<?xml version="1.0" encoding="UTF-8" ?>

<!DOCTYPE manualpage SYSTEM "/style/manualpage.dtd">

<?xml-stylesheet type="text/xsl" href="/style/manual.en.xsl"?>

<manualpage metafile="rewritemap.xml.meta">

<parentdocument href="./">Rewrite</parentdocument>

<title>Using RewriteMap</title>

<summary>

<p>This document supplements the <module>mod_rewrite</module>

<a href="/mod/mod_rewrite.html">reference documentation</a>. It describes

the use of the <directive module="mod_rewrite">RewriteMap</directive> directive,

and provides examples of each of the various <code>RewriteMap</code> types.</p>

<note type="warning">Note that many of these examples won't work unchanged in your

particular server configuration, so it's important that you understand

them, rather than merely cutting and pasting the examples into your

configuration.</note>

</summary>

<seealso><a href="/mod/mod_rewrite.html">Module documentation</a></seealso>

<seealso><a href="intro.html">mod_rewrite introduction</a></seealso>

<seealso><a href="remapping.html">Redirection and remapping</a></seealso>

<seealso><a href="access.html">Controlling access</a></seealso>

<seealso><a href="vhosts.html">Virtual hosts</a></seealso>

<seealso><a href="proxy.html">Proxying</a></seealso>

<seealso><a href="advanced.html">Advanced techniques and tricks</a></seealso>

<seealso><a href="avoid.html">When not to use mod_rewrite</a></seealso>

<section id="introduction"><title>Introduction</title>

<p>The <directive>RewriteMap</directive> directive defines a

<em>Rewriting Map</em> which can be used inside rule

substitution strings by the mapping-functions to

insert/substitute fields through a key lookup. The source of

this lookup can be of various types.</p>

<p>The <a id="mapfunc" name="mapfunc"><em>MapName</em></a> is

the name of the map and will be used to specify a

mapping-function for the substitution strings of a rewriting

rule via one of the following constructs:</p>

<p class="indent">

<strong><code>${</code> <em>MapName</em> <code>:</code>

<em>LookupKey</em> <code>}</code><br />

<code>${</code> <em>MapName</em> <code>:</code>

<em>LookupKey</em> <code>|</code> <em>DefaultValue</em>

<code>}</code></strong>

</p>

<p>When such a construct occurs, the map <em>MapName</em> is

consulted and the key <em>LookupKey</em> is looked-up. If the

key is found, the map-function construct is substituted by

<em>SubstValue</em>. If the key is not found then it is

substituted by <em>DefaultValue</em> or by the empty string

if no <em>DefaultValue</em> was specified.</p>

<p>For example, you might define a

<directive>RewriteMap</directive> as:</p>

<example>

RewriteMap examplemap txt:/path/to/file/map.txt

</example>

<p>You would then be able to use this map in a

<directive>RewriteRule</directive> as follows:</p>

<example>

RewriteRule ^/ex/(.*) ${examplemap:$1}

</example>

<p>The following combinations for <em>MapType</em> and

<em>MapSource</em> can be used:</p>

</section>

<section id="txt"><title>txt: Plain text maps</title>

<p>MapType: <code>txt</code>, MapSource: Unix filesystem

path to valid regular file</p>

<p>This is the standard rewriting map feature where the

<em>MapSource</em> is a plain ASCII file containing

either blank lines, comment lines (starting with a '#'

character) or pairs like the following - one per

line.</p>

<p class="indent">

<strong><em>MatchingKey</em>

<em>SubstValue</em></strong>

</p>

<example><title>Example</title>

<pre>

##

## map.txt -- rewriting map

##

Ralf.S.Engelschall rse # Bastard Operator From Hell

Mr.Joe.Average joe # Mr. Average

</pre>

</example>

<example>

RewriteMap real-to-user txt:/path/to/file/map.txt

</example>

</section>

<section id="rnd"><title>Randomized Plain Text</title>

<p>MapType: <code>rnd</code>, MapSource: Unix filesystem

path to valid regular file</p>

<p>This is identical to the Standard Plain Text variant

above but with a special post-processing feature: After

looking up a value it is parsed according to contained

``<code>|</code>'' characters which have the meaning of

``or''. In other words they indicate a set of

alternatives from which the actual returned value is

chosen randomly. For example, you might use the following map

file and directives to provide a random load balancing between

several back-end server, via a reverse-proxy. Images are sent

to one of the servers in the 'static' pool, while everything

else is sent to one of the 'dynamic' pool.</p>

<p>Example:</p>

<example><title>Rewrite map file</title>

<pre>

##

## map.txt -- rewriting map

##

static www1|www2|www3|www4

dynamic www5|www6

</pre>

</example>

<example><title>Configuration directives</title>

RewriteMap servers rnd:/path/to/file/map.txt<br />

<br />

RewriteRule ^/(.*\.(png|gif|jpg)) http://${servers:static}/$1

[NC,P,L]<br />

RewriteRule ^/(.*) http://${servers:dynamic}/$1 [P,L]

</example>

</section>

<section id="dbm"><title>DBM Hash File</title>

<p>MapType:

<code>dbm[=<em>type</em>]</code>, MapSource: Unix filesystem

path to valid regular file</p>

<p>Here the source is a binary format DBM file containing

the same contents as a <em>Plain Text</em> format file, but

in a special representation which is optimized for really

fast lookups. The <em>type</em> can be sdbm, gdbm, ndbm, or

db depending on <a href="/install.html#dbm">compile-time

settings</a>. If the <em>type</em> is omitted, the

compile-time default will be chosen.</p>

<p>To create a dbm file from a source text file, use the <a

href="/programs/httxt2dbm.html">httxt2dbm</a> utility.</p>

<example>

$ httxt2dbm -i mapfile.txt -o mapfile.map

</example>

</section>

<section id="int"><title>Internal Function</title>

<p>

MapType: <code>int</code>, MapSource: Internal Apache httpd

function</p>

<p>Here, the source is an internal Apache httpd function.

Currently you cannot create your own, but the following

functions already exist:</p>

<ul>

<li><strong>toupper</strong>:<br />

Converts the key to all upper case.</li>

<li><strong>tolower</strong>:<br />

Converts the key to all lower case.</li>

<li><strong>escape</strong>:<br />

Translates special characters in the key to

hex-encodings.</li>

<li><strong>unescape</strong>:<br />

Translates hex-encodings in the key back to

special characters.</li>

</ul>

</section>

<section id="prg"><title>External Rewriting Program</title>

<p>

MapType: <code>prg</code>, MapSource: Unix filesystem

path to valid regular file

</p>

<p>Here the source is a program, not a map file. To

create it you can use a language of your choice, but

the result has to be an executable program (either

object-code or a script with the magic cookie trick

'<code>#!/path/to/interpreter</code>' as the first

line).</p>

<p>This program is started once, when the Apache httpd server

is started, and then communicates with the rewriting engine

via its <code>stdin</code> and <code>stdout</code>

file-handles. For each map-function lookup it will

receive the key to lookup as a newline-terminated string

on <code>stdin</code>. It then has to give back the

looked-up value as a newline-terminated string on

<code>stdout</code> or the four-character string

``<code>NULL</code>'' if it fails (<em>i.e.</em>, there

is no corresponding value for the given key).</p>

<p>This feature utilizes the <code>rewrite-map</code> mutex,

which is required for reliable communication with the program.

The mutex mechanism and lock file can be configured with the

<directive module="core">Mutex</directive> directive.</p>

<p>External rewriting programs are not started if they're defined in a

context that does not have <directive>RewriteEngine</directive> set to

<code>on</code></p>.

<p>A trivial program which will implement a 1:1 map (<em>i.e.</em>,

key == value) could be:</p>

<example>

<pre>

#!/usr/bin/perl

$| = 1;

while (&lt;STDIN&gt;) {

# ...put here any transformations or lookups...

print $_;

}

</pre>

</example>

<p>But be very careful:</p>

<ol>

<li>``<em>Keep it simple, stupid</em>'' (KISS).

If this program hangs, it will cause Apache httpd to hang

when trying to use the relevant rewrite rule.</li>

<li>A common mistake is to use buffered I/O on

<code>stdout</code>. Avoid this, as it will cause a deadloop!

``<code>$|=1</code>'' is used above, to prevent this.</li>

</ol>

</section>

<section id="dbd"><title>SQL Query</title>

<p>MapType: <code>dbd</code> or <code>fastdbd</code>,

MapSource: An SQL SELECT statement that takes a single

argument and returns a single value.</p>

<p>This uses <module>mod_dbd</module> to implement a rewritemap

by lookup in an SQL database. There are two forms:

<code>fastdbd</code> caches database lookups internally,

<code>dbd</code> doesn't. So <code>dbd</code> incurs a

performance penalty but responds immediately if the database

contents are updated, while <code>fastdbd</code> is more

efficient but won't re-read database contents until server

restart.</p>

<p>If a query returns more than one row, a random row from

the result set is used.</p>

<example>

<title>Example</title>

RewriteMap myquery "fastdbd:SELECT destination FROM rewrite WHERE source = %s"

</example>

</section>

<section id="summary"><title>Summary</title>

<p>The <directive>RewriteMap</directive> directive can occur more than

once. For each mapping-function use one

<directive>RewriteMap</directive> directive to declare its rewriting

mapfile. While you cannot <strong>declare</strong> a map in

per-directory context it is of course possible to

<strong>use</strong> this map in per-directory context. </p>

<note><title>Note</title> For plain text and DBM format files the

looked-up keys are cached in-core until the <code>mtime</code> of the

mapfile changes or the server does a restart. This way you can have

map-functions in rules which are used for <strong>every</strong>

request. This is no problem, because the external lookup only happens

once!

</note>

</section>

</manualpage>