Re: GerH source distro at hebbut.net: HOW TO compile/compare/merge for Bill et al

-------- Original-Nachricht --------
> Datum: Thu, 27 Mar 2008 03:58:20 +0100
> Von: "Ger Hobbelt" <[email protected]>
> An: [email protected]
> Betreff: Re: [Crm114-general] GerH source distro at hebbut.net: HOW TO	compile/compare/merge for Bill et al

> On Thu, Mar 27, 2008 at 1:47 AM, Steve <[email protected]> wrote:
> >  [snip]
> >  >   ./configure && make && make test
> >  >
> >  >
> >  Heureka! I did it! Got now a full working ebuild on Gentoo using your
> new build system :)
> >
> >  Had to patch src/crm114_sysincludes.h and src/crmregex_tre.c in order
> to get everything working as expected. I did not fiddle long around about
> the error in regex_t not having a value member. I just removed the (anyway
> informational) part of the code:
> 
> Excellent!
> 
> Q: You said you had to fiddle sysincludes.h? Can I see what you did?
> 
> Note: the source tree to be published next (ETA 1 hour, so multiply
> that by 24 ;-) ) should have a slightly changed TRE detection which
> _MIGHT_ <cross-fingers /> work out okay on your box. Nevertheless,
> it's a long shot, but if you are willing to spend some time on testing
> this for me later, I'll buy you a beer next time I cross the border
> (assuming you live somewhere in Germany/Austria, or at least Europe,
> given your usage of German and GMX ;-) )
> 
> As before, I'm VERY interested in those config.* files I meantioned:
> those contain the log of what ./configure found on your sandbox, and
> it might provide me with that much needed bit of insight.
> 
> 
> 
> 
> >  --- src/crmregex_tre.c  2008-03-25 17:24:38.000000000 +0100
> >  +++ src/crmregex_tre.c.new      2008-03-26 22:19:25.999327250 +0100
> >  @@ -281,8 +281,8 @@
> >          memcpy(preg, ppreg_temp, rtsize);
> >          if (internal_trace)
> >              fprintf(stderr, " About to return:\n"
> >  -                            " preg->re_nsub=%d preg->value=%p, i=%d,
> status=%d\n",
> >  -                (int)preg->re_nsub, preg->value, i,
> regex_cache[i].status);
> >  +                            " preg->re_nsub=%d, i=%d, status=%d\n",
> >  +                (int)preg->re_nsub, i, regex_cache[i].status);
> >          return regex_cache[i].status;
> >      }
> >   #endif //CRM_REGEX_CACHESIZE == 0
> >
> >
> >
> >  The ebuild is still "hackerish" but at least I can build and install :)
> >
> >  Thanks for that!
> 
> 
> You're welcome!
> 
> Note that I'll wait with applying this patch (useful in itself so
> others will have an easier time building and using this package on
> their systems) after I get a chance to get the configure script TRE
> detection working properly. I hate to know there's a bug in there
> without the option of getting to kill it.
> 
> 
> 
> >  I almost forgot! GCC version:
> [...]
> >  Because of the gcc version I had as well to patch normalizemime in
> order to be able to compile it with gcc 4.3.1:
> 
> heh. Sorry, I don't support normalizemime  ;-P
> 
> They should add sys/types.h detection to their configure script
> though, as it's not available on all boxes. So something like
> 
> #ifdef HAVE_SYS_TYPES_H
> #include <sys/types.h>
> #endif
> 
> would be the clean way to go - in conjunction with an adjusted
> configure.ac script.
> 
> 
> 
> I
> >  --- normalizemime.cc    2007-03-20 06:46:18.000000000 +0100
> >  +++ normalizemime.cc.new        2008-03-24 14:56:09.820275709 +0100
> >  @@ -58,6 +58,8 @@
> >   #include <map>
> >   #include <string>
> >
> >  +#include <sys/types.h>
> >  +
> >   #include <regex.h>
> >   #include <assert.h>
> >   #include <ctype.h>
> [...]
> 
> 
> >  > NOTE: I have NOT tested 'make install' as I consider that a security
> >  > issue: I run an *experimental* crm-filter setup and development on a
> >  > single box, so I got to be careful there. Sorry.
> >  >
> >  I did quickly. On my Gentoo box 'make install' is violating the sandbox
> by trying to write to /usr/share/emacs/site-lisp/crm114-mode.el even if I
> told make install to use DESTDIR=....
> 
> Argh. Cruft leftover from where-ever.
> 
> Thanks for testing this for me!
> 
> 
> 
> Take care and thanks again for all the the testing and reporting!
> 
I think I found another problem. When using --with-regex-libraries=<path> then "-l<path>" gets added instead of "-L<path>". Could that be?


> -- 
> Met vriendelijke groeten / Best regards,
> 
> Ger Hobbelt
> 
// Steve

> --------------------------------------------------
> web: http://www.hobbelt.com/
>  http://www.hebbut.net/
> mail: [email protected]
> mobile: +31-6-11 120 978
> --------------------------------------------------
> 
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> Crm114-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/crm114-general

-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace