Breaking Changes:
- Deprecated all `vim.lsp.util.{*diagnostics*}()` functions.
- Instead, all functions must be found in vim.lsp.diagnostic
- For now, they issue a warning ONCE per neovim session. In a
"little while" we will remove them completely.
- `vim.lsp.callbacks` has moved to `vim.lsp.handlers`.
- For a "little while" we will just redirect `vim.lsp.callbacks` to
`vim.lsp.handlers`. However, we will remove this at some point, so
it is recommended that you change all of your references to
`callbacks` into `handlers`.
- This also means that for functions like |vim.lsp.start_client()|
and similar, keyword style arguments have moved from "callbacks"
to "handlers". Once again, these are currently being forward, but
will cease to be forwarded in a "little while".
- Changed the highlight groups for LspDiagnostic highlight as they were
inconsistently named.
- For more information, see |lsp-highlight-diagnostics|
- Changed the sign group names as well, to be consistent with
|lsp-highlight-diagnostics|
General Enhancements:
- Rewrote much of the getting started help document for lsp. It also
provides a much nicer configuration strategy, so as to not recommend
globally overwriting builtin neovim mappings.
LSP Enhancements:
- Introduced the concept of |lsp-handlers| which will allow much better
customization for users without having to copy & paste entire files /
functions / etc.
Diagnostic Enhancements:
- "goto next diagnostic" |vim.lsp.diagnostic.goto_next()|
- "goto prev diagnostic" |vim.lsp.diagnostic.goto_prev()|
- For each of the gotos, auto open diagnostics is available as a
configuration option
- Configurable diagnostic handling:
- See |vim.lsp.diagnostic.on_publish_diagnostics()|
- Delay display until after insert mode
- Configure signs
- Configure virtual text
- Configure underline
- Set the location list with the buffers diagnostics.
- See |vim.lsp.diagnostic.set_loclist()|
- Better performance for getting counts and line diagnostics
- They are now cached on save, to enhance lookups.
- Particularly useful for checking in statusline, etc.
- Actual testing :)
- See ./test/functional/plugin/lsp/diagnostic_spec.lua
- Added `guisp` for underline highlighting
NOTE: "a little while" means enough time to feel like most plugins and
plugin authors have had a chance to refactor their code to use the
updated calls. Then we will remove them completely. There is no need to
keep them, because we don't have any released version of neovim that
exposes these APIs. I'm trying to be nice to people following HEAD :)
Co-authored: [Twitch Chat 2020](https://twitch.tv/teej_dv)
- remove redundant autocmd list
This "grouped" list is useless, it only gets in the way when searching
for event names.
- intro.txt: cleanup
- starting.txt: update, revisit
- doc: `:help bisect`
- mbyte.txt: update aliases 1656367b90. closes#11960
- options: remove 'guifontset'. Why:
- It is complicated and is used by almost no one.
- It is unlikely to be implemented by Nvim GUIs (complicated to parse,
specific to Xorg...).
* Fix some small doc issues
* doc: fixup
* doc: fixup
* Fix lint and rebase
* Remove bad advice
* Ugh, stupid mpack files...
* Don't let people include these for now until they specifically want to
* Prevent duplicate tag
This makes it possible to restore the working directory of :terminal
buffers when reading those buffers from a session file.
Fixes#11288
Co-authored-by: Justin M. Keyes <justinkz@gmail.com>
* 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`.
- make parameters_doc a dict intead of a list
BEFORE:
"parameters_doc": [
{
"buffer": "Buffer handle, or 0 for current buffer"
}
],
AFTER:
"parameters_doc": {
"buffer": "Buffer handle, or 0 for current buffer"
},
- make "return", "seealso", lists instead of strings
Changes the generated msgpack result values in the runtime/doc/*.mpack
files to be formatted like this (instead of being formatted like Vim help text):
[
'nvim_win_get_var': {
'signature': 'nvim_win_get_var({window}, {name}, {err})',
'parameters': [('Window', 'window'), ('String', 'name'), ('Error *', 'err')],
'parameters_doc': [{'window': 'Window handle, or 0 for current window', 'name': 'Variable name'}],
'doc': ['Gets a window-scoped (w:) variable'],
'return': ['Variable value'],
'seealso': []
}
},
...
]
The following script is cut out from vim-patch.sh:
```sh
#!/usr/bin/env bash
BASENAME=vim-patch.sh
printf "\nInstructions:
To port one of the above patches to Neovim, execute this script with the patch revision as argument and follow the instructions, e.g.
'%s -p v8.0.1234', or '%s -P v8.0.1234'
NOTE: Please port the _oldest_ patch if you possibly can.
You can use '%s -l path/to/file' to see what patches are missing for a file.
" "${BASENAME}" "${BASENAME}" "${BASENAME}"
```
The code itself should be correct, but shellcheck 0.7.0 says:
```
In /tmp/test.sh line 5:
printf "\nInstructions:
^-- SC2183: This format string has 2 variables, but is passed 3 arguments.
```
We also had a problem before that a `%s` was added, but the accompanying
argument to printf was forgotten. Using a heredoc is less error-prone, since we
insert variables directly.
Judging class definitions in the form "string.find (line, '=% s * class% (')"
must force writing class definitions in this format, but such a mechanism is Absent.
Also, Lua has no formal class in the language specification, and implements inheritance with setmetadable.
To detect this, we should have a parser for it, not a simple regular expression.
Convenient for API clients who want to reuse the API docs in their own
docs. Could be used e.g. to eliminate nvim.net's own doxygen parser:
3a736232a4/src/NvimClient.APIGenerator/Docs
TODO: currently the result values are formatted as Vim help docs. We
should change the values to have structure, something like this:
[{
'nvim_win_get_var': [
'line1,
'line2',
[ 'item1', 'item2', ... ]
],
'nvim_win_set_var': [
...
],
...
}]
close#11296
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.
* 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
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>
Replace "src/nvim/" => "src/".
Replace ".*/.vim-src/" => "".
This allows to use tab completion based on existing files in Neovim's
source, or via .vim-src.
Previously you would have to typically remove the "nvim/" part manually
yourself.
Align matchit.vim with upstream Vim. We don't want to maintain a fork of
matchit.vim; our small changes should be sent to
https://github.com/chrisbra/matchit
* scripts/vim-patch.sh: fall back to "origin" for remote
Without this, it would fail e.g. with a locally cloned repo of Neovim.
* scripts/vim-patch.sh: assign_commit_details: handle tags [ci skip]
- Handle "v" prefix from Vim tags.
- Exit in case of error therein already.
* -l/-L: pass through git-log options [ci skip]
This allows for only listing missing patches for a given Vim file:
> scripts/vim-patch.sh -L src/edit.c
`vim-patch.sh -L`: down to ~0.5s from ~85s.
`vim-patch.sh -l`: down to ~6s from ~90s.
% diff old new:
193c193
< • v8.0.1366
---
> • v8.0.1367
354d353
< • v8.0.1738
This is due to duplicate tags for Vim commits
(https://github.com/vim/vim/issues/4510):
- vim/vim@1ad022a9b: tagged as v8.0.1367, v8.0.1366
- vim/vim@5d69da462: tagged as v8.0.1739, v8.0.1738
- render_node() is now the main rendering function: it traverses a node
and builds the Vim help text recursively.
- render_para() is weird and ugly, it is the entry-point for rendering
the help text for one docstring'd function.
- Any long symbol is intentional and should never be hardwrapped.
- Vim help tags are often hyphenated, and hardwrapping on hyphens breaks
the Vim help syntax parser.
Reverts previous experiment. PVS root is working correctly, one can
observe this in the PVS-studio.err file, for example:
/usr/local/clang-7.0.0/lib/clang/7.0.0/include/stddef.h:51:1: warning: V677 ...
/usr/local/clang-7.0.0/lib/clang/7.0.0/include/stddef.h:132:1: warning: V677 ...
./src/nvim/fileio.c:1382:1: warning: V1026 ...
./src/nvim/fileio.c:1388:1: warning: V1026 ...
The "./src/nvim/…" paths are correctly rooted, yet PVS somehow still
thinks it should analyze "/usr/local/clang-7.0.0/…".
See also: https://stackoverflow.com/q/44906903
Use uv_set_vterm_state() to override libuv's guess.
See https://github.com/libuv/libuv/pull/1873/ for discussion.
This commit uses a terminal-detection approach based on
GetProcessImageFileNameW(...), which will be reverted in the following
commit. The approach was intended to handle the case of running in
winpty (:terminal), but we will add $NVIM env var for that.
Also add some support for ConEmu, cygwin.
Comparing `vimCommand` from Vim's runtime/syntax/vim.vim, one can see
that "augroup" and similar commands are conspicuously missing. They are
handled specially (`vimAugroupKey`, `vimAutoCmd`).
Excluding them from the generated `vimCommand` keyword list, fixes their
highlighting.
closes#9327
I have `merge.ff = no` in my Git config to not use fast-forward merges
by default, but when updating the Vim sources it should not cause a
merge commit.
[ci skip]
In the case of nested repos, such as when "neovim/" repo contains
"neovim/.vim-src/" repo,
git rev-parse --show-toplevel
returns the fullpath to the "neovim/" repo, which failed the condition.
ref #8875
ref c05d7153d3
Longer summary lines for commits may be wrapped when downloading the
commit from GitHub. Unwrapping the lines ensures they can be compared
with the raw Vim commits.
The difference before and after this patch using:
./scripts/vim-patch.sh -r 8684
Before:
✘ Wrong commit message.
Expected:
vim-patch:8.0.1464: completing directory after :find does not add slash
Problem: Completing directory after :find does not add slash.
Solution: Adjust the flags for globpath(). (Genki Sky)
vim/vim@8a37b03
Actual:
vim-patch:8.0.1464: completing directory after :find does not
add slash
Problem: Completing directory after :find does not add slash.
Solution: Adjust the flags for globpath(). (Genki Sky)
After:
✔ Found expected commit message.
[ci skip]
This ensures that special characters in the variables are not
interpreted as escapes/format characters in the printf string, as was
seen with upstream patch 8.0.0615.
$ ./scripts/vim-patch.sh -p 8.0.0615
Updating Vim sources: /$HOME/src/neovim/.vim-src
✔ Updated Vim sources.
✔ Found Vim revision 'bf15b8d78b22661db8b19d662b62bb9a061cdd37'.
Creating patch...
Pre-processing patch...
✔ Saved patch to '/$HOME/src/neovim/vim-8.0.0615.patch'.
Fetching 'origin/master'.
✔ From https://github.com/neovim/neovim
* branch master -> FETCH_HEAD
Creating new branch 'vim-8.0.0615' based on 'origin/master'.
✔ Switched to a new branch 'vim-8.0.0615'
Branch 'vim-8.0.0615' set up to track remote branch 'master' from 'origin'.
Creating empty commit with correct commit message.
./scripts/vim-patch.sh: line 40: printf: `w': invalid format character
✔ [vim-8.0.0615 db7fbb016] vim-patch:8.0.0615: using ./scripts/vim-patch.sh: line 44: printf: `w': invalid format character
✘ [vim-8.0.0615 db7fbb016] vim-patch:8.0.0615: using
[ci-skip]
plog-converter behaviour is not the best one when creating fullhtml report and
directory already exists: it puts report inside an existing directory. Not sure
what exactly it does if inside exists as well, but if I am not mistaking report
will not be created.
This will allow users to use AppImageUpdate to update their AppImage.
It requires publishing the created zsync file alongside the appimage
file for the releases.
vim-patch:8.0.1206: no autocmd for entering or leaving the command line
(commit a4f6cec7a3)
NA patches:
vim-patch:8.0.0320: warning for unused variable with small build
The hardcoded URL breaks very often, this confuses people.
Instead, auto-detect if no URL is provided.
Also auto-detect if the script is invoked with no arguments.
vimrc_example.vim is not relevant to Nvim. Anything worth having in
there should be made an actual default.
.gitignore:
- remove *.orig ... super annoying
- Prefer "TUI" where possible to refer to the host terminal.
- Remove obsolete tags and ancient TTY exposition.
- Establish "terminal" to consistently mean "terminal emulator" in all
Nvim documentation. This removes the need for verbose qualifiers in
tags and prose.
References #6280
References #6803
This was added from the beginning to submit_pr in 775a16b0b, but I
cannot see why that is useful - in contrast, it will mess with the local
branch in case "origin" cannot be pushed to (i.e. when it points to
neovim/neovim itself).
Needed as argument list is growing large and I absolutely do not find short
options provided by getopts being particularly readable for unfamiliar commands.
[ci skip]
Contains unfinished attempt to integrate IWYU (ref #549). To finish it different
job should be done, specifically:
- Instead of feeding IWYU with modified file a mirror source tree should be
created with the help of CMake which will contain modified sources. This
solves the problem with IWYU thinking that `*.generated.h` headers should be
included in place of `*` headers.
- Build IWYU as all other third-party utilities.
- Make modified sources avoid problems with `nvim/func_attr.h` includes and
various related tricks.
Current script may only be used for manual checks like this:
./scripts/check-includes.py \
--generated-includes-dir build/include \
--generated-includes-dir build/src/nvim/auto \
--file src/nvim/eval/typval.c \
-- -Isrc -Ibuild/include -Ibuild/src/nvim/auto \
-DINCLUDE_GENERATED_DECLARATIONS
(it is also somewhat fine with `--file src/nvim/eval/typval.h`). I have no idea
why (I mean, why developer think that these lines are needed, why they are
suggested is pretty obvious: because there is typedef which mentions them before
structs are defined), but for typval.h it reports, among other things, that it
should add lines
struct dictvar_S;
struct listitem_S;
struct listvar_S;
struct listwatch_S;
Now it checks functions also after every semicolon and closing figure brace,
possibly preceded by whitespaces (tabs and spaces). This should make messing
with declarations in macros not needed.
Reasoning; currently INTERNAL_CALL is mostly used to determine whether it is
needed to deal with NL-used-as-NUL problem. This code is useful for nvim_… API
calls done from VimL, but not for API calls done from lua, yet lua needs to
supply something as channel_id.
During testing found the following bugs:
1. msgpack-gen.lua script is completely unprepared for Float values either in
return type or in arguments. Specifically:
1. At the time of writing relevant code FLOAT_OBJ did not exist as well as
FLOATING_OBJ, but it would be used by msgpack-gen.lua should return type
be Float. I added FLOATING_OBJ macros later because did not know that
msgpack-gen.lua uses these _OBJ macros, otherwise it would be FLOAT_OBJ.
2. msgpack-gen.lua should use .data.floating in place of .data.float. But it
did not expect that .data subattribute may have name different from
lowercased type name.
2. vim_replace_termcodes returned its argument as-is if it receives an empty
string (as well as _vim_id*() functions did). But if something in returned
argument lives in an allocated memory such action will cause double free:
once when freeing arguments, then when freeing return value. It did not cause
problems yet because msgpack bindings return empty string as {NULL, 0} and
nothing was actually allocated.
3. New code in msgpack-gen.lua popped arguments in reversed order, making lua
bindings’ signatures be different from API ones.
No tests yet, no documentation update, no :lua* stuff, no vim module.
converter.c should also work with typval_T, not Object.
Known problem: luaeval("1", {}) results in
PANIC: unprotected error in call to Lua API (attempt to index a nil value)
Ref #3823
It's important that users have a single, easy-to-remember place for
reading about the API. So this commit changes gen_api_vimdoc.py so that
the generated section is appended to api.txt instead of creating
a separate document.
Also remove the section numbering and ToC: it's a maintenance cost, and
it will be unnecessary when #5169 is integrated.
