[cgiapp] Re: CAP::DebugScreen and ErrorDocument

[Mark Stosberg <suppressed>]

On 2005-12-24, RA Jones <suppressed> wrote:
>
> [error] [client 127.0.0.1] [Sat Dec 24 16:56:58 2005] myapp.cgi: \t(in 
> cleanup) Can't call method "commit" on unblessed reference at 
> C:/usr/local/site/lib/CGI/Session/Driver/DBI.pm line 122 during global 
> destruction., referer: 
> http://127.0.0.1/cgi-bin/MyApp/myapp.cgi?rm=display_record;id=4
>
> [error] [client 127.0.0.1] DESTROY created new reference to dead object 
> 'CGI::Session::Driver::mysql' during global destruction., referer: 
> http://127.0.0.1/cgi-bin/MyApp/myapp.cgi?rm=display_record;id=4

I also use CGI::Session and get messages like this too...but I was
ignoring them. :( . 

> These do not appear if 'use CAP::DebugScreen' is commented out. And I 
> still get the 500 Internal Server Error page rather than 
> CAP::DebugScreen functionality if I deliberately introduce a run-time 
> fatality (eg calling non-existent template in a runmode).

You should get the 500 error for compile time issues, but the
DebugScreen for run time issues. That part works for me. 

> CGI::Carp::DebugScreen works fine in the same application if used in 
> place of CAP::DebugScreen. Hope it's not just me.

It's not. There's not a whole lot benefit of the CAP integrated version 
anyway, besides that it integrates with ViewCode. It's a nice feature,
but doesn't sound like it is worth these other problems. 

    Mark

-- 
http://mark.stosberg.com/ 


---------------------------------------------------------------------
Web Archive:  http://www.mail-archive.com/suppressed/
              http://marc.theaimsgroup.com/?l=cgiapp&r=1&w=2
To unsubscribe, e-mail: suppressed
For additional commands, e-mail: suppressed