Commit Graph

95 Commits

Author SHA1 Message Date
Brian Anderson
ce750a7dbc Box AST idents 2012-06-13 11:30:45 -07:00
Tim Chevalier
88f03743b0 Comments only: annotate FIXMEs 2012-06-07 15:13:35 -07:00
Michael Sullivan
d542e67827 Do some cleanup of the allocation code. 2012-06-06 18:26:47 -07:00
Niko Matsakis
01b5777c8b prohibit type parameters in native fns and other minor fixes
trans now can safely assert that it never sees a type param
2012-06-03 20:03:08 -07:00
Brian Anderson
80dc2e11a1 rustc: Remove the last use of upcall_shared_malloc 2012-05-30 21:23:34 -07:00
Brian Anderson
c6a23cddfb rustc: Make unique closures use the new unique allocator 2012-05-30 21:23:34 -07:00
Brian Anderson
6d37c90ce6 rustc: Remove tuplify_cbox_ty. Unused 2012-05-30 21:23:33 -07:00
Niko Matsakis
1f4b3cfc36 zero out alloca for preserving values
conservatively do it always for now
2012-05-30 11:26:15 -07:00
Niko Matsakis
9e6a068034 replace last_use with liveness info 2012-05-25 14:37:30 -07:00
Brian Anderson
9aa18c2852 rustc: Move ast_map to the syntax crate 2012-05-21 23:39:27 -07:00
Niko Matsakis
5be8bf1de7 use -Z to distinguish internal debugging options 2012-05-18 10:38:18 -07:00
Tim Chevalier
8affc78e8a Turn constants in back::abi into uints, and propagate types
This means GEPi now takes a list of uints. Apologies in advance
if this is hard to rebase against, but it gets rid of many a cast :-)

Also modernized some for loops here and there.
2012-05-03 22:31:38 -07:00
Tim Chevalier
9d274ec5f2 Re-rename option functions
get_with_default (nee from_maybe) => get_default
with_option (nee maybe) => map_default
with_option_do (nee may) => iter

As per discussion of 21be1379d5
2012-04-06 12:20:13 -07:00
Tim Chevalier
21be1379d5 Rename some core::option functions
from_maybe => get_with_default
maybe => with_option
may => with_option_do

I know these names are kind of ridiculous, but it's the best I could think of.
Feel free to bikeshed. Closes #2081
2012-04-02 16:12:49 -07:00
Marijn Haverbeke
b5a4fa9456 Move some code over to iterator-for to see how it performs. 2012-03-27 15:46:33 +02:00
Marijn Haverbeke
064f82d68d Support returning from loop blocks
The code is somewhat invasive, but it seems hard to do this in a
clean way, since the design itself involves a bunch of 'action
at a distance'.

Issue #1819
2012-03-27 12:33:13 +02:00
Marijn Haverbeke
f6e3738b9c Support an alternate for syntax that calls a higher-order function
The last argument of the call must be a block, and the type of this
argument must a function returning bool. `break` and `cont` are
supported in the body of the block, and return `false` or `true` from
the function. When the end of the function is reached, `true` is
implicitly returned.

    for vec::all([1, 2, 3]) {|elt|
        if elt == 2 { break; }
        log(error, elt);
    }

Issue #1619
2012-03-27 09:27:31 +02:00
Tim Chevalier
8bbd78075f Handle self correctly when translating classes
This change uses the same code for handling the "self" reference for
classes as is already used for impls/ifaces. This allows removing the
extra maybe_self_id argument (which was just for classes) to trans_closure
that I added before. I also rewrote the translation for class ctors so
that it doesn't generate new AST nodes (instead translating directly).

Also changed visit so that it visits class ctors correctly with visit_fn,
and changed typestate to not do return-checking when visiting a class ctor.
2012-03-23 20:37:50 -07:00
Marijn Haverbeke
1b81c5112a Remove last vestiges of old-style intrinsics
Closes #2048
2012-03-23 16:08:01 +01:00
Marijn Haverbeke
a84b56bc70 Remove more needless context threading in trans 2012-03-23 14:45:47 +01:00
Marijn Haverbeke
52d618a99a Revert removal of intrinsics
Oops. We can't do this yet until the next snapshot.
2012-03-23 12:51:20 +01:00
Marijn Haverbeke
f5024692d4 Remove support for the old-style intrinsics
Closes #2042
Closes #1981
2012-03-23 12:21:55 +01:00
Marijn Haverbeke
ea2e379e4f Implement built-in native modules as an alternative to intrinsics
Issue #1981
2012-03-23 10:49:47 +01:00
Graydon Hoare
b224bcf6e1 Add an LLVM-instruction-counting mode to trans.
Pipe to xdu to see a trans call graph of generated insns.
2012-03-22 13:44:20 -07:00
Niko Matsakis
b653a18416 add mut decls to rustc and make them mandatory 2012-03-21 21:04:14 -07:00
Graydon Hoare
56828d49fd Remove unused "flav" parameter from function registration paths. 2012-03-20 15:16:12 -07:00
Marijn Haverbeke
22bef74b55 Remove shared tydescs
All tydescs are static now, there's no need to worry about
marshalling them between threads anymore.
2012-03-16 15:38:42 +01:00
Marijn Haverbeke
15a325f267 Support binding of methods off boxed iface values
Closes #435
2012-03-16 12:58:26 +01:00
Marijn Haverbeke
3ab9978b9b Remove support for dynamically-sized types from translation code 2012-03-15 15:08:31 +01:00
Marijn Haverbeke
a4d75a4ed1 Remove GEP_tup_like 2012-03-15 15:08:31 +01:00
Marijn Haverbeke
e4cbd43c43 Huge kludge to get intrinsics' type parameters passed 2012-03-15 09:41:21 +01:00
Marijn Haverbeke
9aa78e34e4 Never pass tydesc to functions
My assumption that native generics needed them was wrong, so tydescs
can be eliminated from function signatures completely.
2012-03-15 09:40:52 +01:00
Marijn Haverbeke
8f84d4c8b6 Remove a large part of the tydesc-passing code 2012-03-15 09:31:34 +01:00
Marijn Haverbeke
83c9f58534 Rename dict to vtable throughout the compiler
The difference went away.
2012-03-15 09:28:50 +01:00
Marijn Haverbeke
4511f936b1 Hugely simplify iface handling
With the assumption of monomorphization
2012-03-15 09:26:54 +01:00
Marijn Haverbeke
168398bb3d Stop generating generic versions of generic functions
Monomorphic instances are generated on demand.
2012-03-15 09:00:21 +01:00
Patrick Walton
273c5e5f11 rustc: Lift the @ from the type definition of crate_ctxt into its uses
This will make it easier to convert crate_ctxt into a region pointer, since
there are functions that return crate contexts. There would be no way to type
these functions if crate_ctxt had to be an inferred region pointer.
2012-03-14 17:31:16 -07:00
Brian Anderson
b968c8e6cd Name types after their modules instead of 't' 2012-03-13 15:14:17 -07:00
Graydon Hoare
6f5853f5a1 Libc/os/run/rand/io reorganization. Close #1373. Close #1638.
- Move io, run and rand to core.
 - Remove incorrect ctypes module (use libc).
 - Remove os-specific modules for os and fs.
 - Split fs between core::path and core::os.
2012-03-12 20:08:29 -07:00
Patrick Walton
c9375fed8d stdlib: Stop incurring vtable dispatch costs when hashmaps are used
This required changing almost all users of hashmaps to import the hashmap interface first.

The `size` member in the hashmap structure was renamed to `count` to work around a name conflict.
2012-03-07 17:35:13 -08:00
Patrick Walton
c245d9e980 Revert "stdlib: Stop incurring vtable dispatch costs when hashmaps are used"
This reverts commit f0250a23d3.
2012-03-07 16:56:45 -08:00
Patrick Walton
f0250a23d3 stdlib: Stop incurring vtable dispatch costs when hashmaps are used
This required changing almost all users of hashmaps to import the hashmap interface first.

The `size` member in the hashmap structure was renamed to `count` to work around a name conflict.
2012-03-07 16:51:31 -08:00
Brian Anderson
0ee96de4ee rustc: Lower case error messages 2012-03-05 17:05:20 -08:00
Tim Chevalier
e2fa6f03f5 Translate simple classes
Programs using classes with fields only (no methods) compile and run,
    as long as nothing refers to a class in a different crate (todo).

    Also changed the AST representation of classes to have a separate
    record for constructor info (instead of inlining the fields in the
    item_class node), and fixed up spans and pretty-printing for
    classes.
2012-03-04 12:43:17 -08:00
Graydon Hoare
87c14f1e3d Move src/comp to src/rustc 2012-03-02 18:46:13 -08:00