aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
...
* | | clang/'Dead nested assignment': don't assign unused value to precision #15012dundargoc2021-07-07
|/ /
* | clang/'Dead assignment': ignore warning #15000dundargoc2021-07-05
| | | | | | | | This assignment is informative/documenting though technically unnecessary. https://github.com/neovim/neovim/pull/11900#discussion_r381860165 .
* | signs: Change b_signcols_max -> b_signcols_validLewis Russell2021-07-04
| | | | | | | | | | b_signcols_max isn't used to store the max value, it is use to invalidate b_signcols.
* | vim-patch:8.2.3097: crash when using "quit" at recovery promptJan Edmund Lazo2021-07-04
| | | | | | | | | | | | | | | | Problem: Crash when using "quit" at recovery prompt and autocommands are triggered. Solution: Block autocommands when creating an empty buffer to use as the current buffer. (closes vim/vim#8506) https://github.com/vim/vim/commit/1d97efce0ce31ddf0dd4c1ae2228ef4caee6880c
* | vim-patch:8.2.3093: tablabel_tooltip test fails with AthenaJan Edmund Lazo2021-07-04
| | | | | | | | | | | | Problem: tablabel_tooltip test fails with Athena. (Dominique Pellé) Solution: Skip the test when using Athena. (closes vim/vim#8508) https://github.com/vim/vim/commit/fb773a3e0aa9466bb953c29bad0bf4ba34461a40
* | vim-patch:8.2.3088: with 'virtualedit' set to "block" Visual highlight is wrongJan Edmund Lazo2021-07-04
| | | | | | | | | | | | | | Problem: With 'virtualedit' set to "block" Visual highlight is wrong after using "$". (Marco Trosi) Solution: Do not set w_old_cursor_lcol to MAXCOL. (closes vim/vim#8495) https://github.com/vim/vim/commit/9cee4a1c9c69542ccd73bcd2db05920150856361
* | vim-patch:8.2.3087: Gemtext files are not recognizedJan Edmund Lazo2021-07-04
| | | | | | | | | | | | Problem: Gemtext files are not recognized. Solution: Recognize .gmi and .gemini files. (closes vim/vim#8427) https://github.com/vim/vim/commit/fcde67c99fbb6d7294e01b9e87d8d2ae7af31255
* | vim-patch:8.2.3085: JSONC files are not recognizedJan Edmund Lazo2021-07-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: JSONC files are not recognized. Solution: Recognize .jsonc files. (Izhak Jakov, closes vim/vim#8500) https://github.com/vim/vim/commit/acbb4b5720fc27b8b07357b763ed4e5a944ae1da N/A patches for version.c: vim-patch:8.2.3081: cannot catch errors in a channel command Problem: Cannot catch errors in a channel command. Solution: Instead of skipping the error make it silent. (closes vim/vim#8477) https://github.com/vim/vim/commit/11a632d60bde616feb298d180108819ebb1d04a0 vim-patch:8.2.3082: a channel command "echoerr" does not show anything Problem: A channel command "echoerr" does not show anything. Solution: Do not use silent errors when using an "echoerr" command. (closes vim/vim#8494) https://github.com/vim/vim/commit/b836f631dba2534efd314a8f77439cebc75acd4e vim-patch:8.2.3094: Test_popup_atcursor_pos() fails without the conceal feature Problem: Test_popup_atcursor_pos() fails without the conceal feature. Solution: Add a check for the conceal feature. (Dominique Pellé, closes vim/vim#8505) https://github.com/vim/vim/commit/c60e959cba03fae9cff9b2674fab646cc70819fc vim-patch:8.2.3098: popup window test is flaky on MS-Windows with GUI Problem: Popup window test is flaky on MS-Windows with GUI. Solution: Skip the check in this situation. https://github.com/vim/vim/commit/999db2346b61e6f6e6fde2d0b56f4646a46b727d
* | fix(clint): fix link to style guide. (#14975)dundargoc2021-07-03
| |
* | refactor(memfile_test): replace long_u type with size_t. #14974dundargoc2021-07-03
| | | | | | | | long_u was a custom integer type that is no longer used. More information can be found in https://github.com/neovim/neovim/issues/459.
* | PVS/V1051: Ignore "modulo by 1" warning for ARRAY_SIZE. (#14943)dundargoc2021-07-03
| | | | | | | | | | The warning is triggered by the second division when arr is a char array since sizeof(char) is 1. A modulo by 1 may normally indicate a logic error, but in this case it's intended.
* | doc/startup: cleanup; assign ENN to "conflicing configs" msg (#14971)Justin M. Keyes2021-07-03
| |
* | ex_cmds: verify invalid command nameBjörn Linse2021-07-02
| |
* | Merge pull request #14864 from seandewar/get-config-zindexBjörn Linse2021-07-02
|\ \ | | | | | | fix(api/win_get_config): include z-index
| * | fix(api/win_get_config): include z-indexSean Dewar2021-06-19
| | |
* | | Revert "tui: improve support for tmux, GNU Screen" (#14786)Érico Nogueira Rolim2021-06-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Revert "tui: improve support for tmux, GNU Screen" This reverts commit c266c2f36ea56220268d2b19364e28815e338c2f. This commit broke neovim under tmux, where it "types" the background string as input to the terminal. Fixing this with neovim as-is requires changing tmux keybindings or TERM settings, which has to be done by every single user, therefore a fix should happen on neovim's side. Fixes #14298 * tui: fix code lints.
* | | vim-patch:8.2.1905: the wininfo list may contain stale entries (#14884)Daniel Steinberg2021-06-30
| | | | | | | | | | | | | | | | | | Problem: The wininfo list may contain stale entries. Solution: When closing a window remove any other entry where the window pointer is NULL. https://github.com/vim/vim/commit/4882d983397057ea91c584c5a54aaccf15016d18
* | | vim-patch:8.2.3072: "zy" does not work well when "virtualedit' is "block"Jan Edmund Lazo2021-06-29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: The "zy" command does not work well when 'virtualedit' is set to "block". (Johann Höchtl) Solution: Make endspaces zero. (Christian Brabandt, closes vim/vim#8468, closes vim/vim#8448) https://github.com/vim/vim/commit/7d7bcc6ba01454c1cbb1dcbf5f11c3a4051535d6 N/A patches for version.c: vim-patch:8.2.3075: xxd always reports an old version string Problem: Xxd always reports an old version string. (Åsmund Ervik) Solution: Update the version string with the last known change date. (Jürgen Weigert, closes vim/vim#8475) https://github.com/vim/vim/commit/80b2ba3e9602e09ab523e78946fbb26da311b68b
* | | 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.
generated by cgit (git 2.34.1) at 2025-06-29 00:01:44 +0000