Problem:
Empty dictionaries are converted into typed tables of the form `{ [true]
= 6}` instead of an empty dictionary representation `{}`. This leads to
incorrect table representation, along with failure in JSON encoding of
such tables as currently tables with only string and number type keys
can be encoded.
Solution:
The typed table logic has been removed from `nlua_push_Dictionary`. The
typed table logic is required only for float value conversions which is
already handled in `nlua_push_Float`. So, it is(was) no longer required
here.
Fixesneovim/neovim#29218
This patch replaces fswatch with inotifywait from inotify-toools:
https://github.com/inotify-tools/inotify-tools
fswatch takes ~1min to set up recursively for the Samba source code
directory. inotifywait needs less than a second to do the same thing.
https://github.com/emcrisostomo/fswatch/issues/321
Also it fswatch seems to be unmaintained in the meantime.
Signed-off-by: Andreas Schneider <asn@cryptomilk.org>
This is a breaking change which will make refactor of typval and shada
code a lot easier. In particular, code that would use or check for
v:msgpack_types.binary in the wild would be broken. This appears to be
rarely used in existing plugins.
Also some cases where v:msgpack_type.string would be used to represent a
binary string of "string" type, we use a BLOB instead, which is
vimscripts native type for binary blobs, and already was used for BIN
formats when necessary.
msgpackdump(msgpackparse(data)) no longer preserves the distinction
of BIN and STR strings. This is very common behavior for
language-specific msgpack bindings. Nvim uses msgpack as a tool to
serialize its data. Nvim is not a tool to bit-perfectly manipulate
arbitrary msgpack data out in the wild.
The changed tests should indicate how behavior changes in various edge
cases.
Problem: with a single `context.options` there is no way for user to
force which scope (local, global, both) is being temporarily set and
later restored.
Solution: replace single `options` context with `bo`, `go`, `wo`, and
`o`. Naming and implementation follows how options can be set directly
with `vim.*` (like `vim.bo`, etc.).
Options are set for possible target `win` or `buf` context.
Problem: `vim._with()` has many different use cases which are not
covered with tests.
Solution: cover with tests. Some (many) test cases are intentionally
marked as "pending" because they cover cases which don't work as
expected at the moment (and fixing them requires specific knowledge of
C codebase). Use them as a reference for future fixes.
Also some of "can be nested" tests currently might pass only because
the tested context doesn't work.
This change fixes an issue where glob patterns like `{a,ab}` would not
match `ab` because the first option `a` matches, then the end of the
string is expected but `b` is found, and LPeg does not backtrack to try
the next option `ab` which would match. The fix here is to also append
the rest of the pattern to the generated LPeg pattern for each option.
This changes a glob `{a,ab}` from being parsed as
("a" or "ab") "end of string"
to
("a" "end of string" or "ab" "end of string")
Here, matching against `ab` would try the first option, fail to match,
then proceed to the next option, and match.
The sacrifice this change makes is dropping support for nested `{}`
conditions, which VSCode doesn't seem to support or test AFAICT.
Fixes#28931
Co-authored-by: Sergey Slipchenko <faergeek@gmail.com>
It's a function to perform operations in their own sealed context,
similar to pythons `with`. This helps ease operations where you need to
perform an operation in a specific context, and then restore the
context.
Marked as private for now as it's not ready for public use. The current
plan is to start using this internally so we can discover and fix any
problems. Once this is ready to be exposed it will be renamed to
`vim.with`.
Usage:
```lua
local ret = vim._with({context = val}, function()
return "hello"
end)
```
, where `context` is any combination of:
- `buf`
- `emsg_silent`
- `hide`
- `horizontal`
- `keepalt`
- `keepjumps`
- `keepmarks`
- `keeppatterns`
- `lockmarks`
- `noautocmd`
- `options`
- `sandbox`
- `silent`
- `unsilent`
- `win`
(except for `win` and `buf` which can't be used at the same time). This
list will most likely be expanded in the future.
Work on https://github.com/neovim/neovim/issues/19832.
Co-authored-by: Lewis Russell <lewis6991@gmail.com>
Problem: The default commentstring for C/C++ can lead to invalid code
when commenting and does not match the Nvim codebase.
Solution: Change commentstring to `// %s` as used by Neovim. Also
set all commentstrings that derive from the default C string explicitly
(and correctly).
Deprecate vim.diagnostic.goto_prev() and vim.diagnostic.goto_next() in
favor of a unified vim.diagnostic.jump() interface.
We cannot name the function "goto()" because some of our tooling
(luacheck and stylua) fail to parse it, presumably because "goto" is a
keyword in newer versions of Lua.
vim.diagnostic.jump() also allows moving to a specific diagnostic and
moving by multiple diagnostics at a time (useful for creating mappings
that use v:count).
On Windows, '{' is currently not treated as a wildcard char, so another
wildcard char is needed for the pattern to be treated as a wildcard.
It may be worth trying to make '{' always a wildcard char in the future,
but that'll be a bit harder as it'll be necessary to make sure '{' is
escaped at various places.
For many small/simple functions (like those found in shared.lua), the
runtime of vim.validate can far exceed the runtime of the function
itself. Add an "overload" to vim.validate that uses a simple assertion
pattern, rather than parsing a full "validation spec".
Deprecation with vim.deprecate is currently too noisy. Show the
following warning instead:
[function] is deprecated. Run ":checkhealth vim.deprecated" for more information.
The important part is that the full message needs to be short enough to
fit in one line in order to not trigger the "Press ENTER or type command
to continue" prompt.
The full information and stack trace for the deprecated functions will
be shown in the new healthcheck `vim.deprecated`.
Problem: Currently comment detection, addition, and removal are done
by matching 'commentstring' exactly. This has the downside when users
want to add comment markers with space (like with `-- %s`
commentstring) but also be able to uncomment lines that do not contain
space (like `--aaa`).
Solution: Use the following approach:
- Line is commented if it matches 'commentstring' with trimmed parts.
- Adding comment is 100% relying on 'commentstring' parts (as is now).
- Removing comment is first trying exact 'commentstring' parts with
fallback on trying its trimmed parts.
The optimizations that vim.iter uses for array-like tables don't require
that the source table has no holes. The only thing that needs to change
is the determination if a table is "list-like": rather than requiring
consecutive, integer keys, we can simply test for (positive) integer
keys only, and remove any holes in the original array when we make a
copy for the iterator.
Problem:
- The test for vim.deprecate() has a "mock" which is outdated because
vim.deprecate() no longer uses that.
- The tests get confused after a version bump.
Solution:
Make the tests adapt to the current version.
Problem:
The nvim_win_xx_ns function family introduced in ba0370b1d7
needs more bake-time. Currently it's narrowly defined for windows, but
other scopes ("buffer") and features are likely in the future.
Solution:
- Rename the API with double-underscore to mark it as EXPERIMENTAL.
TODO/FUTURE:
- Rename and change the signature to support more than just "window"
scope, and for other flexibility.
- Open question: we could choose either:
- "store scopes on namespaces", or
- "store namespaces on scopes (w:/b:/…)"
reverts e0d92b9cc2#28502
Problem:
`vim.ui.open()` has a `pcall()` like signature, under the assumption
that this is the Lua idiom for returning result-or-error. However, the
`result|nil, errmsg|nil` pattern:
- has precedent in:
- `io.open`
- `vim.uv` (`:help luv-error-handling`)
- has these advantages:
- Can be used with `assert()`:
```
local result, err = assert(foobar())
```
- Allows LuaLS to infer the type of `result`:
```
local result, err = foobar()
if err then
...
elseif result then
...
end
```
Solution:
- Revert to the `result|nil, errmsg|nil` pattern.
- Document the pattern in our guidelines.
Problem: Calling :redraw from vim.ui_attach() callback results in
recursive cmdline/message events.
Solution: Avoid recursiveness where possible and replace global "call_buf"
with separate, temporary buffers for each event so that when a Lua
callback for one event fires another event, that does not result
in invalid memory access.
Follow-up to #28490
Problem:
The new behaviour of goto_next/prev() of navigating to the next highest
severity doesn't work well when diagnostic providers have different
interpretations of severities. E.g. the user may be blocked from
navigating to a useful LSP warning, due to some linter error.
Solution:
The behaviour of next highest severity is now a hidden option
`_highest = true`. We can revisit how to integrate this behaviour
during the 0.11 cycle.
Reverts parts of https://github.com/neovim/neovim/pull/27674
LSP snippets typically do include tabs or spaces to add extra
indentation and don't rely on the client using `autoindent`
functionality.
For example:
public static void main(String[] args) {\n\t${0}\n}
Notice the `\t` after `{\n`
Adding spaces or tabs independent of that breaks snippets for languages
like Haskell where you can have snippets like:
${1:name} :: ${2}\n${1:name} ${3}= ${0:undefined}
To generate:
name ::
name = undefined
Problem: when line is blank link then there will got an invalid column number in math.min compare.
Solution: make sure the min column number is 0 not an illegal number.
Problem: Enabling ext_messages claims to set 'cmdheight' to zero, but
only does so for the current tabpage.
Solution: Set stored 'cmdheight' value to zero for all tabpages.
Problem:
vim.iter has both `rfind()` and various `*back()` methods, which work
in "reverse" or "backwards" order. It's inconsistent to have both kinds
of names, and "back" is fairly uncommon (rust) compared to python
(rfind, rstrip, rsplit, …).
Solution:
- Remove `nthback()` and let `nth()` take a negative index.
- Because `rnth()` looks pretty obscure, and because it's intuitive
for a function named `nth()` to take negative indexes.
- Rename `xxback()` methods to `rxx()`.
- This informally groups the "list-iterator" functions under a common
`r` prefix, which helps discoverability.
- Rename `peekback()` to `pop()`, in duality with the existing `peek`.
Problem:
`vim.ui.open` unnecessarily invents a different success/failure
convention. Its return type was changed in 57adf8c6e0, so we might as
well change it to have a more conventional form.
Solution:
Change the signature to use the `pcall` convention of `status, result`.
vim.fs.root() is a function for finding a project root relative to a
buffer using one or more "root markers". This is useful for LSP and
could be useful for other "projects" designs, as well as for any plugins
which work with a "projects" concept.
Specifically, functions that are run in the context of the test runner
are put in module `test/testutil.lua` while the functions that are run
in the context of the test session are put in
`test/functional/testnvim.lua`.
Closes https://github.com/neovim/neovim/issues/27004.
Problem: `vim.deprecate()` can be relatively significantly slower than
the deprecated function in "Nvim" plugin.
Solution: Optimize checks for "Nvim" plugin. This also results into not
distinguishing "xxx-dev" and "xxx" versions when doing checks, which
is essentially covered by the deprecation logic itself.
With this rewrite I get the times from #28459: `{ 0.024827, 0.003797, 0.002024, 0.001774, 0.001703 }`.
For quicker reference:
- On current Nightly it is something like `{ 3.72243, 0.918169, 0.968143, 0.763256, 0.783424 }`.
- On 0.9.5: `{ 0.002955, 0.000361, 0.000281, 0.000251, 0.00019 }`.
Problem:
We need to establish a pattern for `enable()`.
Solution:
- First `enable()` parameter is always `enable:boolean`.
- Update `vim.diagnostic.enable()`
- Update `vim.lsp.inlay_hint.enable()`.
- It was not released yet, so no deprecation is needed. But to help
HEAD users, it will show an informative error.
- vim.deprecate():
- Improve message when the "removal version" is a *current or older* version.
Problem:
vim.diagnostic.enable() does not match the signature of vim.lsp.inlay_hint.enable()
Solution:
- Change the signature so that the first 2 args are (bufnr, enable).
- Introduce a 3rd `opts` arg.
- Currently it only supports `opts.ns_id`.
Problem:
`vim.diagnostic.is_disabled` and `vim.diagnostic.disable` are unnecessary
and inconsistent with the "toggle" pattern (established starting with
`vim.lsp.inlay_hint`, see https://github.com/neovim/neovim/pull/25512#pullrequestreview-1676750276
As a reminder, the rationale is:
- we always need `enable()`
- we always end up needing `is_enabled()`
- "toggle" can be achieved via `enable(not is_enabled())`
- therefore,
- `toggle()` and `disable()` are redundant
- `is_disabled()` is a needless inconsistency
Solution:
- Introduce `vim.diagnostic.is_enabled`, and `vim.diagnostic.enable(…, enable:boolean)`
- Note: Future improvement would be to add an `enable()` overload `enable(enable:boolean, opts: table)`.
- Deprecate `vim.diagnostic.is_disabled`, `vim.diagnostic.disable`
Problem:
vim.ui.open "locks up" Nvim if the spawned process does not terminate. #27986
Solution:
- Change `vim.ui.open()`:
- Do not call `wait()`.
- Return a `SystemObj`. The caller can decide if it wants to `wait()`.
- Change `gx` to `wait()` only a short time.
- Allows `gx` to show a message if the command fails, without the
risk of waiting forever.
Also close Nvim instance before removing log file, otherwise the Nvim
instance will still write to the log file.
Also adjust log level in libuv_process_spawn(). Ref #27660
Design
- Enable commenting support only through `gc` mappings for simplicity.
No ability to configure, no Lua module, no user commands. Yet.
- Overall implementation is a simplified version of 'mini.comment'
module of 'echasnovski/mini.nvim' adapted to be a better suit for
core. It basically means reducing code paths which use only specific
fixed set of plugin config.
All used options are default except `pad_comment_parts = false`. This
means that 'commentstring' option is used as is without forcing single
space inner padding.
As 'tpope/vim-commentary' was considered for inclusion earlier, here is
a quick summary of how this commit differs from it:
- **User-facing features**. Both implement similar user-facing mappings.
This commit does not include `gcu` which is essentially a `gcgc`.
There are no commands, events, or configuration in this commit.
- **Size**. Both have reasonably comparable number of lines of code,
while this commit has more comments in tricky areas.
- **Maintainability**. This commit has (purely subjectively) better
readability, tests, and Lua types.
- **Configurability**. This commit has no user configuration, while
'vim-commentary' has some (partially as a counter-measure to possibly
modifying 'commentstring' option).
- **Extra features**:
- This commit supports tree-sitter by computing `'commentstring'`
option under cursor, which can matter in presence of tree-sitter
injected languages.
- This commit comments blank lines while 'tpope/vim-commentary' does
not. At the same time, blank lines are not taken into account when
deciding the toggle action.
- This commit has much better speed on larger chunks of lines (like
above 1000). This is thanks to using `nvim_buf_set_lines()` to set
all new lines at once, and not with `vim.fn.setline()`.
`exec_lua` makes code slighly harder to read, so it's beneficial to
remove it in cases where it's possible or convenient.
Not all `exec_lua` calls should be removed even if the test passes as it
changes the semantics of the test even if it happens to pass.
From https://github.com/neovim/neovim/pull/28155#discussion_r1548185779:
"Note for tests like this, which fundamentally are about conversion, you
end up changing what conversion you are testing. Even if the result
happens to be same (as they often are, as we like the rules to be
consistent if possible), you are now testing the RPC conversion rules
instead of the vim script to in-process lua conversion rules."
From https://github.com/neovim/neovim/pull/28155#discussion_r1548190152:
"A test like this specifies that the cursor is valid immediately and not
after a separate cycle of normal (or an other input-processing) mode."
Backslashes are valid characters in unix style paths.
Fix the conversion of backslashes to forward slashes in several `vim.fs`
functions when not on Windows. On Windows, backslashes will still be converted
to forward slashes.
When the edited file is a symlink, the unexpanded file name is needed to
to achieve the same behavior as the autocommand pattern matching in Vim.
Neither args.file nor args.match are guaranteed to be unexpanded, so use
bufname() instead.
A lot of functions in move.c only worked for curwin, alternatively
took a `wp` arg but still only work if that happens to be curwin.
Refactor those that are needed for update_topline(wp) to work
for any window.
fixes#27723fixes#27720
Problem: Text properties are wrong after "cc". (Axel Forsman)
Solution: Pass the deleted byte count to inserted_bytes(). (closesvim/vim#10412,
closesvim/vim#7737, closesvim/vim#5763)
d0b1a09f44
Co-authored-by: LemonBoy <thatlemon@gmail.com>
Problem: Crash on exit with EXITFREE and using win_execute().
Solution: Also save and restore tp_topframe. (issue vim/vim#9374)
dab17a0689
Couldn't repro the crash in the test, but I only care about this patch so
switch_win sets topframe properly for win_split_ins in nvim_open_win and
nvim_win_set_config.
Add a test using nvim_win_call and :wincmd, as I couldn't repro the issue via
nvim_open_win or nvim_win_set_config (though it's clear they're affected by this
patch).
That said, at that point, could just use {un}use_tabpage inside switch_win
instead, which also updates tp_curwin (though maybe continue to not set it in
restore_win). That would also fix possible inconsistent behaviour such as:
:call win_execute(w, "let curwin_nr1 = tabpagewinnr(1)")
:let curwin_nr2 = tabpagewinnr(1)
Where it's possible for curwin_nr1 != curwin_nr2 if these commands are run from
the 1st tabpage, but window "w" is in the 2nd (as the 1st tabpage's tp_curwin
may still be invalid). I'll probably PR a fix for that later in Vim.
Co-authored-by: Bram Moolenaar <Bram@vim.org>
Problem:
vim._watch.watchdirs has terrible performance.
Solution:
- On linux use fswatch as a watcher backend if available.
- Add File watcher section to health:vim.lsp. Warn if watchfunc is
libuv-poll.
Problem: Some LSP servers return `textDocument/documentLink` responses
containing file URIs with line/column numbers in the fragment.
`vim.uri_to_fname` returns invalid file names for these URIs.
Solution: Remove the URI fragment from file URIs.
Currently, highlight.on_yank() does buffer-local highlighting, this PR
makes it window scoped.
Also fix the problem that when yanking in a buffer, moving to another
buffer, and yanking before the original buffer highlight disappears, the
original buffer highlight won't disappear on timeout.
and for return value of nlua_exec/nlua_call_ref, as this uses
the same family of functions.
NB: the handling of luaref:s is a bit of a mess.
add api_luarefs_free_XX functions as a stop-gap as refactoring
luarefs is a can of worms for another PR:s.
as a minor feature/bug-fix, nvim_buf_call and nvim_win_call now preserves
arbitrary return values.
Problem: Loading `vim.fs` via the `vim.loader` Lua package loader will
result in a stack overflow due to a cyclic dependency. This may happen
when the `vim.fs` module isn't byte-compiled, i.e. when `--luamod-dev`
is used (#27413).
Solution: `vim.loader` depends on `vim.fs`. Therefore `vim.fs` should
be loaded in advance.
Problem:
Processing non-fast events during SystemObj:wait() may cause two pieces
of code to interfere with each other, and is different from jobwait().
Solution:
Don't process non-fast events during SystemObj:wait().
