>>
> 
> Hmmm, perhaps the Makefile WITH_DB is not added as -DWITH_DB for some 
> obscure make(1) reason.

Both removing the ifdef, or passing -DWITH_DB at compile time works for me.

> 
>> Of course now I'm having other problems in the perl directory:
>>
>> cc -c  -I../. -I../build -I../lib -I../popt -I../rpmdb -I../rpmio 
>> -I../rpmconstant -I/usr/local/include -I/usr/local/include/python2.4 
>> -I/usr/local/include/beecrypt -I/usr/local/ssl/include 
>> -DRPM_OS_SOLARIS=020900 -I/usr/local/include/neon   -g -xs 
>> -D_GNU_SOURCE -D_REENTRANT -D_LARGEFILE64_SOURCE -DNE_LFS   -g   
>> -DVERSION=\"5.0a3\" -DXS_VERSION=\"5.0a3\" -KPIC 
>> "-I/usr/perl5/5.6.1/lib/sun4-solaris-64int/CORE"   RPM_Dependencies.c
>> "../rpmdb/rpmtag.h", line 104: warning: enumerator value overflows 
>> INT_MAX (2147483647)
> 
> Hmm, this is likely
>         RPM_MASK_RETURN_TYPE        = 0xffff0000
> 
> A simpler
>          #define RPM_MASK_RETURN_TYPE    0xffff0000
> outside of the enum likely fixes.
> 

I was focused on the beecrypt issue, but this works great, thanks for 
the fix.


"../lib/rpmcli.h", line 157: warning: integer overflow detected: op "<<"
> 
> Hmmm, this one is a bit harder, I'm out of bits.
> 
>     RPMVERIFY_LGETFILECONFAIL   = (1 << 31)     /*!< lgetfilecon failed */
> 
> You can simply nuke for the moment, its a SELinux peculier failure bit.
> 
> 

If it's SELinux then you're right, we don't care much about it ;)


>> "RPM_Spec.xs", line 69: warning: implicit function declaration: _newspec
>> "RPM_Spec.xs", line 182: warning: implicit function declaration: 
>> _specbuild
>> "../config.h", line 706: warning: macro redefined: _FILE_OFFSET_BITS
> 
> Hmmm, I'm not sure what this is. Smells like config.h included twice,
> likely in RPM_Files.xs.
> 
>> "/usr/local/include/beecrypt/gnu.h", line 39: warning: no explicit 
>> type given
>> "/usr/local/include/beecrypt/gnu.h", line 39: syntax error before or 
>> at: bc_thread_t
>> "/usr/local/include/beecrypt/gnu.h", line 39: warning: old-style 
>> declaration or incorrect type for: bc_thread_t
>> "/usr/local/include/beecrypt/api.h", line 59: warning: macro 
>> redefined: assert
>> "../rpmio/rpmio_internal.h", line 288: syntax error before or at: :
>> "../rpmio/rpmio_internal.h", line 291: cannot recover from previous 
>> errors
>> cc: acomp failed for RPM_Files.c
>>
> 
> Yah, RPM_Files.xs is the problem.
> 
> There's likely a few messy defines that RPM_Files.xs is awkwardly trying to
> identify from config.h because rpmio_internal.h. needs.
> 
> I added these lines in the past couple of days
> 
> /* XXX artifacts needed to include "rpmio_internal.h" w/o "system.h" */
> void * vmefail(size_t size);
> #define xstrdup(_str)   (strcpy((malloc(strlen(_str)+1) ? : 
> vmefail(strlen(_str)+1)), (_str)))
> 
> mostly because I was too lazy to wrestle with rpmio_internal.h right then.
> 
> Likely all that is needed by RPM_File.xs from config.h is this:
> 
> #ifdef HAVE_BEECRYPT_API_H
> #include <beecrypt/api.h>
> #endif
> 
> 73 de Jeff
> 


Worked great again! Outside of the small issue I had with linking in 
rpmconstant, the package finally finishes building... only took me till 
alpha three. haha...

I look forward to actually trying out the functionality, especially when 
it's officially beta. Thanks again for your help with these bugs Jeff. 
Solaris (especially older releases like 9) always have has its own 
quirks, but actually this was a pretty painless build considering most 
of the fixes are trivial.

-Dave