I'm still trying to get Crypt::Rijndael to compile correctly everywhere *and* pass all the tests. I've had some reports of it compiling just fine on Solaris (using gcc and suncc), but then not doing the right thing. As such, I've added a test to encrypt a string then decrypt it to see if it comes out the same way. I need some people with access to Solaris (any version) to run Crypt::Rijndael-1.04_02 through CPAN.pm with CPAN::Reporter (and some other thing that reports to CPAN Testers).
The latest development release doesn't show any failures yet, but there is only one test for Solaris (2.9 using gcc).
Previously I was testing this with the Sourceforge compiler farm, but that's no longer available (and how much more sucky can Sourceforge get?). I've been looking for a nice VMWare image of Solaristoo
Here's what I see.... (Score:1)
Re: (Score:2)
UltraSparc T1 (Niagara) (Score:1)
All the tests passed but I got a few warnings.
Rijndael.xs: In function 'XS_Crypt__Rijndael_new':
Rijndael.xs:120: warning: pointer targets in passing argument 3 of 'rijndael_set up' differ in signedness
Rijndael.xs: In function 'XS_Crypt__Rijndael_encrypt':
Rijndael.xs:162: warning: pointer targets in passing argument 4 of '(ix) != 0l ?
Re: (Score:1)
Solaris 8 (Score:1)
I worked with you on getting Crypt::Rijndael to build on Solaris using the Sun Studio 10 C Compiler a while back so I figured I might part of your target audience. I'm not familiar with CPAN Testers (and don't have the time to familiarize myself with it right now), but I'm hoping this is still useful to you:
[user@unixbox1:/home/user/temp/Crypt-Rijndael-1.04_02]> uname -a
SunOS unixbox1 5.8 Generic_117350-49 sun4u sparc SUNW,Ultra-250
[user@unixbox1:/home/user/temp]> gunzip -cd Crypt-Rijndael-
Pure Perl (Score:2)
Re: (Score:2)
Re: (Score:2)
This also led me to figure out the other signedness warnings I've seen on some platforms. It compiles cleanly now (so far), but we'll see what the testers see now.
Thanks,
Solaris vmware and qemu images (Score:1)
It works (Score:2)