Ticket #314 (closed todo: fixed)
.HLL_map doesn't do what we want
| Reported by: | whiteknight | Owned by: | coke |
|---|---|---|---|
| Priority: | normal | Milestone: | 1.1 |
| Component: | core | Version: | |
| Severity: | medium | Keywords: | imcc hll |
| Cc: | Language: | ||
| Patch status: | Platform: | all |
Description
See #216 for more background information. The .HLL_map directive doesn't really do what we want it to do: It can't be used to map HLL types defined at runtime, only HLL types that are defined at compile time. As a general workaround, you can use the $P0.'hll_map'() method on the ParrotInterpreter PMC to map classes defined at runtime.
I don't think it's a good idea to have a .HLL_map directive that silently fails depending on where the class is defined. I think there are two solutions to this:
1) Deprecate .HLL_map, and use ParrotInterpreter.'hll_map'() exclusively. 2) Work on some kind of cache that .HLL_map can use to store mapped type names that aren't known at compile time, and get resolved to a class type at runtime after that type has been properly defined.
Either way, the current implementation of .HLL_map is too confusing and unhelpful for what we need, and we need to get it right by 1.0.
