Also, kudos to @zsugabubus for fixing a related issue in #12417
This also prevents any sorting of the paths from man. We need to
respect the order we get from it otherwise you end up loading
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/share/man/man1/ls.1
on MacOS instead of /usr/share/man/man1/ls.1
Also cleaned it up a little and made it faster.
Closes#9159 and #9271
Also changes man#extract_sect_and_name_ref to only return a single
section at a time. This fixes a bug in its usage in man#goto_tag
where get_paths would be called with multiple sections and it does
not support that.
I noticed that our tagfunc doesn't obey b:man_default_sects and
I'll fix that next.
When UV_OVERLAPPED_PIPE was used for the pipe passed to the child process, a
problem occurred with the standard input of the .Net Framework application
(#11809). Therefore, add the overlapped option to jobstart() and change it so
that it is set only when necessary
* take wrapping into account when computing float height
* factor out size calculation
* add test
* accept and pass through opts.wrap_at in floating_preview
* make padding configurable
* slightly refactor fancy_floating_markdown to make use of make_position
* padding using string.format
* move trim and pad to separate function
* nit
Co-authored-by: Hirokazu Hata <h.hata.ai.t@gmail.com>
* remove mention of backward compat
* make lint happy
Co-authored-by: Hirokazu Hata <h.hata.ai.t@gmail.com>
* Add check for typeDefinition support in vim lsp server
* Check for typeDefinitionProvider in server
* Check for declarationProvider in server
* Add check for client support
* Fix typo
Old behavior is: foo(${placeholder: bar, ...)
with lots of random garbage you'd never want inserted.
New behavior is: foo(bar, baz)
(which maybe is good, maybe is bad [depends on user], but definitely better than it was).
-----
* Implement rudimentary snippet parsing
Add support for parsing and discarding snippet tokens from the completion items.
Fixes#11982
* Enable snippet support
* Functional tests for snippet parsing
Add simplified real-world snippet text examples to the completion items
test
* Add a test for nested snippet tokens
* Remove TODO comment
* Return the unmodified item if the format is plain text
* Add a plain text completion item
Problem: Terminal debugger: can't re-open source code window.
Solution: Add the :Source command. Also create the window if needed when
gdb stops at a source line.
c4b533e1e9
It's confusing because vim-lsp already has the same name as the plugin name that predates this built-in lsp.
Also, since "vim.fn.stdpath" is used, adding the prefix "nvim-" is redundant, so just "lsp.log" will suffice.
The `:lcd -` command, and `:tcd -` which returns to a previous directory, can be done with `:lc -` and `:tc -` respectively.
Basically, the d is optional, so I updated the documentation to indicate this in the traditional format.
* Change uri_to_fname to not convert non-file URIs
A URI with a scheme other than file doesn't have a local file path.
* fixup! Change uri_to_fname to not convert non-file URIs
* fixup! fixup! Change uri_to_fname to not convert non-file URIs
The callbacks for `textDocument/documentSymbol` and `workspace/symbol`
never received the `bufnr` argument because the logic that adds error
validation and logging swallowed the argument.
* add lua function to highlight yanked region
* extract namespace, better naming, default values
* add default for event argument
* free timer
* factor out mark to position calculation
* d'oh
* make sure timer stops before callback (cf. luv example)
* factor out timer, more documentation
* fixup
* validate function argument for schedule
* fix block selection past eol
* correct handling of multibyte characters
* move arguments around, some cleanup
* move utility functions to vim.lua
* use anonymous namespaces, avoid local api
* rename function
* add test for schedule_fn
* fix indent
* turn hl-yank into proper (hightlight) module
* factor out position-to-region function
mark extraction now part of highlight.on_yank
* rename schedule_fn to defer_fn
* add test for vim.region
* todo: handle double-width characters
* remove debug printout
* do not shadow arguments
* defer also callable table
* whitespace change
* move highlight to vim/highlight.lua
* add documentation
* add @return documentation
* test: add check before vim.defer fires
* doc: fixup
If the LSP sends an end line that is larger than what nvim considers to be the
last line, you get an Index out of bounds error when fetching the line from
nvim, a change that was introduced in #12223. This change removes the strict
indexing and checks the return value from nvim_buf_get_lines.
Problem: Cannot delete a match from another window. (Paul Jolly)
Solution: Add window ID argument to matchdelete(), clearmatches(),
getmatches() and setmatches(). (Andy Massimino, closesvim/vim#4178)
aff749145e
At least the `gopls` language server seems to return nil/null if no code
actions are available. Currently this results in an error:
> Error executing vim.schedule lua callback: shared.lua:199: Expected table, got nil
* Add textDocument/codeAction
* Add callback for workspace/executeCommand
* Escape newlines in codeAction titles
* Return empty list in get_line_diagnostics if no buffer diagnostics
* Add stub documentation
* Validate context parameter in code_action
* Add support for edit in CodeAction responses
* Group diagnostics by line in vim.lsp.util.get_line_diagnostics()
* Advertise code action literal support
* Make apply_text_edits non-ASCII safe
Use `vim.str_byteindex` to correct starting and ending positions for text edits if the line contains non-ASCII characters.
Fixes#12221
* text_edit may be applied to other buffers
* make sure the buffer is loaded
* add comments
* add test for non-ASCII edits
* lsp: handle kinds not specified in protocol
fix: #12200
If the client set "symbolKind.valueSet",
the client must handle it properly even if it receives a value outside the specification.
* test: add lsp.util.{get_completion_item_kind_name, get_symbol_kind_name} test case
* lsp: make lsp.util.{get_completion_item_kind_name, get_symbol_kind_name} private
Besides the special-case in get_scrolloff_value(), it makes sense for
'scrolloff' and 'sidescrolloff' to reflect the correct values (for
plugins, scripts, …).
ref 53d607af9c53accfd634435908fb79061f1212b9
ref #11915
ref #12230
* lsp: add workspace/symbol
* refactor symbol callback
* set hierarchical symbol support to true
* add documentation and default mapping
Co-authored-by: Hirokazu Hata <h.hata.ai.t@gmail.com>
This commit creates 4 new highlight groups:
- LspDiagnosticsErrorSign
- LspDiagnosticsWarningSign
- LspDiagnosticsInformationSign
- LspDiagnosticsHintSign
These highlight groups are linked to their corresponding LspDiagnostics
highlight groups by default.
This lets users choose a different color for their sign columns and
virtualtext diagnostics.
According to the spec there is the possibility that when a
VersionedTextDocumentIdentifier is used in a TextEdit the value may be
null. Currently we don't check for this and always assume that it's set.
So currently if a TextEdit comes in for a rename for example with the
version null, it fails as we are comparing the bufnumber with nil.
https://microsoft.github.io/language-server-protocol/specification#versionedTextDocumentIdentifier
The method with the name 'textDocument/peekDefinition' is not part of
the official language server protocol specification. Therefore no
language server can/will support this. Thereby all related code and
documentation as been removed.
Expose `vim.lsp.buf.diagnostics_by_buf`
This makes it easier to customize the diagnostics behavior. For example
to defer the update they can override the
`textDocument/publishDiagnostics` callback to only call
`buf_diagnostics_save_positions` and then defer the other actions to a
autocmd event.
In contrast to other callbacks for LSP requests like
`textDocument/documentSymbols`, does the one for references not open the
quickfix window after the quickfix list was filled. This left the user
in a situation he don't know what or if something had happen.
Related to: neovim/neovim#12170
Problem: The "last used" info of a buffer is under used.
Solution: Add "lastused" to getbufinfo(). List buffers sorted by last-used
field. (Andi Massimino, closesvim/vim#4722)
52410575be
Before this commit, the LSP client would throw errors when messages
without severity would be sent by the server. We make severity default
to `Error` as a kludge before proper heuristics to discover the severity
of a message are found.
[RFC] Update treesitter runtime
Get rid of our utf8proc dependency
Note that we unconditionnally escape treesitter queries, might need to be revisited.
Fix#12122
>Error executing vim.schedule lua callback: /usr/local/share/nvim/runtime/lua/vim/lsp/util.lua:560: Expected lua string
Some lsp server(e.g. https://github.com/bash-lsp/bash-language-server) not have kind in reference, reference["kind"] is nil
On some versions of Windows, WSL is unable to execute symbolic links to
Windows executables (microsoft/WSL#3999). As a workaround for that problem
this changes to use resolve() on WSL if win32yank was a symbolic link.
fixes#12113.
According to the LSP specification, showMessage is what is displayed and logMessage is what is stored.
Since these are different things, I devide the callback into those that match.
Problem: Can execute shell commands in rvim through interfaces.
Solution: Disable using interfaces in restricted mode. Allow for writing
file with writefile(), histadd() and a few others.
8c62a08faf
Problem: A plugin cannot easily expand a command like done internally.
Solution: Add the expandcmd() function. (Yegappan Lakshmanan, closesvim/vim#4514)
80dad48c50
Problem: :echomsg and :echoerr do not handle List and Dict like :echo does.
(Daniel Hahler)
Solution: Be more tolerant about the expression result type.
461a7fcfce
Add lua functional tests for :echo,:echon,:echomsg,:echoerr
because nvim did not port "test_" functions from Vim
that modify internal state.
Testing :echoerr via try/catch is sufficient.
This changes the `textDocument/references' callback to annotate the
locations returned by the server with the content present at the
locations range.
The quickfix list then looks as follows:
cr8/insert_fake_data.py|17 col 32| from .misc import parse_table, parse_version
cr8/insert_fake_data.py|43 col 15| version = parse_version(r['rows'][0][0])
cr8/java_magic.py|8 col 22| from cr8.misc import parse_version
cr8/java_magic.py|30 col 19| version = parse_version(fst)
cr8/java_magic.py|33 col 16| return parse_version(version_str)
Instead of:
cr8/insert_fake_data.py|17 col 32|
cr8/insert_fake_data.py|43 col 15|
cr8/java_magic.py|8 col 22|
cr8/java_magic.py|30 col 19|
cr8/java_magic.py|33 col 16|
haskell-ide-engine sends `hover` payloads as follows:
{
contents = {
kind = "markdown",
value = "\n```haskell\nfoo :: Either String (Integer, Text)\n```\n`foo` *local*"
},
range = {
end = {
character = 5,
line = 57
},
start = {
character = 2,
line = 57
}
}
}
`value` starts with `\n`. The logic in `convert_input_to_markdown_lines`
threw away the whole information.
Allow duplicates so that in languages with overloaded functions it will
show all signatures.
E.g. instead of having a single (last one wins)
add(int index, String element)
It shows all signatures:
add(String e) : boolean
add(int index, String element) : void
fix: #11826
Some lanuguage servers return complementary candidates whose prefixes do not match are also returned.
So we exclude completion candidates whose prefix does not match.
ex) Microsoft python-language-server, rust-analyzer
cpanm cannot look for Perl modules from root directories
without sudo so it creates '~/perl5/' and look for Perl modules in there.
Whether this directory existed before running cpanm or not,
cpanm returns a warning to advice the user to setup local::lib
in order to use modules in '~/perl5/' and exits with error code 0.
Each line in the warning always starts with '!'.
Display this warning to the user.
Continue parsing the version number if the warning can be ignored
because lines that are not prefixed with '!' are valid output.
Fix#11858
cpanm outputs a warning that suggest to use 'sudo' or use local::lib.
cpanm exits with 0 so nvim thinks that the command worked.
cpanm output that starts with "!" is likely an error.
Close#11858
Problem: MS-Windows: Cannot interrupt gdb when program is running.
Solution: Add debugbreak() and use it in the terminal debugger.
Respect 'modified' in a prompt buffer.
4551c0a9fc
Problem: Terminal debugger only works with the terminal feature.
Solution: Make it also work with a prompt buffer. Makes it possible to use
on MS-Windows. Various other improvements. (closesvim/vim#3012)
b3307b5e7e
Use the logic explained in the softtabstop help section for defining
the tabSize parameter in formatting requests. This means that:
- if softtabstop is 0, tabstop is used
- if softtabstop < 0, shiftwidth is used
- if softtabstop > 0, softtabstop is used
When inserting spaces instead of tabs, softtabstop is used in vim.
Therefor it would be more logical to use it when formatting instead
of the current tabstop.
Using `vim.lsp.buf.rename()` can result in receiving a TextEdit that
affects a file for which there is no active or loaded buffer.
In that case `api.nvim_buf_get_lines(...)` returned an empty result,
leading to an error.
Closes https://github.com/neovim/neovim/issues/11790
These had broken encodings, set it to UTF-8. All remianing Neovim
non-ASCII documentation files are UTF-8 encoded. And so are their Vim
original versions.
fix#11753close#11781
The virtualenv troubleshooting in the Python provider health checks is
supposed to help the user determine whether running Python from Neovim
(as in `system('python')` or `system(exepath('python'))`) will use the
correct executable when a virtualenv is active. Currently however, it
issues spurious warnings in legitimate setups, and conversely, fails to
warn about potentially problematic ones.
See https://github.com/neovim/neovim/issues/11753#issuecomment-578715584
for a more detailed analysis, but at a high level, this is due to two
things:
- the virtualenv check is part of the Python provider check defined in
`s:check_python`, which uses a roundabout and sometimes erroneous way of
determining the Python executable
- more generally, it shouldn't be part of the provider check at all,
because it's not really related to the Python *provider*, i.e. the
Python executable which can communicate with Neovim via `pynvim`, but to
the Python the user is editing source files for, which typically
shouldn't even have `pynvim` installed
This patch reimplements the virtualenv check and factors it out into its
own separate function, which is however still kept in
`health/provider.vim` alongside the rest of the Python troubleshooting,
since troubleshooting all Python-related stuff in one place is probably
a good idea in order to alleviate any potential confusion (e.g. users
who run only provider checks might be left wondering whether their
virtualenv Python was properly detected if the report only shows their
global Python as the provider used by Neovim).
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>
The old `:Man` implementation would take either the word under
the cursor, or the argument passed in, and load that as a man page.
Since we now use 'tagfunc' and look for all relevant man-pages, if
your system has several (i.e. same name, different sections), we return
several, giving the user an option.
This works for most tag commands except `:tjump`, which will
fail if there's multiple tags to choose from. This just happens to
be what the cscope code uses (it actually attempts to prompt the
user, but this fails).
Problem: The Normal highlight is not defined when compiled with GUI.
Solution: Always define Normal. (Christian Brabandt, closesvim/vim#4072)
f90b6e03a9
In a multi-window scenario, it is possible to return focus to the last
accessed window via n_CTRL-W_p. However, in the case of a multi-tab
scenario, there was previously no way to return focus to the last
accessed *tab*. Here, that ability is added via n_g<tab>.
Additionally, the index of the previous tab is exposed via
tabpagenr('#'), mirroring the existing functionality of winnr('#').
test_registers.vim can fail even if a clipboard manager is running.
If a clipboard manager is not running, this test always fails with xclip.
Use xsel as a workaround.
https://github.com/astrand/xclip/issues/20 suggests closing stdout
when sending input via stdin.
Environment
- Ubuntu Xenial
- Vim 7.4 (any app with broken clipboard code will do)
- Neovim nightly
Steps to reproduce:
0. Start the clipboard manager.
1. Open a file in Vim on Linux.
Vim should have +clipboard enabled.
'set clipboard='
2. Yank some text to the clipboard register.
3. Quit Vim.
4. Run 'cd /path/to/neovim/repo/'
5. Run 'make oldtest'.
Do not run any individual tests.
They likely pass with or without this fix.
Before fix: test_registers.vim can fail.
After fix: test_registers.vim always passes.
Close https://github.com/neovim/neovim/issues/7958https://wiki.ubuntu.com/ClipboardPersistence#The_state_of_things
Any uses of `has("python3")` will cause the `g:loaded_python3_provider`
variable to be set if the system path does not have a Python with Neovim
support. Subsequent assignments to `g:python3_host_program` will
therefore not cause the provider to be activated.
Reduce API surface. We don't need so many variations of functions. Too
many functions means verbose, largely redundant documentation, tests,
and cognitive burden.
Problem: Not possible to hide a balloon.
Solution: Hide the balloon when balloon_show() is called with an empty
string or list. Add balloon_gettext().
be0a2597ae
Problem: Having 'pumwidth' default to zero has no merit.
Solution: Make the default 15, as the actual default value.
42443c7d7f
Includes 'pumwidth' documentation changes from 8.0.1531.
Sort 'pum*' option in alphabetical order.
Problem: The minimum width of the popup menu is hard coded.
Solution: Add the 'pumwidth' option. (Christian Brabandt, James McCoy,
closesvim/vim#2314)
a8f04aa275
Traditionally, when navigating to a specific location from the middle of
the jumplist results in shifting the current location to the bottom of
the list and adding the new location after it. This behavior is not
desireable to all users--see, for example
https://vi.stackexchange.com/questions/18344/how-to-change-jumplist-behavior.
Here, another jumplist behavior is introduced. When jumpoptions (a new
option set added here) includes stack, the jumplist behaves like the
tagstack or like history in a web browser. That is, when navigating to
a location from the middle of the jumplist
2 first
1 second
0 third <-- current location
1 fourth
2 fifth
to a new location the locations after the current location in the jump
list are discarded
2 first
1 second
0 third
<-- current location
The result is that when moving forward from that location, the new
location will be appended to the jumplist:
3 first
2 second
1 third
0 new
If the new location is the same
new == second
as some previous (but not immediately prior) entry in the jumplist,
2 first
1 second
0 third <-- current location
1 fourth
2 fifth
both occurrences preserved
3 first
2 second
1 third
0 second (new)
when moving forward from that location.
It would be desireable to go farther and, when the new location is the
same as the location that is currently next in the jumplist,
new == fourth
make the result of navigating to the new location by jumping (e.g. 50gg)
be the same as moving forward in the jumplist
2 first
1 second
0 third
1 new <-- current location
2 fifth
and simply increment the jumplist index. That change is NOT part of
this patch because it would require passing the new cursor location to
the function (setpcmark) from all of its callees. That in turn would
require those callees to know *before* calling what the new cursor
location is, which do they do not currently.
Having the cursor change column can be surprising.
Force startofline in functional and old tests.
Remove the functional breakindent test, as it's a subset of the oldtest one.
Having the cursor change column can be surprising.
Force startofline in functional and old tests.
Remove the functional breakindent test, as it's a subset of the oldtest one.
- We already find ourselves renaming nvim_execute_lua in tests and
scripts, which suggests "exec" is the verb we actually want.
- Add "exec" verb to `:help dev-api`.
This addresses a minor quality problem with the recent `'tagfunc'`
changes for `man.vim` (see [link]).
Currently, with the cursor on a parenthese, hitting `K` will jump us to
the man page of the next mentioned entry, instead of the one to which
the parenthese (or section number) belongs.
```
pcrepattern(3), terminfo(5), glob(7), regex(7).
e.g. ^ e.g. ^
```
Adding the parentheses to `'iskeyword'` means we correctly handle these cases too.
[link]: https://github.com/neovim/neovim/pull/11280#discussion_r348342357
These options were previously global. A global-local window option
behaves closer to a global option "per default" (i e with :set),
but still supports local behavior via :setl
Also this restores back-compat for nvim_set_option("fcs", ...)
which are currently broken on 0.4.x but worked in earlier versions
- In the process, refactored focusable_preview to a util function.
- Add text for locations_to_items of the current line.
- Improve location callback to handle multiple return values by using
set_qflist.
- Remove update_tagstack and leave note for future travelers.
Add set_qflist and set_loclist.
- Also add locations_to_items, which calculates byte offsets for
character positions in files and avoids unnecessary operations.
- Add vim variable meta accessors: vim.env, vim.{g,v,w,bo,wo}
- Redo gen_char_blob to generate multiple blobs instead of just one
so that multiple Lua modules can be inlined.
- Reorder vim.lua inclusion so that it can use previously defined C
functions and utility functions like vim.shared and vim.inspect things.
- Inline shared.lua into nvim, but also keep it available in runtime.
* Clear 'cc' in nvim_open_win 'minimal' style #11361
Add 'colorcolumn' to the list of options that should be cleared when creating
a 'minimal'-style floating window.
- Use correct implementation of text_edits.
- Send indent options to rangeFormatting and formatting.
- Remove references to vim bindings and filetype from lsp.txt
- Add more examples to docs.
- Add before_init to allow changing initialize_params.
- Implements textDocument/formatting, textDocument/rangeFormatting,
workspace/applyEdit.
TODO:
- still has edge cases around replacement probably. Only tested with
inserts on the same position.
- Rename builtin_callbacks to default_callbacks and slightly change its
semantics:
- No longer contains the default implementations. Instead, any
default_callbacks will be used in preference for our .buf methods.
- Add this to the docs.
Problem: Using a full path is supported for 'directory' but not for
'backupdir'. (Mikolaj Machowski)
Solution: Support 'backupdir' as well. (Christian Brabandt, closesvim/vim#179)
b782ba475a
* Display ✓ or ✗ based on the line user sees
* Add vim-tutor-mode expectations to the lines marked with an arrow.
* Fix some existing expectations to behave predictably.
Mainly configuration and RPC infrastructure can be considered "done". Specific requests and their callbacks will be improved later (and also served by plugins). There are also some TODO:s for the client itself, like incremental updates.
Co-authored by at-tjdevries and at-h-michael, with many review/suggestion contributions.
We often want to do type checking of public function arguments.
- test: Rename utility_function_spec.lua to vim_spec.lua
- .luacov: Map lua module names
Problem: If spellfile is missing, then "set spell" in modeline/sandbox
fails with a non-obvious error.
Solution: Check for sandbox early and give a meaningful error.
Fixes#11359
Test case:
# test.latex has `% vim: set spelllang=hu:`
# no spell file for `hu` yet!
nvim -u NORC --cmd 'autocmd FileType tex setlocal spell' --cmd 'set modeline' test.latex
Before:
No spell file for "hu" in utf-8
Download it?
Downloading hu.utf-8.spl...
Error detected while processing /usr/local/share/nvim/runtime/autoload/netrw.vim:
line 583:
E12: Command not allowed from exrc/vimrc in current dir or tag search: au WinEnter *^Iif &ft == "netrw"|call s:NetrwInsureWinVars()|endif
Error detected while processing function spellfile#LoadFile[60]..spellfile#Nread[13]..netrw#NetRead[4]..<SNR>67_NetrwOptionsSave:
line 66:
E171: Missing :endif
Error detected while processing function spellfile#LoadFile[60]..spellfile#Nread:
line 13:
E171: Missing :endif
Error detected while processing function spellfile#LoadFile:
line 60:
E171: Missing :endif
Error detected while processing modelines:
line 1:
E12: Command not allowed from exrc/vimrc in current dir or tag search
After:
Error detected while processing function spellfile#LoadFile:
line 5:
E605: Exception not caught: Cannot download spellfile in sandbox/modeline. Try ":set spell" from the cmdline.
Error detected while processing modelines:
line 1:
E12: Command not allowed from exrc/vimrc in current dir or tag search
Problem: Cannot navigate through errors relative to the cursor.
Solution: Add :cabove, :cbelow, :labove and :lbelow. (Yegappan Lakshmanan,
closesvim/vim#4316)
3ff33114d7
sys.path.remove("") raises ValueError if the item is missing.
https://docs.python.org/3/library/functions.html#filter:
> filter(function, iterable) is equivalent to the generator expression (item
> for item in iterable if function(item))
fixes#11293
compared to vim.api.|nvim_call_function|, this fixes some typing issues
due to the indirect conversion via the API. float values are preserved
as such (fixes#9389) as well as empty dicts/arrays.
Ref https://github.com/norcalli/nvim.lua for the call syntax
In order to find if there was already an open man page, the :Man command
would cycle through each window to see if &ft=='man'. This triggers
autocmds, e.g. BufEnter, unnecessarily and can have unexpected
side-effects.
Change the logic to check each window's ft without switching to it
unless it is actually a man window.
Signed-off-by: Joshua Rubin <me@jawa.dev>
Problem: Cannot get the window id associated with the location list.
Solution: Add the "filewinid" argument to getloclist(). (Yegappan
Lakshmanan, closesvim/vim#3202)
c9cc9c78f2
It's a bit cumbersome for us to add an export target every time we define a new function.
It's also cumbersome to care about the order of definition when creating a new function by referring to other functions in the module.
