[Xitami] Re: SSI Implementation: buggy?

[Paul Reid]

> "Is Xitami's SSI implimentation buggy or incomplete?", 

Are we talking 2.4 or 2.5?

2.4's SSI is stated to be incomplete, but is just a Perl script which
you could edit to add any feature you need. 

> I have reviewed several SSI tutorials from the net... the set command is not described very often,

I'm no SSI expert, but is it possible that SSI set is something invented
by Apache? Considering Apache's dominance in the webserver world,
tutorials may not bother to mention if something works only on Apache. 

For what it's worth: "set" is NOT documented on the NCSA site's SSI
page, the birthplace of CGI and SSI. 

http://hoohoo.ncsa.uiuc.edu/docs/tutorials/includes.html

The NCSA site is referenced in Xitami 2.4's docs at /xitami/INDEX6.HTM,
which does not list "set" but adds: "With a small knowledge of Perl, you
can easily modify the Perlssi filter to add support for other tags." 

Obviously if you don't want to hack some Perl, and nobody here has a
pre-tested PerlSSI "set" modification you can use, that's no use. 

Xitami 2.5 has in-built SSI, which I have not looked at. If "set" really
is a commonly used extension of SSI, maybe Ewen should know. 

> I recognize that Apache also is not the be-all and end-all of 
webserver software

Well, but it IS the dominant server by a good margin. Each of us has our
own reasons for picking a server; I'd say that a need for 99.99% Apache
compatibility is a good reason to use Apache. They have ports to most
platforms. All free. 

They say the Windows port may be unstable, but it appears to be as
stable as anything on Windows can be. I think they disclaim it just so
they don't tarnish the reliability reputation Apache has running on
clean unix boxes. Certainly WinApache should be stable enough for hard
development duty. And it isn't significantly bigger than Xitami; the
main drawback is arcane configuration files to edit. But I had the
basics working pretty easy. 

  -PRR