Commit Graph

5218 Commits

Author SHA1 Message Date
zeertzjq
9381d08e29 test(tui_spec): use RPC request to setup autocommands
This avoids changing cmdline and fixes a warning.
2023-02-18 10:44:35 +08:00
zeertzjq
9b9f8dfcc4 test: make {MATCH:} behave less unexpectedly in screen:expect()
Include the rest of the line and allow multiple {MATCH:} patterns.
2023-02-18 10:44:35 +08:00
Justin M. Keyes
09b3432eaf
fix(api): allow empty Lua table for nested dicts #22268
Problem:
The Lua-API bridge allows Dict params to be empty Lua (list) tables at
the function-signature level. But not for _nested_ Dicts, because they
are not modeled:
fae7540732/src/nvim/api/keysets.lua (L184)
Some API functions like nvim_cmd check for kObjectTypeDictionary and
don't handle the case of empty Lua tables (treated as "Array").

Solution:
Introduce VALIDATE_T_DICT and use it in places where
kObjectTypeDictionary was being checked directly.

fixes #21005
2023-02-16 07:07:18 -08:00
Ghjuvan Lacambre
d34c64e342
feat: $NVIM_APPNAME #22128
This commit implements the ability to control all of the XDG paths
Neovim should use. This is done by setting an environment variable named
NVIM_APPNAME. For example, setting $NVIM_APPNAME makes Neovim look for
its configuration directory in $XDG_CONFIG_HOME/$NVIM_APPNAME instead of
$XDG_CONFIG_HOME/nvim.

If NVIM_APPNAME is not set or is an empty string, "nvim" will be used as
default.

The usecase for this feature is to enable an easy way to switch from
configuration to configuration. One might argue that the various $XDG
environment variables can already be used for this usecase. However,
setting $XDG environment variables also affects tools spawned by Neovim.
For example, while setting $XDG_CONFIG_HOME will enable Neovim to use a
different configuration directory, it will also prevent Git from finding
its "default" configuration.

Closes https://github.com/neovim/neovim/issues/21691
2023-02-16 04:15:02 -08:00
zeertzjq
968cd1ed93 vim-patch:9.0.1309: scrolling two lines with even line count and 'scrolloff' set
Problem:    Scrolling two lines with even line count and 'scrolloff' set.
Solution:   Adjust how the topline is computed. (closes vim/vim#10545)

1d6539cf36

Cherry-pick test_scroll_opt.vim changes from patch 8.2.1432.

Co-authored-by: Bram Moolenaar <Bram@vim.org>
2023-02-16 08:09:41 +08:00
Justin M. Keyes
1539f71363
Merge #22214 move init_default_autocmds to lua 2023-02-14 19:13:12 -05:00
zeertzjq
05faa8f30a
test: make expect_unchanged() less confusing (#22255)
Problem:
The sleep before collecting the initial screen state is confusing and
may lead to unexpected success if it comes after a blocking RPC call.

Solution:
Remove that sleep and add an "intermediate" argument.
2023-02-15 07:26:55 +08:00
Justin M. Keyes
556f8646c0
refactor(api): consistent VALIDATE messages #22262
Problem:
Validation messages are not consistently formatted.
- Parameter names sometimes are NOT quoted.
- Descriptive names (non-parameters) sometimes ARE quoted.

Solution:
Always quote the `name` value passed to a VALIDATE macro _unless_ the
value has whitespace.
2023-02-14 11:19:28 -08:00
bfredl
9a9a4d38a8
Merge pull request #11756 from bfredl/crashfix
custom statusline crash containing unprintable unicode
2023-02-14 18:51:56 +01:00
Björn Linse
39f8aaeb81 fix(status): handle unprintable chars in the statusline 2023-02-14 15:56:14 +01:00
Lewis Russell
00c0a52f94
Merge pull request #22108 from luukvbaal/statuscolumn
perf(column): only build fold/sign column when present in 'statuscolumn'
2023-02-14 14:36:12 +00:00
Justin M. Keyes
ff3d04b75b
refactor(api): VALIDATE macros #22256
- VALIDATE() takes a format string
- deduplicate check_string_array
- VALIDATE_RANGE
- validate UI args
2023-02-14 05:07:38 -08:00
zeertzjq
3a6a7add57
test: remove unused field ext_float (#22243) 2023-02-14 18:26:22 +08:00
Justin M. Keyes
46a87a5d2b
refactor(api): VALIDATE macros #22187
Problem:
- API validation involves too much boilerplate.
- API validation errors are not consistently worded.

Solution:
Introduce some macros. Currently these are clumsy, but they at least
help with consistency and avoid some nesting.
2023-02-14 02:19:04 -08:00
luukvbaal
5396808267
fix(folds): cursorline highlight is not always applied on closed folds (#22242)
Problem:    The cursorline highlight logic checks for `w_cursor.lnum`
            which may be different from the line number passed to
            `win_line()` even when the cursor is actually on that line.
Solution:   Update cursor line highlight logic to check for the line
            number of the start of a closed fold if necessary.
2023-02-14 15:52:22 +08:00
zeertzjq
820430dc0b fix(tui): exit on input eof 2023-02-14 07:49:31 +08:00
zeertzjq
5a8039a0cb test(tui_spec): remove unnecessary arguments for remote UI 2023-02-14 07:49:31 +08:00
Jonas Strittmatter
b518aceaa8
feat(filetype): fall back to file extension when matching from hashbang (#22140)
If nothing matched in match_from_hashbang, also check the file extension table.
For a hashbang like '#!/bin/env foo', this will set the filetype to 'fooscript'
assuming the filetype for the 'foo' extension is 'fooscript' in the extension
table.
2023-02-13 16:04:16 -07:00
Lewis Russell
d359f7a533
Merge pull request #22191 from lewis6991/feat/playground_imp
feat(treesitter): playground improvements
2023-02-13 10:54:03 +00:00
zeertzjq
c2375efe56 fix(ui): make sure screen is valid after resizing
Problem:
When not inside an Ex command, screen_resize() calls update_screen(),
which calls screenclear() and set the screen as valid. However, when
inside an Ex command, redrawing is postponed so update_screen() screen
doesn't do anything, and the screen is still invalid after the resize,
causing ui_comp_raw_line() to be no-op until update_screen() is called
on the main loop.

Solution:
Restore the call to screenclear() inside screen_resize() so that the
screen is invalid after screen_resize(). Since screenclear() changes
redraw type from UPD_CLEAR to UPD_NOT_VALID, it is called at most once
for each resize, so this shouldn't change how much code is run in the
common (not inside an Ex command) case.
2023-02-13 08:50:54 +08:00
dundargoc
f573fcbc0d
build: don't check environment variable to detect CI (#22234)
Instead use the cmake option, which should act as the definitive source
to determine whether we use CI or not.
2023-02-12 14:42:00 +01:00
Jonas Strittmatter
9668c166e8
fix(filetype): make vim.filetype.match() work with contents only (#22181)
Co-authored-by: Gregory Anders <greg@gpanders.com>
2023-02-11 08:08:33 -07:00
glacambre
c5b34fa554 refactor: move init_default_autocmds to lua
The original motivation for this change came from developping
https://github.com/neovim/neovim/pull/22159, which will require adding
more autocommand creation to Neovim's startup sequence.

This change requires lightly editing a test that expected no autocommand
to have been created from lua.
2023-02-11 14:02:17 +01:00
zeertzjq
9437800d28 vim-patch:9.0.1298: inserting register on the cmdline does not trigger incsearch
Problem:    Inserting a register on the command line does not trigger
            incsearch or update hlsearch.
Solution:   Have cmdline_insert_reg() return CMDLINE_CHANGED when appropriate
            and handle it correctly. (Ken Takata, closes vim/vim#11960)

c4b7dec382

Co-authored-by: K.Takata <kentkt@csc.jp>
2023-02-11 19:04:27 +08:00
zeertzjq
7d58de11f4
fix(rpc)!: preseve files when stdio channel is closed (#22137)
BREAKING CHANGE: Unsaved changes are now preserved rather than discarded
when stdio channel is closed.
2023-02-11 18:25:01 +08:00
bfredl
0837980db4 fix(client): wait for session to exit
This replicates the old native.pid_wait(self._pid)
call, except using the proper libuv pattern (run loop unitil exit
callback)
2023-02-10 20:19:04 +01:00
bfredl
f8f82901cd fix(tests): fixes for using vim.mpack and more ASAN 2023-02-10 20:19:04 +01:00
bfredl
d6279f9392 refactor(tests): move lua-client into core and use it for functionaltests
Eliminates lua-client and non-static libluv as test time dependencies

Note: the API for a public lua-client is not yet finished.
The interface needs to be adjusted to work in the embedded loop
of a nvim instance (to use it to talk between instances)
2023-02-10 20:19:04 +01:00
Lewis Russell
8a985d12dd fix(treesitter): don't trample parsers when filetype!=lang
This allows vim.treesitter.show_tree() to work on buffers where the
filetype does not match the parser language name e.g, bash/sh.
2023-02-10 16:15:56 +00:00
zeertzjq
1b379ce430
test(exit_spec): make sure that autocommands are triggered (#22188)
Previously, if the autocommands are not triggered, the tests may still
pass because no assertion is done. Add an assertion so that the tests
will fail if the autocommands aren't triggered.
2023-02-09 19:48:17 +08:00
zeertzjq
b2b82ff142
fix(rpc): ignore redraw events when exiting (#22184)
When a TUI client has already stopped, handling UI events will cause a
heap-use-after-free, so ignore them.
2023-02-09 14:36:17 +08:00
zeertzjq
ecc40660d1
fix(rpc): ignore redraw events when not in UI client (#21892)
Otherwise it will crash.
2023-02-09 10:53:47 +08:00
dundargoc
ffd216e869
build: create test/CMakeLists.txt and move test-related code (#22179)
Having a clear separation between build code and test code makes it
easier to get a higher-level understanding of how the neovim build
works.
2023-02-08 23:46:39 +01:00
Lewis Russell
228684d2fb
fix(decoration): don't show signcolumn for non-sign_text extmark (#22135)
Fixes: #22127
2023-02-05 23:49:43 +00:00
figsoda
e1d5ad1cb8 feat(treesitter): add metadata option for get_node_text 2023-02-04 21:15:03 -05:00
figsoda
bb8845340b feat(treesitter): allow capture text to be transformed
Co-authored-by: Lewis Russell <lewis6991@gmail.com>
2023-02-04 21:04:45 -05:00
zeertzjq
5c4b503d3c
vim-patch:9.0.1279: display shows lines scrolled down erroneously (#22126)
Problem:    Display shows lines scrolled down erroneously. (Yishai Lerner)
Solution:   Do not change "wl_lnum" at index zero. (closes vim/vim#11938)

61fdbfa1e3

Co-authored-by: Bram Moolenaar <Bram@vim.org>
2023-02-05 09:37:12 +08:00
Luuk van Baal
08fb3b5309 perf(column): only build fold/sign column when present in 'statuscolumn'
Problem:    The fold and sign column is built and stored regardless of
            whether the corresponding item is present in 'statuscolumn'.
Solution:   Since the 'statuscolumn' parses itself, we can defer
            building the columns until the corresponding item is
            actually encountered.
2023-02-04 11:04:43 +01:00
bfredl
cbf9199d65
Merge pull request #21331 from LiadOz/LiadOz/prompt-insert-extmark
fix(extmarks): adjust extmarks when inserting prompt prefix
2023-02-02 15:23:09 +01:00
luukvbaal
64fa75a86a
fix(column): estimate 'statuscolumn' width appropriately
Problem:    The 'statuscolumn' width is being estimated without the
            proper context. In particular, this resulted in the fact
            that a custom fold column could be included in the estimated
            `number_width()`, and doubly added when actually drawing the
            statuscolumn due to `win_col_off()` also adding the
            `'foldcolumn'` width. Resulting in a status column that is
            `'foldcolumn'` cells wider than necessary.
Solution:   Estimate 'statuscolumn' width in `get_statuscol_str()` when
            a buffer's line count has changed.
2023-02-02 09:35:51 +00:00
zeertzjq
2c5906b55b
fix(exit): skip unnecessary steps in TUI preserve_exit() (#21897)
This prevents the TUI from doing unexpected things when receiving a
deadly signal or running out of memory.
2023-02-02 10:05:03 +08:00
Lewis Russell
9ce44a750c
fix(man): use italics for <bs>_ (#22086)
fix(man): use italics for <bs>_

Even though underline is strictly what this should be. <bs>_ was used by
nroff to indicate italics which wasn't possible on old typewriters so
underline was used. Modern terminals now support italics so lets use
that now.

See:
- https://unix.stackexchange.com/questions/274658/purpose-of-ascii-text-with-overstriking-file-format/274795#274795
- https://cmd.inp.nsk.su/old/cmd2/manuals/unix/UNIX_Unleashed/ch08.htm
2023-02-01 17:21:42 +00:00
bfredl
8376486e8f
Merge pull request #17537 from bfredl/neolua
refactor(tests): run unittests using main nvim binary - delete separate nvim-test build
2023-01-31 19:58:33 +01:00
bfredl
13aa23b62a refactor(tests): run unittests using main nvim binary in interpreter mode
This allows us to get rid of the separate "nvim-test" target
2023-01-31 19:28:26 +01:00
bfredl
42999a8d64 fix(test): fix issues detected by running unittests in ASAN/UBSAN 2023-01-31 18:28:29 +01:00
Christian Clason
8b11cf5092
Merge pull request #22005 from clason/sanitize-lang
fix(treesitter): validate language name
2023-01-31 17:28:32 +01:00
bfredl
50b256d515 fix(tests): use -l mode for lsp tests
This fixes "fake server" from leaking memory, which makes ASAN very
upset, except on current ASAN CI for some reason.
2023-01-31 13:15:23 +01:00
zeertzjq
860fea1a3f
fix(highlight): properly deal with underline mask when listing (#22057) 2023-01-31 07:08:23 +08:00
Christian Clason
c032e83b22 fix(treesitter): validate language name
Problem: Some injections (like markdown) allow specifying arbitrary
language names for code blocks, which may be lead to errors when
looking for a corresponding parser in runtime path.

Solution: Validate that the language name only contains alphanumeric
characters and `_` (e.g., for `c_sharp`) and error otherwise.
2023-01-28 11:28:52 +01:00
Anton Kriese
3544082f46
test: exepath() returns correct path with cmd.exe, powershell #21928
test(exepath): test if exepath returns correct path with multiple
Windows shells

This test covers the changes from #21175 where exepath() is set to
prefer file extensions in powershell.exe aswell as in cmd.exe.

In both shells, the file with a valid extension should be returned
instead of the extensionless file.
2023-01-26 03:06:29 -08:00