Now stop to discuss this problem. If there is anything left not merged
in HEAD then please send a patch. If not, then the problem is solved,
end of the discussion.There will be, because obviously that code is working in 5_3.
Sorry to harp on about this, but I still can't break it here. No idea why it
should have failed in HEAD on the snaps box. The file was 100% identical to
the 5_3 version until you committed your changes, and the 5_3 snaps aren't
broken. And the point of reported breakage is even weirder - it comes on an
equivalent line to AC_DEFINE().
Does anyone know the conditions that broke the script in HEAD, e.g. which
extension/what was missing? Even throwing away the libxml libraries doesn't
cause an error here.
Thanks,
- Steph
Now stop to discuss this problem. If there is anything left not merged
in HEAD then please send a patch. If not, then the problem is solved,
end of the discussion.There will be, because obviously that code is working in 5_3.
Sorry to harp on about this, but I still can't break it here. No idea why it
should have failed in HEAD on the snaps box. The file was 100% identical to
the 5_3 version until you committed your changes, and the 5_3 snaps aren't
broken. And the point of reported breakage is even weirder - it comes on an
equivalent line to AC_DEFINE().Does anyone know the conditions that broke the script in HEAD, e.g. which
extension/what was missing? Even throwing away the libxml libraries doesn't
cause an error here.
Easiest way to reproduce the runtime error in confutil is to empty the
lib/ dir and leave only resolv's files.
Cheers,
Pierre
Does anyone know the conditions that broke the script in HEAD, e.g. which
extension/what was missing? Even throwing away the libxml libraries
doesn't
cause an error here.Easiest way to reproduce the runtime error in confutil is to empty the
lib/ dir and leave only resolv's files.
Heh...
Checking for arpa\nameser.h ... ..\bindlib_w32
Checking for library resolv_a.lib ... <not found>
Checking for library resolv.lib ... <not found>
ERROR: We really need that arpa\nameser.h file - it is part of the bindlib
package
Actually with that fixed, the thing still works for me in HEAD. Sounds
like a good idea but... doesn't help here.
I'm going to drop this for now and investigate when I've time. But it really
wasn't a nasty merging accident (unlike the other one), it's just a weird
and - for me at least - unreproducible error.
- Steph
Does anyone know the conditions that broke the script in HEAD, e.g. which
extension/what was missing? Even throwing away the libxml libraries
doesn't
cause an error here.Easiest way to reproduce the runtime error in confutil is to empty the
lib/ dir and leave only resolv's files.Heh...
Checking for arpa\nameser.h ... ..\bindlib_w32
Checking for library resolv_a.lib ... <not found>
Checking for library resolv.lib ... <not found>
ERROR: We really need that arpa\nameser.h file - it is part of the bindlib
packageActually with that fixed, the thing still works for me in HEAD. Sounds
like a good idea but... doesn't help here.I'm going to drop this for now and investigate when I've time. But it really
wasn't a nasty merging accident (unlike the other one), it's just a weird
and - for me at least - unreproducible error.
It was broken, it works now, problem solved, end of the discussion.
Waiting for your patches if you need something else in HEAD.
--
Pierre
http://blog.thepimp.net | http://www.libgd.org
It was broken, it works now, problem solved, end of the discussion.
Well not really, because HEAD and 5_3 are now different and the fact that we
don't know why it failed in HEAD means we don't know if there's a problem
in 5_3.
Waiting for your patches if you need something else in HEAD.
Oh please. I missed ONE line in ONE merge. That is not a reason to play this
game.
- Steph