<!DOCTYPE HTML>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Werkzeug Documentation</title>
<link rel="stylesheet" href="_static/style.css" type="text/css">
<link rel="stylesheet" href="_static/print.css" type="text/css" media="print">
<link rel="stylesheet" href="_static/pygments.css" type="text/css">
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: '#',
VERSION: '0.6.1'
};
</script>
<script type="text/javascript" src="_static/jquery.js"></script>
<script type="text/javascript" src="_static/interface.js"></script>
<script type="text/javascript" src="_static/doctools.js"></script>
<script type="text/javascript" src="_static/werkzeug.js"></script>
<link rel="contents" title="Global table of contents" href="contents.html">
<link rel="index" title="Global index" href="genindex.html">
<link rel="search" title="Search" href="search.html">
<link rel="top" title="Werkzeug v0.6.1 documentation" href="index.html">
<link rel="next" title="Dealing with Request Data" href="request_data.html">
<link rel="prev" title="Important Terms" href="terms.html">
</head>
<body>
<div class="page">
<div class="header">
<h1 class="heading"><a href="index.html"
title="back to the documentation overview"><span>Werkzeug</span></a></h1>
</div>
<ul class="navigation">
<li class="indexlink"><a href="index.html">Overview</a></li>
<li><a href="terms.html">« Important Terms</a></li>
<li class="active"><a href="#">Unicode</a></li>
<li><a href="request_data.html">Dealing with Request Data »</a></li>
</ul>
<div class="body">
<div id="toc">
<h3>Table Of Contents</h3>
<div class="inner"><ul>
<li><a class="reference external" href="#">Unicode</a><ul>
<li><a class="reference external" href="#unicode-in-python">Unicode in Python</a></li>
<li><a class="reference external" href="#unicode-in-http">Unicode in HTTP</a></li>
<li><a class="reference external" href="#error-handling">Error Handling</a></li>
<li><a class="reference external" href="#request-and-response-objects">Request and Response Objects</a></li>
</ul>
</li>
</ul>
</div>
</div>
<div class="section" id="unicode">
<span id="id1"></span><h1>Unicode<a class="headerlink" href="#unicode" title="Permalink to this headline">¶</a></h1>
<span class="target" id="module-werkzeug"></span><p>Since early Python 2 days unicode was part of all default Python builds. It
allows developers to write applications that deal with non-ASCII characters
in a straightforward way. But working with unicode requires a basic knowledge
about that matter, especially when working with libraries that do not support
it.</p>
<p>Werkzeug uses unicode internally everywhere text data is assumed, even if the
HTTP standard is not unicode aware as it. Basically all incoming data is
decoded from the charset specified (per default <cite>utf-8</cite>) so that you don’t
operate on bytestrings any more. Outgoing unicode data is then encoded into
the target charset again.</p>
<div class="section" id="unicode-in-python">
<h2>Unicode in Python<a class="headerlink" href="#unicode-in-python" title="Permalink to this headline">¶</a></h2>
<p>In Python 2 there are two basic string types: <cite>str</cite> and <cite>unicode</cite>. <cite>str</cite> may
carry encoded unicode data but it’s always represented in bytes whereas the
<cite>unicode</cite> type does not contain bytes but charpoints. What does this mean?
Imagine you have the German Umlaut <cite>ö</cite>. In ASCII you cannot represent that
character, but in the <cite>latin-1</cite> and <cite>utf-8</cite> character sets you can represent
it, but they look differently when encoded:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="s">u'ö'</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s">'latin1'</span><span class="p">)</span>
<span class="go">'\xf6'</span>
<span class="gp">>>> </span><span class="s">u'ö'</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s">'utf-8'</span><span class="p">)</span>
<span class="go">'\xc3\xb6'</span>
</pre></div>
</div>
<p>So an <cite>ö</cite> might look totally different depending on the encoding which makes
it hard to work with it. The solution is using the <cite>unicode</cite> type (as we did
above, note the <cite>u</cite> prefix before the string). The unicode type does not
store the bytes for <cite>ö</cite> but the information, that this is a
<tt class="docutils literal"><span class="pre">LATIN</span> <span class="pre">SMALL</span> <span class="pre">LETTER</span> <span class="pre">O</span> <span class="pre">WITH</span> <span class="pre">DIAERESIS</span></tt>.</p>
<p>Doing <tt class="docutils literal"><span class="pre">len(u'ö')</span></tt> will always give us the expected “1” but <tt class="docutils literal"><span class="pre">len('ö')</span></tt>
might give different results depending on the encoding of <tt class="docutils literal"><span class="pre">'ö'</span></tt>.</p>
</div>
<div class="section" id="unicode-in-http">
<h2>Unicode in HTTP<a class="headerlink" href="#unicode-in-http" title="Permalink to this headline">¶</a></h2>
<p>The problem with unicode is that HTTP does not know what unicode is. HTTP
is limited to bytes but this is not a big problem as Werkzeug decodes and
encodes for us automatically all incoming and outgoing data. Basically what
this means is that data sent from the browser to the web application is per
default decoded from an utf-8 bytestring into a <cite>unicode</cite> string. Data sent
from the application back to the browser that is not yet a bytestring is then
encoded back to utf-8.</p>
<p>Usually this “just works” and we don’t have to worry about it, but there are
situations where this behavior is problematic. For example the Python 2 IO
layer is not unicode aware. This means that whenever you work with data from
the file system you have to properly decode it. The correct way to load
a text file from the file system looks like this:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">f</span> <span class="o">=</span> <span class="nb">file</span><span class="p">(</span><span class="s">'/path/to/the_file.txt'</span><span class="p">,</span> <span class="s">'r'</span><span class="p">)</span>
<span class="k">try</span><span class="p">:</span>
<span class="n">text</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="s">'utf-8'</span><span class="p">)</span> <span class="c"># assuming the file is utf-8 encoded</span>
<span class="k">finally</span><span class="p">:</span>
<span class="n">f</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
</pre></div>
</div>
<p>There is also the codecs module which provides an open function that decodes
automatically from the given encoding.</p>
</div>
<div class="section" id="error-handling">
<h2>Error Handling<a class="headerlink" href="#error-handling" title="Permalink to this headline">¶</a></h2>
<p>With Werkzeug 0.3 onwards you can further control the way Werkzeug works with
unicode. In the past Werkzeug ignored encoding errors silently on incoming
data. This decision was made to avoid internal server errors if the user
tampered with the submitted data. However there are situations where you
want to abort with a <cite>400 BAD REQUEST</cite> instead of silently ignoring the error.</p>
<p>All the functions that do internal decoding now accept an <cite>errors</cite> keyword
argument that behaves like the <cite>errors</cite> parameter of the builtin string method
<cite>decode</cite>. The following values are possible:</p>
<dl class="docutils">
<dt><cite>ignore</cite></dt>
<dd>This is the default behavior and tells the codec to ignore characters that
it doesn’t understand silently.</dd>
<dt><cite>replace</cite></dt>
<dd>The codec will replace unknown characters with a replacement character
(<cite>U+FFFD</cite> <tt class="docutils literal"><span class="pre">REPLACEMENT</span> <span class="pre">CHARACTER</span></tt>)</dd>
<dt><cite>strict</cite></dt>
<dd>Raise an exception if decoding fails.</dd>
</dl>
<p>Unlike the regular python decoding Werkzeug does not raise an
<tt class="xref py py-exc docutils literal"><span class="pre">UnicodeDecodeError</span></tt> if the decoding failed but an
<a title="werkzeug.exceptions.HTTPUnicodeError" class="reference external" href="exceptions.html#werkzeug.exceptions.HTTPUnicodeError"><tt class="xref py py-exc docutils literal"><span class="pre">HTTPUnicodeError</span></tt></a> which
is a direct subclass of <cite>UnicodeError</cite> and the <cite>BadRequest</cite> HTTP exception.
The reason is that if this exception is not caught by the application but
a catch-all for HTTP exceptions exists a default <cite>400 BAD REQUEST</cite> error
page is displayed.</p>
<p>There is additional error handling available which is a Werkzeug extension
to the regular codec error handling which is called <cite>fallback</cite>. Often you
want to use utf-8 but support latin1 as legacy encoding too if decoding
failed. For this case you can use the <cite>fallback</cite> error handling. For
example you can specify <tt class="docutils literal"><span class="pre">'fallback:iso-8859-15'</span></tt> to tell Werkzeug it should
try with <cite>iso-8859-15</cite> if <cite>utf-8</cite> failed. If this decoding fails too (which
should not happen for most legacy charsets such as <cite>iso-8859-15</cite>) the error
is silently ignored as if the error handling was <cite>ignore</cite>.</p>
<p>Further details are available as part of the API documentation of the concrete
implementations of the functions or classes working with unicode.</p>
</div>
<div class="section" id="request-and-response-objects">
<h2>Request and Response Objects<a class="headerlink" href="#request-and-response-objects" title="Permalink to this headline">¶</a></h2>
<p>As request and response objects usually are the central entities of Werkzeug
powered applications you can change the default encoding Werkzeug operates on
by subclassing these two classes. For example you can easily set the
application to utf-7 and strict error handling:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="kn">from</span> <span class="nn">werkzeug</span> <span class="kn">import</span> <span class="n">BaseRequest</span><span class="p">,</span> <span class="n">BaseResponse</span>
<span class="k">class</span> <span class="nc">Request</span><span class="p">(</span><span class="n">BaseRequest</span><span class="p">):</span>
<span class="n">charset</span> <span class="o">=</span> <span class="s">'utf-7'</span>
<span class="n">encoding_errors</span> <span class="o">=</span> <span class="s">'strict'</span>
<span class="k">class</span> <span class="nc">Response</span><span class="p">(</span><span class="n">BaseResponse</span><span class="p">):</span>
<span class="n">charset</span> <span class="o">=</span> <span class="s">'utf-7'</span>
</pre></div>
</div>
<p>Keep in mind that the error handling is only customizable for all decoding
but not encoding. If Werkzeug encounters an encoding error it will raise a
<tt class="xref py py-exc docutils literal"><span class="pre">UnicodeEncodeError</span></tt>. It’s your responsibility to not create data that is
not present in the target charset (a non issue with all unicode encodings
such as utf-8).</p>
</div>
</div>
<div style="clear: both"></div>
</div>
<div class="footer">
© Copyright 2008 by the <a href="http://pocoo.org/">Pocoo Team</a>,
documentation generated by <a href="http://sphinx.pocoo.org/">Sphinx</a>
</div>
</div>
</body>
</html>
distrib
>
Fedora
>
14
>
i386
>
media
>
os
>
by-pkgid
>
5c4f8358fd6fdc210fb0d926bd25802c
>
files
>
266
