468 Commits

Author SHA1 Message Date
Marijn Haverbeke
67cc89f38d Rewrite exhaustiveness checker
Issue #352
Closes #1720

The old checker would happily accept things like 'alt x { @some(a) { a } }'.
It now properly descends into patterns, checks exhaustiveness of booleans,
and complains when number/string patterns aren't exhaustive.
2012-02-15 15:47:42 +01:00
Marijn Haverbeke
16e5760564 Fix bug in safe-reference checker
At some point, a refactor broke the code that handled local declarations
to no longer descend into the initializer expressions.

Closes #1846
2012-02-15 11:53:32 +01:00
Kevin Atkinson
d6d7134c37 When parsing a source string, fail when the entire string is not parsed.
(For now only fail when parse_from_source_str is used to avoid
possible compatibility problems; parse_expr_from_source_str still
does not check.)
2012-02-14 19:37:33 -08:00
Kevin Atkinson
0e44133e7c Do a better job of reporting source location for files (i.e. filemap)
that are really a substr of another file.
2012-02-14 19:37:33 -08:00
Brian Anderson
0a503228f6 rustc: Typecheck crust functions as *u8 2012-02-14 11:24:09 -08:00
Niko Matsakis
520c3f5f79 correct error message and add to test 2012-02-13 07:33:16 -08:00
Niko Matsakis
4eb52f69a9 do not fail if iface/impl method have different number of parameters 2012-02-13 06:50:58 -08:00
Niko Matsakis
a7641e99e8 remove leading ::, new tests 2012-02-13 06:50:58 -08:00
Brian Anderson
005a3efb8b test: Fix cros-crate-glob-collision 2012-02-11 18:20:02 -08:00
Niko Matsakis
3f15a28b16 correct fully qualified type names to include the crate; add tests (take 2)
This reverts commit fc023d91c77f468211ba9f7681b4d2a6469aac74.
2012-02-11 14:09:52 -08:00
Niko Matsakis
fc023d91c7 Revert prior commit until I have time to track down the valgrind error
This reverts commit 2ec3a0b60833736a3c0d805768885b4392c36100.
2012-02-11 09:49:02 -08:00
Niko Matsakis
2ec3a0b608 correct fully qualified type names to include the crate; add tests
fixes #1745
2012-02-10 21:53:17 -08:00
Brian Anderson
17110fb28b test: Add tests for #1802 2012-02-10 15:43:35 -08:00
Marijn Haverbeke
7f1ea3ef6a Don't allow binding patterns to bind keywords
Closes #1586
2012-02-10 15:54:47 +01:00
Marijn Haverbeke
74d4e2a32e Add compile-fail tests for interfaces/impls
Closes #1475
2012-02-10 13:31:33 +01:00
Brian Anderson
80cd6aa96e test: Update and un-xfail compile-fail/tag-that-dare-not... 2012-02-05 18:06:54 -08:00
Niko Matsakis
9f0239ce9c add a test where mode inference ought to fail 2012-02-05 13:59:09 -08:00
Kevin Atkinson
5ef53382ae Add support for parsing quasi-quotes, doesn't do anything useful yet. 2012-02-03 20:23:49 -07:00
Marijn Haverbeke
c976b7c623 Fix tests for str::le patch 2012-02-03 14:32:23 +01:00
Kevin Cantu
8f367ebfeb Rename (again) str::unsafe::slice -> str::unsafe::slice_bytes and
str::unsafe::safe_slice -> str::unsafe::slice_bytes_safe_range
2012-02-01 21:56:53 -08:00
Kevin Cantu
c71667d9d2 Propagating unsafe::slice 4 2012-02-01 21:56:53 -08:00
Marijn Haverbeke
856a544d0c Remove native types from stdlib 2012-02-01 12:23:13 +01:00
Tim Chevalier
e5d095d67e Change option::t to option
Now that core exports "option" as a synonym for option::t, search-and-
replace option::t with option.

The only place that still refers to option::t are the modules in libcore
that use option, because fixing this requires a new snapshot
(forthcoming).
2012-01-31 17:05:20 -08:00
Tim Chevalier
fba35e1a3c Require alts to be exhaustive
middle::check_alt does the work. Lots of changes to add default cases
into alts that were previously inexhaustive.
2012-01-31 10:08:24 -08:00
Brian Anderson
53dbde6cc2 rustc: Make 'attempted access of field' error non-fatal 2012-01-27 16:58:27 -08:00
Marijn Haverbeke
e48bf6f3f4 Make occurs check in ty::fixup_vars more reliable
It wouldn't detect cycles that went through several type vars before.

Closes #1464
2012-01-27 18:58:52 +01:00
Marijn Haverbeke
dc55c06aa3 Properly check enum types for recursion
Closes #742
2012-01-27 16:03:43 +01:00
Brian Anderson
28fbb19664 rustc: Switch the --no-core switch to a #[no_core] attribute 2012-01-26 16:43:33 -08:00
Brian Anderson
5610205363 rustc: Move core injection to its own module 2012-01-26 16:23:48 -08:00
Marijn Haverbeke
888262b337 Allow operator overloading of the indexing operator
The method `op_index` (which takes a single argument) is used for
this.

Issue #1520
2012-01-26 15:23:11 +01:00
Marijn Haverbeke
87b064b249 First stab at operator overloading
When no built-in interpretation is found for one of the operators
mentioned below, the typechecker will try to turn it into a method
call with the name written next to it. For binary operators, the
method will be called on the LHS with the RHS as only parameter.

Binary:

    +   op_add
    -   op_sub
    *   op_mul
    /   op_div
    %   op_rem
    &   op_and
    |   op_or
    ^   op_xor
    <<  op_shift_left
    >>  op_shift_right
    >>> op_ashift_right

Unary:

    -   op_neg
    !   op_not

Overloading of the indexing ([]) operator isn't finished yet.

Issue #1520
2012-01-26 14:25:06 +01:00
Niko Matsakis
5e13d19cc0 s/block()/fn()/g 2012-01-23 19:06:33 -08:00
Brian Anderson
7d5d1a311e test: Add no-reformat directive to compile-fail/issue-1362 2012-01-23 17:37:15 -08:00
Kevin Atkinson
664d9cc86f Add regression test for issue #1362.
Although its not really needed.  Without that fix, reported spans will
likely be bogus if the error is within the first couple of lines
(probable around 5) of that file.  Thus, many of the compile-fail
tests will fail due to incorrect location.
2012-01-23 17:37:15 -08:00
Kevin Atkinson
5e60facf78 Add regresion tests for issue #1448 and #1387. 2012-01-23 17:37:15 -08:00
Tim Chevalier
6db688e893 Check that the names mentioned in tag exports are actually types (or variants)
Check that in export foo{}, foo is an enum type, and that in export
foo{bar, quux}, foo is an enum type and bar and quux are variants belonging
to foo.
2012-01-23 15:48:08 -08:00
Tim Chevalier
9dc59e1506 Export all enum variants by default; new syntax for selectively exporting variants
See issue 1426 for details. Now, the semantics of "export t;" where t is a tag are
to export all of t's variants as well. "export t{};" exports t but not its
variants, while "export t{a, b, c};" exports only variants a, b, c of t.

To do:
- documentation
- there's currently no checking that a, b, c are actually variants of t in the
 above example
- there's also no checking that t is an enum type, in the second two examples above
- change the modules listed in issue 1426 that should have the old export
semantics to use the t{} syntax

I deleted the test export-no-tag-variants since we're doing the opposite now,
and other tests cover the same behavior.
2012-01-23 15:48:08 -08:00
Tim Chevalier
ba5cc236f7 WIP on issue 1426 (exporting all tags)
Support Lenny222's proposed syntax for exporting a tag without
its variants, or selected tags from a variant, in the AST and parser.
No support further down the line yet. Tests are xfailed.
2012-01-20 19:48:33 -08:00
Tim Chevalier
e36df0f6c8 Handle fail after return correctly in typestate
Previously, typestate would conclude that this function was
correctly diverging:

fn f() -> ! { ret; fail; }

even though it always returns to the caller. It wasn't handling the
i_diverge and i_return bits correctly in the fail case. Fixed it.

Closes #897
2012-01-20 17:26:31 -08:00
Graydon Hoare
7b1a8f0a91 Additional ; to , changes, disable "tag" and ";" in parser. Close #1430. Close #1428. 2012-01-19 19:29:21 -08:00
Patrick Walton
59ebe6af18 rustc: Make the pretty printer output commas after enum variants. Update all tests accordingly. 2012-01-19 18:41:06 -08:00
Kevin Atkinson
d388daa996 Remove incorrect comments from a few test cases. 2012-01-19 17:31:36 -07:00
Patrick Walton
d97783e3e5 test: "tag" -> "enum" in compile-fail 2012-01-19 16:01:47 -08:00
Patrick Walton
1731f5d709 test: Fix error patterns in name-clash-nullary 2012-01-19 14:47:43 -08:00
Niko Matsakis
882bea5b6d Rename fn*() to fn() as originally planned. 2012-01-19 07:11:50 -08:00
Haitao Li
327a15d58c rustc: Add a usage pass to collect one-off analyses
This patch starts from move the analysis which checkes of probably
incorrectly usage of `int|uint` in native fn.

Issue #1543
2012-01-19 17:27:44 +08:00
Tim Chevalier
5b028f527f Remove support for the '.' after a nullary tag in a pattern
(Commit also includes lots of changes to remove '.'s that a git
merge messed up, or else it was monkeys.)
2012-01-19 01:04:59 -08:00
Patrick Walton
5ac57d1891 test: Don't test that tags with zero variants are supported 2012-01-18 12:43:01 -08:00
Haitao Li
f03eb96f39 rustc: Warn when int or uint is used in a native type decl
Issue #1403
2012-01-19 02:10:36 +08:00
Marijn Haverbeke
1375b31c1c Store type names in crate metadata
Improves type error messages referring to external types.

Issue #1507
2012-01-16 12:08:00 +01:00