Problem:
Since 2f06413dfb#13042, "ESC+c" sequence is treated as "ESC c"
instead of "M-c" (ALT/META+c) when not mapped, aka "fallthrough"
behavior. But "isolated" (non-ALT/META) mappings to ESC and c were not
resolved. This behavior is especially confusing for the TUI.
Solution:
Resolve isolated ESC, c mappings when there is no M-c mapping.
Change ins_char_typebuf() to escape CSI, K_SPECIAL.
fixes#13086fixes#15869
fix(runtime): add compressed {&packpath}/start/*/pack/*[/after] representation to &rtp
by suggestion by at-tpope
Summary:
We can add XDG_DATA_DIR/nvim/site/pack/*/start/* (et al) as an unexpanded wildchar to &rtp which keeps it both short and explicit and still supporting globpath(&rtp, ...).
ref #15101
* refactor: format with uncrustify
* fixup(dundar): fix functions comments
* fixup(dundar): remove space between variable and ++/--
* fixup(dundar): better workaround for macro attributes
This is done to be able to better use uncrustify rules for macros
* fixup(justin): make preprocessors follow neovim style guide
Continuation of https://github.com/neovim/neovim/pull/15202
A plugin like telescope could override it with a fancy implementation
and then users would get the telescope-ui within each plugin that
utilizes the vim.ui.select function.
There are some plugins which override the `textDocument/codeAction`
handler solely to provide a different UI. With custom client commands and
soon codeAction resolve support, it becomes more difficult to implement
the handler right - so having a dedicated way to override the picking
function will be useful.
Initially, we planned to set the NVIM environment variable to detect that
neovim is running in the neovim built-in terminal. At the time this code
was written, there was no way for a parent to set environment variables
for a program running in an embedded terminal. Later it was implemented in
#12937, but the code to set the NVIM was not added. #11390 now uses ConPTY
instead of winpty when possible, so it is no longer necessary to force the
use of win32con even when running inside an embedded terminal. Therefore,
we now do not need this code.
site packages must be sourced before user config
NOTE: we only consider dirs exactly matching "after" to be an AFTER dir.
vim8 considers all dirs like "foo/bar_after", "Xafter" etc, as an
"after" dir in SOME codepaths not not in ALL codepaths.
* add vim.json.encode and vim.json.decode
* use vim.NIL instead of cjson.null
* resolve strict-prototypes warnings
* The following benchmark shows an approximately 2.5x (750 ms vs 300 ms) improvement in deserialization performance over
vim.fn.json_decode on a medium package.json
```lua
local uv = vim.loop
local function readfile(path)
return
end
local json_url = "https://raw.githubusercontent.com/rust-analyzer/rust-analyzer/b24c8d5c89ee93d1172b4127564f5da3b0c88dad/editors/code/package.json"
io.popen(string.format('curl -v -f -L -O %q &> /dev/null', json_url))
local json_string = io.open('package.json'):read '*a'
uv.update_time()
local start = uv.hrtime()
for i = 1,1000 do
vim.fn.json_decode(json_string)
end
uv.update_time()
print(string.format("Deserialization time vim.fn.json_decode: %s ms", (uv.hrtime() - start) * (1e-6)))
uv.update_time()
local start = uv.hrtime()
for i = 1,1000 do
vim.json.decode(json_string)
end
uv.update_time()
print(string.format("Deserialization time vim.json.decode: %s ms", (uv.hrtime() - start) * (1e-6)))
```
Co-Authored-By: Björn Linse <bjorn.linse@gmail.com>
Problem:
This crashes Nvim:
tabedit
call nvim_win_set_option(1000, 'statusline', 'status')
split
wincmd J
wincmd j
Solution:
- Change `no_display` parameter value to be the same as in matching
`restore_win_noblock` call. In case of different values `topframe`
isn't restored to `curtab->tp_topframe`.
- Call `restore_win_noblock` if `switch_win_noblock` returns `FAIL`
(`switch_win` must always have matching `restore_win`)
- Change `switch_win`/`restore_win` to `_noblock` versions to allow
autocommands.
fixes#14097fixes#13577
https://pvs-studio.com/en/docs/warnings/v507/
"Pointer to local array 'sourcing_name_buf' is stored outside the scope
of this array. Such a pointer will become invalid."
False positive: `sourcing_name = save_sourcing_name` before returning
from this scope.
Problem: setreg() does not always clear the register.
Solution: Clear the register if the dict argument is empty. (Andy Massimino,
closesvim/vim#3370)
7633fe595e
Do not getdigits for block_len strictly. For example, a user could
previously abort Nvim using:
:call setreg("0", "abort!", "\<C-V>999999999999999999")
or, after v8.2.0924's port:
:call setreg("0", #{regcontents: ["abort!"],
\ regtype: "\<C-V>999999999999999999"})
Instead, default to 0 so block_len is -1, which acts like the selection
width was omitted (defaults to length of longest line).
Problem: ":call setreg('"', [])" reports an internal error.
Solution: Make the register empty. (Yasuhiro Matsumoto)
659c94d483
Required for v8.2.1035.
Note that setreg('"', []) didn't cause an internal error for us, but
didn't clear the register properly either...
Problem: Cannot save and restore a register properly.
Solution: Add getreginfo() and make setreg() accept a dictionary. (Andy
Massimino, closesvim/vim#3370)
bb861e293e
Cherry-pick eval.txt changes for getreginfo() from:
6aa57295cf207f009326
Problem: Test_eval is old style.
Solution: Change some tests to a new style test.
90455cfa87
Cherry-pick Test_setreg_basic changes from v8.2.0610.
Note that the old-style version of the tests (and the cherry-picked
changes) exist in legacy/eval_spec.lua; keep them as they've already
been Lua'd.
Required for v8.2.0924.
Since the `State` is global, other scripts are unexpectedly affected during the
'inccommand' preview. This commit introduces a new flag for `do_cmdline`, in
order to ignore trailing '|'-separated commands only for the command invoking
the preview.
fix#8796, update #7494
