[Resurect] Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-08-07 Thread Philip M. Gollucci
Which of these was the final verdict? I talked to Stas a while ago and he was for committing it which I'd like to do to resolve this issue. Stas Bekman wrote: > Index: xs/Apache2/SubRequest/Apache2__SubRequest.h > === > --- xs/Apac

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-09 Thread Scott Penrose
:-) Stas, that looks like it fixes it... (slight change of code) if (r->main) { modperl_config_req_t *rcfg = modperl_config_req_get(r->main); + if (rcfg->wbucket) MP_RUN_CROAK(modperl_wbucket_flush(rcfg->wbucket, FALSE), "Apache2::SubReq

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-07 Thread Stas Bekman
Scott Penrose wrote: [...] (gdb) bt #0 0xb7bef73d in modperl_wbucket_flush (wb=0x0, add_flush_bucket=0) at modperl_filter.c:236 #1 0xb7a7efaf in mpxs_ap_run_sub_req (r=0x84e1708) at Apache2__SubRequest.h:25 #2 0xb7a7f55b in XS_Apache2__SubRequest_run (cv=0x8351e78) at SubRequest.xs:83

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-06 Thread Scott Penrose
You have two choices, 'modperl' OR 'perl-script'. 'default' is part of httpd core not a mod_perl extension. Thanks but that is correct and my intention. You have many choices, modperl and perl-script is only for perl, default-handler is for the built in default handler (basically files). T

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-05 Thread Philip M. Gollucci
Scott Penrose wrote: > > SetHandler modperl > PerlResponseHandler Zaltana::ExpandTest::response > PerlOutputFilterHAndler Zaltana::ExpandTest > > > SetHandler default > PerlOutputFilterHAndler Zaltana::ExpandTest > > It should be noted that the Subrequest can be anything -

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-05 Thread Scott Penrose
After banging my head all day I have narrowed this down just a little... I had a terrible time working out WHY the Test code works but nothing else does. And I have finally solved that, but not why. The answer is which Handler for the original content being filtered. I added a simple handler

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-05 Thread Scott Penrose
You can't test mod_perl modules from the command line, Scott. Only running under a mod_perl interpreter. Dooh. I didn't know that. Mine have always worked - I guess I just have not come across the thing that does not work :-) Until now. Right, that worked in mod_perl - still seg faults, so

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-05 Thread Stas Bekman
Scott Penrose wrote: On 05/07/2006, at 5:47, Stas Bekman wrote: http://perl.apache.org/docs/2.0/user/handlers/ filters.html#HTTP_Request_vs__Connection_Filters OK Some very weird and interesting problems here. I think they are bugs, but I am not certain. Firstly - if you write the code..

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-04 Thread Scott Penrose
On 05/07/2006, at 5:47, Stas Bekman wrote: http://perl.apache.org/docs/2.0/user/handlers/ filters.html#HTTP_Request_vs__Connection_Filters OK Some very weird and interesting problems here. I think they are bugs, but I am not certain. Firstly - if you write the code... (straight from the exa

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-04 Thread Stas Bekman
Scott Penrose wrote: On 04/07/2006, at 20:53, Scott Penrose wrote: On 04/07/2006, at 17:30, Stas Bekman wrote: Scott Penrose wrote: Is this a request or a connection filter? It won't work as a connection filter (r is not available there). It is an ouptut filter. But what output filter

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-04 Thread Scott Penrose
On 04/07/2006, at 20:53, Scott Penrose wrote: On 04/07/2006, at 17:30, Stas Bekman wrote: Scott Penrose wrote: Is this a request or a connection filter? It won't work as a connection filter (r is not available there). It is an ouptut filter. But what output filter? There are two: http://per

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-04 Thread Scott Penrose
On 04/07/2006, at 17:30, Stas Bekman wrote: Scott Penrose wrote: Is this a request or a connection filter? It won't work as a connection filter (r is not available there). It is an ouptut filter. But what output filter? There are two: http://perl.apache.org/docs/2.0/user/handlers/ filter

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-04 Thread Stas Bekman
Scott Penrose wrote: Is this a request or a connection filter? It won't work as a connection filter (r is not available there). It is an ouptut filter. But what output filter? There are two: http://perl.apache.org/docs/2.0/user/handlers/filters.html#HTTP_Request_vs__Connection_Filters Wha

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-03 Thread Scott Penrose
Is this a request or a connection filter? It won't work as a connection filter (r is not available there). It is an ouptut filter. What's the core's backtrace? I will generate this tomorrow (don't have access to the machine today). Thanks Scott P.S. I have done a temporary work around wh

Re: SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-07-02 Thread Stas Bekman
Scott Penrose wrote: Hey Guys I wrote a simple Filter which calls a subrequest - and started by doing this very simply along the SSI style lines... Please note - that although I could do this on the fly rather than buffer, the end result for my code requires capture of the content, so ju

SubRequest in Handler vs SubRequest in Filter - and Segmentation Fault

2006-06-18 Thread Scott Penrose
Hey Guys I wrote a simple Filter which calls a subrequest - and started by doing this very simply along the SSI style lines... Please note - that although I could do this on the fly rather than buffer, the end result for my code requires capture of the content, so just imagine that bit is