On 12/23/05, Cees Hek <ceeshek-Re5JQEeQqe8AvxtiuMwx3w@xxxxxxxxxxxxxxxx> wrote:
> Web servers and web browsers don't need to know about the query
> string. It is just the application that parses the query string that
> needs to support this format (ie your perl scripts).
Yes, I meant application, not server.
> And I am fairly sure that most of the common perl modules on CPAN that
> parse query strings can handle this (at least URI, CGI.pm CGI::Simple
> and CGI::Minimal do). What it comes down to is splitting the query
> string on [&;] instead of just &, so it is not really a major effort
> to support these URLs in a query parser.
Well, right, I know it's not *hard* to do, the question is whether
it's done. I've experimented with this before, in a variety of
languages, and had mixed results. Although a quick test indicates
that (at least) Python and Ruby's standard CGI libraries do the right
thing with ; in addition to &.
So I'll add a package global $Template::Plugin::URL::JOINER and have
it default to ';'. Comments?
--
(darren)
|
