id stringlengths 8 11 | text stringlengths 32 472 | label stringclasses 5
values | label_id int64 0 4 |
|---|---|---|---|
rust_1285 | make on windows - mingw 20111118 fails | I am trying to build rust on mingw - windows. make gives me this error: make: **\* No rule to make target `src/llvm/test/CodeGen/X86/fastcc-sret.ll/toy.cpp', needed by`/c/Users/salil.wadnerkar/rust/llvm/i686-pc-mingw32/Release+Asserts/bin/llvm-config'. Stop. The complete log is ... | low | 3 |
rust_1286 | crash: Assertion cx.fcx.llobjfields.contains_key(did.node) failed at src/comp/middle/trans.rs:2914 | The following code crashes rustc f05eaa4a650132174a43113e7bf1bce293515428 (2011-12-09): ``` use std; import std::option; import std::ptr; import std::str; import std::vec; #[link_name = "pcre"] native mod _native { type... | medium | 2 |
rust_1287 | Tutorial: please clarify "mutable vector" | http://www.rust-lang.org/doc/tutorial/syntax.html reads **"[mutable T] Mutable vector type"**, which i find confusing. - I would read "[mutable T]" as "a vector of mutable T". - But maybe i am wrong and it is meant to read "a mutable vector of T". I'd be happy if someone coul... | low | 3 |
rust_1289 | getopts should use result::t | getopts::getopts returns it's own internally defined type result. Instead, it should be changed to use the result::t type to be more consistent. | labels: E-easy, C-cleanup | low | 3 |
rust_1290 | 'make all' only builds stage 1 | It's not obvious from the makefile why this is | medium | 2 |
rust_1291 | build broken on os x | It appears to me that the build is broken on os x lion 10.7.2 (with xcode 4.1 (4B110)) ``` cd build ../configure <..> configure: configure: CFG_SRC_DIR := ../ configure: CFG_BUILD_DIR := /opt/rust/src/rust/build configure: CFG_OSTYPE := apple-darwin configure: CFG_CPUTYPE := x86_ | labels: O-maco... | medium | 2 |
rust_1295 | Decouple #[link_name = ""] and #[link_args] | When link_name == "" we don't try to link the native library, which we primarily use to link to libc. As a side-effect this causes any link_args to not be emitted. These two things should not be coupled, since one might want to suppress the automatic -L argument to the link... | low | 3 |
rust_1296 | Make a new attribute for the #[link_name = ""] special case | # [link_name = ""] is used to tell rustc not to issue the -L linker flag. This is a special case of the link_name flag so let's replace it with its own attribute, like #[nolink]. | labels: E-easy, C-cleanup, A-linkage | low | 3 |
rust_1297 | Investigate a segmentation fault in link::link_binary | Commit 76f13c95a0d1b1f33805719ccde9600b000e4418 worked around a segmentation fault. Investigate the root cause. | labels: I-crash | high | 1 |
rust_1298 | Add testing supports tests those need multiple compile commands | For example, I need to test the re-export issue (#1115). Steps are: - compile a crate as library, which re-exports a function. - call that re-exported function from another crate. - verify compile-pass or run-pass, etc. This requires executing rustc twic... | question | 4 |
rust_1299 | invalid format string causes leaked memory in rust main loop | The following program; use std; import std::io; fn main(arg : [str]) { std::io::println(#fmt("%r", "Goat")); } fails to compile with the following internal error: herugrim::~/code/markov2 $ rustc markov.rs markov.rs:6:26: 6:30 error: unknown type in convers... | medium | 2 |
rust_1300 | Link with ld instead of gcc | It seems silly to have this runtime dependency on gcc just for calling the linker. Calling ld will make #[link_args] and --link_args more accurate. Right now they would be better described as gcc_args. | labels: A-linkage | medium | 2 |
rust_1301 | Allow link_args attributes on crates | Right now link_args attributes only apply to native modules. We are soon adding a --link_args command line option as well. A crate-level attribute makes sense too. | labels: E-easy, A-linkage | low | 3 |
rust_1302 | Disable locking in memory_region when RUSTRT_TRACK_ALLOCATIONS=0 | The exchange heap uses a lock around every malloc and free, i.e. every vector and string acquires at least two locks during its lifetime. It looks like when we are not tracking allocations (now the default) there should be no need for a lock here. | lab... | low | 3 |
rust_1303 | Add #[link_type="object"] to link against object files | I need to write some wrapper functions in C for my zeromq bindings. It seems a bit heavyweight to create a shared library just to link these functions in, so I'm cheating and putting my .o files in the link_args for the zmq native mod. I'd rather put these functi... | low | 3 |
rust_1304 | Put all the trans_* modules under an umbrella trans module | Trans has slowly been divided up into smaller pieces, but the structure is a bit odd. Instead of a trans module that contains a bunch of smaller modules we have a bunch of smaller modules that are peers of each other and named trans_something. We should refac... | low | 3 |
rust_1305 | rustc crash with upcall fail "option none" | I'm getting this exception: ``` rust: upcall fail 'option none', ../src/libcore/option.rs:34 rust: domain main @0x102029600 root task failed ``` With this reduced (and invalid) code: ``` use std; import std::str; obj a() { fn b() { c({ |d| }); } } ``` | medium | 2 |
rust_1306 | Should `5.f` parse as a float literal, or expr_field(int literal, "f")? | We can make the lexer not consider a dot part of a number when followed immediately by a letter. This makes `5.f`, `10.f64`, and similar invalid, but the dot is already redundant there anyway. This allows us to do `10.some_method_name()`. That'd ... | medium | 2 |
rust_1308 | OS X generates warning about compact debug info for __morestack | This is probably not a problem, since the unwinder falls back to dwarf for frames without compact info, but we should figure out a way to suppress it at least. Note that there are other errors in this output. ``` (stdcore=843d93) rustcrypto $ ls README.m... | medium | 2 |
rust_1310 | Add caching to tag_variant | This function allocates tons of memory and the results are cacheable. | labels: E-easy | low | 3 |
rust_1311 | llvm-config sometimes doesn't provide the correct include directory | llvm-config --cxxflags looks like this for me: ``` -I/home/banderson/Dev/rust/src/llvm/include -I/home/banderson/Dev/rust/build/llvm/x86_64-unknown-linux-gnu/include -D_DEBUG -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMI... | medium | 2 |
rust_1313 | deal with .dSYM on os x | When building shared libs on os x gcc generates separate .dSYM directories that contain out-factored symbol information. This is a bit ugly and different from the other platforms. We need to investigate/decide what to do with it: leave it, delete it, somehow generate normal libs | labels: O-m... | medium | 2 |
rust_1315 | Go over std and core and add pure declarations | Noticed some fns in std and core could be annotated as pure. This should be done so that they may be used from predicates. This probably needs to happen twice, once now and once later, after mutation of local variables is possible in pure functions. It would also help if... | low | 3 |
rust_1316 | Type casts should be considered as pure where possible | Noticed this when seeing a u8 as uint cast | labels: A-type-system | medium | 2 |
rust_1322 | Return address at the bottom of stack should be 0 | We have 0xdeadbeef as our final return address. The unwinder expects to see 0 here afaict. | labels: A-runtime | medium | 2 |
rust_1323 | Use __morestack to switch to the C stack | This would allow gdb to backtrace through the stack switch. Should be doable by writing an assembly function that follows the __morestack protocol. | labels: A-runtime | medium | 2 |
rust_1324 | Remove stack alignment from core::task::unsafe_spawn_inner | Pretty sure that aligning here has no effect since the setup in context::call does its own alignment. | labels: A-runtime, E-easy, C-cleanup | low | 3 |
rust_1325 | Rewrite upcall_new_stack/del_stack to follow the standard upcall convention | These upcalls are used by __morestack and have a different calling convention than other upcalls, but there's no reason that has to be the case. | labels: A-runtime, C-cleanup | low | 3 |
rust_1326 | #[link_name = ""] should be an error even when #[nolink] | # [link_name = ""] now results in an error, but to support the transition to the #[nolink] attribute, it does not when the #[nolink] attribute is also applied. After we snapshot we should make #[link_name = ""] always an error. | labels: E-easy, A-linkage | question | 4 |
rust_1327 | Fix failing tests on windows | There's at least one test that fails on windows and isn't run by check-fast | labels: A-testsuite, O-windows | medium | 2 |
rust_1328 | Use quotes to improve error messages | Since Haskell's [1 .. 10] is not (yet) implemented in Rust and i read about ranges, i tried ``` let a = [1 to 10]; ``` this gives the baffling error message ``` error: expecting ,, found to ``` A more readable version would be ``` error: expecting "," but found "to" ``` | labels: ... | low | 3 |
rust_1330 | Importing a module that does not exist fails hard | ``` rust import std::vec; use std; fn main() { } ``` ``` godot:build jdm$ x86_64-apple-darwin/stage0/bin/rustc --target=x86_64-apple-darwin ~/src/resolve_test.rs rust: upcall fail 'option none', /Users/administrator/src/rustbot/workspace-snap-stage3-i386-unknown-darwi... | medium | 2 |
rust_1332 | Add valgrind guard bytes to the end of the stack again | I previously removed them thinking they had no practical effect, but I was wrong since the first few words of the stack are things that shouldn't be scribbled on. | labels: A-runtime | medium | 2 |
rust_1333 | fix setenv and getenv (especially on os x) | I digged a bit into setenv/getenv not always working on os x and found two issues: - Sadly the man page says: ``` Successive calls to setenv() or putenv() assigning a differently sized value to the same name will result in a memory leak. The FreeBSD seman- tics for these fun... | medium | 2 |
rust_1335 | from_strs should return option::t<T> | Since from_str functions will be mainly used to parse user input, they should return optional types instead of failing on malformed input. | labels: E-easy | low | 3 |
rust_1337 | Add debug representation of strings | labels: A-debuginfo | medium | 2 |
rust_1338 | Add debug representation of objects | labels: A-debuginfo | medium | 2 |
rust_1339 | Add debug representation of enums | labels: A-debuginfo | medium | 2 |
rust_1340 | Fix debug representation of arrays of structs | "Up-to-date" test case: ``` rust fn main() { struct moo {a: float}; let x = [moo{a: 1.5}, moo{a: 2.3}]; io::println(fmt!("%?", x)); } ``` (build with `rustc -Z debug-info`, then load it up in gdb and set a breakpoint on `main`.) Original reported test case follows ``` rus... | medium | 2 |
rust_1341 | Add debug representation of function types | labels: A-debuginfo | medium | 2 |
rust_1342 | Add debug representation of resources | labels: A-debuginfo | medium | 2 |
rust_1343 | Building with debugging causes test failures on i686-unknown-linux-gnu | This command: ``` make check-stage1-T-i686-unknown-linux-gnu-H-x86_64-unknown-linux-gnu-rfail DEBUG=1 ``` Results in tests where the instruction pointer ends up in bizarre places or the stack is misaligned. I don't know if it's related to the rece... | medium | 2 |
rust_1344 | Name mangling mistakes with -g | This is a backtrace from compiletest built with -g ``` #11 0x00007ffff7aee3ff in context::call_shim_on_c_stack (this=0x43d700, args=0x445e80, fn_ptr=0x7ffff7af2f48) at ../src/rt/arch/x86_64/context.h:58 #12 0x00007ffff7af4066 in call_upcall_on_c_stack (args=0x445e80, fn_ptr=0x7ff | labe... | low | 3 |
rust_1345 | Rust snapshot binaries aren't recognized by RHEL5 | The elf we generate uses the `Linux` ABI, which is a recent invention, older Linuxes having used the `System V` ABI. On Scientific Linux 5 this results in the snapshots being unrunnable: ``` libcore-14bd852465126fe7-0.1.so: ELF file OS ABI invalid ``` See http://stack... | question | 4 |
rust_1346 | os x / make check-fast bug that seems to effect invoke-external-native | https://gist.github.com/1500793 looks like this is concurrency related, cannot reproduce this safely on the machine where i see it and am logging this here make check-fast CFG_DISABLE_VALGRIND=1 run-pass [stage2]: test/run-pass/invoke-external-nat... | high | 1 |
rust_1347 | "char": make "to_lower"/"to_upper" work for non-ASCII characters | This probably should be implemented by means of the non-exported unicode.rs in libcore. | labels: C-enhancement, A-Unicode | low | 3 |
rust_1348 | "char": add a type and function for Unicode Character Categories | For Unicode Character Categories see http://www.fileformat.info/info/unicode/category/index.htm Haskell implements the type "GeneralCategory" and a function to determine a character's "GeneralCategory". Their implementation goes like this: - the script ... | low | 3 |
rust_1351 | change map::new_*_hash to mk_*_hash | It seems that the prefix `mk_` is used more often than `make_`, so we should swap the function names before 0.1. | labels: E-easy | low | 3 |
rust_1352 | swap arg order of vec::init_fn | This allows for the block syntax sugar. | labels: E-easy | low | 3 |
rust_1353 | cargo: auto-fetch default sources file from rust-lang.org | We should be publishing a "default" local-sources.json file on rust-lang.org (or, at minimum, our github repo for now). When you run cargo for the first time it should fetch this if it hasn't got a local-sources.json file, and there should be a "cargo init" co... | low | 3 |
rust_1354 | cargo: list packages matching name | "cargo list foo" or "cargo search foo" should return a list of packages matching foo, including installation-status and (if installed locally) where they're located in the search path, and possibly the short package desc / license / version. | medium | 2 |
rust_1355 | cargo: shortcut for adding a source | It's a minor UI convenience, but "cargo add-source <source-uri>" is a nice way to save people the activation-energy of manually editing a json file. | labels: C-enhancement | low | 3 |
rust_1356 | cargo: dependency-resolution | Cargo packages will depend on other packages. Add dependency-graph-resolving logic and recursive dependency-installation. Dependencies can be expressed in the central json file. Cyclic dependency is an error condition. Also differentiate which installed packages are installed as | labels... | low | 3 |
rust_1357 | cargo: extract json from crate files | The json describing a crate to a cargo source should be automatically extracted from the crate file. | labels: C-enhancement | low | 3 |
rust_1358 | cargo: refresh source index from remote crates | There should be a cargo command (say "cargo index crate.rc") that takes a crate file and updates (in-place) a sources file with fresh metadata extracted from the crate file. This is something a person operating a public sources-list will want to run when registering a ne... | low | 3 |
rust_1359 | cargo: add $HOME/.cargo/lib to rustc default search path | If it exists, the cargo lib dir should be added to the rustc default linking path. Figure out what counts as most-useful $HOME on windows. | labels: E-easy, A-driver | low | 3 |
rust_1362 | Bogus span identified with unmatched import | Compiling the rustray.rc crate here https://github.com/brson/rustray/tree/bogus-error results in this error: ``` -:6:14265: 6:14272 error: unresolved import: option -:6:14272: 6:14277 error: unresolved import: math error: aborting due to previous errors ``` The error is act... | medium | 2 |
rust_1364 | Make the initial obstack size very small | This looks to be the reason at the moment that we can't scale up past ~10,000 tasks. Right now the size of each shadow stack segment is 500K. We probably need to start this as something very small and double it each segment, like the main stack. | labels: A-runtime | medium | 2 |
rust_1366 | JSON: add support for "null" | "json.rs" currently has no support for the JSON value **null** (http://en.wikipedia.org/wiki/JSON) Please - add support for the JSON value "null" to libstd/json.rs - encoding: either add "Null" to "tag json" or "make to_str()" accept an "option::t<json>". - decoding: should be fi | question | 4 |
rust_1369 | stdlib/json.rs is written in a very inefficient style | It takes a very Haskell approach to parsing, cutting off piece by piece of the parsed string, and thus constantly allocating new strings. The functional style can be preserved by passing around a (@str, uint) pair with the offset as well as the string. But it'd pr... | low | 3 |
rust_1370 | std::unicode::icu: add libicu character category functions | Add bindings to libicu's (http://icu-project.org/apiref/icu4c/uchar_8h.html) functions: - UBool u_isdigit (UChar32 c) - UBool u_islower (UChar32 c) - UBool u_isspace (UChar32 c) - UBool u_isupper (UChar32 c) - UChar32 u_tolower (UChar32 c) - UChar32 u_toupper... | medium | 2 |
rust_1371 | "char": implement "is_whitespace" and "maybe_digit" using "std::unicode::icu" | "is_whitespace" and "maybe_digit" should be implemented using the libicu binding in module "std::unicode::icu". libicu is battle tested, we should not replicate their effort. See #1370 | medium | 2 |
rust_1372 | #fmt float broken on Ubuntu Precise | Assert failed on Ubuntu Precise x86_64. Works fine on another Ubuntu 11.04 (i386). ``` java use std; fn main() { std::io::println(#fmt("%f", 5.82)); assert(#fmt("%f", 5.82) == "5.82"); } ``` ``` java 0 rust: upcall fail 'Assertion "" + extfmt::rt::conv_float({flags: [extfmt::rt:: |... | medium | 2 |
rust_1373 | Libcore: move "io" to libcore | In libcore-land we we can perform calculations and string operations, but we need to discard them eventually, because there are no means to exchange them with the world. Please move the module "io" to libcore. | medium | 2 |
rust_1374 | Unwinding can still sometimes leak memory | The issue seems to be related to by-copy parameters (thus, they are probably something that I broke). I'll link a patch in a second that, when reverted, triggers the issue. | medium | 2 |
rust_1375 | #fmt uses wrong default precision for float | # fmt("%f", 5.82) yields "5.82". Other printf's yield "5.820000". | labels: E-easy | low | 3 |
rust_1376 | fix machine type equality | float, int, uint should be equal to their machine type equivalents. This is currently not the case and originally was the reason for the introduction of trivial casts. Now that these are gone, this needs fixing, i.e. let f: float = fXY::sin(1.0) should compile. | medium | 2 |
rust_1379 | Expose floating point environment and handle fp exceptions | libmath provides a rich interface to deal with numerical exceptions, rounding modes, and the like. We need to figure, if and how much of it we want to (e.g. as a resource that clears on destruction) Useful references and any comments on this very welcome Addi... | low | 3 |
rust_1380 | New logging macros #info and #debug do not work | Some handling for these got lost somewhere - currently unknown. | labels: A-runtime, E-easy | low | 3 |
rust_1381 | Prominent instructions on how to register a package with cargo-central | We'd like people to be registering their packages with cargo-central pretty much from the start. This involves writing up a small howto or FAQ entry on the matter before release. | labels: E-easy | low | 3 |
rust_1383 | Type unification error messages excessively mystifying | Rust is rapidly gaining a very expressive type and kind system for pointers, but the errors seem to be decreasingly clear; for instance, I came across the following errors today when updating some code: ``` ./fftw.rs:52:8: 52:25 error: mismatched types: expected ... | medium | 2 |
rust_1384 | Request for source addition | I'm developing some stuff off and on that ends up landing in https://github.com/jwise/rust-packages. Can this be added to the default sources.list on rust-lang.org? | labels: E-easy | low | 3 |
rust_1386 | split stdtest into stdtest/coretest | Stuff in core should not be tested in stdtest | labels: E-easy, C-cleanup, A-testsuite | low | 3 |
rust_1387 | invalid location with #debug type error | This code highlights an invalid code location: ``` fn main() { let x = [1,2,3]; #debug(x); } ``` Returns this error message: ``` foo.rs:4:186: 4:187 error: first argument to #fmt must be a string literal. foo.rs:4 } ^ rust: upcall fail 'explicit failure', ../src/comp/driver/ses... | medium | 2 |
rust_1388 | odd issue with #fmt and floats | I've run into a rather weird error with #fmt and floats: ``` use std; import std::io; fn main() { let f = -0.550153; let s = #fmt("%f", f); io::println(s); } ``` Incorrectly prints out `0`. But if we instead write: ``` use std; import std::io; fn main() { let f = -0.550153; let s | medium | 2 |
rust_1390 | kind checker does not recognize implicit returns | This shouldn't compile without annotating the parameter as `copy` kind, but it does: ``` fn id<T>(x: T) -> T { x } ``` If you return by `ret`, the problem is detected. | labels: A-type-system | medium | 2 |
rust_1391 | make rustdoc handle imports and exports | rustdoc should copy docs from imported symbols. this would, for example, help float get docs from f32/f64. Probably this should wait/ties in with copying default docs from interfaces when typeclasses become available. | labels: T-rustdoc | low | 3 |
rust_1393 | support explicit discriminant numbers on tag variants | Currently tags always use 0 ... N, where N is the number of tag variants, as the discriminant values. This makes them useless for modeling C enums, which they are the natural analogue of. We should support a form of variant-declarator that provides a discriminant ... | question | 4 |
rust_1394 | rustc, rustdoc: give nicer feedback when no arguments are given | A "how to use" information instead of the "upcall fail" messages would be nice: ``` $ rustc error: No input filename given. rust: upcall fail 'explicit failure', ../src/comp/driver/driver.rs:570 rust: domain main @0x10202aa00 root task failed $ rustdoc r... | low | 3 |
rust_1395 | make check and install should depend on all | Running 'make check && make install' isn't quite enough to get all targets built. Depend on 'all'. | medium | 2 |
rust_1396 | change "when" to "if" in pattern guards | I suggest we replace the word "when" in pattern guards with the word "if". This is the only place that "when" appears in our grammar and so it would seem more consistent: ``` alt foo { pat(a, b) if cond { ... } pat(a, b) { ... } } ``` | labels: A-frontend, E-easy | low | 3 |
rust_1398 | Allow tuples and other irrefutable patterns in parameter specifications | Currently we can write ``` for (a, b) in vec::zip(as, bs) { ... } ``` but not ``` vec::iter(vec::zip(as, bs)) { |(a, b)| ... } ``` nor ``` fn proc((a: A, b: B)) { ... } vec::iter(vec::zip(as, bs), proc) ``` It would be nice to have these irrefuta... | medium | 2 |
rust_1399 | Last use doesn't consider closure bodies properly | The following test case will crash (once fixed, something similar should be added to the test suite): ``` fn invoke_several_times(f: lambda()) { f(); f(); f(); } fn bycopy<K>(+k: K) { log(error,k); } fn byref<K>(&&k: K) { log(error,k); } fn testfn(+k: ~int) { bycopy(k... | high | 1 |
rust_1400 | trans.rs assertion error compiling code | I've reduced this down to about bare minimum: ``` use std; native mod foo { fn bar(x: int); } obj new(i: int) { fn f() { str::as_buf("", {|_x| foo::bar(i) }); } } ``` Errors out with: ``` rust: upcall fail 'Assertion cx.fcx.llobjfields.contains_key(did.node) failed', ../src/com... | medium | 2 |
rust_1401 | make all no longer works after make clean | At least for me, running make clean seems to then prevent me from ever building again. I get lots of linking errors related to LLVM (see below). I presume this is related to 7e611366bb07f1c26cf5692bccfdfa4fb1d96d25. ``` g++ -dynamiclib -lpthread -framework CoreServices -Wl,-n... | question | 4 |
rust_1402 | passing rust records by value to native functions broken | I thought rust records were supposed to be compatible with c structs, but it seems there is an issue passing them by value to a native module. You can find a example of the bug here: https://gist.github.com/1543358 | labels: E-hard, A-codegen | medium | 2 |
rust_1403 | warn when int or uint is used in a native type decl | Perhaps we should issue a warning---or even an error?---when `int` or `uint` is used in a native type declaration? It is a very common mistake to assume they are the same as the C types `int` and `unsigned`. We could suggest the use of `ctypes::c_int` or `ctypes::lo... | low | 3 |
rust_1405 | rustc: errors should not display runtime boilerplate logging | Whenever I have a filename typo when running rustc, I always think I'm tripping over some compiler bug before I notice the `error: error opening ...` line at the top. Could rustc catch that the file doesn't exist earlier and cleanly exit instead? ``` % rust... | low | 3 |
rust_1406 | Add fixed sized arrays for native module records | Some c libraries, like zeromq and libclang, return structures by value that contain fixed size arrays. We can't bind to these functions though because there is no way to express a fixed size array in a structure in rust. It would greatly simplify binding to these libra... | question | 4 |
rust_1410 | Set up FreeBSD bot | FreeBSD support is almost here (#1409). We'll need a bot. | question | 4 |
rust_1411 | All non-www URLs redirect to front page | http://rust-lang.org/doc/tutorial/index.html should redirect to http://www.rust-lang.org/doc/tutorial/index.html but instead it redirects to http://rust-lang.org/ | medium | 2 |
rust_1412 | Standard Library documentation: sort function names | I think the list of functions e.g. on http://www.rust-lang.org/doc/core/files/str-rs.html would be much easier to browse, if the function names were sorted. Is this created by rustdoc? | labels: T-rustdoc | medium | 2 |
rust_1413 | Standard Library documentation output: missing record return type | See http://doc.rust-lang.org/doc/core/files/str-rs.html#str.char_range_at "char_range" is of type ``` fn char_range_at(s: str, i: uint) -> {ch: char, next: uint} ``` but the output only shows ``` fn char_range_at(s: str, i: uint) -> ``` | labels: T-rus... | medium | 2 |
rust_1415 | Stop passing spans around in trans | These spans are only used in output generated when hitting an unexpected circumstance (compiler bug). Passing them around is cumbersome, and already being done very poorly (the association of spans with block contexts is completely busted), so I think the code is better off witho | ... | low | 3 |
rust_1416 | Revert llvm branch commit | https://github.com/brson/llvm/commit/72b5218a095dc390cba26f75baf4f616369ca295 is incorrect and unnecessary. | labels: A-LLVM | medium | 2 |
rust_1417 | do something more careful than eval_const_expr | Eval_const_expr is used in the front end in several places. This code quasi-duplicates stuff that trans should probably be doing (in Graydon's words). Any evaluation of compile-time values should likely be moved to a pass somewhere in the middle end. I assume there is a ... | low | 3 |
rust_1418 | Tutorial is out of date online | This change from November 2011 https://github.com/graydon/rust/commit/6168f8f15b8ad86e5b0670e39d6fcc0a212acb59#doc/tutorial/test.md is not yet online: http://www.rust-lang.org/doc/tutorial/test.html Can we just point to the GitHub Markdown files? | medium | 2 |
rust_1419 | Stop building stage3 on normal test cycles | We're not comparing stage2 and stage3 at the moment, and the amount of bugs caught by that extra stage that wouldn't be caught in stage2 is probably very small. We could shave off some cycle time by only going to stage2, and doing a stage3 check only on checkpoints. | labels... | low | 3 |
rust_1420 | rustlang.org still points to github | rustlang.org points to https://github.com/graydon/rust whereas rust-lang.org is an actual Web site. rustlang.org should probably be an alias for rust-lang.org. | labels: E-easy | low | 3 |
rust_1422 | Allow modifications to local variables in pure functions | I know this issue is on someone's plate, but i could not find a related issue. Currently we can not iterate over strings, vectors, lists etc using pure functions. That means that functions like `str::is_whitespace()` can not be declared pure either, even though... | low | 3 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.