aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
...
* | | Merge pull request #14930 from clason/ts-ifdef-fixupBjörn Linse2021-06-29
|\ \ \ | | | | | | | | fixup(ts): put ts_query_cursor_set_match_limit behind feature guard
| * | | fixup(ts): put ts_query_cursor_set_match_limit behind feature guardChristian Clason2021-06-29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a fixup for #14915, which used the above-mentioned call to restore the behavior of the pre-release version of tree-sitter to that of 0.19.5. However, this function was introduced after 0.19.5, breaking distro builds that link against 0.19.5 instead of the tag specified in neovim's build script. Now the function should only be called when it is available _and_ needed. Once tree-sitter is bumped to 0.19.6 (when this is released), this guard can be removed again. Fixes #14923 (among others)
* | | | Merge pull request #14779 from tjdevries/tjdevries/vim_opt_fixesTJ DeVries2021-06-29
|\ \ \ \ | | | | | | | | | | An assortment of various vim.opt fixups.
| * | | | fix(vim.opt): #14708 Now lets you put duplicate values in wildmodeTJ DeVries2021-06-29
| |/ / /
* | | | Merge pull request #14335 from chentau/extmark_delbytesBjörn Linse2021-06-29
|\ \ \ \ | |/ / / |/| | | Extmarks: manually zero out `curbuf->deleted_bytes2` on substitute and join
| * | | Manually zero out deleted_bytes2 when substituting and joining lineschentau2021-06-29
| | |/ | |/|
* | | Merge pull request #14915 from clason/ts-match-limitBjörn Linse2021-06-28
|\ \ \ | | | | | | | | fix(treesitter): set match limit for query cursors
| * | | fix(treesitter): set match limit for query cursorsChristian Clason2021-06-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Upstream tree-sitter raised the number of pending matches for a query cursor from 32 to 64k in <https://github.com/tree-sitter/tree-sitter/commit/ 78010722a49ed6224c773c22b0d25a8c9fbde584>, which severely impacted performance for some highlighting queries. This uses the `ts_query_cursor_set_match_limit` function introduced in <https://github.com/tree-sitter/tree-sitter/commit/ cd96552448a6e0d4eb27fc54b27cb5130c4b6f76> to manually set this back to the old default of 32. Fixes #14897
* | | | [RDY] Add buffer information to tabline_update (#12481)John Gehrig2021-06-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Add buffer information to tabline_update Most terminal implementations of the tabline display buffer and tab information. Many neovim-qt users disable GuiTabline because it lacks functionality provided in the terminal implementation. The tabline_update event should include buffer information too, so client GUIs can display rich useful tabs.
* | | | vim-patch:8.2.2954: short file name extension for Scala not recognizedJan Edmund Lazo2021-06-26
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Short file name extension for Scala not recognized. Solution: Recognize *.sc. (closes vim/vim#8337) https://github.com/vim/vim/commit/6db7b6375a3ea3afef5295b1366896902012e640
* | | | vim-patch:8.2.3049: JSON patch file not recognizedJan Edmund Lazo2021-06-26
| | | | | | | | | | | | | | | | | | | | | | | | Problem: JSON patch file not recognized. Solution: Recognize json-patch as json. (Kevin Locke, closes vim/vim#8450) https://github.com/vim/vim/commit/6582e230a0f6592287b1123c5fc3807d6fed997e
* | | | vim-patch:8.2.3050: cannot recognize elixir filesJan Edmund Lazo2021-06-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Cannot recognize elixir files. Solution: Recognize Elixir-specific files. Check if an .ex file is Euphoria or Elixir. (Austin Gatlin, closes vim/vim#8401, closes vim/vim#8446) https://github.com/vim/vim/commit/f3caeb63d62c08b579e9b5f40b35e8bf64dde87a
* | | | vim-patch:8.2.3018: 'quickfixtextfunc' formatting is lost when switching ↵Yorick Peterse2021-06-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | buffers (#14865) Problem: Formatting using quickfixtextfunc is lost when updating location lists for different buffers. (Yorick Peterse) Solution: Use the right window for the locaiton list. (Yegappan Lakshmanan, closes vim/vim#8400, closes vim/vim#8403) https://github.com/vim/vim/commit/ad52f96a2d3169cb1b915c1d4a6ba26ba6e5bd0a
* | | | vim-patch:8.1.0897: can modify a:000 when using a reference (#14902)Jan Edmund Lazo2021-06-26
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Can modify a:000 when using a reference. Solution: Make check for locked variable stricter. (Ozaki Kiichi, closes vim/vim#3930) https://github.com/vim/vim/commit/05c00c038bc16e862e17f9e5c8d5a72af6cf7788
* | | | Fix `<afile>` getting prepended with a slash on WinClosed (#14515)Ghjuvan Lacambre2021-06-26
|/ / / | | | | | | | | | | | | | | | It is wrong to assume that you can't shorten a path if it's in `/`: you can always shorten it by removing the forward slash. Closes #14512
* | | vim-patch:8.2.3033: no error when using alpha delimiter with :globalJan Edmund Lazo2021-06-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: No error when using alpha delimiter with :global. Solution: Check the delimiter like with :substitute. (closes vim/vim#8415) https://github.com/vim/vim/commit/419a40ac9657e39646b2e0f3f71d7736b0c459d1 N/A patches for version.c: vim-patch:8.1.2391: cannot build when __QNXNTO__ is defined Problem: Cannot build when __QNXNTO__ is defined. (Ian Wayne Larson) Solution: Move the check for "qansi". (Ken Takata, closes vim/vim#5317) https://github.com/vim/vim/commit/c95e8d649045add4e77b423bc159dc0d845af559 vim-patch:8.2.2294: VMS: a few remaining problems Problem: VMS: a few remaining problems. Solution: Add VMS specific changes. Add Lua support. (Zoltan Arpadffy) https://github.com/vim/vim/commit/82c38fe508155c11a904e6111b5bfb6adde3fb9a vim-patch:8.2.3034: installing packages on github CI sometimes fails Problem: Installing packages on github CI sometimes fails. Solution: Update package information first. (Christian Brabandt, closes vim/vim#8432) https://github.com/vim/vim/commit/ef7be8348fd830e409504a442f3f3ad0931cfbc3 vim-patch:8.2.3037: configure reports libcanberra when checking for libsodium Problem: Configure reports libcanberra when checking for libsodium. Solution: Adjust the message. (Ozaki Kiichi, closes vim/vim#8435) https://github.com/vim/vim/commit/8ce3ca8961c5968a02de0a0b98f906695331f254 vim-patch:8.2.3038: Amiga built-in version string doesn't include build date Problem: Amiga built-in version string doesn't include build date. Solution: Add the build date if available. (Ola Söder, closes vim/vim#8437) https://github.com/vim/vim/commit/cc6504098605f894b557109b618e88913a89914b vim-patch:8.2.3043: Amiga: cannot get the shell size on MorphOS and AROS Problem: Amiga: cannot get the shell size on MorphOS and AROS. Solution: Use control sequences. (Ola Söder, closes vim/vim#8438) https://github.com/vim/vim/commit/d415d269130b233c0f198c75cc159bee721aa55c
* | | vim-patch:8.2.2185: BufUnload is not triggered for the quickfix dummy bufferJan Edmund Lazo2021-06-23
| | | | | | | | | | | | | | | | | | | | | Problem: BufUnload is not triggered for the quickfix dummy buffer. Solution: Do trigger BufUnload. (Pontus Leitzler,closes vim/vim#7518, closes vim/vim#7517) Fix white space around "=". https://github.com/vim/vim/commit/1cfb9bb5c06c07f14475f39c4eb57fea1f0dfb69
* | | vim-patch:8.2.1255: cannot use a lambda with quickfix functionsJan Edmund Lazo2021-06-23
| | | | | | | | | | | | | | | | | | Problem: Cannot use a lambda with quickfix functions. Solution: Add support for lambda. (Yegappan Lakshmanan, closes vim/vim#6499) https://github.com/vim/vim/commit/d43906d2e5969288f239df851f5ad7b1dc2c7251
* | | vim-patch:8.1.1437: code to handle callbacks is duplicatedJan Edmund Lazo2021-06-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Code to handle callbacks is duplicated. Solution: Add callback_T and functions to deal with it. https://github.com/vim/vim/commit/3a97bb3f0f8bd118ae23f1c97e55d84ff42eef20 Port Vim's put_callback() as callback_put() because Neovim's naming convention is {type}_{action}, not {action}_{type}. Renaming put_callback type as PutCallback. https://neovim.io/develop/style-guide.xml#Type_Names
* | | Merge pull request #14868 from shadmansaleh/patch_verbose_for_luaBjörn Linse2021-06-23
|\ \ \ | | | | | | | | fix(runtime): Fix bugs regarding lua runtime files
| * | | fix(source): Source giving E484 & parsing error at line 1 for lua filesshadmansaleh2021-06-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It's happening because do_source is only expected to return FAIL when it was unable to open file . But `nlua_exec_file` returns fail for parsing and execution error too . Those errors are emitted through `nlua_error`. So now return value of nlua_exec_file is ignored like do_cmdline. It now only returns fail when it was unable to open file that check is done before calling nlua_exec_file or do_cmdline. Errors in nlua_exec_file are still directly emitted through nlua_error like before.
| * | | BugFix: Fix inconsistent verbose messageshadmansaleh2021-06-21
| | |/ | |/| | | | | | | | | | | | | | | | When a keymap is set from lua currently verbose message says it's set from line 1. That's incorrect because we don't really know when it was set. So until proper :verbose support isn't added for sourceing lua it shouldn't say where it was set at.
* | | extmark: fix deletable nodes in MarkTree sometimes getting skippedsnezhniylis2021-06-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As per #14236, performing extmark cleanup in a certain namespace does not guarantee removing all the extmarks inside given namespace. The issue resides within the tree node removal method and results in a couple of rare edge cases. To demonstrate what causes this bug, I'll give an example covering one of the edge cases. === AN EXAMPLE === (A) (B) (C) (D) (E) --------- --------- --------- --------- --------- <0, 1> <0, 1> <0, 1> <0, 1> <0, 1> <0, 2> <0, 2> <0, 2> <0, 2> <0, 2> <0, 3> <0, 3> <0, 3> <0, 3> <0, 3> <0, 4> <0, 4> <0, 4> <0, 4> <0, 4> <0, 5> <0, 5> <0, 5> <0, 5> <0, 5> <0, 6> <0, 6> <0, 6> <0, 6> <0, 6> <0, 7> <0, 7> <0, 7> <0, 7> <0, 7> <0, 8> <0, 8> <0, 8> <0, 8> <0, 8> <0, 9> <0, 9> * * <0, 9> * <0, 9> [0, 10] * [0, 10] <0, 9> [0, 11] [0, 11] [0, 11] [0, 11] [0, 11] [0, 12] [0, 12] * [0, 12] [0, 12] [0, 12] [0, 13] [0, 13] [0, 13] [0, 13] [0, 13] [0, 14] [0, 14] [0, 14] [0, 14] [0, 14] [0, 15] [0, 15] [0, 15] [0, 15] [0, 15] [0, 16] [0, 16] [0, 16] [0, 16] [0, 16] [0, 17] [0, 17] [0, 17] [0, 17] [0, 17] [0, 18] [0, 18] [0, 18] [0, 18] [0, 18] [0, 19] [0, 19] [0, 19] [0, 19] [0, 19] [0, 20] [0, 20] [0, 20] [0, 20] [0, 20] DIAGRAM EXPLANATION * Every column is a state of the marktree at a certain stage. * To make it simple, I don't draw the whole tree. What you see are 2 leftmost parent nodes ([0, 10], [0, 20]) and their children placed in order `MarkTreeIter` would iterate through. From top to bottom. * Numbers on this diagram represent extmark coordinates. Relative positioning and actual mark IDs used by the marktree are avoided for simplicity. * 2 types of brackets around coordinates represent 2 different extmark namespaces (`ns_id`s). * '*' shows iterator position. ACTUAL EXPLANATION Let's assume, we have two sets of extmarks from 2 different plugins: * Plugin1: <0, 1-9> * Plugin2: [0, 10-20] 1. Plugin2 calls `vim.api.nvim_buf_clear_namespace(buf_handle, ns_id, 0, -1)` to clear all its extmarks which results in `extmark_clear` call. 2. The iteration process goes on ignoring extmarks with irrelevant `ns_id` from Plugin1, until it reaches [0, 10], entering state (A). 3. At the end of cleaning up process, `marktree_del_itr` gets called. This function is supposed to remove given node and, if necessary, restructure the tree. Also, move the iterator to the next node. The bug occurs in this function. 4. The iterator goes backwards to the node's last child, to put it in the place of its deleted parent later. (B) 5. The parent node is deleted and replaced with its child node. (C) 6. Since now this node has 8 children, which is less than `MT_BRANCH_FACTOR - 1`, it get's merged with the next node. (D) 7. Finally, since at (B) the iterator went backward, it goes forward twice, skipping [0, 11] node, causing this extmark to persist, causing the bug. (E) ANALYSIS AND SOLUTION The algorithm works perfectly when the parent node gets replaced by its child, but no merging occurs. I.e. the exact same diagram, but without the (D) stage. If not for (D), it would iterate to <0, 9> and then to [0, 11]. So, iterating twice makes sense. The actual problem is in (C) stage, because the iterator index isn't adjusted and still pointing to no longer existent node. So my solution is to adjust iterator index after removing the child node. More info: https://github.com/neovim/neovim/pull/14719
* | | vim-patch:8.2.3020: unreachable code (#14866)jimman20032021-06-20
|/ / | | | | | | | | Problem: Unreachable code. Solution: Remove the code. (closes vim/vim#8406) https://github.com/vim/vim/commit/2fb749568662c86992aea3b596458b9e470f223d
* | Merge pull request #14824 from vigoux/extmarks-ts-perfBjörn Linse2021-06-18
|\ \ | | | | | | perf(extmarks): allow ephemeral extmarks past EOF to remove O(strlen^2) cost of tree-sitter
| * | perf(extmarks): allow ephemeral extmarks past EOFThomas Vigouroux2021-06-16
| | |
* | | vim-patch:8.2.3012: when 'rightleft' is set the line number is drawn ↵Jan Edmund Lazo2021-06-17
|/ / | | | | | | | | | | | | | | | | reversed (#14839) Problem: When 'rightleft' is set the line number is sometimes drawn reversed. Solution: Adjust how space is handled. (Christian Brabandt, closes vim/vim#8389, closes vim/vim#8391) https://github.com/vim/vim/commit/29f0dc3689eafcf7888e06d57d1cf79e62c5c148
* | Merge pull request #14454 from eltociear/patch-1Thomas Vigouroux2021-06-15
|\ \ | | | | | | screen: fix typo in screen.c
| * | screen: fix typo in screen.cIkko Ashimine2021-04-29
| | | | | | | | | accomodate -> accommodate
* | | option: fix typo in option.cIkko Ashimine2021-06-16
| | | | | | | | | seperated -> separated
* | | Merge pull request #14804 from clason/rounded-bordersMichael Lingelbach2021-06-14
|\ \ \ | | | | | | | | [RDY] feat(float): add rounded borders preset
| * | | feat(float): add rounded borders presetChristian Clason2021-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add `borders = "rounded"` preset for `nvim_open_win`, equivalent to border = {"╭", "─", "╮", "│", "╯", "─", "╰", "│"} Also add undocumented "solid" preset to docs.
* | | | vim-patch:8.2.2990: Jupyter Notebook files are not recognizedJan Edmund Lazo2021-06-13
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Jupyter Notebook files are not recognized. Solution: Recognize *.ipynb. (closes vim/vim#8375) https://github.com/vim/vim/commit/2e66b0d1373891f40e2561ccd2d3369de1614bcd
* | | | vim-patch:8.2.0936: some terminals misinterpret the code for getting cursor ↵Jan Edmund Lazo2021-06-13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | style Problem: Some terminals misinterpret the code for getting cursor style. Solution: Send a sequence to the terminal and check the result. (IWAMOTO Kouichi, closes vim/vim#2126) Merged with current code. https://github.com/vim/vim/commit/a45551a53557dba98973fdb3ff737dea2fffcda3 Cherry-pick Test_cwindow_highlight() from patch v8.1.2040 because it is skipped.
* | | | vim-patch:8.2.1354: test 59 is old styleJan Edmund Lazo2021-06-13
|/ / / | | | | | | | | | | | | | | | Problem: Test 59 is old style. Solution: Convert into a new style test. (Yegappan Lakshmanan, closes vim/vim#6604) https://github.com/vim/vim/commit/aa970abd0a987de96321d33db82f70bbceac931b
* | | Merge pull request #14788 from shadmansaleh/fix/lua_runtime1Björn Linse2021-06-13
|\ \ \ | | | | | | | | fixup(runtime): Fix lua runtime files not listed in :scriptnames
| * | | refactor(startup): Load init.lua with do_sorceshadmansaleh2021-06-13
| | | | | | | | | | | | | | | | This was init.lua will be logged in startuptime
| * | | fixup(runtime): Fix lua runtime files not listedshadmansaleh2021-06-13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | lua runtime files weren't listed in :scriptname & profiler. This fixes that. * Add tests * Small doc tweeks
* | | | vim-patch:8.2.2896: spellfile functionality not fully testedJan Edmund Lazo2021-06-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Spellfile functionality not fully tested. Solution: Add tests for CHECKCOMPOUNDPATTERN and COMMON. (Dominique Pellé, closes vim/vim#8270) https://github.com/vim/vim/commit/dc3275a1ac73b6c4d0c9d2e238ea80b477705b6f
* | | | vim-patch:8.2.0945: cannot use "z=" when 'spell' is offJan Edmund Lazo2021-06-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Cannot use "z=" when 'spell' is off. Solution: Make "z=" work even when 'spell' is off. (Christian Brabandt, Gary Johnson, closes vim/vim#6227) https://github.com/vim/vim/commit/152e79e94bb935e75b866bd55479648cde11066a
* | | | vim-patch:8.2.0046: tests for spell suggestions are slowJan Edmund Lazo2021-06-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Tests for spell suggestions are slow. Solution: Use shorter words. Test with latin1 and utf-8 to cover more code. (Dominique Pelle, closes vim/vim#5399) https://github.com/vim/vim/commit/767340574b5a0c697e650b3bbc3a4af10e51cb89
* | | | vim-patch:8.2.0039: memory access error when "z=" has no suggestionsJan Edmund Lazo2021-06-12
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Memory access error when "z=" has no suggestions. Solution: Check for negative index. https://github.com/vim/vim/commit/569fea2c312126dd5a542c4b1aa51095136a2c0d
* | | | vim-patch:8.2.0038: spell suggestions insufficiently testedJan Edmund Lazo2021-06-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Spell suggestions insufficiently tested. Solution: Add spell suggestion tests. (Dominique Pelle, closes vim/vim#5398) https://github.com/vim/vim/commit/e9a8d1f9adaf4599b5a7923f8db8e207ed6e7eca Requires latest en.utf-8.spl from https://ftp.nluug.nl/pub/vim/runtime/spell/. Include the following patch because patch v8.2.0946 was merged: vim-patch:8.2.0948: spell test fails Problem: Spell test fails. Solution: Adjust expected text of the prompt. https://github.com/vim/vim/commit/d281b7c227bc4c78813fdc297ccee4b2cad7e605
* | | | vim-patch:8.1.1865: spellrare and spellrepall in the wrong orderJan Edmund Lazo2021-06-12
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Spellrare and spellrepall in the wrong order. Solution: Put spellrare below spellrepall. (closes vim/vim#4820) https://github.com/vim/vim/commit/a3891681f72fd9efdea6444620d787358850d823
* | | | vim-patch:8.1.1838: there is :spellwrong and :spellgood but not :spellrareJan Edmund Lazo2021-06-12
|/ / / | | | | | | | | | | | | | | | Problem: There is :spellwrong and :spellgood but not :spellrare. Solution: Add :spellrare. (Martin Tournoij, closes vim/vim#4291) https://github.com/vim/vim/commit/08cc374dabd2a02785129fa1c0100f7745c244ad
* | | fix(ui): Fix pum incorrect position in multigrid modeSerg Tereshchenko2021-06-12
| | | | | | | | | | | | Refs #12985
* | | Merge pull request #14761 from janlazo/vim-8.2.2966Jan Edmund Lazo2021-06-11
|\ \ \ | | | | | | | | vim-patch:8.2.{1702,2422,2966,2971,2974}
| * | | vim-patch:8.2.2974: Greek spell checking uses wrong case foldingJan Edmund Lazo2021-06-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Greek spell checking uses wrong case folding. Solution: Fold capital sigma depending on whether it is at the end of a word or not. (closes vim/vim#299) https://github.com/vim/vim/commit/4f135275984722c1b1e9ace72eeeb7ce7e4ec983
| * | | vim-patch:8.2.2971: cannot yank a block without trailing spacesJan Edmund Lazo2021-06-10
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Cannot yank a block without trailing spaces. Solution: Add the "zy" command. (Christian Brabandt, closes vim/vim#8292) https://github.com/vim/vim/commit/544a38e44db0f25ec4fa7a2a4666cf28a2336f33
| * | | vim-patch:8.2.2422: crash when deleting with line number out of rangeJan Edmund Lazo2021-06-10
| | | | | | | | | | | | | | | | | | | | | | | | Problem: Crash when deleting with line number out of range. (Houyunsong) Solution: Avoid using a negative line number. https://github.com/vim/vim/commit/1d859e24218635c57c09801840ff159cb845ae6a
generated by cgit (git 2.34.1) at 2025-05-01 06:26:55 +0000