On Sun, Apr 28, 2002 at 10:08:01PM -0400, Jeff wrote: > And a long time coming... > > An assembler supporting keyed aggregates. Because of the current lack of > macro support (that will be added in a few days), it's not built by > default and isn't invoked in the test suite.
Yay! Thank you. > Since keyed aggregate support was waiting on this step, key constructs > will not yet be supported, but this is the first major step in getting > there. Macro support will be added over the next few days, along with a > different backend that the language compilers will be able to use that > will support compilation straight to Parrot bytecode, without a .pasm > file as an intermediary. Hopefully we'll be able to use this for the > tests as well, which should speed things up by quite a bit. Well, one problem -- there's no way to create an aggregate right now, or any other PMC. The assembler needs to be taught about the PMC constants: ./newasm t/pmc/perlhash6.pasm > ph6.pbc Undefined label PerlHash used at line 1 That comes from the line new P0, PerlHash I know, I can replace that with a constant integer, or stop whining and add the support myself, but my parrot time right now is going towards finishing off the memory management problems in the hashtable implementation.
