| 3c4f517 | bors[bot] | 03 December 2018, 09:24:52 UTC | Merge #30224 30224: Remove incorrect return in load_path_expand r=fredrikekre a=fredrikekre This fixes a problem where Julia does not find an active project when the `~/.julia/environments/v#.#` folder exist but does not contain a `Project.toml` file. Co-authored-by: Fredrik Ekre <ekrefredrik@gmail.com> | 03 December 2018, 09:24:52 UTC |
| 7614cec | Keno Fischer | 30 November 2018, 14:19:00 UTC | Fix code paths in anticodegen.c | 02 December 2018, 23:15:17 UTC |
| d851fa5 | Keno Fischer | 30 November 2018, 14:15:53 UTC | Fix "duplicate definition of jl_typestack_t" error Certain versions of clang seem to dislike doing the typedef twice, so use a forward declaration and only one typedef. | 02 December 2018, 23:15:17 UTC |
| 413fc47 | Matt Bauman | 01 December 2018, 19:14:48 UTC | Detangle Slice and fix mixed-dimension in-place reductions (#28941) * Detangle Slice between whole dimensions and axes We use axes in many downstream computations that may not re-index directly into the original array, so we add a second type parameter to `Slice` that is turned on when converting `:` in indexing expressions -- and really only `SubArray` cares about. * Reduce allocations for in-place reductions and fix mixed-dimensionality edge-cases Alleviates #28928 but does not completely remove allocations due to the allocation of the view that gets passed to `map!`. * Introduce a whole new IdentityUnitRange that we will encourage offset array implementations to use instead of Base.Slice * Test that maximum! allocates less | 01 December 2018, 19:14:48 UTC |
| ecd7291 | Keno Fischer | 12 September 2018, 20:21:41 UTC | Initialize the union selector of undefined upsilon nodes While we're guaranteed never to look at undefined values, we're not guaranteed that the type information can't change, which in the case of a union generates code to change the layout. As a result, we need to make sure to have the union selectors be valid (the actual data will be junk and no semantic operation will ever look at it, but it shouldn't be ouf of bounds). Ref #29152 | 01 December 2018, 14:27:26 UTC |
| 5b827f5 | Andy Ferris | 01 December 2018, 04:59:11 UTC | Merge pull request #30076 from bramtayl/patch-1 Slight refactor to collection widening | 01 December 2018, 04:59:11 UTC |
| d54f40d | Elliot Saba | 01 December 2018, 01:36:00 UTC | Update libgfortran name matching to include SOVERSION (#30135) This fixes the annoying re-copying of `libgfortran` every time on machines that have an SOVERSIONed filename. | 01 December 2018, 01:36:00 UTC |
| f57466c | Fredrik Ekre | 01 December 2018, 00:31:33 UTC | Remove incorrect return in load_path_expand This fixes a problem where Julia does not find and active project when the ~/.julia/environments/v#.# folder exist but does not contain a Project.toml file. | 01 December 2018, 00:59:09 UTC |
| 3dd678f | woclass | 30 November 2018, 22:20:24 UTC | fix typo (#30209) - ccall :GetClipboardData failed should throw a `GetClipboardData` error | 30 November 2018, 22:20:24 UTC |
| 055da33 | Tim Holy | 30 November 2018, 22:12:09 UTC | Merge pull request #30215 from blegat/ReverseEmptyStepRangeLen Fix reverse with empty StepRangeLen | 30 November 2018, 22:12:09 UTC |
| 245bc6d | Jeff Bezanson | 30 November 2018, 22:08:00 UTC | fix #21233, start REPL when script run with `-i` throws an error (#30199) | 30 November 2018, 22:08:00 UTC |
| 25f7537 | Katharine Hyatt | 30 November 2018, 21:16:15 UTC | Merge pull request #30213 from JuliaLang/ksh/alphaomega Test for first/lastness of reversed iterators | 30 November 2018, 21:16:15 UTC |
| 9d8fe1a | Katharine Hyatt | 30 November 2018, 21:16:02 UTC | Merge pull request #30214 from JuliaLang/ksh/zipcount Tests for size(Zip) and countfrom() | 30 November 2018, 21:16:02 UTC |
| b409238 | Benoît Legat | 30 November 2018, 17:07:46 UTC | Fix reverse with empty StepRangeLen | 30 November 2018, 17:07:46 UTC |
| ac0e75c | Katharine Hyatt | 30 November 2018, 16:47:03 UTC | Tests for size(Zip) and countfrom() | 30 November 2018, 16:47:03 UTC |
| dd7d2f5 | Katharine Hyatt | 30 November 2018, 16:33:46 UTC | Test for first/lastness of reversed iterators | 30 November 2018, 16:33:46 UTC |
| 80820ac | Valentin Churavy | 30 November 2018, 14:26:40 UTC | Merge pull request #30019 from sebasguts/sg/ccall_docs Extended ccall documentation | 30 November 2018, 14:26:40 UTC |
| b3cf152 | Jameson Nash | 30 November 2018, 14:24:06 UTC | reliable line numbers in code info printing (#29893) * code_llvm annotations: format prettier Use box-drawing characters and indentation to make the output readable more rapidly. * code_typed/code_lowered: format line-info similarly to code_llvm/code_native For consistency of user experience, reduce the variance in our IR printing across formats. This also now shows inlining and line number information even if the output device might not support color, which was previously impossible (a regression since v0.6). * code formatting: reduce inlining indentation Here we make the observation that it's somewhat common to have chains of methods of one function that recursively handle arguments in some fashion (for example, map-tuple or +). However, since they all have the same method name, it's possible to represent these on a single line. * code show: thread debuginfo verbosity parameter throughout APIs These levels are intended to roughly correspond approximately to compiler options -g0 (none) through -g2 (including local variables). Currently we do not implement the ability to represent local variable information, so the default is to only show line number information suitable for identifying the source line of a given statement (but not the types). * code formatting: enable color usage Only for Julia IR, for now. Later, this should be threaded into code_llvm and enabled there also. I chose green for new nodes (like new life), and used yellow for metadata (because the strong contrast helps to hide it visually, and similar to sticky notes), and used grey for basic blocks (because they are minor information), and used cyan/gray/Base.warn_color() for type information (because that's what we've always used). | 30 November 2018, 14:24:06 UTC |
| ef496b0 | Max Horn | 30 November 2018, 09:55:02 UTC | Two documentation updates (#30049) * `require` takes to arguments * `jl_new_uniontype` does not exist anymore | 30 November 2018, 09:55:02 UTC |
| 795591f | Tim Holy | 30 November 2018, 09:05:23 UTC | Merge pull request #30164 from JuliaLang/teh/bcast_axis_promotion Use only safe axis types for Broadcast.combine_axes | 30 November 2018, 09:05:23 UTC |
| a87860a | Elliot Saba | 30 November 2018, 08:22:36 UTC | Revert "Update libosxunwind to v0.0.4 (#30153)" (#30203) This reverts commit 5221fedef9320d5aebac8dffdef5e249c84298be. A bad interaction between libosxunwind's Makefiles and Julia's makefiles broke this for newer macs, see https://github.com/JuliaLang/julia/pull/30153 for more details. | 30 November 2018, 08:22:36 UTC |
| 6bfe947 | Jeff Bezanson | 30 November 2018, 02:56:26 UTC | avoid jl_arrayunset in dicts with bitstypes; add some more @inbounds (#30113) * avoid jl_arrayunset in dicts with bitstypes; add some more @inbounds * also skip jl_arrayunset for isbitsunion * amend typo | 30 November 2018, 02:56:26 UTC |
| 67bba81 | Jeff Bezanson | 30 November 2018, 02:51:26 UTC | add `@locals` macro for obtaining dictionary of local vars and values (#29733) implements #29366 | 30 November 2018, 02:51:26 UTC |
| 7ba6c82 | Max Horn | 30 November 2018, 02:49:16 UTC | Implement get! and get for IdDict with a callable default (#30187) Fixes #30165 | 30 November 2018, 02:49:16 UTC |
| e4f408e | Jeff Bezanson | 29 November 2018, 21:59:51 UTC | fix #29429, make local-sparam conflict an error instead of warning (#30184) | 29 November 2018, 21:59:51 UTC |
| 4b270b1 | Curtis Vogt | 29 November 2018, 21:49:41 UTC | Fix issue with Git Credential Manager for Windows (#30195) - Ignores additional whitespace when reading in credentials. - Warn when encountering unknown attributes. Should allow for more graceful failures in the future if the protocol is extended. | 29 November 2018, 21:49:41 UTC |
| 3f8c311 | Katharine Hyatt | 29 November 2018, 20:42:13 UTC | Tests for float conversions with many leading zeros (#30020) | 29 November 2018, 20:42:13 UTC |
| 2b688c3 | Katharine Hyatt | 29 November 2018, 20:39:59 UTC | Tests for prev/this/nextind with Integer args (#30193) | 29 November 2018, 20:39:59 UTC |
| 5221fed | Alex Arslan | 29 November 2018, 20:22:58 UTC | Update libosxunwind to v0.0.4 (#30153) Fixes #29912. | 29 November 2018, 20:22:58 UTC |
| 8012ba1 | Jeff Bezanson | 29 November 2018, 20:17:59 UTC | fix #29781, disallow spaces in broadcast calls (#30182) | 29 November 2018, 20:17:59 UTC |
| 15f649c | Jeff Bezanson | 29 November 2018, 20:16:52 UTC | fix #29968, qualify macro names in parsing int literals and cmds (#30181) | 29 November 2018, 20:16:52 UTC |
| 9cc759f | Jeff Bezanson | 29 November 2018, 20:14:54 UTC | add some 1.1 NEWS (#30183) [ci skip] [av skip] | 29 November 2018, 20:14:54 UTC |
| 4ee464e | Simon Byrne | 29 November 2018, 19:43:36 UTC | fix Rational -> BigFloat rounding (#30189) | 29 November 2018, 19:43:36 UTC |
| caea73d | Tamas K. Papp | 29 November 2018, 19:35:11 UTC | Make randperm and randcycle use the argument's type. (#29670) Remove tests for eltype being `Int`. This is potentially controversial as it may break things, but I did not see it documented anywhere as part of the API. | 29 November 2018, 19:35:11 UTC |
| 89a2c08 | Jeff Bezanson | 29 November 2018, 19:16:49 UTC | don't print `Type{Vararg}` in type errors, since that expr is invalid (#30180) | 29 November 2018, 19:16:49 UTC |
| 2ec5560 | Jameson Nash | 29 November 2018, 17:01:08 UTC | inference: fix a mistake with ssa_def_expr usage (#30179) Uses of this function previously assumed that all Slots were SSA. Fortunately, we now just have one user (post #29935) and it has very simple requirements so we can check conservatively for this case without loss of (much) accuracy. | 29 November 2018, 17:01:08 UTC |
| a50d5af | Jeff Bezanson | 28 November 2018, 20:23:27 UTC | fix #29885, make grisu digit buffer task-local (#29907) | 28 November 2018, 20:23:27 UTC |
| d4c6d16 | Kim Schmider | 28 November 2018, 20:17:04 UTC | Allow boolean parsing of strings containing 0 and 1 (#29997) | 28 November 2018, 20:17:04 UTC |
| f1a41e2 | Tim Besard | 28 November 2018, 19:28:12 UTC | Merge pull request #30166 from JuliaLang/tb/restore_globalunique Don't restore globalUnique. | 28 November 2018, 19:28:12 UTC |
| a2a440e | Jeff Bezanson | 28 November 2018, 18:09:35 UTC | fix #30062, need front-end error for invalid multiple update lhs (#30069) | 28 November 2018, 18:09:35 UTC |
| f24711c | Takafumi Arakaki | 28 November 2018, 17:47:21 UTC | Do not allow THIN SPACE U+2009 as an operator suffix (#30158) | 28 November 2018, 17:47:21 UTC |
| 437d87c | Keno Fischer | 23 November 2018, 14:39:30 UTC | [AllocOpt] Track deleted instructions It was possible for the instruction contained in first_safepoint to be delted, causing use-after-free problems later. Use a WeakVH rather than a raw Instruction* to null-out the entry when the Instruction gets deleted. In the future we may want to pay closer attention to live ranges to avoid this problem entirely. Fixes #30116 | 28 November 2018, 15:30:45 UTC |
| 065d0f5 | Tim Holy | 28 November 2018, 13:53:45 UTC | Merge pull request #30149 from JuliaLang/teh/fix_28611 Use ReshapedArray for isbitsunion arrays (fixes #28611) | 28 November 2018, 13:53:45 UTC |
| 402c747 | Milan Bouchet-Valat | 28 November 2018, 11:41:41 UTC | Merge pull request #30102 from JuliaLang/nl/findminmax Fix findmin/findmax to return cartesian indices for BitMatrix | 28 November 2018, 11:41:41 UTC |
| 076ea45 | Andreas Noack | 28 November 2018, 10:04:02 UTC | Fix Matrix(QRSparseQ) constructor (#26392) * Fix Matrix(QRSparseQ) constructor Fixes #26367 * Make sure that size(Q, i) definition covers HessenbergQ | 28 November 2018, 10:04:02 UTC |
| 9c8560f | Eric Davies | 28 November 2018, 08:29:41 UTC | Add news for #30159 (#30169) | 28 November 2018, 08:29:41 UTC |
| a9161d9 | Milan Bouchet-Valat | 20 November 2018, 21:37:46 UTC | Fix findmin/findmax to return cartesian indices for BitMatrix For consistency with other AbstractArray types. The call to keys is optimized out by the compiler for vectors thanks to @inbounds. | 28 November 2018, 08:26:55 UTC |
| 40c2d89 | Milan Bouchet-Valat | 20 November 2018, 21:34:34 UTC | Move findmin and findmax BitArray methods from LinearAlgebra to Base There is no reason to have them in LinearAlgebra. | 28 November 2018, 08:26:55 UTC |
| 27ee7a3 | Tim Besard | 28 November 2018, 07:34:11 UTC | Just get rid of globalUnique restore. | 28 November 2018, 07:34:25 UTC |
| aec7f2a | Tim Besard | 27 November 2018, 14:46:02 UTC | Fix restoring globalUnique. | 28 November 2018, 07:34:25 UTC |
| 4f8a7b9 | Jeff Bezanson | 27 November 2018, 23:21:45 UTC | fix #29269, type intersection bug in union parameters with typevars (#29406) also fixes #25752 | 27 November 2018, 23:21:45 UTC |
| 6778bef | Eric Davies | 27 November 2018, 21:33:35 UTC | Set TimeZones as scalars for broadcasting (#30159) | 27 November 2018, 21:33:35 UTC |
| 7acacfc | Jameson Nash | 27 November 2018, 16:29:40 UTC | inference: add missing limit-cycle poisoning (#30098) Seems to have been missing since the original implementation of IPO constant-prop. It's impressive how long we got away with just poisoning the cache with `Any` if there was a self-recursive call, but it requires a fairly convoluted setup to trigger this case. Fix #29923 | 27 November 2018, 16:29:40 UTC |
| 1884cb4 | Tim Holy | 27 November 2018, 12:01:12 UTC | Use only safe axis types for Broadcast.combine_axes #30074 used the wrong notion of consistency since `OneTo(1)` is consistent (wrt broadcasting) with any range, but `OneTo` cannot handle `Slice(-1:1)`. | 27 November 2018, 12:01:12 UTC |
| 37d2639 | Tim Holy | 25 November 2018, 22:24:06 UTC | Fix typos in devdocs on isbitsunion arrays | 27 November 2018, 11:59:56 UTC |
| abaa2ea | Tim Holy | 25 November 2018, 09:51:25 UTC | Use ReshapedArray when reshaping isbitsunion Arrays (fixes #28611) | 27 November 2018, 11:59:52 UTC |
| 58c0ed5 | Simon Byrne | 26 November 2018, 20:39:09 UTC | Add pid to malloc log (#29969) | 26 November 2018, 20:39:09 UTC |
| a836064 | Ben Arthur | 26 November 2018, 17:52:24 UTC | document tab completion for dictionary keys (#30147) | 26 November 2018, 17:52:24 UTC |
| ce82406 | Bogumił Kamiński | 26 November 2018, 17:51:22 UTC | Add news for new Char broadcasting (#30024) | 26 November 2018, 17:51:22 UTC |
| 79b5b2c | Stefan Karpinski | 24 November 2018, 16:01:45 UTC | check_new_version: more consistent error messages | 26 November 2018, 16:17:34 UTC |
| abd0ceb | Stefan Karpinski | 21 November 2018, 21:52:11 UTC | check_new_version: disallow 0.0.0 initial versions | 26 November 2018, 16:17:34 UTC |
| 3e52032 | Stefan Karpinski | 21 November 2018, 21:36:03 UTC | check_new_version: handle new version less than existing ones | 26 November 2018, 16:17:34 UTC |
| 37447e4 | Stefan Karpinski | 21 November 2018, 21:23:25 UTC | check_new_version: allow unsorted existing versions | 26 November 2018, 16:17:34 UTC |
| ef50e25 | Jameson Nash | 26 November 2018, 15:17:20 UTC | stdlib-git: improve integration with git-external (#30075) Previously some of the work was shifted, leading to awkward extraneous copies and preventing usage of DEPS_GIT=1 mode (the main point of this script). There also seemed to be unexpected failure modes in the current usage where the files in stdlib/Pkg would routinely go missing for me. I don't know the source of those, but I hope that this will fix it (because it makes more of the state immutable). Also noticed some issues with existing scripts (expanding variables at unintended points, not rebuilding on some changes), and fixed those. | 26 November 2018, 15:17:20 UTC |
| 024bcb7 | Simon Byrne | 25 November 2018, 19:31:14 UTC | Fix indexing with Union{} array (#30146) | 25 November 2018, 19:31:14 UTC |
| f4c7779 | Fredrik Ekre | 23 November 2018, 21:42:38 UTC | Fix show test when Sockets are loaded in main. (#30119) | 23 November 2018, 21:42:38 UTC |
| cd447cf | bramtayl | 23 November 2018, 19:35:56 UTC | Slight code simplifications | 23 November 2018, 19:35:56 UTC |
| bc34b99 | YonghyunRyu | 23 November 2018, 18:18:07 UTC | varinfo() is moved to `InteractiveUtils` (#30130) When we want to call `varinfo()` on the remote processes, we have to call InteractiveUtils.varinfo() instead. #25751 | 23 November 2018, 18:18:07 UTC |
| f10530e | Takafumi Arakaki | 23 November 2018, 10:57:46 UTC | Fix method ambiguities in SparseArrays (#30120) * Remove unused struct CapturedScalars * Fix method ambiguities in SparseArrays * Fix HigherOrderFns._copy(f) | 23 November 2018, 10:57:46 UTC |
| 5ea2258 | bramtayl | 22 November 2018, 22:39:40 UTC | Renaming, abstractarrays | 22 November 2018, 22:39:40 UTC |
| 038d866 | Keno Fischer | 21 November 2018, 17:54:45 UTC | Fix deadlock caused by REPL blinking The REPL beeping (blinking) code made the assumption that the prompt prefix is always a String. That's not necessarily the case, and in fact it can be an arbitrary funcion to be called when the prompt is printed (e.g. TerminalExtensions uses this for the iTerm integration). However, even worse, this code acquired a lock that it never released on error, causing a deadlock. Fix the original assumption and also add a try/catch/finally block such that future errors print the error rather than deadlocking the terminal. Fixes https://github.com/Keno/TerminalExtensions.jl/issues/40 | 22 November 2018, 11:11:51 UTC |
| f1071a7 | Keno Fischer | 20 November 2018, 11:39:18 UTC | Add test for llvmcall Function* interface | 22 November 2018, 11:11:28 UTC |
| bc2b548 | Keno Fischer | 19 November 2018, 15:08:04 UTC | Put boxed arguments in llvmcall into the correct addrspace I suspect Cxx is the only user at the moment, but absent this you either get an incorrect assert in debug builds (if you're setting addrspaces correctly in the called function), or missing GC roots (if you don't, to avoid the assert). Fix that. | 22 November 2018, 11:11:28 UTC |
| d7198bf | chethega | 22 November 2018, 00:25:41 UTC | amend typo | 22 November 2018, 00:25:41 UTC |
| 72fdb29 | chethega | 21 November 2018, 21:42:36 UTC | also skip jl_arrayunset for isbitsunion | 21 November 2018, 21:42:36 UTC |
| 36d0116 | Elliot Saba | 21 November 2018, 20:37:03 UTC | Install `ccache` on OSX Travis (#30111) | 21 November 2018, 20:37:03 UTC |
| 6d94a87 | chethega | 21 November 2018, 19:27:20 UTC | avoid jl_arrayunset in dicts with bitstypes; add some more @inbounds | 21 November 2018, 19:27:20 UTC |
| a43a536 | bramtayl | 21 November 2018, 16:49:10 UTC | Added back || typeof(el) === T | 21 November 2018, 16:49:10 UTC |
| ee27e91 | cormullion | 21 November 2018, 15:57:23 UTC | error msg typo: "delcaration" -> "declaration" (#30106) | 21 November 2018, 15:57:23 UTC |
| e5e6b64 | Michele Zaffalon | 21 November 2018, 15:43:59 UTC | Clarify the role of pivoting in QR factorization (#29976) * Clarify the role of pivoting in QR factorization Ref: https://discourse.julialang.org/t/left-matrix-division-with-singular-matrix/15503 * Update qr.jl | 21 November 2018, 15:43:59 UTC |
| e3bcfde | Klaus Crusius | 21 November 2018, 15:26:18 UTC | added message to NEWS.md (#30090) | 21 November 2018, 15:26:18 UTC |
| 3813837 | Simon Byrne | 21 November 2018, 15:01:23 UTC | CHOLMOD cleanup (#29462) * CHOLMOD cleanup Fixes #29367, and cleans up by removing a lot of unneeded code. * Use regular (non-long) suitesparse calls on 32-bit This keeps it so that `itype` will be `INT` on 32-bit and `LONG` on 64-bit. * make itype consistent across all types * allow pattern factors, use nothing for default permutation * fix naming error, remove more code, don't return new object from change_factor! to avoid potential memory leaks * fix check logic * simplify promotion * fix function call * update tests * tests now throw different errors * Hopefully fix 32-bit issues by switching back to SuiteSparse_long * Don't expand index type * add test for ldlt | 21 November 2018, 15:01:23 UTC |
| 464397e | Chris Foster | 20 November 2018, 07:23:58 UTC | Sort method signatures before precompile to avoid order dependence. This should help in making ordering-dependent bugs such as #29923 (an error in caching of inferred methods) deterministic, which will make them much easier to find in the future. | 21 November 2018, 05:14:33 UTC |
| a646c26 | Mus | 20 November 2018, 21:48:24 UTC | Don't bundle busybox with Windows (#30022) | 20 November 2018, 21:48:24 UTC |
| 7125a7f | Mus | 20 November 2018, 21:47:35 UTC | Include latest 7zip version v18.05 (#30035) | 20 November 2018, 21:47:35 UTC |
| 6e1464b | Matt Bauman | 20 November 2018, 16:51:10 UTC | Merge pull request #30074 from JuliaLang/teh/bcast_axes_infer Make `broadcast_axes` inferrable even if axes are of different types | 20 November 2018, 16:51:10 UTC |
| 3371caf | bramtayl | 20 November 2018, 13:09:56 UTC | added returns, removed redundant code | 20 November 2018, 13:09:56 UTC |
| 9f43871 | Keno Fischer | 20 November 2018, 05:28:49 UTC | Make T.instance inferable (#30086) | 20 November 2018, 05:28:49 UTC |
| 69fabf7 | Jeff Bezanson | 20 November 2018, 00:51:01 UTC | fix #30030, assignment expr as named tuple element value (#30045) | 20 November 2018, 00:51:01 UTC |
| 1c5cde2 | Simon Byrne | 19 November 2018, 22:09:43 UTC | Fix RoundNearestTiesAway (#29700) Fixes #29698. | 19 November 2018, 22:09:43 UTC |
| 1952bec | Matt Bauman | 19 November 2018, 22:09:29 UTC | Merge pull request #29167 from anders-dc/patch-1 Update arrays.md for element-wise assignment | 19 November 2018, 22:09:29 UTC |
| 77ed51f | Matt Bauman | 19 November 2018, 20:41:29 UTC | Merge pull request #30060 from JuliaLang/ksh/aa Three missing abstractarray tests | 19 November 2018, 20:41:29 UTC |
| 02d8da2 | Valentin Churavy | 19 November 2018, 16:52:50 UTC | Merge pull request #30079 from mikhail-j/libgit2-cacert Update Mozilla CA certificate store to latest (10-17-2018) for libgit2 SSL. | 19 November 2018, 16:52:50 UTC |
| ec3ef78 | bramtayl | 19 November 2018, 15:33:09 UTC | Update array.jl | 19 November 2018, 15:33:09 UTC |
| a1aa542 | lewisl | 19 November 2018, 15:22:30 UTC | Update the Regex section of strings.md. (#29921) * Update the Regex section of strings.md. This adds a brief description and example of using the Regex constructor to create regex strings programmatically. * Update doc/src/manual/strings.md Co-Authored-By: lewisl <lewis@neilson-levin.org> * Update doc/src/manual/strings.md Co-Authored-By: lewisl <lewis@neilson-levin.org> * Update doc/src/manual/strings.md Co-Authored-By: lewisl <lewis@neilson-levin.org> | 19 November 2018, 15:22:30 UTC |
| be98372 | Benoît Legat | 19 November 2018, 14:59:15 UTC | Add sh color syntaxing in README (#30081) | 19 November 2018, 14:59:15 UTC |
| c4f2a85 | Katharine Hyatt | 16 November 2018, 19:22:31 UTC | Three missing abstractarray tests | 19 November 2018, 14:42:33 UTC |
| 3900658 | Mathieu Besançon | 19 November 2018, 14:31:48 UTC | Remove "slave" variable (#30058) * replace master, slave -> pty_master pty_slave | 19 November 2018, 14:31:48 UTC |
| 9a7c79c | Simon Byrne | 19 November 2018, 10:19:19 UTC | Make precision a keyword argument, improve handling of rounding mode. (#29157) * Improve handling of precision and rounding mode arguments in BigFloat constructors. This makes a few changes to avoid unnecessarily calling `setrounding` and `setprecision`. It also changes `precision` to be a keyword argument (though does not yet deprecate the old behaviour). | 19 November 2018, 10:19:19 UTC |
| f55d517 | Benjamin Born | 19 November 2018, 10:17:04 UTC | Update README.md (#30080) Current stable release is 1.0.2 | 19 November 2018, 10:17:04 UTC |
