If you think a new Crypt::Rijndael might break your application, please test it with 0.06_09. Barring anyone complaining, it's going to be an official release after the holidays.
When I took over Crypt::Rijndael I didn't ealize how many people were using it, and I don't want to break any applications with an unintended upgrade some of you might perform.
I think I've fixed most of the architecture issues, and am waiting for more results from CPAN Testers. I'm especially interested in Solaris reports for this release.
In file included from Rijndael.xs:14:
rijndael.h:57: error: conflicting types for 'UINT32'
C:/perl/mingw/bin/../lib/gcc/mingw32/3.4.5/../../../../include/basetsd .h:56: error: previous declaration of 'UINT32' was here
Re:Still getting UINT32 problems on Win32-Strawber
brian_d_foy on 2006-12-17T20:27:57
Could you give me any ideas how to fix those? Was it already declared in basetsd.h?
Although I do want to fix the module for everyone, I'm not going to hold things up for Strawberry Perl since its still experimental. I will try to fix it though.Re:Still getting UINT32 problems on Win32-Strawber
dagolden on 2006-12-18T01:59:49
Do you feel the patch in this post is too much of a hack?
Re:Still getting UINT32 problems on Win32-Strawber
Alias on 2006-12-18T09:15:04
Experimental? Well the packaging mechanisms we are using are still alpha, but the Perl itself and the compiler are bog standard 5.8.8 and MinGW...
There's nothing particularly special about them at all.Re:Still getting UINT32 problems on Win32-Strawber
brian_d_foy on 2006-12-18T17:16:53
I don't know much about Windows, but when I looked at the Strawberry Perl page, it says "Strawberry Perl is currently an alpha release and is not recommended for production purposes." That's what I mean by "experimental".
I'm happy to apply patches, but I don't know why MinGW is acting differently from GNU on unix.Re:Still getting UINT32 problems on Win32-Strawber
BinGOs on 2006-12-19T10:36:13
[22:24] <@tsee> BinGOs, since you have VC++, can you test Crypt::Rijndael 0.06_09 for me after replacing "typedef unsigned long UINT32;" with "typedef unsigned int UINT32;" in the MINGW ifdef in
rijndael.h? Or rather for brian d foy who maintains it now.I did this and it compiles and tests fine on Strawberry Perl with MinGW and dmake.
I also built and tested with Activestate Perl and VC++ toolkit and that compiled fine and tested okay.
