using a %f specifier to od_printf()

(* or ?) chars are present.  Doing so prevents re-writing of the path
parameter, so behaviour is different between the two.

This would break something like:
int munlink(char *glob)
{
        char    this[MAX_PATH+1];

        strcpy(this,glob);
        while(fexistcase(this)) {
                unlink(this);
                strcpy(this,glob);
        }
}

Does Synchronet rely on the old behaviour anywhere?

ng.

the file if it doesn't exist (previously, it would always create it).
External JavaScript (XJS) and CGI handler lists are now stored in memory
(using "named string lists") rather than reading/parsing the .ini files for
each request. Using command read_ini_list() function for reading both handler
and mime_types .ini files.

the file if it doesn't exist (previously, it would always create it).

Tracker1's request.
This should allow for competing dynamic HTML solutions based on server-side
JavaScript. Tip: use http_request.real_path for the path to file-to-be-parsed.
Bugfix: SSJS error files used hard-coded extension (instead of configured SSJS
file extension).

regarding Synchronet v3.

echo error messages to status output (default: stderr) as well.

JS_SetThreadStackLimit API function), default is 32K bytes. This should
keep jsexec from crashing due to the ircd recursion bug. It'll log a JS
exception/error and exit (or loop) instead. Use "-S0" to set an unlimited
thread stack size.

cound go negative remaining... which would be returned as a huge
positive.

performance fix was made in write_template().

performance fix was made in write_template().

REPEAT/END REPEAT sections.

Sacrificed some easy customization for a HUGE increase in performance.  On a PIII-450 page rendering went from an average of 10+ seconds to 1 - 1.5 seconds.  (oops synchronet.css was sent on previous comit by mistake)

Sacrificed some easy customization for a HUGE increase in performance.  On a PIII-450 page rendering went from an average of 10+ seconds to 1 - 1.5 seconds.

compile.