Hello, I have been working on a series of cleanup patches that lay the ground towards improving the install target, and feature enhancing .include, and load_bytecode.
The patches have been accepted and up to this point do not change parrot's behavior for existing code. I have laid out a two-step of proposal, in in rt # 41908 ( please read the ticket for the proposals ). This mail is a pointer to that RT. link to a text/plain of the message body below: http://rt.perl.org/rt3/Ticket/Attachment/230303/107114/ The first proposal regards an environment variable PARROT_LOAD_PREFER that can be accepted or rejected independent of the second which is a much larger scale change, and slightly more intrusive. I would like feedback at this point before moving forward on the proposals. I am not welded to these proposals, and even for my own learning I would like to learn how other dev's would approach the problems I am working on. a closing section expands on what is needed to insulate the src/library.c API properly, and fix a minor race. This insulation is not linked to the proposals technically. My knowledge of parrot is still very limited, to limited to gauge the impact across the rest of the tree, so I will hold fire before making up patches. Cheers, Mike Mattie ([EMAIL PROTECTED])
signature.asc
Description: PGP signature
