| 5b8bba6 | Jameson Nash | 27 January 2021, 19:43:18 UTC | Force specialization on the broadcasteds | 12 April 2021, 18:54:09 UTC |
| 637a2f2 | Matt Bauman | 30 April 2020, 21:37:35 UTC | Remove inline annations from broadcast kernels | 12 April 2021, 18:51:57 UTC |
| c649ebb | Jeff Bezanson | 12 April 2021, 18:36:46 UTC | fix #39982, reduce cursor artifacts in REPL after #39538 (#40411) | 12 April 2021, 18:36:46 UTC |
| c3c8ffa | Jeff Bezanson | 11 April 2021, 20:39:33 UTC | fix #32843, codegen attempting `typeof(NULL)` for some union tags (#40365) also fixes part of #40065 | 11 April 2021, 20:39:33 UTC |
| 8fc38a4 | DilumAluthgeBot | 11 April 2021, 17:18:58 UTC | [automated] Bump the Pkg stdlib from 74098fe7 to 879c21d9 (#40422) | 11 April 2021, 17:18:58 UTC |
| 0a6ea9f | jz2903 | 11 April 2021, 15:08:45 UTC | fix some typos (#40434) Fix typos in comments | 11 April 2021, 15:08:45 UTC |
| d29126a | Ikko Ashimine | 11 April 2021, 11:54:40 UTC | doc: fix typo in io.jl (#40433) | 11 April 2021, 11:54:40 UTC |
| 6844259 | Sebastian Stock | 11 April 2021, 11:52:37 UTC | Preserve Hessenberg shape when possible (#40039) * Preserve Hessenberg shape when possible * Remove obsolete comments * Add NEWS * Support for unitful Hessenberg matrices * Mark some tests for unitful matrices as broken | 11 April 2021, 11:52:37 UTC |
| 5b7f4c5 | Sheehan Olver | 11 April 2021, 05:42:03 UTC | Fix eigvals(:::SymTridagonal) with longer off-diagonal vector (#40165) * Fix eigvals(:::SymTridagonal) with longer off-diagonal vector * Update tridiag.jl * Update lapack.jl * Revert "Update lapack.jl" This reverts commit 5c4047689c42a6fb4292aa3e6ceff3506e770142. * Update lapack.jl | 11 April 2021, 05:42:03 UTC |
| c87d85f | Logan Kilpatrick | 10 April 2021, 17:58:39 UTC | Update contributing guide with new video links (#40426) * Update contributing guide with new links * Update CONTRIBUTING.md | 10 April 2021, 17:58:39 UTC |
| f9c6d98 | Viral B. Shah | 10 April 2021, 17:55:41 UTC | Add LBT checksums and bump versions in all the places. (#40429) * Add LBT checksums * remove old checksums * Change LBT version in various places | 10 April 2021, 17:55:41 UTC |
| e260f2b | Kristoffer Carlsson | 10 April 2021, 16:32:45 UTC | fix a typo in TOML error message (#40428) * fix a typo in TOML error message | 10 April 2021, 16:32:45 UTC |
| 008275c | kimikage | 10 April 2021, 10:18:38 UTC | doc: fix examples with invalid MIME type (#40427) | 10 April 2021, 10:18:38 UTC |
| fc69c9a | Alexey Stukalov | 10 April 2021, 01:59:56 UTC | Allow any f for sum/minimum/maximum(f, v::AbstractSparseVector) (#29884) * sum/minimum/maximum(f, v::AbstractSparseVector) generalize sum/minimum/maximum(abs/abs2, v::AbstractSparseVector) to arbitrary f * add broken sum(f, [])==0 tests for reference | 10 April 2021, 01:59:56 UTC |
| bb608e5 | Thibaut Lienart | 09 April 2021, 23:17:33 UTC | regex: remove error when accessing an unset group (#40391) closes #31456 | 09 April 2021, 23:17:33 UTC |
| 722f90f | Michael Hatherly | 09 April 2021, 23:13:07 UTC | [REPL] Implement basic pager using TerminalMenus (#38956) | 09 April 2021, 23:13:07 UTC |
| 21eb1b6 | felix | 09 April 2021, 23:09:35 UTC | add some assorted subtyping tests (#20627) | 09 April 2021, 23:09:35 UTC |
| 2f0156a | DilumAluthgeBot | 09 April 2021, 22:26:20 UTC | [automated] Bump the Tar stdlib from ac4d442 to 86825ec (#40421) Co-authored-by: Dilum Aluthge <dilum@aluthge.com> | 09 April 2021, 22:26:20 UTC |
| 54c7002 | quildtide | 09 April 2021, 18:18:09 UTC | Improve flow of documentation for non-standard strings (#40057) | 09 April 2021, 18:18:09 UTC |
| 4996445 | Jonas Schulze | 09 April 2021, 14:19:28 UTC | Add U+2AEA (⫪) and U+2AEB (⫫) to binary operators (#39403) | 09 April 2021, 14:19:28 UTC |
| b838cdf | Arvind Ganesh | 09 April 2021, 14:16:03 UTC | Implemented bitwise nand and nor (#40339) | 09 April 2021, 14:16:03 UTC |
| 90bea03 | Jameson Nash | 09 April 2021, 11:19:38 UTC | remove unnecessary `@generated` from literal_pow (#40408) This was added before the Compiler could constant-prop, now it will just get in the way of general inferability. | 09 April 2021, 11:19:38 UTC |
| 752e459 | Benoît Legat | 09 April 2021, 10:19:05 UTC | Introduce ArrayContainer in performance tips (#36542) | 09 April 2021, 10:19:05 UTC |
| 40e57aa | Matt Bauman | 09 April 2021, 03:48:12 UTC | Dot-broadcasting for short-circuiting ops .&& and .|| (#39594) I have long wanted a proper fix for issue #5187. It was the very first Julia issue I filed. This is a shot at such a fix. This PR: * Enables parsing for `.&&` and `.||`. They are parsed into `Expr(:call, :.&&, ...)` expressions at the same precedence as their respective `&&` and `||`: ```julia-repl julia> Meta.show_sexpr(:(a .&& b)) (:call, :.&&, :a, :b) ``` * Unlike all other dotted operators `.op` (like `.+`), the `op`-alone part (`var"&&"`) is not an exported name from Base. As such, this effectively lowers to `broadcasted((x,y)->x && y, ...)`, but instead of using an anonymous function I've named it `Base.andand` and `Base.oror`: ```julia-repl julia> Meta.@lower a .&& b :($(Expr(:thunk, CodeInfo( @ none within `top-level scope' 1 ─ %1 = Base.broadcasted(Base.andand, a, b) │ %2 = Base.materialize(%1) └── return %2 )))) ``` * I've used a named function to enable short-circuiting behavior _within the broadcast kernel itself_. In the case that the second argument is a part of the same fused broadcast kernel, it will only evaluate if required: ```julia-repl julia> mutable struct F5187; x; end julia> (f::F5187)(x) = (f.x += x) julia> (iseven.(1:4) .|| (F5187(0)).(ones(4))) 4-element Vector{Real}: 1.0 true 2.0 true ``` Co-authored-by: Simeon Schaub <simeondavidschaub99@gmail.com> | 09 April 2021, 03:48:12 UTC |
| c2ec70c | Jameson Nash | 09 April 2021, 03:10:57 UTC | inference: limit single-level nested `Type` signature (#40379) Previously `type_more_complex` returns `false` for `Type{Type{...}}` compared against `Type{...}`. Per comment there, this should return `true`, but the code was ordered incorrect (it was also missing support for Vararg, after the recent changes). Fixes #40336 Co-authored-by: Shuhei Kadowaki <aviatesk@gmail.com> | 09 April 2021, 03:10:57 UTC |
| b539198 | Stefan Karpinski | 08 April 2021, 21:08:24 UTC | docs: more scope lawyering (re #40238) (#40394) | 08 April 2021, 21:08:24 UTC |
| 7d6dfe1 | Oscar Smith | 08 April 2021, 20:04:14 UTC | faster expm1 (#37440) Faster and slightly more accurate (and all Julia). | 08 April 2021, 20:04:14 UTC |
| 1defc11 | Jeff Bezanson | 08 April 2021, 19:59:27 UTC | remove method definition for `Int` in test/core.jl (#40380) | 08 April 2021, 19:59:27 UTC |
| d3b012b | Jameson Nash | 08 April 2021, 19:43:43 UTC | Merge pull request #40364 from JuliaLang/jn/searchsorted-unsigned-bug searchsorted ranges: simplify code, fix bug with Unsigned needle | 08 April 2021, 19:43:43 UTC |
| 5b6edad | Jameson Nash | 08 April 2021, 19:43:01 UTC | avoid calling dlerror unless absolutely necessary (#40392) Most platforms implement this sensibly, but POSIX does not require that, so FreeBSD does not implement this sensibly. But we can mostly avoid the problems on FreeBSD by avoiding calling it unless we absolutely must. Fixes #39582 | 08 April 2021, 19:43:01 UTC |
| f021c67 | Fredrik Ekre | 08 April 2021, 17:45:20 UTC | Fix temp install of BugReporting.jl (#40403) Fixes #40398 | 08 April 2021, 17:45:20 UTC |
| 41857b5 | jaakkor2 | 08 April 2021, 17:07:31 UTC | Bump MACOSX_VERSION_MIN from 10.9 to 10.10 (#40400) Fixes #39363 (BinaryBuilder already made this change) | 08 April 2021, 17:07:31 UTC |
| b373a9a | Federico Stra | 08 April 2021, 16:58:18 UTC | Fix signed integer overflow in `mapreduce_impl` (#38643) This PR solves #38627 as regards the code in reducedim.jl, reduce.jl and reinterpretarray.jl. It does not yet solve all the relevant issues with missing.jl: in the code there for _mapreduce and mapreduce_impl, there are still several instances of the construct: ``` while i <= ilast ... i += 1 end ``` This overflows when ilast == typemax(Int). More seriously, inside the while loops there is often `@inbounds A[i]`, which is then an out-of-bounds read! | 08 April 2021, 16:58:18 UTC |
| 9814bd6 | Andreas Noack | 08 April 2021, 16:54:19 UTC | Avoid AbstractString in GitError struct to make extraction inferred. (#40402) This significantly reduces the number of invalidated methods when loading ProgressLogging. | 08 April 2021, 16:54:19 UTC |
| 35157d9 | Jeffrey Sarnoff | 08 April 2021, 16:53:44 UTC | doc: clarify example returning a tuple (#40353) | 08 April 2021, 16:53:44 UTC |
| 3060edb | jmert | 08 April 2021, 16:46:02 UTC | fix #33413, crash with `cglobal(dlsym(...))` (#37910) | 08 April 2021, 16:46:02 UTC |
| fe09ad9 | Samuel Ainsworth | 08 April 2021, 16:34:12 UTC | Add comments to runtests.jl to explain the `net_on`/loopback situation (#40359) | 08 April 2021, 16:34:12 UTC |
| 0869969 | Elliot Saba | 08 April 2021, 16:05:57 UTC | Bump OpenBLAS_jll to `v0.3.13+5` (#40396) * Bump OpenBLAS_jll to `v0.3.13+5` This should fix the illegal instruction errors older processors are running into. * Fix doctests * Update checksums Co-authored-by: Viral B. Shah <viral@mayin.org> | 08 April 2021, 16:05:57 UTC |
| cc1e70a | Jeff Bezanson | 08 April 2021, 15:37:31 UTC | fix #40337, no error info from serialized TaskFailedException (#40395) | 08 April 2021, 15:37:31 UTC |
| 027bd4a | Benoît Legat | 08 April 2021, 04:56:34 UTC | Replace _iszero by iszerodefined && iszero (#38789) This PR defines an iszerodefined function that custom types implements in case iszero is well-defined even if zero is not defined. Take 2 of #36194 after it was reverted in #38764. What was missing in #36194 is that in case iszero is not working for a given type, we should drop the structure and fall back to an Array. | 08 April 2021, 04:56:34 UTC |
| b05f94d | ro-ble | 08 April 2021, 04:54:35 UTC | doc: hint on how to type the ÷ sign (#38762) The ÷ sign is the only operator in this chapter which has no dedicated key on the keyboard. With the Unicode input being at the very end of the documentation, newcomers probably do not know how to write the integer division sign. Therefore, I added this short hint on how to type it. | 08 April 2021, 04:54:35 UTC |
| 6117388 | Sukera | 08 April 2021, 02:58:13 UTC | Widen type signature of bytes2hex (#39710) and make it slightly faster! Also improves error message on hex2bytes! when passing a non-ASCII string, And ADD compat notice, fix implementation to be more generic in regards to AbstractString. | 08 April 2021, 02:58:13 UTC |
| 549380d | Ian Butterworth | 08 April 2021, 02:13:55 UTC | If module is missing a docstring show readme if available (#39093) * print available readme if no docstring for module * tweak search & print * handle pathless modules * limit readme print to first 200 lines * review changes * print message on single line * handle modules in sysimage that are missing src files * tweak formatting * fix docstring test | 08 April 2021, 02:13:55 UTC |
| 0da808d | Ian Butterworth | 08 April 2021, 01:11:05 UTC | add option to `compilecache` so that Pkg precompilation can not skip different versions of loaded modules (#40345) * allow Pkg precomp to not skip loaded modules * rename to ignore_loaded_modules * arg alignment | 08 April 2021, 01:11:05 UTC |
| 35a186b | Ian Butterworth | 08 April 2021, 01:07:31 UTC | Only print a single `LoadError:` prefix for nested LoadErrors during showerror (#38725) * fold repeated `LoadError :` prints during showerror * Only print a single LoadError: prefix | 08 April 2021, 01:07:31 UTC |
| 263451e | Ian Butterworth | 08 April 2021, 00:58:21 UTC | Allow setting lower bound on Experimental.@optlevel (#39536) * allow setting lower bound on Experimental.at-optlevel * rename arg to min-optlevel * Apply suggestions from code review Co-authored-by: Jeff Bezanson <jeff.bezanson@gmail.com> * more suggestions from code review Co-authored-by: Jeff Bezanson <jeff.bezanson@gmail.com> | 08 April 2021, 00:58:21 UTC |
| 76c1d1d | Mosè Giordano | 07 April 2021, 18:48:19 UTC | [Test] Define `broken`/`skip` keyword argument for `@test` (#39322) * [Test] Define `broken`/`skip` keyword argument for `@test` They make it less tedious having tests failing consistently only under specific conditions. * [Test] Use new `@test ... broken=cond` in some tests | 07 April 2021, 18:48:19 UTC |
| d02aba4 | Elliot Saba | 07 April 2021, 16:11:12 UTC | Bump libblastrampoline to v3.0.4 (#40381) This should fix some armv7l issues, as libblastrampoline had the same issues in its armv7l trampolines as Julia itself had. It also fixes some errors in armv7l interface autodetection. | 07 April 2021, 16:11:12 UTC |
| 5a0d827 | Valentin Churavy | 07 April 2021, 15:54:08 UTC | Merge pull request #40378 from JuliaLang/cv/deserialize-msg-error Provide an error message for `Distributed.deserialize_msg` | 07 April 2021, 15:54:08 UTC |
| 8218480 | Pramodh Gopalan V | 07 April 2021, 05:29:20 UTC | Add docs for _ as a variable (#40183) | 07 April 2021, 05:29:20 UTC |
| 7352320 | Marco | 06 April 2021, 21:48:26 UTC | Remove a few redundant diagonal `ldiv!` methods (#38484) | 06 April 2021, 21:48:26 UTC |
| df48fb6 | hsm207 | 06 April 2021, 21:47:51 UTC | test: simple tests for hash_64_32 and hash_32_32 (#35520) Tests to help improve coverage | 06 April 2021, 21:47:51 UTC |
| 70333da | Steven G. Johnson | 06 April 2021, 21:41:07 UTC | don't use realpath in Sys.which (#40233) As [discussed recently on discourse](https://discourse.julialang.org/t/weird-sys-which-function/58070), some programs require a certain executable name to function correctly, and behave badly if you call `realpath` to expand symbolic links (potentially changing the program name). c.f. discussion in #26559, implemented `Sys.which` | 06 April 2021, 21:41:07 UTC |
| d9f0f04 | Tommy-Xavier Robillard | 06 April 2021, 21:39:38 UTC | Add `lock` wrapper for Channels (#39312) Added lock(f, c::Channel) utility function | 06 April 2021, 21:39:38 UTC |
| 2e1de7d | Logan Kilpatrick | 06 April 2021, 21:39:03 UTC | doc: update reducedim.jl to fix a few doc strings (#40351) Closes #40300, and fixes other docs discrepancies | 06 April 2021, 21:39:03 UTC |
| 336cdf8 | Elliot Saba | 06 April 2021, 21:10:05 UTC | [build] Add `symlink_system_library` calls for CSL (#40217) * [build] Add `symlink_system_library` calls for CSL We differentiate between a `USE_BINARYBUILDER_CSL=0` configuration (where we copy the CSL libraries into the Julia build tree) versus a `USE_SYSTEM_CSL=1` configuration (where we symlink them in a manner similar to other system libararies) * Expand `JL_PRIVATE_LIBS-1` for versioned library names as well Without this, we fail to install things like `libgcc_s.so.1` when installed into the private libdir by `symlink_system_libraries`. | 06 April 2021, 21:10:05 UTC |
| 54b610b | Carsten Bauer | 06 April 2021, 20:35:00 UTC | Show methods for LBTConfig and LBTLibraryInfo (#40357) * show methods for LBTConfig and LBTLibraryInfo * add interface information to LBTConfig printing * interface information in show fallback of LBTConfig and LBTLibraryInfo * LBTConfig(...) instead of LBTConfig() for n>3 libs | 06 April 2021, 20:35:00 UTC |
| 79e198b | Jeff Bezanson | 06 April 2021, 19:35:18 UTC | fix #40249, reshaping SharedArray on another process (#40286) | 06 April 2021, 19:35:18 UTC |
| e34a904 | WorldsOkayestAirman | 06 April 2021, 18:30:31 UTC | Allow enabling multiple external profilers in build (#38741) | 06 April 2021, 18:30:31 UTC |
| e3c0dec | Jameson Nash | 06 April 2021, 17:59:29 UTC | errorshow: fix bad test from #40304 (#40376) This conflicted with a test in core.jl for callable numbers | 06 April 2021, 17:59:29 UTC |
| 9f8cc3d | Curtis Vogt | 06 April 2021, 16:46:54 UTC | Remove unnecessary multiple returns | 06 April 2021, 16:46:54 UTC |
| 4489288 | Curtis Vogt | 06 April 2021, 16:46:23 UTC | Provide an error message for deserialize_msg | 06 April 2021, 16:46:23 UTC |
| 2fc32f2 | Lyndon White | 06 April 2021, 15:19:22 UTC | Update Emoji Completions (#39111) * Update Emoji Completions * =new should overwrite old | 06 April 2021, 15:19:22 UTC |
| bda6b94 | Jameson Nash | 06 April 2021, 15:07:15 UTC | ensure require_one_based_indexing can constant-fold for ranges (#40358) Refs performance issue noted in #38527 | 06 April 2021, 15:07:15 UTC |
| fedefe9 | Martin Holters | 06 April 2021, 03:03:45 UTC | Normalize (simplify) `UnionAll`s when used as type parameter (#36211) | 06 April 2021, 03:03:45 UTC |
| 40d4f4a | Jameson Nash | 06 April 2021, 00:30:25 UTC | searchsorted ranges: simplify code, fix bug with Unsigned needle This avoids computing `length` unnecessarily, which often requires a division, and is therefore often expensive. | 06 April 2021, 01:14:02 UTC |
| 454fc51 | Jameson Nash | 06 April 2021, 01:13:58 UTC | argmin ranges bug: you cannot pass an unsorted array to searchsorted | 06 April 2021, 01:14:02 UTC |
| b780905 | Seth Axen | 05 April 2021, 17:37:56 UTC | Fix real matrix sqrt and log for edge cases (#40144) * Add failing tests * Improve 2x2 real sqrt for scalar diagonal * Avoid sylvester if zero is a solution * Avoid unnecessary sqrt * Use correct variable names * Avoid erroring due to NaNs * Exactly adapt Higham's algorithm, with hypot * Remove unreachable branch * Add failing tests * Handle overflow/underflow * Invert instead of dividing * Apply d=0 constraint from standardized real schur form * Handle underflow * Avoid underflow/overflow in log diagonal blocks * Add tests for log underflow/overflow | 05 April 2021, 17:37:56 UTC |
| 58fba2b | Viral B. Shah | 04 April 2021, 23:29:44 UTC | Use 0.3.13+4 openblas binaries to fix #40279 (#40343) * Use 0.3.13+4 openblas binaries to fix #40279 * Add tests and doctest fixes for #40343 (#40346) * Update checksums * changed a doctest to have nicer output Co-authored-by: Pablo San-Jose <lekand@gmail.com> | 04 April 2021, 23:29:44 UTC |
| 89fff18 | Mustafa M | 04 April 2021, 09:54:36 UTC | Fix relpath when path and startpath are in the same drive (#40323) * Fix relpath when path and startpath are in the same drive When startpath == ".", assume the startpath is in the same drive. This subsequently required tweaking the existing logic to then canonicalize the drive casing in instances where the drive casing differs. | 04 April 2021, 09:54:36 UTC |
| 0596d51 | Jameson Nash | 04 April 2021, 05:39:57 UTC | remove UndefVarError deprecation from v0.5 (#40333) Looks like this got lost in deprecation removal? | 04 April 2021, 05:39:57 UTC |
| 3c0ad21 | Jameson Nash | 03 April 2021, 22:55:19 UTC | test: fix bad copy in #40304 (#40334) | 03 April 2021, 22:55:19 UTC |
| d0438c0 | Jameson Nash | 03 April 2021, 22:53:20 UTC | Revert "Fix some inferability issues in SparseArrays" (#40332) | 03 April 2021, 22:53:20 UTC |
| b0b4a48 | Simeon Schaub | 03 April 2021, 18:43:43 UTC | fix inconsistent BroadcastFunction behavior (#40330) fixes #40309 | 03 April 2021, 18:43:43 UTC |
| 595b0e6 | Steven G. Johnson | 03 April 2021, 18:40:13 UTC | correct reverseind example for Unicode (#40156) | 03 April 2021, 18:40:13 UTC |
| c90e8e1 | Bogumił Kamiński | 03 April 2021, 18:39:47 UTC | Clean up of Boolean range indexing (#39993) This PR references the following comments: https://github.com/JuliaLang/julia/pull/31829/files#r592887561 https://github.com/JuliaLang/julia/pull/31829/files#r592887003 https://github.com/JuliaLang/julia/pull/39962/files#r592888379 | 03 April 2021, 18:39:47 UTC |
| 12b8948 | fp4code | 03 April 2021, 18:24:06 UTC | Fixed segfault when adding OffsetArray and UniformScaling (#38544) | 03 April 2021, 18:24:06 UTC |
| 8b6a778 | Benjamin Galliot | 03 April 2021, 18:18:03 UTC | Add test for Unicode named subpatterns and property mixes of scripts and classes. (#40328) | 03 April 2021, 18:18:03 UTC |
| d8b7302 | Debraj Bhal | 03 April 2021, 17:22:48 UTC | doc: Add example for NaN comparison (#38953) | 03 April 2021, 17:22:48 UTC |
| 78d3ff4 | Harmen Stoppels | 03 April 2021, 17:15:20 UTC | Use dot and views in QR code (#38973) The QR-code is quite ancient, and was written without using other linalg routines because views were expensive then. This lets LinearAlgebra to use dot and norm and such, which are likely optimized for special number types. | 03 April 2021, 17:15:20 UTC |
| ad513ca | Nicholas Bauer | 03 April 2021, 17:12:10 UTC | docs: Reference functions instead of methods (#38986) By referencing methods sometimes and not just functions, there was some duplication and confusion. For instance, we had `get(::Any, ::Any, ::Any)` method and also a `get` function, the latter of which duplicated the documentation for the former on the same page. | 03 April 2021, 17:12:10 UTC |
| ff316b2 | Jameson Nash | 03 April 2021, 17:06:23 UTC | Merge pull request #39207 from JuliaLang/mh/sparse-type-stability Fix some inferability issues in SparseArrays | 03 April 2021, 17:06:23 UTC |
| 03ec87f | Navid C. Constantinou | 03 April 2021, 16:55:01 UTC | Adds method to summary() for functions (#39173) | 03 April 2021, 16:55:01 UTC |
| 28e63dc | Jose Antonio Pereiro Morejon | 03 April 2021, 16:08:50 UTC | Add register_error_hint for UndefVarError (#39850) * add show_error_hints for UndefVarError Hi, I just include the show_error_hints part from #39089 Further relevant discussion in [discourse](https://discourse.julialang.org/t/help-with-register-error-hint/56140/4) * Update errorshow.jl * Update Base.jl Co-authored-by: Simeon Schaub <simeondavidschaub99@gmail.com> | 03 April 2021, 16:08:50 UTC |
| 9c50643 | Arsh Sharma | 03 April 2021, 16:08:35 UTC | Add hint in case of missing operator (#40304) * Add hint for missing operators * Move hint to it's own error type * Clear the hint handler after displaying hint * Don't clear the already empty array in tests * Don't pop the hint handler * Rename handler | 03 April 2021, 16:08:35 UTC |
| 2ee8c87 | Atsushi Sakai | 03 April 2021, 15:39:59 UTC | doc: add cross references in subset-related operators (#39750) | 03 April 2021, 15:39:59 UTC |
| aaf2b9f | Moritz Schauer | 03 April 2021, 12:43:09 UTC | Verbalize definition of isapprox in docstring (#39765) * Verbalize definition of isapprox * More examples for isapprox | 03 April 2021, 12:43:09 UTC |
| 8d1f052 | Shuhei Kadowaki | 03 April 2021, 12:36:55 UTC | make fuzzy scoring more robust against whitespaces (#39932) | 03 April 2021, 12:36:55 UTC |
| 78b629f | Jishnu Bhattacharya | 03 April 2021, 12:35:26 UTC | Add a method to similar with Integers (#39933) | 03 April 2021, 12:35:26 UTC |
| 872954e | Steven G. Johnson | 03 April 2021, 12:19:26 UTC | document rtol and atol for ≈ in at-test macro (#40061) * document rtol and atol for ≈ in at-test macro * Update stdlib/Test/docs/src/index.md Co-authored-by: Stefan Karpinski <stefan.karpinski@post.harvard.edu> * Update stdlib/Test/docs/src/index.md Co-authored-by: Stefan Karpinski <stefan.karpinski@post.harvard.edu> Co-authored-by: Jameson Nash <vtjnash@gmail.com> | 03 April 2021, 12:19:26 UTC |
| bf1ab0e | Kenta Sato | 03 April 2021, 12:18:10 UTC | make radical operators juxtaposable (#40173) * make radical operators juxtaposable * fix test * update NEWS.md [skip ci] Co-authored-by: Jameson Nash <vtjnash@gmail.com> | 03 April 2021, 12:18:10 UTC |
| 8fc052f | Adam A. Oppenheimer | 03 April 2021, 12:16:45 UTC | Update noteworthy-differences.md (#40326) They to there | 03 April 2021, 12:16:45 UTC |
| 4076e87 | Kristoffer Carlsson | 03 April 2021, 06:20:36 UTC | remove mention that non-underscored symbols denote public API (#40145) | 03 April 2021, 06:20:36 UTC |
| c4d66c8 | Harry Slatyer | 03 April 2021, 04:21:34 UTC | Fix `eltype` example in Methods section of manual (#40017) The code as written doesn't work when passing `eltype(Array{T} where T <: Integer)`, because `Array{T} where T <: Integer` is not a subtype of `AbstractArray{Any}` (but it *is* a subtype of `AbstractArray`). This was confusing when trying to understand why `Base.eltype(Array{T} where T <: Integer)` returned `Any`, so this re-worded the correct example, to make a little more clear where the fallback is defined. Removed the two other "correct" examples, and replaced them with another warning about a possible mistake (namely for types without supertypes). | 03 April 2021, 04:21:34 UTC |
| fbd718f | srz_zumix | 03 April 2021, 03:40:32 UTC | cgmemmgr: check rlimit before using map_size_inc to map more memory (#40136) | 03 April 2021, 03:40:32 UTC |
| 9033877 | Henri Dehaybe | 03 April 2021, 03:33:36 UTC | adding default argument value to `reset!(stateful)` (#40158) I think that most of the time when one calls `reset!(::Stateful)` it will be to simply restart from the begin without changing the iterated collection. So I figured that would avoid `reset!(sf, sf.itr)`. I also added the use of `reset!` to the documentation examples because it's not documented anywhere. This way users know of its existence. | 03 April 2021, 03:33:36 UTC |
| e1f6a64 | Jishnu Bhattacharya | 03 April 2021, 03:26:55 UTC | Fix mapreduce docstring (#40214) | 03 April 2021, 03:26:55 UTC |
| 03abc45 | Sheehan Olver | 03 April 2021, 03:25:14 UTC | Fix reduce(vcat,...) type inference (#40277) (#40294) | 03 April 2021, 03:25:14 UTC |
| c3d84ce | Kenta Sato | 03 April 2021, 02:50:51 UTC | docs: fix typos in AbstractDict (#40325) | 03 April 2021, 02:50:51 UTC |
| 9c7bb03 | Andrey Oskin | 03 April 2021, 02:48:56 UTC | in printstyled, support underline and other flags (#40288) | 03 April 2021, 02:48:56 UTC |
