eda8005 | Keno Fischer | 07 January 2019, 19:29:04 UTC | Add a hook for interpreting foreigncall | 07 January 2019, 19:31:15 UTC |
40cdae2 | Raghvendra Gupta | 07 January 2019, 14:26:26 UTC | Fixed norm (#30481) * Fixed norm * Modified and added tests for norm * Added AbstractFloat for Quaternion Type * Updated norm * Modified norm(::Number) to return float * Removed -p test for norm * Updated tests * Minor change in tests | 07 January 2019, 14:26:26 UTC |
49b8e91 | AzamatB | 07 January 2019, 14:20:39 UTC | Use new Array constructor syntax in offset-arrays.md (#30626) | 07 January 2019, 14:20:39 UTC |
0bce3df | byronbest | 07 January 2019, 14:06:42 UTC | dev docs: fix struct size descriptions (#30620) example sizes are 8 and 16 bytes; should be 1 and 2 for uint8 and int16, resp. | 07 January 2019, 14:06:42 UTC |
dffe119 | jmert | 07 January 2019, 08:51:07 UTC | Sparsity-preserving outer products (#24980) * Add indtype and nnz definitions for SparseColumnView * Handle sparse outer products specially in broadcast * Add specialized kron for sparse outer products * Add tests * Support unitful types * Address review comments. * Change is_specialcase_sparse_broadcast -> can_skip_sparsification. * Lift parent(y) to one function earlier for clarify * Simply call _copy instead of passing through the broadcast machinery again | 07 January 2019, 08:51:07 UTC |
f8f2045 | Keno Fischer | 05 January 2019, 00:35:28 UTC | Fix use counts for mutable struct SROA PR #28478 moved the computation of the use counts before the finish call. to fix #28444. However, the early parts of the finish call fixes up phi node arguments, which fail to get counted if we look at use counts before that fixup is performed. This causes #30594 where the only non-trivial use is on the backedge of the phi and would thus incorrectly fail to get accounted for. Fix that by taking the use count after phi fixup but before dce. | 06 January 2019, 21:04:48 UTC |
efe367d | Keno Fischer | 31 December 2018, 06:00:04 UTC | Vendor newer config.sub for emscripten support in dependencies Upstream config.sub supports emscripten as of three hours ago [1]. However, this file is generally bundled in the tarballs of our dependencies, so it'll be some time until they have caught up. In the meantime, we'll have to patch in the newer config.sub to be able to build our dependencies against emscripten. Since all autotools-based projects can use the same emscripten, we distribute the file itself rather than trying to maintain patch files for whatever version of config.sub our dependencies happen to ship. It's likely that some of the other dependencies will also need this, but for now I've only added it to pcre/gmp/mpfr, which have been confirmed to work when built with emscripten. We can add this patch to other dependencies as they get enabled on the emscripten port. [1] http://git.savannah.gnu.org/cgit/config.git/commit/?id=b3914f8d5588d1f9ce2089cf711119653fe469fc) | 06 January 2019, 21:03:57 UTC |
2931072 | petershintech | 06 January 2019, 20:39:28 UTC | Make sure to call jl_islayout_inline() even in the case assertion is deactivated (#30615) | 06 January 2019, 20:39:28 UTC |
6d2425b | Tim Holy | 06 January 2019, 18:49:50 UTC | Merge pull request #30616 from AzamatB/patch-1 Update outdated syntax in offset-arrays.md | 06 January 2019, 18:49:50 UTC |
d242c2d | AzamatB | 06 January 2019, 05:45:39 UTC | Update offset-arrays.md | 06 January 2019, 05:45:39 UTC |
a25945a | Ashley Sommer | 06 January 2019, 04:06:38 UTC | fix typo in string search api docs (#30600) | 06 January 2019, 04:06:38 UTC |
ffb6f1e | Alex Arslan | 06 January 2019, 00:48:31 UTC | Update OpenBLAS to v0.3.5 (#30583) * Update OpenBLAS to v0.3.5 * Bump BB OpenBLAS bundle * Update `OpenBLAS` checksums to `v0.3.5` * Remove debugging `set -x` in `contrib./refresh_bb_tarballs.sh` * Add a NEWS entry | 06 January 2019, 00:48:31 UTC |
c0721ad | Alex Arslan | 05 January 2019, 22:22:33 UTC | Update libgit2 to v0.27.7 (#30584) * Update libgit2 to v0.27.7 * Remove problematic portions of the libgit2 patches These patches make changes to libgit2's .travis.yml file which was removed when libgit2 switched to Azure Pipelines for CI. Since they are non-functional changes for us, we can simply remove them from the diff. | 05 January 2019, 22:22:33 UTC |
a4b3261 | Alfredo Braunstein | 05 January 2019, 16:38:58 UTC | Faster and simpler sprand for SparseMatrixCSC (#30494) * simpler and faster sprand * doctests * shaving about half the time in _sprand * avoid problems with T<:Integer, T!=Int * add tests clarifying the output distribution of sprand | 05 January 2019, 16:38:58 UTC |
862fe08 | Alex Arslan | 05 January 2019, 06:45:30 UTC | Use XCode 8.3 for macOS on Travis (#30599) This corresponds to macOS 10.12 Sierra. XCode 8 covers El Capitan and Sierra, so if Travis is giving us XCode 8.x for x < 3, we're on El Cap. Homebrew supports only three versions of macOS at a time, which means that El Cap (10.11) is no longer support. This is likely why our Mac builds are trying to build GCC from source; a bottle might not be available. | 05 January 2019, 06:45:30 UTC |
5290bba | ederag | 04 January 2019, 21:35:12 UTC | clarify that the tuple is related to the full signature in dev docs (PR #30556) | 04 January 2019, 21:35:12 UTC |
cf404cf | Jeff Bezanson | 04 January 2019, 20:10:47 UTC | fix type annotation of dynamic `new` expressions in codegen (#30574) | 04 January 2019, 20:10:47 UTC |
8025bab | Jeff Bezanson | 04 January 2019, 20:05:55 UTC | fix #30505, `fieldtype` of Unions where some types would error (#30546) | 04 January 2019, 20:05:55 UTC |
a153fea | Jeff Bezanson | 04 January 2019, 17:53:25 UTC | optimize identity splat of a tuple, `(t...,)` (#30571) | 04 January 2019, 17:53:25 UTC |
0b7f689 | Jeff Bezanson | 04 January 2019, 17:16:24 UTC | don't treat `continue` as incomplete input in the REPL (#30570) | 04 January 2019, 17:16:24 UTC |
fb47697 | Jeff Bezanson | 04 January 2019, 16:56:18 UTC | make the test for #19957 more rigorous (#30578) closes #19957 | 04 January 2019, 16:56:18 UTC |
4d8a968 | Viral B. Shah | 04 January 2019, 05:05:38 UTC | Add the scaled identity matrix to a random matrix to avoid getting a singular matrix (#30576) Fix #30572 | 04 January 2019, 05:05:38 UTC |
43faeac | Viral B. Shah | 03 January 2019, 23:36:11 UTC | Update the community links. (#30540) [ci skip] | 03 January 2019, 23:36:11 UTC |
84a83ab | Jeff Bezanson | 03 January 2019, 23:09:38 UTC | fix `lambda-optimize-vars!` with complex assignment RHSs (#30564) fixes #30563 | 03 January 2019, 23:09:38 UTC |
8b189ec | Elliot Saba | 03 January 2019, 22:03:38 UTC | Fix `bb-install` naming conventions, add hashes (#30535) * Fix `bb-install` naming conventions, add hashes * Set `DEP_LIBS` to include `openblas` on Appveyor * When guessing BB libc, default to `glibc` on Linux * Fix bb-install bash parsing failure Quote to avoid bash freaking out from spurious `)` from compilers that have fancy version strings such as `(Red Hat 7.3.1-5)`, which gets `lastword`'ed down to `7.3.1-5)`. * Add `contrib/refresh_bb_tarballs.sh` to aid in batch-grabbing BB hashes | 03 January 2019, 22:03:38 UTC |
5233ec0 | petershintech | 03 January 2019, 21:59:57 UTC | Fix a typo on a Korean letter. Not "한민êµ" but "대한민êµ" (#30567) Fix for #30486 | 03 January 2019, 21:59:57 UTC |
76e7421 | Jameson Nash | 03 January 2019, 18:14:55 UTC | gf: fix jl_isa_compileable_sig again (#30458) The last attempts were pretty good, but still missed a lot. But this is what you find when you actually try to test it. This is not too important, but it can reduce compilation performance in some case, so it is not ideal. | 03 January 2019, 18:14:55 UTC |
041c214 | Jameson Nash | 03 January 2019, 18:12:50 UTC | llvm: fix target triple (#30554) broken by their move to cmake causing a switch away from the standard --host/--build autoconf fix #28046 | 03 January 2019, 18:12:50 UTC |
a5a5882 | Jeff Bezanson | 03 January 2019, 03:45:58 UTC | use type inference world in `return_type` (#30470) | 03 January 2019, 03:45:58 UTC |
9731ee7 | Jan Weidner | 02 January 2019, 23:00:48 UTC | Fix #30462, out of bounds access in maximum/minimum for certain-sized arrays (#30484) * fix #30462 * fix * more tests | 02 January 2019, 23:00:48 UTC |
5a8ae94 | Matt Bauman | 02 January 2019, 21:48:05 UTC | Merge pull request #30480 from JuliaLang/nl/isa Fix performance of broadcast and collect with Union{T, Missing} | 02 January 2019, 21:48:05 UTC |
356ceee | Andreas Noack | 02 January 2019, 20:06:42 UTC | Add custom deserialize method for UmfpackLU to avoid memory leak (#30425) Fixes #15450 | 02 January 2019, 20:06:42 UTC |
3e6f607 | Jeff Bezanson | 02 January 2019, 20:00:08 UTC | don't use abstract_iteration for types with special cases in _apply (#30483) | 02 January 2019, 20:00:08 UTC |
7ba11d1 | Katharine Hyatt | 02 January 2019, 16:18:49 UTC | Tests for iswritable for TCPSocket (#30550) * Tests for iswritable for TCPSocket | 02 January 2019, 16:18:49 UTC |
2c64220 | Katharine Hyatt | 02 January 2019, 12:07:01 UTC | Two missing tests for sprand (#30551) | 02 January 2019, 12:07:01 UTC |
c154215 | Alfredo Braunstein | 02 January 2019, 10:42:50 UTC | implements copy!(dst::SparseVector, src) (fixes #30443) (#30531) * implements copy!(dst::SparseVector, src) (fixes #30443) * Adjust error message * more efficient copy! * fix typo * fix return value * add an assert and an explicit return | 02 January 2019, 10:42:50 UTC |
f713a4b | Harmen Stoppels | 02 January 2019, 08:59:13 UTC | Add test macro for relaxed inference of small unions (#27516) | 02 January 2019, 08:59:13 UTC |
61f9412 | Elliot Saba | 02 January 2019, 00:25:37 UTC | Fix comparison in `getipaddrs()` test with multiple IPs (#30547) | 02 January 2019, 00:25:37 UTC |
8221c87 | Stephen Eglen | 31 December 2018, 18:52:01 UTC | align superscripts in comment for tan(y) (#30522) | 31 December 2018, 18:52:01 UTC |
2e91c5e | Alex Arslan | 30 December 2018, 07:21:48 UTC | Merge pull request #30533 from JuliaLang/aa/housekeeping Add license headers and fix links in docs | 30 December 2018, 07:21:48 UTC |
3f01ca3 | Alex Arslan | 29 December 2018, 23:44:19 UTC | Update and fix links in the documentation | 30 December 2018, 00:23:05 UTC |
f9e8449 | Alex Arslan | 29 December 2018, 22:35:37 UTC | Add license headers to files | 30 December 2018, 00:21:25 UTC |
121e814 | Ben Arthur | 29 December 2018, 13:48:42 UTC | cluster manager fixes (#30172) * kill workers which don't launch properly * don't emit spurious error messages * document how to asynchronously launch workers | 29 December 2018, 13:48:42 UTC |
87c18d8 | Elliot Saba | 29 December 2018, 07:40:15 UTC | Add BB downloading for OpenBLAS (#30497) * Auto-detect binarybuilder triplet * Add OpenBLAS BinaryBuilder installation scaffolding Also make it easier to add more BB-cached versions of dependencies in the future * Enable `fixup-libgfortran.sh` to directly ask `$FC` for paths * Tell Appveyor and Travis to use BinaryBuilder OpenBLAS Also allow the build system to auto-guess the triplet | 29 December 2018, 07:40:15 UTC |
ba04451 | Keno Fischer | 28 December 2018, 03:30:33 UTC | Add target defines for wasm/emscripten This is perhaps a bit of an overeager commit, given that most of the code to support wasm is not yet upstream, but I think it's important to get our terminology clear early. This attempts to define platform defines in both our C code and at the Julia level. For background, wasm is a general standard for secure execution of untrusted code arising out of the Javascript world (many web browsers these days ship wasm support in their Javascript engine). However, wasm is not tied to JS, so it would be good to avoid hardcoding any such assumptions into our platform defines (e.g. people are considering wasm for kernel space vms, etc.). Emscripten on the other hand is the original C-to-JS compiler, as well as corresponding runtime libraries, simulatenously fulfulling the role of C library and operating system kernel (for some features at least - e.g. providing file system emulation). As such, my suggestion is the following: - Sys.OS gets set to `Emscripten`, which is also what the emulated `uname` syscall returns when emulated by Emscripten, so is consistent with our documentation for what that global means. - A new `isjsvm` function can be used to determine if we're running inside a JavaScript VM (I'd expect this check to be quite common, e.g. to call out to the native RegEx engine). - `isunix` is true for Emscripten, but I don't envision it being true generally for either `isjvm` platforms or wasm platforms. - I didn't currently add a specific check for wasm, since I'd suspect this to be rather rare (e.g. only if you wanted to do something like emit inline wasm, or have a specific implementation for wasm targets). However, you can of course always check Sys.MACHINE for wasm32-*-* or wasm64-*-*. If it turns out to be useful, we can add it. The C code follows the same reasoning and defines _OS_EMSCRIPTEN_ and _CPU_WASM_. I didn't bother defining something for jsvm, since the changes to the C code will have to be quite closely tied to specifics of Emscripten. | 29 December 2018, 05:48:20 UTC |
6dc205a | Steven G. Johnson | 29 December 2018, 03:06:04 UTC | faster mapfoldl for tuples (#30471) | 29 December 2018, 03:06:04 UTC |
002a9f5 | Fredrik Ekre | 27 December 2018, 22:27:32 UTC | Bump Pkg to 1.1.2. (#30521) | 27 December 2018, 22:27:32 UTC |
2ab1405 | Raghvendra Gupta | 27 December 2018, 06:43:32 UTC | Fixed sprand for dimensions of unusual integer type (#30516) Fixes #30502 | 27 December 2018, 06:43:32 UTC |
573cf74 | Fredrik Ekre | 27 December 2018, 02:41:38 UTC | Add Dates as a test dependency to SparseArrays. (#30519) | 27 December 2018, 02:41:38 UTC |
71518a3 | Kenta Sato | 26 December 2018, 19:19:38 UTC | remove unnecessary named let in parse-assignment (#30511) | 26 December 2018, 19:19:38 UTC |
b73a746 | Raghvendra Gupta | 26 December 2018, 17:26:06 UTC | Fix DimensionMismatch in SparseMatrixCSC assignment (#30507) Use _setindex! to fix DimensionMismatch in sparse assignment Fixes #28963 | 26 December 2018, 17:26:06 UTC |
74bf25c | lizz | 25 December 2018, 16:35:56 UTC | Update README.md (#30508) 1.0.2->1.0.3 | 25 December 2018, 16:35:56 UTC |
94993e9 | Alfredo Braunstein | 25 December 2018, 13:06:40 UTC | faster circshift! for SparseMatrixCSC (#30317) * implement circshift! for SparseMatrixCSC * factor helper function shifter!, implement efficient circshift! for SparseVector * add some @inbounds for improved performance * remove allocations completely, giving a large improvement for small matrices * some renaming to avoid polluting the module namespace * remove useless reallocation and fix bug with different in/out types, better tests * avoid action if iszero(r) and/or iszero(c), move sparse vector shifting helpers to sparsevector.jl * Make shift amounts deterministic in tests, move sparse vector tests into sparsevector.jl * comment fix * for some reason, copy!(a::SparseVector, b::SparseVector) does not work | 25 December 2018, 13:06:40 UTC |
8eca27d | JonasIsensee | 24 December 2018, 05:28:09 UTC | Add link to `Symbol` in docstring of `:` (#30468) * Add link to `Symbol` in docstring of `:` I believe it would be helpful to add a reference to `Symbol` to the docstring of `:`. ATM searching for `?:` in the REPL does not mention the use of `:` in symbols. * Update base/range.jl Co-Authored-By: JonasIsensee <jonas.isensee@web.de> | 24 December 2018, 05:28:09 UTC |
341e031 | DanielChen | 22 December 2018, 21:47:20 UTC | Fix issue ref typo in comments of `doc/make.jl` (#30490) In the doc/make.jl, there is a code block for making links for stdlib. It mentioned JuliaDocs/Documenter.jl#522 in the comment but JuliaDocs/Documenter.jl#522 seems not relevant to this and it should be JuliaDocs/Documenter.jl#552. Fix #30489. | 22 December 2018, 21:47:20 UTC |
817d97d | Harrison Grodin | 22 December 2018, 19:05:15 UTC | Ignore compactness in show_delim_array (#30442) Also remove show_comma_array | 22 December 2018, 19:05:15 UTC |
c387c4b | Jeff Bezanson | 22 December 2018, 19:00:58 UTC | fix HTML display of method static parameters (#30472) This was still using pre-0.7 syntax. | 22 December 2018, 19:00:58 UTC |
da1df1e | Samuel Massinon | 22 December 2018, 04:25:40 UTC | Add getipaddrs (#30349) * Add getipaddrs getipaddr would only return the first ip address it would find. getipaddrs will return all the ip addresses it finds. * Add getipaddrs to docs * Update NEWS for `getipaddrs` * Add Note To `getipaddrs` Mention that the function is available in Julia 1.2 * Allow `getipaddrs` to Return lo IPs * Add document on `include_lo` | 22 December 2018, 04:25:40 UTC |
7acb991 | Tim Besard | 21 December 2018, 23:33:20 UTC | Only use llvm-config for library selection when building against system LLVM. (#30459) | 21 December 2018, 23:33:20 UTC |
7c3904a | Nathan Daly | 21 December 2018, 20:43:21 UTC | rm `doc/NEWS-update.jl` step from CONTRIBUTING.md (#30482) Update CONTRIBUTING.md to stop recommending that people run `./julia doc/NEWS-update.jl` after every `NEWS.md` change in every PR. | 21 December 2018, 20:43:21 UTC |
f503e80 | Valentin Churavy | 21 December 2018, 19:50:26 UTC | Merge pull request #30477 from raghav9-97/repeat Tests for repeat with 0-dimensional Array | 21 December 2018, 19:50:26 UTC |
3648f4c | Patrick Kofod Mogensen | 21 December 2018, 19:45:23 UTC | Port helper functions,_ldexp_exp and _frexp_exp, for hyperbolic functions (#29499) | 21 December 2018, 19:45:23 UTC |
163b019 | Valentin Churavy | 21 December 2018, 15:38:17 UTC | Merge pull request #30392 from JuliaLang/tb/help_hidden Add a CLI help mode to show hidden opts. | 21 December 2018, 15:38:17 UTC |
6646d1e | Jarrett Revels | 19 December 2018, 15:53:42 UTC | add extra convenience kwargs to CustomParams and code_typed | 21 December 2018, 15:22:54 UTC |
184fbc4 | Milan Bouchet-Valat | 21 December 2018, 11:35:30 UTC | Fix performance of broadcast and collect with Union{T, Missing} Use the same pattern as in collect_to_with_first! (which is used when size is known). | 21 December 2018, 12:57:54 UTC |
e5b5a01 | raghav9-97 | 21 December 2018, 06:08:40 UTC | Tests for repeat with 0-dimensional Array | 21 December 2018, 06:08:40 UTC |
c8c48a2 | Nathan Daly | 21 December 2018, 01:59:34 UTC | Fix NEWS.md by running `NEWS-update.jl` (#30474) Fixes the out-of-date hyperlinks on NEWS.md. | 21 December 2018, 01:59:34 UTC |
851ae46 | Alex Arslan | 20 December 2018, 22:33:59 UTC | Use a tier system for platform support (#30447) This makes the level to which we support various platforms more specific and adds a bit of formalism that's typical of other projects, e.g. Rust. The tier system is explained in the document, but to recap: 1. Builds and passes tests, have binaries and CI 2. Builds, may pass tests, may have binaries 3. May build 4. Hardcore broken All of our listed platforms have been assigned a tier, which is pretty obvious for most of them. I've also added an entry in the support table for Linux with musl libc. Julia builds and passes most tests on Alpine, but that isn't verified regularly so I've assigned that tier 3 for now. [ci skip] | 20 December 2018, 22:33:59 UTC |
8b1bf4c | Alex Arslan | 20 December 2018, 20:11:02 UTC | Disable a couple of errorshow tests on FreeBSD (#30428) These two specific errorshow tests work on FreeBSD debug builds, like what we run on CI, but they fail with regular FreeBSD builds. The buildbots that upload nightly binaries ensure that the tests pass prior to uploading, and the fact that this doesn't pass on regular builds is what's keeping us from having FreeBSD nightly binaries. See issue 30233. | 20 December 2018, 20:11:02 UTC |
8604233 | Raghvendra Gupta | 20 December 2018, 16:55:40 UTC | Fix #18742: Added check for `/` DimensionMismatch error (#30446) * Added check for / DimensionMismatch error * Added tests | 20 December 2018, 16:55:40 UTC |
e7e33cc | Klaus Crusius | 20 December 2018, 16:46:07 UTC | test cases and NEWS for #30372 (sparse matmul) (#30433) | 20 December 2018, 16:46:07 UTC |
258e08a | Markus Kuhn | 20 December 2018, 07:34:18 UTC | Base.worker_timeout() mention in manual (#30439) The manual mentions at https://docs.julialang.org/en/v1/manual/environment-variables/#JULIA_WORKER_TIMEOUT-1 a function Base.worker_timeout() but the implementation has instead only a function Distributed.worker_timeout() | 20 December 2018, 07:34:18 UTC |
735d755 | Jameson Nash | 20 December 2018, 04:38:37 UTC | coverage: support output in LCOV tracefile format (#30381) Also ensure that `julia_cmd` is forwarding all desirable options, and provide a general framework for specifying formatted filenames for similar such options. | 20 December 2018, 04:38:37 UTC |
072ad7d | Dominique | 20 December 2018, 01:42:28 UTC | generalize sparse matrix slicing to integer types (#30319) | 20 December 2018, 01:42:28 UTC |
744ead4 | Jan Weidner | 20 December 2018, 01:33:45 UTC | fix maximum on generic orders (#30441) Close #30436 Address the non-numeric values case that is visible only with long arrays reported in #30320 | 20 December 2018, 01:33:45 UTC |
99efc91 | Stephan Hilb | 19 December 2018, 23:40:45 UTC | stdlib/SparseArrays: fix scalar setindex! for vector eltype (#29331) fix #29034 | 19 December 2018, 23:40:45 UTC |
21f7a1e | Fredrik Ekre | 19 December 2018, 21:39:11 UTC | Add link to the pdf version of the documentation to the html pages, fix #28604. (#30449) | 19 December 2018, 21:39:11 UTC |
2c71488 | Jeff Bezanson | 19 December 2018, 16:48:13 UTC | make `ndigits` more generic (#30384) | 19 December 2018, 16:48:13 UTC |
eddcbd7 | Keno Fischer | 30 November 2018, 21:06:42 UTC | Completely remove undefined behavior | 18 December 2018, 21:59:25 UTC |
78159b9 | Jameson Nash | 19 March 2018, 17:01:01 UTC | MurmerHash: avoid undef-behavior in C This pointer is not necessarily properly aligned, making this undefined behavior. All modern compilers should optimize this to the proper sequence for unaligned access. (confirmed with objdump that Apple LLVM version 9.0.0 (clang-900.0.38)) gives exactly the same asm for hashing.o as before) fix #26512 | 18 December 2018, 21:59:25 UTC |
92ac90e | Jarrett Revels | 14 December 2018, 01:08:29 UTC | attempt to refine return type when it could be improved via PartialTuple | 18 December 2018, 20:54:56 UTC |
3380501 | Jarrett Revels | 21 April 2018, 18:36:02 UTC | log when a worker begins a test so that it is easier to debug when a worker hangs due to test failures | 18 December 2018, 20:17:13 UTC |
e6938a0 | Jeff Bezanson | 18 December 2018, 17:26:19 UTC | fix #30124, broadcast regression due to removed pure annotation (#30420) | 18 December 2018, 17:26:19 UTC |
7b2db8e | Jan Weidner | 18 December 2018, 17:03:43 UTC | Fix #28849: faster maximum, minimum (#30320) | 18 December 2018, 17:03:43 UTC |
8d8b3d9 | Samikshya Chand | 18 December 2018, 16:46:17 UTC | Fix `repr` on `DateTime` (#30200) * Add print * Add show function * modify tests * Adding changes in docs * readability * add compact case * revert compact case * compact case * typo * add news * Update NEWS.md | 18 December 2018, 16:46:17 UTC |
a0b7a76 | Jameson Nash | 18 December 2018, 02:23:03 UTC | Condition/RecursiveLock: add ability to handle threads (#30061) This extends Condition to assert that it may only be used in the single-threaded case (co-operatively scheduled), and then adds a thread-safe version of the same: `Threads.Condition`. Additionally, it also upgrades ReentrantLock, etc. to be thread-safe. | 18 December 2018, 02:23:03 UTC |
046755c | Raghvendra Gupta | 17 December 2018, 22:41:42 UTC | Fixes #24214::isvalid(Substring{String}) (#30397) | 17 December 2018, 22:41:42 UTC |
8b35e84 | Jameson Nash | 17 December 2018, 21:22:09 UTC | loading: work on simplifications (and some corrections) of docs (#29946) | 17 December 2018, 21:22:09 UTC |
433ba13 | Matt Bauman | 17 December 2018, 20:51:46 UTC | Try implementing N-dimensional indexing for fast linear SubArrays (#30266) | 17 December 2018, 20:51:46 UTC |
b167bc2 | Jeff Bezanson | 17 December 2018, 16:39:38 UTC | fix #30394, an unsoundness in ml_matches (#30396) This fixes a corner case where a bug is caused, counter-intuitively, by an over-estimated intersection. We have method signatures A and B, with A<B (A is a strict subtype). We have a dispatch tuple X, where X<:B and !(X<:A). However, intersection returns X for intersect(X,A). Since there appears to be a match there and A<B, ml_matches skips the match with B. The fix just requires dispatch tuples to be a subtype of a signature in order to match at all. | 17 December 2018, 16:39:38 UTC |
fae262c | Klaus Crusius | 17 December 2018, 15:58:35 UTC | spmatmul sparse matrix multiplication - performance improvements (#30372) * General performance improvements for sparse matmul Details for the polyalgorithm are in: https://github.com/JuliaLang/julia/pull/30372 | 17 December 2018, 15:58:35 UTC |
b451001 | Raghvendra Gupta | 17 December 2018, 14:52:59 UTC | Fix sparse cholesky to return Vector when the RHS is a Vector (#30416) Fixes #28985 | 17 December 2018, 14:52:59 UTC |
64133f6 | Raghvendra Gupta | 17 December 2018, 14:50:13 UTC | Fix #30006, getindex accessing fields that might not exist (#30405) * Fix #30006, range getindex accessing fields that might not exist * Add tests for #30006 | 17 December 2018, 14:50:13 UTC |
f36ace9 | Viral B. Shah | 17 December 2018, 14:47:09 UTC | Fix #20409: rank for sparse matrices. (#30415) | 17 December 2018, 14:47:09 UTC |
e51a707 | Jameson Nash | 17 December 2018, 02:09:56 UTC | stacktrace: prevent OOB-error in sysimage lookup (#30369) Previously, with a multi-versioned system image, there might be additional entries at the end of the clone list that do not correspond to an actual method (such as jlplt thunks). Also some code cleanup for clarity. fix #28648 | 17 December 2018, 02:09:56 UTC |
36cc20c | Jeff Bezanson | 16 December 2018, 22:17:46 UTC | fix unionall env in showing nested method static parameters (#30395) | 16 December 2018, 22:17:46 UTC |
01fe7c2 | Alex Arslan | 14 December 2018, 21:29:09 UTC | Allow passing a function to extrema (#30323) Currently `minimum` and `maximum` can accept a function argument, but `extrema` cannot. This makes it consistent. | 14 December 2018, 21:29:09 UTC |
43c6b57 | Katharine Hyatt | 14 December 2018, 21:17:19 UTC | Merge pull request #30292 from JuliaLang/ksh/nt A few missing NamedTuple tests | 14 December 2018, 21:17:19 UTC |
6326130 | Tim Besard | 14 December 2018, 16:18:54 UTC | Add a CLI help mode to show hidden opts. | 14 December 2018, 16:25:00 UTC |
c379900 | Fabian Gans | 14 December 2018, 09:41:51 UTC | fix reinterpret for 0-dimensional arrays (#30376) | 14 December 2018, 09:41:51 UTC |