There's an array of strings initialized in "type.c" that's indexed by type tag. The strings are used for printing a value as "#<...>" when the value has no other output form. A string like "<rational-number>" can show up in `dump-memory-stats`, even though exact non-integer numbers don't use that string for printing.
So, as a first cut, a symbol form of each string would probably make sense as a result from the new function. You might want to make some adjustments, such as 'number instead of 'rational-number, or maybe you want to keep the distinction. Beware of trying to allocate an array of symbols on initialization: New type tags can get allocated at run time (though an unsafe API), and allocating symbols after start-up means that they won't be interned in the space that is shared by across places. So, any symbols allocated after startup will need to be in a place-specific array. It may be simplest to fill a place-local array on demand. You'll probably end up changing "schthread.h" to add a new place-local variable. At Sat, 26 Mar 2016 17:34:40 -0700, Alexis King wrote: > Ok, I’m happy to take a crack at it if you think that’s a good approach. > I am quite comfortable with C, but I have basically no idea how Racket’s > internals work, so that’d be the real barrier. I think one of the > less-technical questions would also just be how to expose this > information in Racket. > > I do think using a number is probably too fragile, but even then, it > doesn’t matter too much, since somehow there needs to be a known set of > values for comparisons. Something like (eq? (type-of some-val) > type:list) needs to be possible, but I’m not sure where type:list should > go or what it should actually be called. > > > On Mar 26, 2016, at 16:01, Jay McCarthy <[email protected]> wrote: > > > > We could expose the C virtual machine's type tags and that would be > > O(1) to a unique number. We could define the number as not being > > stable across executions of Racket or versions, whichever we like > > better. > > > > If you're uncomfortable with the number, we could combine it with an > > O(1) hash lookup from the number to a symbol. > > > > I think an extension like this is achievable for someone with a > > cursory knowledge of C. You should do it Alexis. > > > > -- > > > > The numbers are given here in this enum: > > > > https://github.com/racket/racket/blob/master/racket/src/racket/src/stypes.h > > > > The type tag is just called type: > > > > > https://github.com/racket/racket/blob/master/racket/src/racket/include/scheme.h > #L326` > > > > Jay > > -- > 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/034504BF-6A0B-4158-B389-4DCA0DED6F > FD%40gmail.com. > For more options, visit https://groups.google.com/d/optout. -- 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/56f73db6.c1c1420a.b865e.6995SMTPIN_ADDED_MISSING%40mx.google.com. For more options, visit https://groups.google.com/d/optout.
