span class="go">end: element</span>
801 <p>In order to avoid wasting time on uninteresting parts of the tree, the <tt class="docutils literal">iterwalk</tt>
802 iterator can be instructed to skip over an entire subtree with its
803 <tt class="docutils literal">.skip_subtree()</tt> method.</p>
804 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">root</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="s1">'''</span>
805 <span class="gp">... </span><span class="s1"><root></span>
806 <span class="gp">... </span><span class="s1"> <a> <b /> </a></span>
807 <span class="gp">... </span><span class="s1"> <c /></span>
808 <span class="gp">... </span><span class="s1"></root></span>
809 <span class="gp">... </span><span class="s1">'''</span><span class="p">)</span>
811 <span class="gp">>>> </span><span class="n">context</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">iterwalk</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">events</span><span class="o">=</span><span class="p">(</span><span class="s2">"start"</span><span class="p">,</span> <span class="s2">"end"</span><span class="p">))</span>
813 <span class="gp">>>> </span><span class="k">for</span> <span class="n">action</span><span class="p">,</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">context</span><span class="p">:</span>
814 <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">action</span><span class="p">,</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span><span class="p">))</span>
815 <span class="gp">... </span> <span class="k">if</span> <span class="n">action</span> <span class="o">==</span> <span class="s1">'start'</span> <span class="ow">and</span> <span class="n">elem</span><span class="o">.</span><span class="n">tag</span> <span class="o">==</span> <span class="s1">'a'</span><span class="p">:</span>
816 <span class="gp">... </span> <span class="n">context</span><span class="o">.</span><span class="n">skip_subtree</span><span class="p">()</span> <span class="c1"># ignore <b></span>
817 <span class="go">start: root</span>
818 <span class="go">start: a</span>
819 <span class="go">end: a</span>
820 <span class="go">start: c</span>
821 <span class="go">end: c</span>
822 <span class="go">end: root</span>
824 <p>Note that <tt class="docutils literal">.skip_subtree()</tt> only has an effect when handling <tt class="docutils literal">start</tt> or
825 <tt class="docutils literal"><span class="pre">start-ns</span></tt> events.</p>
828 <div class="section" id="python-unicode-strings">
829 <h1>Python unicode strings</h1>
830 <p>lxml.etree has broader support for Python unicode strings than the ElementTree
831 library. First of all, where ElementTree would raise an exception, the
832 parsers in lxml.etree can handle unicode strings straight away. This is most
833 helpful for XML snippets embedded in source code using the <tt class="docutils literal">XML()</tt>
835 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">root</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="sa">u</span><span class="s1">'<test> </span><span class="se">\uf8d1</span><span class="s1"> + </span><span class="se">\uf8d2</span><span class="s1"> </test>'</span> <span class="p">)</span>
837 <p>This requires, however, that unicode strings do not specify a conflicting
838 encoding themselves and thus lie about their real encoding:</p>
839 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">XML</span><span class="p">(</span> <span class="sa">u</span><span class="s1">'<?xml version="1.0" encoding="ASCII"?></span><span class="se">\n</span><span class="s1">'</span> <span class="o">+</span>
840 <span class="gp">... </span> <span class="sa">u</span><span class="s1">'<test> </span><span class="se">\uf8d1</span><span class="s1"> + </span><span class="se">\uf8d2</span><span class="s1"> </test>'</span> <span class="p">)</span>
841 <span class="gt">Traceback (most recent call last):</span>
842 <span class="c">...</span>
843 <span class="gr">ValueError</span>: <span class="n">Unicode strings with encoding declaration are not supported. Please use bytes input or XML fragments without declaration.</span>
845 <p>Similarly, you will get errors when you try the same with HTML data in a
846 unicode string that specifies a charset in a meta tag of the header. You
847 should generally avoid converting XML/HTML data to unicode before passing it
848 into the parsers. It is both slower and error prone.</p>
849 <div class="section" id="serialising-to-unicode-strings">
850 <h2>Serialising to Unicode strings</h2>
851 <p>To serialize the result, you would normally use the <tt class="docutils literal">tostring()</tt>
852 module function, which serializes to plain ASCII by default or a
853 number of other byte encodings if asked for:</p>
854 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
855 <span class="go">b'<test> &#63697; + &#63698; </test>'</span>
857 <span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">'UTF-8'</span><span class="p">,</span> <span class="n">xml_declaration</span><span class="o">=</span><span class="bp">False</span><span class="p">)</span>
858 <span class="go">b'<test> \xef\xa3\x91 + \xef\xa3\x92 </test>'</span>
860 <p>As an extension, lxml.etree recognises the name 'unicode' as an argument
861 to the encoding parameter to build a Python unicode representation of a tree:</p>
862 <div class="syntax"><pre><span></span><span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">'unicode'</span><span class="p">)</span>
863 <span class="go">u'<test> \uf8d1 + \uf8d2 </test>'</span>
865 <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="s2">"test"</span><span class="p">)</span>
866 <span class="gp">>>> </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="s1">'unicode'</span><span class="p">)</span>
867 <span class="go">u'<test/>'</span>
869 <span class="gp">>>> </span><span class="n">subel</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">el</span><span class="p">,</span> <span class="s2">"subtest"</span><span class="p">)</span>
870 <span class="gp">>>> </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="s1">'unicode'</span><span class="p">)</span>
871 <span class="go">u'<test><subtest/></test>'</span>
873 <span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">etree</span><span class="o">.</span><span class="n">ElementTree</span><span class="p">(</span><span class="n">el</span><span class="p">)</span>
874 <span class="gp">>>> </span><span class="n">etree</span><span class="o">.</span><span class="n">tostring</span><span class="p">(</span><span class="n">tree</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">'unicode'</span><span class="p">)</span>
875 <span class="go">u'<test><subtest/></test>'</span>
877 <p>The result of <tt class="docutils literal"><span class="pre">tostring(encoding='unicode')</span></tt> can be treated like any
878 other Python unicode string and then passed back into the parsers.
879 However, if you want to save the result to a file or pass it over the
880 network, you should use <tt class="docutils literal">write()</tt> or <tt class="docutils literal">tostring()</tt> with a byte
881 encoding (typically UTF-8) to serialize the XML. The main reason is
882 that unicode strings returned by <tt class="docutils literal"><span class="pre">tostring(encoding='unicode')</span></tt> are
883 not byte streams and they never have an XML declaration to specify
884 their encoding. These strings are most likely not parsable by other
886 <p>For normal byte encodings, the <tt class="docutils literal">tostring()</tt> function automatically
887 adds a declaration as needed that reflects the encoding of the
888 returned string. This makes it possible for other parsers to
889 correctly parse the XML byte stream. Note that using <tt class="docutils literal">tostring()</tt>
890 with UTF-8 is also considerably faster in most cases.</p>
895 <hr class="footer" />
896 Generated on: 2020-01-29.
projects / platform / upstream / python-lxml.git / blob