On Mon May 07 06:49:42 2007, codermattie wrote: > Hello, > > I noticed while debugging my patches to the loader that the message > for a failure in .load_bytecode was really > horrible. It was something like this: > > immc: Couldn't find file %s > > At the time I didn't know what immc was, and the message didn't really > tell me much more than some > deep guts of parrot couldn't load a file anymore. > > In this patch I try and clarify things a bit by adding to the message > something they can associate > with a parrot op , load_bytecode. > > Hopefully the message will now be: > > immc: "load_bytecode" couldn't find file %s > > Which helps alot for a person that is using the API but isn't familiar > with the architecture. > > The one test that failed repeatably was t/op/load_bytecode.t . I fixed > that (in the patch) > but I still got that mysterious failure in t/perl/Parrot_Docs.t which > I can't reproduce with > prove. > > here is the output: > > t/perl/Parrot_Docs............................... > t/perl/Parrot_Docs...............................ok 1/25 > t/perl/Parrot_Docs...............................ok 2/25 > # Failed test (t/perl/Parrot_Docs.t at line 61) > Use of uninitialized value in pattern match (m//) at > t/perl/Parrot_Docs.t line 65. > > That could be from my copy -R HEAD BASE which is duplicating svn meta- > data the wrong > way ? not sure. > > If your harness is OK, and I haven't broken any error report > convention please apply. > > Cheers, > Mike Mattie - codermattie <!-- x --> at gmail.com
Thanks! Applied as 18451. Note that the patch of the test was based the wrong way around, and this was all that had to be changed to make the patch apply, run and test correctly.
