<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
<html lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Usage [Universal Encoding Detector]</title>
<link rel="stylesheet" href="css/chardet.css" type="text/css">
<link rev="made" href="mailto:mark@diveintomark.org">
<meta name="generator" content="DocBook XSL Stylesheets V1.65.1">
<meta name="keywords" content="character, set, encoding, detection, Python, XML, feed">
<link rel="start" href="index.html" title="Documentation">
<link rel="up" href="index.html" title="Documentation">
<link rel="prev" href="supported-encodings.html" title="Supported encodings">
<link rel="next" href="how-it-works.html" title="How it works">
</head>
<body id="chardet-feedparser-org" class="docs">
<div class="z" id="intro"><div class="sectionInner"><div class="sectionInner2">
<div class="s" id="pageHeader">
<h1><a href="/">Universal Encoding Detector</a></h1>
<p>Character encoding auto-detection in Python. As smart as your browser. Open source.</p>
</div>
<div class="s" id="quickSummary"><ul>
<li class="li1">
<a href="http://chardet.feedparser.org/download/">Download</a> ·</li>
<li class="li2">
<a href="index.html">Documentation</a> ·</li>
<li class="li3"><a href="faq.html" title="Frequently Asked Questions">FAQ</a></li>
</ul></div>
</div></div></div>
<div id="main"><div id="mainInner">
<p id="breadcrumb">You are here: <a href="index.html">Documentation</a> → <span class="thispage">Usage</span></p>
<div class="section" lang="en">
<div class="titlepage">
<div><div><h2 class="title">
<a name="usage" class="skip" href="#usage" title="link to this section"><img src="images/permalink.gif" alt="[link]" title="link to this section" width="8" height="9"></a> Usage</h2></div></div>
<div></div>
</div>
<div class="section" lang="en">
<div class="titlepage">
<div><div><h3 class="title">
<a name="usage.basic" class="skip" href="#usage.basic" title="link to this section"><img src="images/permalink.gif" alt="[link]" title="link to this section" width="8" height="9"></a> Basic usage</h3></div></div>
<div></div>
</div>
<p>The easiest way to use the <span class="application">Universal Encoding Detector</span> library is with the <tt class="function">detect</tt> function.</p>
<div class="example">
<a name="example.basic.detect" class="skip" href="#example.basic.detect" title="link to this example"><img src="images/permalink.gif" alt="[link]" title="link to this example" width="8" height="9"></a> <h3 class="title">Example: Using the <tt class="function">detect</tt> function</h3>
<p>The <tt class="function">detect</tt> function takes one argument, a non-Unicode string. It returns a dictionary containing the auto-detected character encoding and a confidence level from <tt class="constant">0</tt> to <tt class="constant">1</tt>.</p>
<pre class="screen"><tt class="prompt">>>> </tt><span class="userinput"><font color='navy'><b>import</b></font> urllib</span>
<tt class="prompt">>>> </tt><span class="userinput">rawdata = urllib.urlopen(<font color='olive'>'http://yahoo.co.jp/'</font>).read()</span>
<tt class="prompt">>>> </tt><span class="userinput"><font color='navy'><b>import</b></font> chardet</span>
<tt class="prompt">>>> </tt><span class="userinput">chardet.detect(rawdata)</span>
<span class="computeroutput">{'encoding': 'EUC-JP', 'confidence': 0.99}</span></pre>
</div>
</div>
<div class="section" lang="en">
<div class="titlepage">
<div><div><h3 class="title">
<a name="usage.advanced" class="skip" href="#usage.advanced" title="link to this section"><img src="images/permalink.gif" alt="[link]" title="link to this section" width="8" height="9"></a> Advanced usage</h3></div></div>
<div></div>
</div>
<p>If you’re dealing with a large amount of text, you can call the <span class="application">Universal Encoding Detector</span> library incrementally, and it will stop as soon as it is confident enough to report its results.</p>
<p>Create a <tt class="classname">UniversalDetector</tt> object, then call its <tt class="methodname">feed</tt> method repeatedly with each block of text. If the detector reaches a minimum threshold of confidence, it will set <tt class="varname">detector.done</tt> to <tt class="constant">True</tt>.</p>
<p>Once you’ve exhausted the source text, call <tt class="methodname">detector.close()</tt>, which will do some final calculations in case the detector didn’t hit its minimum confidence threshold earlier. Then <tt class="varname">detector.result</tt> will be a dictionary containing the auto-detected character encoding and confidence level (the same as <a href="usage.html#example.basic.detect" title="Example: Using the detect function">the <tt class="function">chardet.detect</tt> function returns</a>).</p>
<div class="example">
<a name="example.multiline" class="skip" href="#example.multiline" title="link to this example"><img src="images/permalink.gif" alt="[link]" title="link to this example" width="8" height="9"></a> <h3 class="title">Example: Detecting encoding incrementally</h3>
<pre class="programlisting python"><font color='navy'><b>import</b></font> urllib
<font color='navy'><b>from</b></font> chardet.universaldetector <font color='navy'><b>import</b></font> UniversalDetector
usock = urllib.urlopen(<font color='olive'>'http://yahoo.co.jp/'</font>)
detector = UniversalDetector()
<font color='navy'><b>for</b></font> line <font color='navy'><b>in</b></font> usock.readlines():
detector.feed(line)
<font color='navy'><b>if</b></font> detector.done: <font color='navy'><b>break</b></font>
detector.close()
usock.close()
<font color='navy'><b>print</b></font> detector.result</pre>
<pre class="screen"><span class="computeroutput">{'encoding': 'EUC-JP', 'confidence': 0.99}</span></pre>
</div>
<p>If you want to detect the encoding of multiple texts (such as separate files), you can re-use a single <tt class="classname">UniversalDetector</tt> object. Just call <tt class="methodname">detector.reset()</tt> at the start of each file, call <tt class="methodname">detector.feed</tt> as many times as you like, and then call <tt class="methodname">detector.close()</tt> and check the <tt class="varname">detector.result</tt> dictionary for the file’s results.</p>
<div class="example">
<a name="advanced.multifile.multiline" class="skip" href="#advanced.multifile.multiline" title="link to this example"><img src="images/permalink.gif" alt="[link]" title="link to this example" width="8" height="9"></a> <h3 class="title">Example: Detecting encodings of multiple files</h3>
<pre class="programlisting python"><font color='navy'><b>import</b></font> glob
<font color='navy'><b>from</b></font> chardet.universaldetector <font color='navy'><b>import</b></font> UniversalDetector
detector = UniversalDetector()
<font color='navy'><b>for</b></font> filename <font color='navy'><b>in</b></font> glob.glob(<font color='olive'>'*.xml'</font>):
<font color='navy'><b>print</b></font> filename.ljust(60),
detector.reset()
<font color='navy'><b>for</b></font> line <font color='navy'><b>in</b></font> file(filename, <font color='olive'>'rb'</font>):
detector.feed(line)
<font color='navy'><b>if</b></font> detector.done: <font color='navy'><b>break</b></font>
detector.close()
<font color='navy'><b>print</b></font> detector.result
</pre>
</div>
</div>
</div>
<div class="footernavigation">
<div style="float: left">← <a class="NavigationArrow" href="supported-encodings.html">Supported encodings</a>
</div>
<div style="text-align: right">
<a class="NavigationArrow" href="how-it-works.html">How it works</a> →</div>
</div>
<hr>
<div id="footer"><p class="copyright">Copyright © 2006, 2007, 2008 Mark Pilgrim · <a href="mailto:mark@diveintomark.org">mark@diveintomark.org</a> · <a href="license.html">Terms of use</a></p></div>
</div></div>
</body>
</html>
