<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html lang="en"><head>
<title>Technical guide: mod_proxy_html</title>
<style type="text/css">
@import url(/index.css) ;
</style>
<style type="text/css">
.v3 { color: red; }
</style>
</head><body>
<div id="apache">
<h1>mod_proxy_html: Technical Guide</h1>
<p><a href="./">mod_proxy_html</a> From Version 2.4 (Sept 2004).
<span class="v3">Updates in Version 3 (Dec. 2006) are highlighted.</span></p>
<h2>Contents</h2>
<ul id="toc">
<li><a href="#url">URL Rewriting</a>
<ul>
<li><a href="#html">HTML Links</a></li>
<li><a href="#event">Scripting Events</a></li>
<li><a href="#cdata">Embedded Scripts and Stylesheets</a></li>
</ul>
</li>
<li><a href="#output">Output Transformation</a>
<ul>
<li><a href="#fpi">FPI (Doctype)</a></li>
<li><a href="#ml">HTML vs XHTML</a></li>
<li><a href="#charset">Character Encoding</a></li>
</ul>
</li>
<li><a href="#meta">meta http-equiv support</a></li>
<li><a href="#misc">Other Fixups</a></li>
<li><a href="#debug">Debugging your Configuration</a></li>
<li><a href="#browser">Workarounds for Browser Bugs</a></li>
</ul>
<h2 id="url">URL Rewriting</h2>
<p>Rewriting URLs into a proxy's address space is of course the primary
purpose of this module. From Version 2.0, this capability has been
extended from rewriting HTML URLs to processing scripts and stylesheets
that <em>may</em> contain URLs.</p>
<p>Because the module doesn't contain parsers for javascript or CSS, this
additional processing means we have had to introduce some heuristic parsing.
What that means is that the parser cannot automatically distinguish between
a URL that should be replaced and one that merely appears as text. It's
up to you to match the right things! To help you do this, we have introduced
some new features:</p>
<ol>
<li>The <code>ProxyHTMLExtended</code> directive. The extended processing
will only be activated if this is On. The default is Off, which gives you
the old behaviour.</li>
<li>Regular Expression match-and-replace. This can be used anywhere,
but is most useful where context information can help distinguish URLs
that should be replaced and avoid false positives. For example,
to rewrite URLs of CSS @import, we might define a rule<br />
<code>ProxyHTMLURLMap url\(http://internal.example.com([^\)]*)\) url(http://proxy.example.com$1) Rihe</code><br />
This explicitly rewrites from one servername to another, and uses regexp
memory to match a path and append it unchanged in $1, while using the
<code>url(...)</code> context to reduce the danger of a match that shouldn't
be rewritten. The <b>R</b> flag invokes regexp processing for this rule;
<b>i</b> makes the match case-insensitive; while <b>h</b> and <b>e</b>
save processing cycles by preventing the match being applied to HTML links
and scripting events, where it is clearly irrelevant.</li>
</ol>
<h3 id="html">HTML Links</h3>
<p>HTML links are those attributes defined by the HTML 4 and XHTML 1
DTDs as of type <strong>%URI</strong>. For example, the <strong>href</strong>
attribute of the <strong>a</strong> element. For a full list, see the
declaration of <code>linked_elts</code> in <code>pstartElement</code>.
Rules are applicable provided the <b>h</b> flag is not set.
<span class="v3">From Version 3, the definition of links to use is
delegated to the system administrator via the <code>ProxyHTMLLinks</code>
directive.</span></p>
<p>An HTML link always contains exactly one URL. So whenever mod_proxy_html
finds a matching <code>ProxyHTMLURLMap</code> rule, it will apply the
transformation once and stop processing the attribute. <span class="v3">This
can be overridden by the <code>l</code> flag, which causes processing
a URL to continue after a rewrite.</span></p>
<h3 id="event">Scripting Events</h3>
<p>Scripting events are the contents of event attributes as defined in the
HTML4 and XHTML1 DTDs; for example <code>onclick</code>. For a full list,
see the declaration of <code>events</code> in <code>pstartElement</code>.
Rules are applicable provided the <b>e</b> flag is not set.
<span class="v3">From Version 3, the definition of events to use is
delegated to the system administrator via the <code>ProxyHTMLEvents</code>
directive.</span></p>
<p>A scripting event may contain more than one URL, and will contain other
text. So when <code>ProxyHTMLExtended</code> is On, all applicable rules
will be applied in order until and unless a rule with the <b>L</b> flag
matches. A rule may match more than once, provided the matches do not
overlap, so a URL/pattern that appears more than once is rewritten
every time it matches.</p>
<h3 id="cdata">Embedded Scripts and Stylesheets</h3>
<p>Embedded scripts and stylesheets are the contents of
<code><script></code> and <code><style></code> elements.
Rules are applicable provided the <b>c</b> flag is not set.</p>
<p>A script or stylesheet may contain more than one URL, and will contain other
text. So when <code>ProxyHTMLExtended</code> is On, all applicable rules
will be applied in order until and unless a rule with the <b>L</b> flag
matches. A rule may match more than once, provided the matches do not
overlap, so a URL/pattern that appears more than once is rewritten
every time it matches.</p>
<h2 id="output">Output Transformation</h2>
<p>mod_proxy_html uses a SAX parser. This means that the input stream
- and hence the output generated - will be normalised in various ways,
even where nothing is actually rewritten. To an HTML or XML parser,
the document is not changed by normalisation, except as noted below.
Exceptions to this may arise where the input stream is malformed, when
the output of mod_proxy_html may be undefined. These should of course
be fixed at the backend: if mod_proxy_html doesn't work as expected,
then neither will browsers in real life, except by coincidence.</p>
<h3 id="fpi">FPI (Doctype)</h3>
<p>Strictly speaking, HTML and XHTML documents are required to have a
Formal Public Identifier (FPI), also know as a Document Type Declaration.
This references a Document Type Definition (DTD) which defines the grammar/
syntax to which the contents of the document must conform.</p>
<p>The parser in mod_proxy_html loses any FPI in the input document, but
gives you the option to insert one. You may select either HTML or XHTML
(see below), and if your backend is sloppy you may also want to use the
"Legacy" keyword to make it declare documents "Transitional". You may
also declare a custom DTD, or (if your backend is seriously screwed
so no DTD would be appropriate) omit it altogether.</p>
<h3 id="ml">HTML vs XHTML</h3>
<p>The differences between HTML 4.01 and XHTML 1.0 are essentially negligible,
and mod_proxy_html can transform between the two. You can safely select
either, regardless of what the backend generates, and mod_proxy_html will
apply the appropriate rules in generating output. HTML saves a few bytes.</p>
<p>If you declare a custom DTD, you should specify whether to generate
HTML or XHTML syntax in the output. This affects empty elements:
HTML <b><br></b> vs XHTML <b><br /></b>.</p>
<p class="v3">If you select standard HTML or XHTML, mod_proxy_html 3 will
perform some additional fixups of bogus markup. If you don't want this,
you can enter a standard DTD using the nonstandard form of
<code>ProxyHTMLDTD</code>, which will then be treated as unknown
(no corrections).</p>
<h3 id="charset">Character Encoding</h3>
<p>The parser uses <strong>UTF-8</strong> (Unicode) internally, and
mod_proxy_html prior to version 3 <em>always</em> generates output as UTF-8.
This is supported by all general-purpose web software, and supports more
character sets and languages than any other charset. <span class="v3">
Version 3 supports, but does not recommend different outputs, using
the <code>ProxyHTMLCharsetOut</code> directive</span>.</p>
<p>The character encoding should be declared in HTTP: for example<br />
<code>Content-Type: text/html; charset=latin1</code><br />
mod_proxy_html has always supported this in its input, and ensured
this happens in output. But prior to version 2, it did not fully
support detection (sniffing) the charset when a backend fails to
set the HTTP Header.</p>
<p>From version 2.0, mod_proxy_html will detect the encoding of its input
as follows:</p>
<ol>
<li>The HTTP headers, where available, always take precedence over other
information.</li>
<li>If the first 2-4 bytes are an XML Byte Order Mark (BOM), this is used.</li>
<li>If the document starts with an XML declaration
<code><?xml .... ?></code>, this determines encoding by XML rules.</li>
<li>If the document contains the HTML hack
<code><meta http-equiv="Content-Type" ...></code>, any charset declared
here is used.</li>
<li>In the absence of any of the above indications, the HTML-over-HTTP default
encoding <b>ISO-8859-1</b> <span class="v3">or the
<code>ProxyHTMLCharsetDefault</code> value</span> is assumed.</li>
<li>The parser is set to ignore invalid characters, so a malformed input
stream will generate glitches (unexpected characters) rather than risk
aborting a parse altogether.</li>
</ol>
<p class="v3">In version 3.0, this remains the default, but
internationalisation support is further improved, and is no longer
limited to the encodings supported by libxml2:</p>
<ul class="v3">
<li>The <code>ProxyHTMLCharsetAlias</code> directive enables server
administrators to support additional encodings by aliasing them to
something supported by libxml2.</li>
<li>When a charset that is neither directly supported nor aliased is
encountered, mod_proxy_html 3 will attempt to support it using Apache/APR's
charset conversion support in <code>apr_xlate</code>, which on most platforms
is a wrapper for the leading conversion utility <code>iconv</code>.
Because of undocumented behaviour of libxml2, this may cause problems
when charset is specified in an HTML <code>META</code> element. This
feature is therefore only enabled when <code>ProxyHTMLMeta</code> is On.</li>
</ul>
<h2 id="meta">meta http-equiv support</h2>
<p>The HTML <code>meta</code> element includes a form
<code><meta http-equiv="Some-Header" contents="some-value"></code>
which should notionally be converted to a real HTTP header by the webserver.
In practice, it is more commonly supported in browsers than servers, and
is common in constructs such as ClientPull (aka "meta refresh").
The <code>ProxyHTMLMeta</code> directive supports the server generating
real HTTP headers from these. However, it does not strip them from the
HTML (except for Content-Type, which is removed in case it contains
conflicting charset information).</p>
<h2 id="misc">Other Fixups</h2>
<p>For additional minor functions of mod_proxy_html, please see the
<code>ProxyHTMLFixups</code> and <code>ProxyHTMLStripComments</code>
directives in the <a href="config.html">Configuration Guide</a>.</p>
<h2 id="debug">Debugging your Configuration</h2>
<p>From Version 2.1, mod_proxy_html supports a <code>ProxyHTMLLogVerbose</code>
directive, to enable verbose logging at <code>LogLevel Info</code>. This
is designed to help with setting up your proxy configuration and
diagnosing unexpected behaviour; it is not recommended for normal
operation, and can be disabled altogether at compile time for extra
performance (see the top of the source).</p>
<p>When verbose logging is enabled, the following messages will be logged:</p>
<ol>
<li>In <strong>Charset Detection</strong>, it will report what charset is
detected and how (HTTP rules, XML rules, or HTML rules). Note that,
regardless of verbose logging, an error or warning will be logged if an
unsupported charset is detected or if no information can be found.</li>
<li>When <code>ProxyHTMLMeta</code> is enabled, it logs each header/value
pair processed.</li>
<li>Whenever a <code>ProxyHTMLURLMap</code> rule matches and causes a
rewrite, it is logged. The message contains abbreviated context information:
<strong>H</strong> denotes an HTML link matched; <strong>E</strong>
denotes a match in a scripting event, <strong>C</strong> denotes a match
in an inline script or stylesheet. When the match is a regexp
find-and-replace, it is also marked as <strong>RX</strong>.</li>
</ol>
<h2 id="browser">Workarounds for Browser Bugs</h2>
<p>Because mod_proxy_html unsets the Content-Length header, it risks
losing the performance advantage of HTTP Keep-Alive. It therefore sets
up HTTP Chunked Encoding when responding to HTTP/1.1 requests. This
enables keep-alive again for HTTP/1.1 agents.</p>
<p>Unfortunately some buggy agents will send an HTTP/1.1 request but
choke on an HTTP/1.1 response. Typically you will see numbers before
and after, and possibly in the middle of, a page. To work around this, set the
<code>force-response-1.0</code> environment variable in httpd.conf.
For example,<br /><code>BrowserMatch MSIE force-response-1.0</code></p>
</div>
<div id="navbar"><a class="internal" href="./" title="Up">Up</a>
*
<a class="internal" href="/" title="WebThing Apache Centre">Home</a>
*
<a class="internal" href="/contact.html" title="Contact WebThing">Contact</a>
*
<a class="external" href="http://www.webthing.com/" title="WebThing Ltd">WebÞing</a>
*
<a class="external" href="http://www.apache.org/" title="Apache Software Foundation">Apache</a></div></body></html>
distrib
>
Mandriva
>
2009.1
>
x86_64
>
media
>
main-release-src
>
by-pkgid
>
a3488e9dc091d3c5f4616db8164cc7c3
>
files
>
5
