Fixes:
% rm -rf .deps
% cmake -DUSE_BUNDLED_LUV=ON -DUSE_BUNDLED=OFF -S third-party -B .deps
% cmake --build .deps
...
[ 50%] Building C object CMakeFiles/luv.dir/src/luv.c.o
/home/daniel/Vcs/neovim/.deps/build/src/luv/src/luv.c:20:10: fatal error: c-api/compat-5.3.h: No such file or directory
20 | #include "c-api/compat-5.3.h"
| ^~~~~~~~~~~~~~~~~~~~
compilation terminated.
This was broken since 78f1eccc8, where apparently the `LUA_COMPAT53_DIR`
was defined in a wrong location.
* scripts/vim-patch.sh: factor out _set_tokens_and_tags
This allows for caching `$tokens` and `$vim_commit_tags`, which will
become relevant with the next commit adding
`list_missing_previous_vimpatches_for_patch`.
Flaky failure (Travis CI, macOS):
[ RUN ] :terminal (with fake shell) works with gf: 10518.41 ms FAIL
test/functional/terminal/ex_terminal_spec.lua:248: Row 1 did not match.
Expected:
|*^ready $ echo "scripts/shadacat.py" |
|* |
|*[Process exited 0] |
|:terminal echo "scripts/shadacat.py" |
Actual:
|*^ |
|*[Process exited 0] |
|* |
|:terminal echo "scripts/shadacat.py" |
To print the expect() call that would assert the current screen state, use
screen:snapshot_util(). In case of non-deterministic failures, use
screen:redraw_debug() to show all intermediate screen states.
stack traceback:
test/functional/ui/screen.lua:579: in function '_wait'
test/functional/ui/screen.lua:361: in function 'expect'
test/functional/terminal/ex_terminal_spec.lua:248: in function <test/functional/terminal/ex_terminal_spec.lua:245>
Follow-up to 8969efca8 (Vim patch 8.1.0723)
NOTE: This changes the main entrypoint for running single oldtest files
to not use/require the ".res" extension anymore. But it is handled for
B/C.
Adds a phony rule to run oldtest by filename.
Not going through "$(MAKE)" avoids GNUmakefile being used then (which I
use for WIP things), and it seems like SINGLE_MAKE should be used anyway
probably.
- Running out of retries, or unexpected screen state should make the
test FAIL, not ERROR.
- Uses levels to report the location of the caller.
- Improve message with retry-failure (formatting).
Before:
[ RUN ] test: 103.53 ms ERR
test/functional/helpers.lua:388:
retry() attempts: 1
test/functional/ui/screen.lua:587: Row 1 did not match.
Expected:
|*X^ |
|{0:~ }|
|{0:~ }|
| |
Actual:
|*^ |
|{0:~ }|
|{0:~ }|
| |
To print the expect() call that would assert the current screen state, use
screen:snapshot_util(). In case of non-deterministic failures, use
screen:redraw_debug() to show all intermediate screen states.
stack traceback:
test/functional/helpers.lua:388: in function 'retry'
test/functional/test_spec.lua:24: in function <test/functional/test_spec.lua:23>
After:
[ RUN ] test: 105.22 ms FAIL
test/functional/test_spec.lua:24: stopping after 1 retry() attempts.
test/functional/test_spec.lua:25: Row 1 did not match.
Expected:
|*X^ |
|{0:~ }|
|{0:~ }|
| |
Actual:
|*^ |
|{0:~ }|
|{0:~ }|
| |
To print the expect() call that would assert the current screen state, use
screen:snapshot_util(). In case of non-deterministic failures, use
screen:redraw_debug() to show all intermediate screen states.
stack traceback:
test/functional/helpers.lua:389: in function 'retry'
test/functional/test_spec.lua:24: in function <test/functional/test_spec.lua:23>
Avoid noise during builds:
> fatal: No annotated tags can describe '417449f468c4ba186954f6295b3338fb55ee7b4a'.
> However, there were unannotated tags: try --tags.
This might be useful in general, but is expected to not happen - and
falling back is OK then. The fallback command would still display
errors then.
It also uses `--first-parent`, which is important for when a release
branch gets merged back.
This is relevant for when using `USE_BUNDLED_LUAJIT=ON` with
`USE_BUNDLED_LUAROCKS=OFF`, and then building without the necessary modules
being installed/activated there yet: it would check the other (system) "lua"
interpreters also, and in case all failed keep the `LUA_PRG` in the cache for
the last failed entry - making it not re-check the previous ones on the next
build (after you might have activated your custom LuaRocks installation).
Only setting LUA_PRG if the check was successful handles the case better where
it is configured already - we should not try to re-configure it then.
The Python provider was special (via [1]), and would continue to do
checks with `0` being set explicitly even.
This was fixed in #11044 (45447e3b6), ref: #11040.
This extends it to use the same method with all providers.
1: https://github.com/neovim/neovim/pull/8047
`tui_terminal_after_startup` gets called right after resuming from
suspending (via `Ctrl-z`) already (not delayed as with the startup
itself), and would set `waiting_for_bg_response` to false then directly.
This results in the terminal response not being processed then anymore,
and leaking into Neovim itself.
This changes it to try 5 times always, which means that it typically
would stop after a few characters of input from the user typically, e.g.
with tmux, which does not send a reply.
While it might be better to have something based on the time (e.g. only
wait for max 1s), this appears to be easier to do.
Fixes regression in 8a4ae3d.
- output "[...]" to indicate throttling is being used, instead of just
an empty line
- go to beginning of line after displaying the pulse, so that following
output is displayed over it
Meant to fix:
[ ERROR ] test/functional/terminal/tui_spec.lua @ 925: TUI FocusGained/FocusLost in terminal-mode
test/functional/ui/screen.lua:587: Row 6 did not match.
Expected:
|{1:r}eady $ |
|[Process exited 0] |
| |
| |
| |
|*gained |
|{3:-- TERMINAL --} |
Actual:
|{1:r}eady $ |
|[Process exited 0] |
| |
| |
| |
|*:terminal |
|{3:-- TERMINAL --} |
To print the expect() call that would assert the current screen state, use
screen:snapshot_util(). In case of non-deterministic failures, use
screen:redraw_debug() to show all intermediate screen states.
stack traceback:
test/functional/ui/screen.lua:587: in function '_wait'
test/functional/ui/screen.lua:370: in function 'expect'
test/functional/terminal/tui_spec.lua:934: in function <test/functional/terminal/tui_spec.lua:925>
I've thought about adding this, but it might not be really relevant, and
slows down the tests a bit (and a warning "warning: Screen test
succeeded immediately" with another test):
```diff
diff --git i/test/functional/terminal/tui_spec.lua w/test/functional/terminal/tui_spec.lua
index ada073c4e..4bc2ab4e0 100644
--- i/test/functional/terminal/tui_spec.lua
+++ w/test/functional/terminal/tui_spec.lua
@@ -818,6 +818,11 @@ describe('TUI FocusGained/FocusLost', function()
..'", "-u", "NONE", "-i", "NONE", "--cmd", "set noswapfile noshowcmd noruler"]')
feed_data(":autocmd FocusGained * echo 'gained'\n")
feed_data(":autocmd FocusLost * echo 'lost'\n")
+ -- Wait for autocommand to be registered.
+ retry(nil, nil, function()
+ feed_data(":autocmd FocusLost\n")
+ screen:expect{any=" echo 'lost'"}
+ end)
feed_data("\034\016") -- CTRL-\ CTRL-N
end)
```
"smglr" was added for TERM=xterm recently to the terminfo database,
which causes display issues with terminals that use `TERM=xterm` by
default for themselves, although not supporting it.
This patch makes "smglr" to be ignored then.
Fixes https://github.com/neovim/neovim/issues/10562
* Fix/keep massaging git-describe result
Ref: https://github.com/neovim/neovim/pull/11117#issuecomment-536416223
* build: revisit generation of version from Git
Fixes "make clean && make", where "auto/versiondef.h" would be missing
since b18b84d - because BYPRODUCTS are apparently removed when cleaning.
This includes the following improvements/changes:
- do not run git-describe during CMake's configure phase just for
reporting
- do not print with changed Git version (too noisy, simplifies code)
* Move to src/nvim (included before config) for easier flow
* fallback to describe always, write empty include file
* update_version_stamp.lua: use prefix always
* handle_background_color: short-circuit if handled already
* Unit tests for handle_background_color
* set waiting_for_bg_response to false in tui_terminal_after_startup
By then it should have been received.
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
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>
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).
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.
* 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 7ed2122622
``
* 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
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
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).
This aligns with `terminfo_stop`, which also flushes the buffer
after disabling things.
This ensures Neovim gets the response to the terminal background query
before exiting (`nvim -u NONE -cq` with e.g. urxvt or kitty).
Caveats:
* With kitty this causes some "flickering", likely since the alternate
screen is being setup with `nvim -u NONE -cq`, whereas it would not be
processed otherwise before quitting (as with the background query).
* tmux after this patch may print ^[[I (CSI I / FocusGained) after
`nvim -u NONE -cq`.
Fixes https://github.com/neovim/neovim/issues/11062
Daniel Hahler