httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ruediger Pluem <rpl...@apache.org>
Subject Re: segfault in dav_validate_request
Date Mon, 31 Dec 2007 14:21:13 GMT


On 12/31/2007 02:26 AM, Nick Kew wrote:
> On Sun, 30 Dec 2007 09:22:28 +0800
> Michael Clark <michael@metaparadigm.com> wrote:
> 
>> I'm getting a segfault here in mod_dav from trunk
> 
> Oops ... more haste less speed:-(
> 
> I attach a revised patch (against current trunk).
> This is completely untested; just posting before going to bed,
> in case anyone feels like picking it up.  If not, I'll try
> and find time to revisit it tomorrow.
> 

I think the following patch against trunk which is based on yours
does it better, because:

- It does not set the ETag header permanently as setting it may not
  be desired for all responses. If we want to do this for more methods
  as we currently do we should do this explicitly in dav_method_* in
  mod_dav.c.
- Handles the case that hooks->getetag returns an empty ETag ("") like
  dav_fs_getetag does for non existing resources.
- Does remove the conditional between calling dav_meets_conditions and
  ap_meets_conditions. To be honest I do not understand why we should not
  call dav_meets_conditions in any case. Especially in the '*' case this
  make sense IMHO as we do not need to do a comparison to an existing
  ETag of the resource in this case. It is only of interest in this case
  whether the resource exists and not what its ETag is.


Passed perl test framework and litmus.

Regards

RĂ¼diger

Mime
View raw message