about summary refs log tree commit diff
path: root/src/rustc
AgeCommit message (Collapse)AuthorLines
2012-07-16improve commentNiko Matsakis-9/+33
2012-07-14infer the scope of borrowsNiko Matsakis-191/+318
2012-07-14remove typestate from code, tests, and docsNiko Matsakis-847/+86
2012-07-14Treat bare vector and string literals as fixed length vecs. Closes #2922.Michael Sullivan-81/+9
2012-07-14Get rid of ast::ty_vstore, which was only used for fixed length.Michael Sullivan-9/+5
2012-07-14Tear out ty_str and ty_vec.Michael Sullivan-191/+66
2012-07-14Get rid of more deprecated strs on non 64-bit linux platforms.Michael Sullivan-15/+15
2012-07-14Fix a bunch of deprecated str/vec errors in code for non 64-bit linux ↵Michael Sullivan-1/+1
platforms...
2012-07-14Merge branch 'vector-reform' into incomingMichael Sullivan-2224/+2268
2012-07-14Make the new world order normative. Closes #2908.Michael Sullivan-1/+1
2012-07-14Move the world over to using the new style string literals and types. Closes ↵Michael Sullivan-2212/+2239
#2907.
2012-07-14Don't emit strings when the destination is ignored.Michael Sullivan-0/+1
2012-07-13Warn on old strsMichael Sullivan-1/+1
2012-07-13Add FIXME marking where the work on #2794 has gotten to, so far.Lindsey Kuper-0/+3
2012-07-13Make an error message more grammatical.Lindsey Kuper-1/+1
2012-07-13Change (hopefully) all of the code that generates strs asts to produce ~strs.Michael Sullivan-6/+15
2012-07-13Change the pretty printer to print vstores for strs in prefix notation.Michael Sullivan-7/+11
2012-07-13Support prefix notation for vstore strings. Closes #2906.Michael Sullivan-84/+87
2012-07-13Change stray `cont` to `again` in a comment.Lindsey Kuper-1/+1
2012-07-13More consistent use of backticks and "expected" in error messages.Lindsey Kuper-49/+49
Got some of the debug messages, here, too. I figure it doesn't hurt to get used to doing this even in places where users won't ever see it.
2012-07-13Refactor how impl self types are storedTim Chevalier-23/+21
In order to avoid a confusing use of the tcache, I added an extra node ID field to trait refs. Now trait refs have a "ref ID" (the one that resolve3 resolves) and an "impl ID" (the one that you look up in the tcache to get the self type). Closes #2434
2012-07-13Better error when rustc fails to write output.Glenn Willen-5/+23
2012-07-13Front-end support for default impls in traits.Lindsey Kuper-36/+105
2012-07-13Consistently use "allowed" rather than "permitted" in error messages.Lindsey Kuper-1/+1
2012-07-13add comments to region inferenceNiko Matsakis-0/+36
2012-07-13change region syntax to &r/T in place of &r.TNiko Matsakis-2/+2
2012-07-14Change cont out of loop error to again out of loopZack Corr-2/+2
2012-07-12Obliterate the callee_id hackTim Chevalier-33/+47
Exprs that could be applications of overloaded operators (expr_unary, expr_binary, expr_index) relied on the previous node ID being "reserved" to carry extra typechecking info. This was incredibly error-prone. Fixed it; now all exprs have two node IDs (which will be wasted in some cases; future work could make this an option instead if the extra int field ends up being a performance problem). Closes #2804
2012-07-12Fix the signature on vec::view.Eric Holk-7/+14
Due to limitations in region inference, this has the effect of making vec::view pretty much entirely unusable.
2012-07-12rustc: Resolve bounds of trait type parametersBrian Anderson-0/+2
2012-07-12Merge remote-tracking branch 'origin/dist-snap' into incomingGraydon Hoare-8/+8
2012-07-12Handle str/~ in patterns.Michael Sullivan-3/+15
2012-07-12Accept prefix notation for writing the types of str/~ and friends.Michael Sullivan-54/+101
2012-07-12rustc: Make all impls even more reachableBrian Anderson-14/+6
With this we write metadata for all impls so that we can properly find reexported impls.
2012-07-12Merge branch 'rt-changes' into incomingMichael Sullivan-99/+60
2012-07-12Make str be treated as str/~.Michael Sullivan-2/+2
2012-07-12Refactor how we convert vstore ast types to ty types.Michael Sullivan-21/+20
2012-07-11infer when types are region parameterized rather than requiring /&Niko Matsakis-196/+424
- removes various fields from various variants in the AST - also update tests not to use this notation
2012-07-11new option to note where purity is req'dNiko Matsakis-1/+11
2012-07-11add ability to report statistics about how borrowck is being usedNiko Matsakis-17/+64
2012-07-11rustc: Treat existential traits as base types for coherence purposesPatrick Walton-10/+11
2012-07-11Remove obsolete type param field from shape information.Michael Sullivan-8/+0
2012-07-11Change the interface of placement new to take a tydesc as part of Issue #2831.Michael Sullivan-20/+15
2012-07-11rustc: Add some more checks to coherencePatrick Walton-9/+197
2012-07-11Fix some version numbers.Graydon Hoare-8/+8
2012-07-10Visit item macros.Eric Holk-2/+2
2012-07-10In resolve, check that an or-pattern has the same number of bindings in each ↵Tim Chevalier-5/+42
disjunct resolve3 wasn't checking this. Added test cases. Also added a helpful informational message in the case where you have a variable binding that you probably think refers to a variant that you forgot to import. This is easier to do in resolve than in typeck because there's code in typeck that assumes that each of the patterns binds the same number of variables.
2012-07-10rustc: Introduce a coherence checker, behind a compile switch for nowPatrick Walton-3/+237
2012-07-10Eliminate some obsolete upcalls.Michael Sullivan-8/+0
2012-07-10Get rid of unused fields in tydescs. Closes #2351.Michael Sullivan-40/+23