On Mon, Aug 08, 2005 at 01:54:08AM -0700, David Blomstrom wrote:
> --- Brian Candler <B.Candler@xxxxxxxxx> wrote:
>
> > (2) using an external rewriting process. Lookup
> > syntax is the same as above,
> > but you write an external script (e.g. in perl or
> > whatever language you're
> > happy with) which performs the lookup in whatever
> > way you like, e.g. from a
> > hash in memory, from a SQL database etc.
>
> So you're suggesting two separate ways of doing this,
> one with Apache mod_rewrite and (2) with a different
> software program, right? About the only related
> software programs I'm familiar with are PHP and MySQL.
> But from your last sentence, it sounds like this
> couuld be done with a MySQL database.
Not exactly - I'm suggesting doing this with mod_rewrite but with an
external rewriting program. Something like this (untested):
RewriteMap mapper prg:/usr/local/bin/mapper
RewriteCond ${mapper:$1} ^(.+)$
RewriteRule /stack/(.+)$ %1 [R,L]
where /usr/local/bin/mapper is a program like this in a language of your
choice:
#!/usr/bin/perl
%mappit = (
"top_predator" => "/articles/wolf",
"canis_lupus" => "/articles/wolf",
);
while (<>) {
chomp;
$ans = $mappit{$_} || "NULL";
print "$ans\n";
}
The RewriteLogic says:
- the URI is matched against /stack/(.+)$
- If it matches, $1 is set to the parenthesised expression and we continue
processing the RewriteCond
- the RewriteCond passes $1 to the mapper, and the result is matched against
the RHS regexp. The regexp .+ matches any non-empty answer. It is captured
by the parentheses as %1
- If that match succeeded, then the RHS rewrite is done, substituting the
URI with %1, turning it into a redirect, and stopping ruleset processing.
Each Apache worker will start up an instance of the mapper program as a
child, and pass messages to it for the ${mapper:...} lookups. Clearly this
program can use whatever lookup mechanism you like: DBI for SQL lookups, for
instance. You might not want the overhead of that on every lookup, but you
could always select the whole table into a Perl hash, and then answer
queries from the hash as above. After changing the SQL table you'd need to
restart Apache to pick up the new values.
Regards,
Brian.
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx
" from the digest: users-digest-unsubscribe@xxxxxxxxxxxxxxxx
For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|