about summary refs log tree commit diff
path: root/src/comp
AgeCommit message (Collapse)AuthorLines
2012-03-13fixup serialize_variantNiko Matsakis-388/+0
2012-03-13get new decorator extensions workingNiko Matsakis-0/+388
2012-03-02Move src/comp to src/rustcGraydon Hoare-54922/+0
2012-03-02Avoid hitting unicode lib per char.Graydon Hoare-2/+9
2012-03-02core: Remove _mut functions from vecBrian Anderson-8/+10
Instead, use vec::to_mut/from_mut to transform vectors in place as needed.
2012-03-02emit mthd data if marked inlineNiko Matsakis-11/+8
2012-03-02restructure to better support method inliningNiko Matsakis-187/+269
2012-03-02retool inline encoding to handle methods, fix testsNiko Matsakis-37/+155
2012-03-02make sure we do not bind unique closures (or blocks, for that matter)Niko Matsakis-10/+11
2012-03-01Use the correct prestate for callsTim Chevalier-2/+2
The prestate for calls was getting set incorrectly to the poststate for the operator in the call. This worked before since most of the time, operator expressions are pure. Issue 1895 shows how this breaks when the operator is a closure that has a move-in capture clause. (I had a several-day, multi-file patch for this that didn't work... and then it turned out to be a one-line fix. The joys of programming.) Closes #1895
2012-03-01Make sure resource item types get written by typecheckerMarijn Haverbeke-5/+7
Closes #1916
2012-02-29optionally enforce local variable mutabilityNiko Matsakis-48/+80
2012-02-29un-xfail test, remove old commentNiko Matsakis-1/+0
2012-02-29report the type of the method as the fty, not the type of the expressionNiko Matsakis-2/+3
using the type of the expression causes us to lose information about when the method was declaring with a generic signature, which leads to producing incorrect code, particularly in the case of returning unit types (which would then get an undefined output ptr)
2012-02-29add the ability to snag the frame so we can verify that we are inliningNiko Matsakis-0/+7
2012-02-29Delete-trailing-whitespaceMarijn Haverbeke-1/+1
2012-02-29Actually use last_use info on closed over var in kind.rsMarijn Haverbeke-4/+11
Issue ##1894
2012-02-29Recognize last uses for copied closed-over variablesMarijn Haverbeke-118/+168
And clean up and fix some bad things in last_use.rs. Closes #1894
2012-02-28libcore: Remove vec::to_ptr in favor of vec::unsafe::to_ptr. Closes #1829.Patrick Walton-22/+23
2012-02-28Also do normal impl method lookup for type parametersMarijn Haverbeke-2/+1
Closes #1909
2012-02-28Bail out of calling tydesc glue in an unreachable contextMarijn Haverbeke-0/+1
Closes #1901
2012-02-28add ability to run multi-crate tests, run tests with --inlineNiko Matsakis-2/+3
2012-02-28enumerate ids for arguments tooNiko Matsakis-1/+5
2012-02-28change def's that are always local to use node_id, add --inline optNiko Matsakis-106/+117
2012-02-28correct freevar encoding, add more debugging outputNiko Matsakis-3/+16
2012-02-25Disallow type parameters in the main() functionTim Chevalier-0/+13
Closes #1900
2012-02-24Add temp cleanups for copy/move mode args when other args fail. Fixes #1374Niko Matsakis-5/+29
2012-02-24Encode/decode AST into metadata, re-instantiate inlined itemsNiko Matsakis-762/+10031
2012-02-23rustc: Don't add duplicate entries to exp_mapBrian Anderson-1/+7
2012-02-23Support basic, unsophisticated, somewhat wrong export-globs.Graydon Hoare-50/+145
2012-02-23Treat export-globs as import-globs for sake of linking.Graydon Hoare-3/+12
2012-02-23Minor renaming to help my comprehension.Graydon Hoare-38/+44
2012-02-23Make sure fold_ty preserves associated def_idsMarijn Haverbeke-0/+4
Closes #1884
2012-02-23Finish cleanup of core::strMarijn Haverbeke-16/+11
Closes #1849
2012-02-23Various cleanups and optimizations in core::strMarijn Haverbeke-10/+9
2012-02-23(core::str) rename substr_len_bytes to substr_len, and delete unused ↵Kevin Cantu-1/+1
byte_index[_from]
2012-02-23(core::str) remove len_bytes aliasKevin Cantu-29/+26
2012-02-23(core::str) replace uses of unsafe::slice_bytes; replace find[_from]_bytes ↵Kevin Cantu-8/+8
with find[_from]
2012-02-23(core::str) stop using index_charsKevin Cantu-2/+2
2012-02-23(core::str) replace byte_index[_from] with index[_from]Kevin Cantu-1/+1
2012-02-23(core::char) rename slice -> slice_charsKevin Cantu-3/+3
2012-02-23(core::str) rename index -> index_charsKevin Cantu-1/+1
2012-02-23(core::str) mostly rename len -> len_charsKevin Cantu-7/+7
2012-02-22rustc: Don't ignore attributes inside empty mods. Closes #1655Brian Anderson-1/+7
2012-02-22rustc: Allow any integral types on rhs of shift opsBrian Anderson-8/+70
2012-02-22Improve error message for use of a non-pure-fn in a predTim Chevalier-3/+3
Closes #1883
2012-02-22Stop normalizing patternsMarijn Haverbeke-320/+285
The check for whether a pat_ident is a variant or a binding is simple and fast. Normalizing patterns again and again is slow and error-prone (several places were forgetting to do it).
2012-02-22Take move captures in account in mutability checkerMarijn Haverbeke-0/+12
Closes #1461
2012-02-22Return a new type var instead of nil when failing to typecheck a fieldMarijn Haverbeke-3/+5
The previous solution was usually causing a second spurious error message.
2012-02-22Make the various from_str functions return optionsMarijn Haverbeke-14/+9
So that they can be used with user input without causing task failures. Closes #1335