Commit Graph

229 Commits

Author SHA1 Message Date
L3MON4D3
6f6286e4f9
fix(lsp): small bugs in snippet-parser #18998
This fixes the following bugs:
`${1:else_text}` -> format with if_text: "else_text"
`${1:-else_text}` -> format with if_text: "else_text"
`${1:}` in `format` (eg. empty else_text) -> error.
`${1:}` (eg. empty placeholder) -> error.

Thanks hrsh7th :)
2022-06-29 09:53:49 -07:00
Justin M. Keyes
b6467dfc23 test(report): formatting, drop dumplog()
Don't need to dumplog() on each failed test because we now have test-ids
that associate log messages with tests.
2022-06-15 19:23:10 -07:00
Mathias Fußenegger
e4df1c9b9e
fix(lsp): fix multi client handling in code action (#18869)
Fixes https://github.com/neovim/neovim/issues/18860
2022-06-05 16:43:32 +02:00
Mathias Fußenegger
c6d747e6a5
feat(lsp): send didChangeConfiguration after init (#18847)
Most LSP servers require the notification to correctly load the
settings and for those who don't it doesn't cause any harm.

So far this is done in lspconfig, but with the addition of vim.lsp.start
it should be part of core.
2022-06-03 18:16:11 +02:00
Javier Lopez
d837b6d50c
fix(checkhealth): skip vim.health #18816
Problem:
https://github.com/neovim/neovim/pull/18720#issuecomment-1142614996

The vim.health module is detected as a healthcheck, which produces spurious errors:

    vim: require("vim.health").check()
    ========================================================================
    - ERROR: Failed to run healthcheck for "vim" plugin. Exception:
      function health#check, line 20
      Vim(eval):E5108: Error executing lua [string "luaeval()"]:1: attempt to call field 'check' (a nil value)
      stack traceback:
      [string "luaeval()"]:1: in main chunk

Solution:
Skip vim.health when discovering healthchecks.
2022-06-01 07:10:10 -07:00
bfredl
58b9b7e7c4 fix(ci): remove 2000ms blocking wait in many plugin/lsp_spec.lua tests
Cuts down typical run time for `plugin/lsp_spec.lua`
from 70 secs to 12 secs in ASAN CI build.

This happens in ASAN/EXIT_FREE builds where nvim waits 2000ms due to
unclosed handled. I wasn't able to pin-point the exact cause.
But these tests ran in nested context where two server/client pairs
were setup for no good reason. Moving these tests out so only one client
is being setup fixed the exit hang.
2022-05-22 14:14:58 +02:00
Gregory Anders
2ffafc7aa9 feat(lsp): add LspAttach and LspDetach autocommands
The current approach of using `on_attach` callbacks for configuring
buffers for LSP is suboptimal:

1. It does not use the standard Nvim interface for driving and hooking
   into events (i.e. autocommands)
2. There is no way for "third parties" (e.g. plugins) to hook into the
   event. This means that *all* buffer configuration must go into the
   user-supplied on_attach callback. This also makes it impossible for
   these configurations to be modular, since it all must happen in the
   same place.
3. There is currently no way to do something when a client detaches from
   a buffer (there is no `on_detach` callback).

The solution is to use the traditional method of event handling in Nvim:
autocommands. When a LSP client is attached to a buffer, fire a
`LspAttach`. Likewise, when a client detaches from a buffer fire a
`LspDetach` event.

This enables plugins to easily add LSP-specific configuration to buffers
as well as enabling users to make their own configurations more modular
(e.g. by creating multiple LspAttach autocommands that each do
something unique).
2022-05-18 11:21:00 -06:00
Arsham Shirvani
f6ba7d69be
fix(man.vim): q in "$MANPAGER mode" does not quit #18443
Problem:
q in "$MANPAGER mode" does not quit Nvim. This is because
ftplugin/man.vim creates its own mapping:
    nnoremap <silent> <buffer> <nowait> q :lclose<CR><C-W>c
which overrides the one set by the autoload file when using :Man!
("$MANPAGER mode")

Solution:
Set b:pager during "$MANPAGER mode" so that ftplugin/man.vim can set the
mapping correctly.

Fixes #18281
Ref #17791

Helped-by: Gregory Anders <8965202+gpanders@users.noreply.github.com>
2022-05-13 07:49:08 -07:00
Fredrik Ekre
a9d25e9472
fix(lsp): perform client side filtering of code actions (#18392)
Implement filtering of actions based on the kind when passing the 'only'
parameter to code_action(). Action kinds are hierachical with a '.' as
the separator, and the filter thus allows, for example, both 'quickfix'
and 'quickfix.foo' when requestiong only 'quickfix'.

Fix https://github.com/neovim/neovim/pull/18221#issuecomment-1110179121
2022-05-12 18:48:02 +02:00
Mathias Fußenegger
55187de115
fix(lsp): fix rename capability checks and multi client support (#18441)
Adds filter and id options to filter the client to use for rename.
Similar to the recently added `format` function.

rename will use all matching clients one after another and can handle a
mix of prepareRename/rename support. Also ensures the right
`offset_encoding` is used for the `make_position_params` calls
2022-05-05 23:56:00 +02:00
William Boman
94eb72cc44
fix(lsp): make sure to always reset active codelens refreshes (#18331)
This fixes issues where subsequent calls to vim.lsp.codelens.refresh()
would have no effect due to the buffer not getting cleared from the
active_refresh table.

Examples of how such scenarios would occur are:
  - A textDocument/codeLens result yielded an error.
  - The 'textDocument/codeLens' handler was overriden in such a way that
    it no longer called vim.lsp.codelens.on_codelens().
2022-05-05 18:50:12 +02:00
Mathias Fußenegger
0344736aa6
fix(lsp): handle textDocumentSync.save bool capability (#18332)
Follow up to https://github.com/neovim/neovim/pull/17814
2022-04-30 22:13:26 +02:00
Mathias Fußenegger
88411613e2
feat(lsp): add async option to vim.lsp.buf.format (#18322)
Deprecates the existing `vim.lsp.buf.formatting` function.
With this, `vim.lsp.buf.format` will replace all three:

- vim.lsp.buf.formatting
- vim.lsp.buf.formatting_sync
- vim.lsp.buf.formatting_seq_sync
2022-04-30 17:23:50 +02:00
Michael Lingelbach
5b04e46d23
feat(lsp): add vim.lsp.buf.format (#18193) 2022-04-30 15:36:40 +02:00
Michael Lingelbach
c618b314c6
chore(lsp): remove capabilities sanitization (#17814)
* feat(lsp)!: remove capabilities sanitization

Users must now access client.server_capabilities which matches the same
structure as the protocol.

https://microsoft.github.io/language-server-protocol/specification

client.resolved_capabilities is no longer used to gate capabilities, and
will be removed in a future release.

BREAKING CHANGE


Co-authored-by: Mathias Fussenegger <f.mathias@zignar.net>
2022-04-30 11:22:30 +02:00
Fredrik Ekre
df09e03cf7
feat(lsp): options to filter and auto-apply code actions (#18221)
Implement two new options to vim.lsp.buf.code_action():
 - filter (function): predicate taking an Action as input, and returning
   a boolean.
 - apply (boolean): when set to true, and there is just one remaining
   action (after filtering), the action is applied without user query.

These options can, for example, be used to filter out, and automatically
apply, the action indicated by the server to be preferred:

    vim.lsp.buf.code_action({
        filter = function(action)
            return action.isPreferred
        end,
        apply = true,
    })

Fix #17514.
2022-04-30 10:14:31 +02:00
ii14
55135cea61
fix(lsp): fix unnecessary buffers being added on empty diagnostics (#18275)
Some language servers send empty `textDocument/publishDiagnostics`
messages after indexing the project with URIs corresponding to unopened buffers.
This commit guards against opening buffers corresponding to empty diagnostics.
2022-04-26 10:00:28 -07:00
Fredrik Ekre
6160973f36
fix(lsp): fix lookup of boolean values in workspace/configuration (#18026) 2022-04-15 11:12:41 +02:00
dundargoc
61205c1def
chore: fix typos (#17755)
Co-authored-by: Jordan Haine <jhaine@securitycompass.com>
2022-03-25 19:57:59 +01:00
Tim Pope
af427dedf6
fix(lsp): set tabSize from 'shiftwidth', not 'softtabstop' (#17787)
The use of 'softtabstop' to set tabSize was introduced in 5d5b068,
replacing 'tabstop'.  If we look past the name tabSize and at the actual
purpose of the field, it's the indentation width used when formatting.
This corresponds to the Vim option 'shiftwidth', not 'softtabstop'.
The latter has the comparatively mundane purpose of controlling what
happens when you hit the tab key (and even this is incomplete, as it
fails to account for 'smarttab').
2022-03-20 10:41:46 -07:00
Michael Lingelbach
3800615da9
fix(lsp): handle insertion of previous line (#17618) 2022-03-06 07:52:11 -08:00
Michael Lingelbach
a5e475fcc2
fix(lsp): start incremental sync range at previous newline character (#17610)
This change forces the start of an incremental sync range to begin always on an existing line.
2022-03-05 09:17:56 -08:00
David Shen
5d6006f9bf
feat(diagnostic): add "code" to the diagnostic structure (#17510) 2022-03-02 18:42:27 -07:00
TJ DeVries
991e472881 feat(lua): add api and lua autocmds 2022-02-27 22:04:55 +01:00
Edmund Cape
300b009f47 fix(healthcheck): handle empty reports 2022-02-09 10:18:22 +08:00
zeertzjq
a87ecf5d08 fix(health): do not run external processes in a shell 2022-02-03 18:38:37 +08:00
Michael Lingelbach
bc722c8a74
fix(lsp): strictly enforce passing offset encoding (#17049)
This removes the "fallback" to utf-16 in many of our helper functions. We
should always explicitly pass these around when possible except in two
locations:

* generating params with help utilities called by buf.lua functions
* the buf.lua functions themselves

Anything that is called by the handler should be passed the offset encoding.
2022-01-13 02:34:04 -08:00
Mathias Fußenegger
e7cd811567
fix(lsp): handle negative activeSignature in signatureHelp (#17064)
omnisharp-roslyn can send negative values:

    {
      activeParameter = 0,
      activeSignature = -1,
      signatures = { {
          documentation = "",
          label = "TestEntity.TestEntity()",
          parameters = {}
        } }
    }

In 3.16 of the specification `activeSignature` is defined as `uinteger`
and therefore negative values shouldn't be allowed, but within 3.15 it
was defined as `number` which makes me think we can be a bit lenient in
this case and handle them.

The expected behavior is quite clear:

    The active signature. If omitted or the value lies outside the
    range of `signatures` the value defaults to zero or is ignored if
    the `SignatureHelp` has no signatures.

Fixes an error:

    util.lua:1685: attempt to get length of local 'lines' (a nil value)
    util.lua:1685: in function 'trim_empty_lines'
    handlers.lua:334: in function 'textDocument/signatureHelp'
2022-01-13 01:47:36 -08:00
hrsh7th
43ef7df22d
fix(lsp): fix applying multiple out-of-range TextEdits (#17037) 2022-01-13 10:28:13 +01:00
Mathias Fußenegger
074b033e7e
refactor(lsp): debounce timer per buf and unify with non-debounce (#17016)
Part of the `pending_change` closure in the `changetracking.prepare` was
a bit confusing because it has access to `bufnr` and `uri` but it could
actually contain pending changes batched for multiple buffers.

(We accounted for that by grouping `pending_changes` by a `uri`, but
it's not obvious what's going on)

This commit changes the approach to do everything per buffer to avoid
any ambiguity.

It also brings the debounce/no-debounce a bit closer together: The
only difference is now whether a timer is used or if it is triggered
immediately
2022-01-11 18:10:29 +01:00
Michael Lingelbach
55a59e56ed
feat(lsp): enable default debounce of 150 ms (#16908) 2022-01-05 08:36:35 -08:00
Michael Lingelbach
1b04da52b3
feat(lsp): add buf_detach_client (#16250)
This allows the user to detach an active buffer from the language
client. If no clients remain attached to a buffer, the on_lines callback
is used to cancel nvim_buf_attach.
2021-12-21 10:53:34 -08:00
matveyt
734fba0d88
feat(runtime): new checkhealth filetype (#16660) 2021-12-18 09:14:42 -07:00
Rishikesh Vaishnav
36c401db24
fix(lsp): correctly align start and end range to codepoints during incremental sync (#16670)
Closes #16624

Fixes two issues with aligning the start position and end position to
codepoints when calculating the start and end range.

When aligning the start position:
* use aligned byte index to calculate character index rather than 
  the unadjusted byte

When aligning the end position:
* do not adjust the end byte if it falls on a UTF-8 codepoint
* align byte to the first byte of the next codepoint rather than the
  last byte of the current codepoint
* compute character character end range on the aligned byte index

This commit also adds additional test coverage, including multibyte operations
that previously failed before this commit.
2021-12-17 18:05:00 -08:00
Michael Lingelbach
a2749482d9
chore(lsp): clean up initialization process (#16369)
* send vim.NIL instead of not sending workspaceFolders
* read fallback rootPath and rootUri from workspaceFolders
* update documentation
2021-11-21 11:39:30 -05:00
Michael Lingelbach
f71be1f87b
fix(lsp): improve incremental sync robustness (#16358)
closes https://github.com/neovim/neovim/issues/16352

* improve handling of multi-byte deletions
2021-11-18 16:49:46 -05:00
Michael Lingelbach
87a053f126
fix(lsp): fix edge cases in incremental sync (#16308) 2021-11-15 08:51:30 -08:00
Mathias Fußenegger
ee3a58d42e
fix(lsp): ensure buffers are re-attached on rename (#16266)
If a LSP server sent a workspace edit containing a rename the buffers
file name changed without the server receiving a close notification for
the old buffer and without the client properly re-attaching on the new
file.

This affected `Move` code-actions in nvim-jdtls, but also
`vim.lsp.buf.rename` on a class level.
2021-11-14 12:55:16 +01:00
Michael Lingelbach
2ecf0a4c61
fix(lsp): rewrite incremental sync (#16252)
* use codeunits/points instead of byte ranges when applicable
* take into account different file formats when computing range and
  sending text (dos, unix, and mac supported)
* add tests of incremental sync
2021-11-09 14:37:48 -08:00
Sebastian Lyng Johansen
16d4af6d2f
feat(ui): add vim.ui.input and use in lsp rename (#15959)
* vim.ui.input is an overridable function that prompts for user input
* take an opts table and the `on_confirm` callback, see `:help vim.ui.input` for more details
* defaults to a wrapper around vim.fn.input(opts)
* switches the built-in client's rename handler to use vim.ui.input by default
2021-11-07 07:13:53 -08:00
Gregory Anders
03b805aee6
feat(lua): enable stack traces in error output (#16228) 2021-11-06 08:26:10 -06:00
Michael Lingelbach
519d8deb08
feat(lsp): add per-client commands (#16101) 2021-11-01 03:14:59 -07:00
Gregory Anders
e921e98ce3
refactor(diagnostic): make display handlers generic (#16137)
Rather than treating virtual_text, signs, and underline specially,
introduce the concept of generic "handlers", of which those three are
simply the defaults bundled with Nvim. Handlers are called in
`vim.diagnostic.show()` and `vim.diagnostic.hide()` and are used to
handle how diagnostics are displayed.
2021-10-29 18:47:34 -07:00
jdrouhard
d1c470957b
feat(lsp): track pending+cancel requests on client object #15949 2021-10-29 05:45:01 -07:00
hrsh7th
bd2f61c6c4
fix(lsp): fix cursor row after textEdits (#16038) 2021-10-18 11:49:33 -07:00
Christian Clason
fbc5b7b090 chore(test): adapt healthcheck test to new css syntax
The runtime file update
2286304cdb
added a `syn keyword` for `css`, which affects (via `html` and
`markdown` syntax files) the highlighting of `:checkhealth` output
(before, `ERROR:` was highlighted with `healthError`; now the colon is
no longer included).
2021-10-16 19:17:28 +02:00
Michael Lingelbach
d288daac2b
fix(lsp): do not invoke handlers for unsupported methods (#15926)
Closes https://github.com/neovim/neovim/issues/15174

Instead of invoking handlers with unsupported methods, pre-compute which
clients support a given method and only notify the user if no clients
support the given method.
2021-10-10 22:32:50 -07:00
Michael Lingelbach
4f4dbfe81c
fix(lsp): update tests using 0.5.0 handler calls (#15969)
Fixes test regression introduced in https://github.com/neovim/neovim/pull/15262
2021-10-08 14:01:55 -07:00
Rishikesh Vaishnav
3f09732195
fix(lsp): expose ContentModified error code to callbacks (#15262) 2021-10-08 11:19:33 -07:00
Javier Lopez
acd5e831b6
fix(checkhealth): mitigate issues with duplicate healthchecks #15919
* fix(runtime/health): mitigate issues with duplicate healthchecks
  Previously if a healthcheck was found as Lua and Vim it was executed
  both times.
  This new implementations prefers Lua, therefore if two are found It only
  runs the Lua one, this way a plugin can mantain both implementations the
  Lua one with the method `check()` and the autoload function `#check()`
  (for none HEAD nvim versions).
  **Note: This will require plugins to use `check()` as the function name,
  since the autoload function that wraps the lua implementation won't be
  called**
* docs(health): use spaces and don't overuse backtics

followup to #15259
2021-10-05 15:37:39 -07:00