Hi folks,
Am 12.01.2014 um 20:25 schrieb ph10@???:
> On Sun, 12 Jan 2014, I wrote:
>> The code in the script is assuming that the existence of a 'fr' locale
>> implies a Windows system, and uses different input files which have
>> "french" screwed into them. I think this must be a hangover from an
>> earlier version where the script couldn't cope with different names for
>> the French locale. The current version edits the test files according to
>> which locale name exists. I will clean it all up and let's see what
>> happens.
>
> I have just committed a patch that extends the locale test (test 3). As
> well as handling the three possible names for the French locale (fr_FR,
> french, and fr) it now has three different output files. The test passes
> if any one of the three output files matches. I hope this will be enough
> to cope with slight differences in different implementations of French
> locales and thereby make test 3 more robust. More outputs can be added
> if any other variations turn up. Now to see if the buildbot is happy…
One more issue is that I did not clean up the build directory between
builds, it was essentially an svn update followed by configure, make,
make check. Now that I do a full clean inbetween everything passes.
Maybe a dependency is missing in the Makefile, but I can happily live
with rebuilding every time.
The build is almost perfect, just the one on Solaris sparcv9 with 64 bit
and JIT enabled dumps core:
https://buildfarm.opencsw.org/buildbot_admin/builders/pcre-jit-solaris10-sparcv9/builds/0
Best regards
— Dago
--
"You don't become great by trying to be great, you become great by wanting to do something,
and then doing it so hard that you become great in the process." - xkcd #896
