>
> That seems like an ok approach. Using scheme_register_process_global() 
> to record the type registration might be better, especially if you're 
> concerned about conflicts. 
>

i dont think so. there still can be race, because  type_names is not 
"protected' by any semaphore.
and even if it will be protected - it will require programmer to know about 
it and write something like
"get lock, call scheme_register_process_global, if it returns null - call 
scheme_make_type, release lock".
not so "user friendly", imho =)
also having dynamicaly growable type_names and static mark_table is... hm. 
"surprising", at least.
and this behavior isn't described anywhere.

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-dev/9992d4cc-38b8-4902-8b1e-b716de6dddf6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to