aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
...
| * screen: don't crash on invalid grid cells being recomposedBjörn Linse2019-09-30
|/
* third-party: upgrade libluv: 1.30.0-0 => 1.30.1-1 (#11092)Daniel Hahler2019-09-30
| | | Changes: https://github.com/luvit/luv/compare/1.30.0-0...1.30.1-1
* provider#pythonx: resolve/expand exe from host var (#11047)Daniel Hahler2019-09-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts part of ade88fe4c [1]. This is required for `let g:python3_host_prog = 'python'` etc, where it should get picked up from PATH. Without this it would show: ``` - INFO: pyenv: Path: /home/user/.pyenv/libexec/pyenv - INFO: pyenv: Root: /home/user/.pyenv - INFO: Using: g:python3_host_prog = "python" - ERROR: "python" was not found. - INFO: Executable: Not found - ERROR: Detected pip upgrade failure: Python executable can import "pynvim" but not "neovim": python - ADVICE: - Use that Python version to reinstall "pynvim" and optionally "neovim". pip3 uninstall pynvim neovim pip3 install pynvim pip3 install neovim # only if needed by third-party software ``` Note that it additionally causes a weird error ("Detected pip upgrade failure"), due to `s:check_bin` emptying `python_exe` (because the non-absolute file not being readable), and `provider#pythonx#DetectByModule('pynvim', a:version)` from 75593e6fce then just getting the value from the host var again (without actual checks). This is implicitly fixed via this patch now (because it is skipped), but could need some improvement in this regard probably. With this patch it resolves it (for a virtualenv where pynvim is not made available intentionally): ``` - INFO: pyenv: Path: /home/daniel/.pyenv/libexec/pyenv - INFO: pyenv: Root: /home/daniel/.pyenv - INFO: Using: g:python3_host_prog = "python" - WARNING: $VIRTUAL_ENV exists but appears to be inactive. This could lead to unexpected results. - ADVICE: - If you are using Zsh, see: http://vi.stackexchange.com/a/7654 - INFO: Executable: /home/daniel/.pyenv/shims/tmp-system-deoplete.nvim-f205aF/python - ERROR: Command error (job=11, exit code 1): `'/home/daniel/.pyenv/shims/tmp-system-deoplete.nvim-f205aF/python' -c 'import sys; sys.path.remove(""); import neovim; print(neovim.__file__)'` (in '/home/daniel/.dotfiles/vim/plugged/deoplete.nvim') Output: Traceback (most recent call last): File "<string>", line 1, in <module>ModuleNotFoundError: No module named 'neovim' Stderr: Traceback (most recent call last): File "<string>", line 1, in <module>ModuleNotFoundError: No module named 'neovim' - INFO: Python version: 3.7.4 - INFO: pynvim version: unable to load neovim Python module - ERROR: pynvim is not installed. Error: unable to load neovim Python module - ADVICE: - Run in shell: pip3 install pynvim ``` Note: this appears to display the error twice via "Output:" and "Stderr:". 1: https://github.com/neovim/neovim/pull/8784
* bundle: upgrade LuaJIT to latest v2.1 (#10321)Daniel Hahler2019-09-30
|
* build: run git-describe for dev version during build (#11117)Daniel Hahler2019-09-30
| | | | | | | | This avoids invoking CMake after a new commit, which might take 15s on some systems. Skipped on CMake < 3.2.0 (missing BYPRODUCTS support). Co-Authored-By: Justin M. Keyes <justinkz@gmail.com>
* Merge pull request #11115 from bfredl/wildmenumodeBjörn Linse2019-09-29
|\ | | | | cmdline: wildmenumode() should be true with wildoptions+=pum
| * cmdline: wildmenumode() should be true with wildoptions+=pumBjörn Linse2019-09-28
| |
* | cmake: use LibFindMacros for utf8proc (#11114)Daniel Hahler2019-09-29
| | | | | | Also update doc.
* | terminfo_start: keep first flushing of ui buffer (#11118)Daniel Hahler2019-09-29
|/ | | | | | | | Initially done in 3626d2107 (#11074, for #11062), it was reverted then in 445f2f409 (#11083, which added flushing later). But it is still required here to avoid the reporting of the background response with urxvt/kitty (`nvim -u NONE -cq`). Apparently I've tested this not enough with 445f2f409 (probably only within tmux).
* Merge pull request #10124 from bfredl/tree-sitter-apiBjörn Linse2019-09-28
|\ | | | | Tree-sitter step 1: vendor runtime lib + add lua API
| * tree-sitter: improve and cleanup testsBjörn Linse2019-09-28
| |
| * tree-sitter: use "module" pattern in lua sourceBjörn Linse2019-09-28
| |
| * tree-sitter: handle node equalityBjörn Linse2019-09-28
| |
| * tree-sitter: start docsBjörn Linse2019-09-28
| |
| * tree-sitter: use "range" instead of "point_range" consistently in lua APIBjörn Linse2019-09-28
| |
| * tree-sitter: simplify editing using the new old_byte_size parameterBjörn Linse2019-09-28
| |
| * tree-sitter: fix lint, delete "demo" plugin (replaced by functional tests)Björn Linse2019-09-28
| |
| * tree-sitter: improve parser API (shared parser between plugins)Björn Linse2019-09-28
| |
| * tree-sitter: cleanup tree refcountingBjörn Linse2019-09-28
| |
| * tree-sitter: inspect languageBjörn Linse2019-09-28
| |
| * tree-sitter: add some more APIBjörn Linse2019-09-28
| |
| * tree-sitter: styleBjörn Linse2019-09-28
| |
| * tree-sitter: use standard luaL_newmetatable and luaL_checkudata patternBjörn Linse2019-09-28
| |
| * tree-sitter: rename tree_sitter => treesitter for consistencyBjörn Linse2019-09-28
| |
| * tree-sitter: add basic testing on ciBjörn Linse2019-09-28
| | | | | | | | build tree-sitter c parser on ci for testing purposes
| * tree-sitter: support pre-registration of languagesBjörn Linse2019-09-28
| |
| * tree-sitter: objectify APIBjörn Linse2019-09-28
| |
| * tree-sitter: split tree-sitter lua interface from demo codeBjörn Linse2019-09-28
| |
| * tree-sitter: cleanup build codeBjörn Linse2019-09-28
| |
| * Create BuildUtf8proc.cmake and FindUtf8proc.cmakeBjörn Linse2019-09-28
| | | | | | | | Using advanced search and replace technology
| * tree-sitter: load parsers as .so filesBjörn Linse2019-09-28
| |
| * tree-sitter: initial tree-sitter supportBjörn Linse2019-09-28
| |
| * tree-sitter: change vendored tree-sitter to use nvim memory managementBjörn Linse2019-09-28
| |
| * tree-sitter: vendor tree-sitter runtimeBjörn Linse2019-09-28
|/ | | | | | | | | tree-sitter/tree-sitter commit 7685b7861ca475664b6ef57e14d1da9acf741275 Included files are: lib/include/tree-sitter/*.h lib/src/*.[ch] LICENSE
* vim-patch:8.1.2083: multi-byte chars do not work properly with "%.*S" in ↵Daniel Hahler2019-09-28
| | | | | | | printf() (#11106) Problem: Multi-byte chars do not work properly with "%.*S" in printf(). Solution: Use mb_ptr2cells(). Daniel Hahler, closes vim/vim#4989) https://github.com/vim/vim/commit/ce0fac28977af31f1dec411d3535b4de2c3169b3
* doc: contrib/local.mk.example: include ENABLE_LTO (#11097)Daniel Hahler2019-09-26
| | | | | | | Using it takes 30+ additional seconds for me with a ccache-enabled build (43s vs. 12s). While it certainly makes sense to use DEBUG during development, bisecting etc, it should be made clearer what causes this.
* Merge pull request #10907 from jszakmeister/add-freebsd-sourcehut-supportJohn Szakmeister2019-09-26
|\ | | | | build: add support for building for FreeBSD under Sourcehut
| * build: add support for building for FreeBSD under Sourcehut [skip ci]John Szakmeister2019-09-25
| |
* | Fix two more flaky tests (#11095)Daniel Hahler2019-09-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * mode_spec: retry with increasing matchtime `matchtime=2` might still be too low (with luacov on AppVeyor). [ ERROR ] test/functional\ui\mode_spec.lua @ 47: ui mode_change event works in insert mode test\functional\ui\screen.lua:587: mode Expected objects to be the same. Passed in: (string) 'insert' Expected: (string) 'showmatch' Hint: full state of "mode": "insert" Followup to fe60013fb. ref #10941 Initially regressed in 7ed212262242 `` * ui/screen_basic_spec: set timeoutlen=10000 This fixes the test on slow CI. Ref: https://ci.appveyor.com/project/neovim/neovim/builds/27600387/job/t468h2b3w9lwtlm5#L10930
* | vim-patch:8.0.0914: highlight attributes are always combined (#10256)Daniel Hahler2019-09-26
| | | | | | | | | | | | | | | | | | Problem: Highlight attributes are always combined. Solution: Add the 'nocombine' value to replace attributes instead of combining them. (scauligi, closes vim/vim#1963) https://github.com/vim/vim/commit/0cd2a94a4030f6bd12eaec44db92db108e33c913 Closes https://github.com/neovim/neovim/pull/10256.
* | Merge pull request #10978 from blueyed/update-libuvDaniel Hahler2019-09-25
|\ \ | | | | | | third-party: update libuv to v1.32.0
| * | test/functional/preload.lua: _set_fmode for WindowsDaniel Hahler2019-09-25
| | |
| * | Revert "win/os_env_exists(): workaround libuv bug #10734"Daniel Hahler2019-09-25
| | | | | | | | | | | | This reverts commit 278c5d452c2cbc436a9cc317407ae6021a226c3a.
| * | third-party: update libuv to v1.31.0Daniel Hahler2019-09-25
| |/
* / paste: fix handling of "<" in cmdline (#11094)Daniel Hahler2019-09-25
|/ | | Fixes https://github.com/neovim/neovim/issues/11088.
* tests: busted: nvim handler: display durations always (#11075)Daniel Hahler2019-09-25
| | | | This shows them also with test failures/errors, where it is useful to see how long the test took (for flaky failures running into timeout).
* cmake/GetCompileFlags: include CMAKE_C_COMPILER_ARG1 (#11091)Daniel Hahler2019-09-25
| | | This is used internally (e.g. on Travis) for 32-bit builds (`-m32`).
* timer_spec: fix/harden flaky tests (#11080)Daniel Hahler2019-09-24
| | | | | | | | | | | | | | | | | | | | | | Those are flaky when using luacov (which causes reproducible slowness). E.g.: [ ERROR ] test/functional\eval\timer_spec.lua @ 105: timers can invoke redraw in blocking getchar() call test\functional\ui\screen.lua:587: Row 3 did not match. Expected: |ITEM 1 | |ITEM 2 | |*{1:~ }| |{1:~ }| |{1:~ }| |^ | Actual: |ITEM 1 | |ITEM 2 | |*ITEM 3 | |{1:~ }| |{1:~ }| |^ |
* tui_spec: improve "TUI paste: exactly 64 bytes" (#11086)Daniel Hahler2019-09-24
| | | | | | | | | | | | | | | | | | | | | | | Doing the screen test first might give insights about a possible (flaky?) failure, where it looks like "feed_data" is processed out of order: [ ERROR ] test/functional/terminal/tui_spec.lua @ 561: TUI paste: exactly 64 bytes #10311 test/functional/helpers.lua:388: retry() attempts: 490 test/functional/terminal/tui_spec.lua:66: Expected objects to be the same. Passed in: (table: 0x44042de8) { *[1] = ' endzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz' } Expected: (table: 0x41d6e568) { *[1] = 'zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz end' } stack traceback: test/functional/helpers.lua:388: in function 'retry' test/functional/terminal/tui_spec.lua:63: in function 'expect_child_buf_lines' test/functional/terminal/tui_spec.lua:569: in function <test/functional/terminal/tui_spec.lua:561> Ref: https://github.com/neovim/neovim/pull/11083#issuecomment-534375201 Build log: https://travis-ci.org/neovim/neovim/jobs/588749739#L5597
* tui: flush ui buffer in tui_terminal_after_startup (#11083)Daniel Hahler2019-09-24
| | | | | | | | | | | | | | This avoids having a dummy event to tickle the main loop. Confirmed using `nvim -u NONE -c 'au FocusGained * q'` in tmux (with `:set -g focus-events on`): without the flushing it would only exit after pressing a key. Moves the flushing done recently in 3626d2107. `nvim -u NONE -cq` is still working (i.e. consuming the response for the terminal background query itself), and the flickering mentioned in 3626d2107 is reduced again. Reverts part of bfb21f3e0 (#7729).
generated by cgit (git 2.34.1) at 2025-07-17 15:24:57 +0000