Jarkko wrote: !> I should point out that in order to get perl@15489 to build I had !> to patch ext/Encode/Encode.xs but I have reason to believe ! !Ummm, care to tell how did you have to patch Encode.xs so that we !can verify that the right patch is in?
The discussion began at: http://www.xray.mpe.mpg.de/mailing-lists/vmsperl/2002-03/msg00117.html where I had a possibly incorrect suggestion of adding a cast. Craig Berry pointed out that Encode 0.99 would have a cast sans the const bit in the message at: http://www.xray.mpe.mpg.de/mailing-lists/vmsperl/2002-03/msg00120.html I have not yet had time to test build without the const... pending (on a slow machine) Jarkko further wrote: !> Since the "all" target is now required so as to avoid the fatal MMS error. ! !Sooo... do you have suggestions of how to unenshrine the bug? !Or is the change really something VMSers have to live with? There should be at least two ways: 0) revert ExtUtils::MakeMaker, MM_VMS, et alia back to a revision when 'all' was not necessary (I suspect, but have not been able to verify that perl@15300 might work on VMS). 1) coerce the output of the new MakeMaker to put the important 'all :' target first. In the meantime I've found a problem with configure.com and am testing a patch to it and ext/Encode/Encode.xs. I cannot promise much headway on the MakeMaker issue today. Peter Prymmer
