Fixes a bug introduced by https://github.com/neovim/neovim/pull/15949
When no supported clients for a given method are available, buf_request
returns early with a nil value. If buf_request_sync is called on a
buffer with no clients that support a given method, the returned
`cancel` method (which is nil), is invoked, resulting in an error.
Solution: return an empty function handle
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.
* This flag allows customizing the time before sending kill -15 to the
server. If set to false, neovim exits immediately after sending
request('shutdown'). Otherwise, polls until all servers have shutdown,
and then kills remaining servers via kill -15 at exit_timeout
duration. Defaults to 500 ms.
closes https://github.com/neovim/neovim/issues/16058
* add client.attached_buffers
* only update client.attached_buffers in on_attach
* use table instead of list for attached_buffers to avoid duplication
The spacing fix drew attention to a couple of places that were using
incorrect formatting such as the key listing for `nvim_open_win`, so
those were fixed too.
'show_line_diagnostics()' and 'show_position_diagnostics()' are
almost identical; they differ only in the fact that the latter also
accepts a column to form a full position, rather than just a line. This
is not enough to justify two separate interfaces for this common
functionality.
Renaming this to simply 'show_diagnostics()' is one step forward, but
that is also not a good name as the '_diagnostics()' suffix is
redundant. However, we cannot name it simply 'show()' since that
function already exists with entirely different semantics.
Instead, combine these two into a single 'open_float()' function that
handles all of the cases of showing diagnostics in a floating window.
Also add a "float" key to 'vim.diagnostic.config()' to provide global
values of configuration options that can be overridden ephemerally.
This makes the float API consistent with the rest of the diagnostic API.
BREAKING CHANGE
It looks a bit off with the extmark going over the cursorline.
(With hl_mode combine it keeps the background of the cursorline under the codelens virtualtext)
Persist configuration settings set with `vim.lsp.with` and
`vim.lsp.diagnostic.on_publish_diagnostics` by setting the config for
the namespace associated with the client.
When using `true` as the value of a configuration option, the option is
configured to use default values. For example, if a user configures
virtual text to include the source globally (using
vim.diagnostic.config) and a specific namespace or producer configures
virtual text with `virt_text = true`, the user's global configuration is
overriden.
Instead, interpret a value of `true` to mean "use existing settings if
defined, otherwise use defaults".
closes https://github.com/neovim/neovim/issues/15261
* normalize uri path to forward slashes on windows
* use a capture group on windows that avoids mistaking drive letters as uri scheme
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.
This is a simple fix for #15899, as it should at least stop calling
`prepareRename` on servers that don't support renaming.
I imagine a better fix would be to inspect the actual value for, but
that requires some plumbing changes on how capabilities are evaluated
before sending requests out.
Co-authored-by: francisco souza <fsouza@users.noreply.github.com>
Users can pass string values for severities that match with the enum
names (e.g. "Warn" or "Info") which are converted to the corresponding
numerical value in `to_severity`. Invalid strings were simply left
as-is, which caused confusing errors later on. Instead, report an
invalid severity string right up front to make the problem clear.
* Add optional second table argument to vim.json.decode which takes
a table 'luanil' which can include the 'object' and/or 'array' keys. These
options use luanil when converting NULL in json objects and arrays
respectively. The default behavior matches the original lua-cjson.
* Remove recursive_convert_NIL function from rpc.lua, use
vim.json.decode with luanil = { object = true } instead. This removes a hotpath
in the json deserialization pipeline by dropping keys with json NULL
values throughout the deserialized table.
- Refactor health.vim to discover lua healthcheck in the runtime
directories lua/**/health{/init}.lua
- Support healthchecks for lua submodules e.g :checkhealth vim.lsp and
also support wildcard "*" at the end for all submodules
:checkhealth vim*
- Refactor health.vim to use variable scope instead of output capturing
- Create health.lua module to wrap report functions and future
extensibility.
- Move away from searching just in the runtimepath, use
`nvim_get_runtime_file` due to #15632
Example:
Plugin linter in rtp can declare it's checkhealts in lua module
`lua/linter/health{/init}.lua` that returns a table with a method
"check" that when executed calls the report functions provided by the
builtin lua module require("health").
The plugin also has a submodule `/lua/linter/providers` in which it
defines `/lua/linter/providers/health{/init}.lua`
This plugin healthcheck can now be run by the ex command:
`:checkhealth linter linter.providers`
Also calling all submodules can be done by:
`:checkhealth linter*
And "linter" and "linter.provider" would be discovered when:
`:checkhealth`
Always make calls to `vim.diagnostic.set` call `vim.diagnostic.show`.
This creates an easier to reason about code path and is also less
surprising when users wish to override override `vim.diagnostic.show`
with custom behavior and `vim.diagnostic.set` is called with empty
diagnostics.
Functionally, the end result is the same: when `show` is called with an
empty diagnostics list, it just calls `hide` and then returns, which is
exactly what `reset` does right now.
N, W, S, E are all inclusive, i.e., always anchor to the exact corner of the
window (including border). This line may also need change in this case (change
0 to -1):
This is most consistent and easiest to reason about, especially with GUIs whose
border do not need to have width/height of 1/1 in cell units.
Fix#15789
Many vim.diagnostic functions expect the user to pass in a namespace id.
This PR allows the user to list active diagnostic namespaces:
```lua
:lua print(vim.inspect(vim.diagnostic.get_namespaces()))
{
[7] = {
name = "vim.lsp.client-1",
opts = {},
sign_group = "vim.diagnostic.vim.lsp.client-1"
}
}
```
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.
In vim.lsp.buf.references, the key vim.type_idx (which evaluates to a
boolean) was set to equal vim.types.dictionary. This resulted in a
boolean key in json which is not allowed by the json spec, and which
lua-cjson fails to serialize.
Rather than relying on the order in which signs are placed to dictate
the order in which they are displayed, explicitly set the priority of
the sign according to the severity of the diagnostic and the value of
severity_sort. If severity_sort is false or unset then all signs use the
same priority.
The `split()` VimL function trims empty items from the returned list by
default, so that, e.g.
split("\nhello\nworld\n\n", "\n")
returns
["hello", "world"]
The Lua implementation of vim.split does not do this. For example,
vim.split("\nhello\nworld\n\n", "\n")
returns
{'', 'hello', 'world', '', ''}
Add an optional parameter to the vim.split function that, when true,
trims these empty elements from the front and back of the returned
table. This is only possible for vim.split and not vim.gsplit; because
vim.gsplit is an iterator, there is no way for it to know if the current
item is the last non-empty item.
Note that in order to preserve backward compatibility, the parameter for
the Lua vim.split function is `trimempty`, while the VimL function uses
`keepempty` (i.e. they are opposites). This means there is a disconnect
between these two functions that may surprise users.
Problem:
Error executing vim.schedule lua callback: ...ovim/HEAD-aba3979/share/nvim/runtime/lua/vim/lsp/buf.lua:502: command: expected string, got
nil
stack traceback:
...ovim/HEAD-aba3979/share/nvim/runtime/lua/vim/lsp/buf.lua:502: in function 'execute_command'
...HEAD-aba3979/share/nvim/runtime/lua/vim/lsp/handlers.lua:151: in function <...HEAD-aba3979/share/nvim/runtime/lua/vim/lsp/handlers.lua:113>
...ovim/HEAD-aba3979/share/nvim/runtime/lua/vim/lsp/buf.lua:465: in function 'callback'
...r/neovim/HEAD-aba3979/share/nvim/runtime/lua/vim/lsp.lua:1325: in function 'handler'
...r/neovim/HEAD-aba3979/share/nvim/runtime/lua/vim/lsp.lua:899: in function 'cb'
vim.lua:281: in function <vim.lua:281>
Solution:
This is a follow-up to the work done in
6c03601e3a.
There are valid situations where a `textDocument/codeAction` is returned
without a command, since a command in optional. For example from Metals,
the Scala language server when you get a code action to add a missing
import, it looks like this:
```json
Result: [
{
"title": "Import \u0027Instant\u0027 from package \u0027java.time\u0027",
"kind": "quickfix",
"diagnostics": [
{
"range": {
"start": {
"line": 6,
"character": 10
},
"end": {
"line": 6,
"character": 17
}
},
"severity": 1,
"source": "bloop",
"message": "not found: value Instant"
}
],
"edit": {
"changes": {
"file:///Users/ckipp/Documents/scala-workspace/sanity/src/main/scala/Thing.scala": [
{
"range": {
"start": {
"line": 6,
"character": 10
},
"end": {
"line": 6,
"character": 17
}
},
"newText": "Instant"
},
{
"range": {
"start": {
"line": 1,
"character": 0
},
"end": {
"line": 1,
"character": 0
}
},
"newText": "\nimport java.time.Instant\n"
}
]
}
}
}
]
```
This change just wraps the logic that grabs the command in a conditional
to skip it if there is no command.
diagnostic_lines() returns a table, so make the early exit condition an
empty table rather than 'nil'. This way, functions that use the input
from diagnostic_lines don't have to do a bunch of defensive nil checking
and can always assume they're operating on a table.
This function isn't compatible with including diagnostic sources when
"source" is "if_many" since it only has access to diagnostics for a
single line. Rather than having an inconsistent or incomplete interface,
make this function private. It is still exported as part of the module
for backward compatibility with vim.lsp.diagnostics, but it can
eventually be made into a local function.
* preserve fields from LSP diagnostics via adding a user_data table to the diagnostic, which can hold arbitrary data in addition to the lsp diagnostic information.
This fixes the handler signature and also prevents n+1 requests firing
if there are multiple clients.
(The first `prepareCallHierarchy` handler is called once per client,
each invocation used `buf_request` to make more requests using *all*
clients)
This is mostly motivated by https://github.com/neovim/neovim/issues/12326
Client side commands might need to access the original request
parameters.
Currently this is already possible by using closures with
`vim.lsp.buf_request`, but the global handlers so far couldn't access
the request parameters.
Some parts of LSP need to use cached diagnostics as sent from the LSP
server unmodified. Rather than fixing invalid line numbers when
diagnostics are first set, fix them when they are displayed to the user
(e.g. in show() or one of the get_next/get_prev family of functions).
* feat(diagnostic): add vim.diagnostic.match()
Provide vim.diagnostic.match() to generate a diagnostic from a string and
a Lua pattern.
* feat(diagnostic): add tolist() and fromlist()
When vim.diagnostic.config() is called, the decorations for diagnostics
are re-displayed to use the new configuration. This should only be done
for loaded buffers.
When severity_sort is true, higher severities should be displayed before
lower severities (e.g. ERROR is displayed over WARN).
Also improved the test case for this.
The recursive implementation of vim.lsp.diagnostic.get() applied
`diagnostic_vim_to_lsp` twice, and the second time gave wrong
results because of the unexpected format.
Fixes https://github.com/neovim/neovim/issues/15689
These links were actually defined backwards: the highlight groups
actually being used for display are the new "Diagnostic*" groups, so
linking the old "LspDiagnostics*" groups to these does absolutely
nothing, since there is nothing actually being highlighted with the
LspDiagnostics* groups.
These links were made in an attempt to preserve backward compatibility
with existing colorschemes. We could reverse the links to maintain this
preservation, but then that disallows us from actually defining default
values for the new highlight groups.
Instead, just remove the links and be done with the old LspDiagnostics*
highlight groups.
This is not technically a breaking change: the breaking change already
happened in #15585, but this PR just makes that explicit.
## Overview
- Move vim.lsp.diagnostic to vim.diagnostic
- Refactor client ids to diagnostic namespaces
- Update tests
- Write/update documentation and function signatures
Currently, non-LSP diagnostics in Neovim must hook into the LSP subsystem. This
is what e.g. null-ls and nvim-lint do. This is necessary because none of the
diagnostic API is exposed separately from the LSP subsystem.
This commit addresses this by generalizing the diagnostic subsystem beyond the
scope of LSP. The `vim.lsp.diagnostic` module is now simply a specific
diagnostic producer and primarily maintains the interface between LSP clients
and the broader diagnostic API.
The current diagnostic API uses "client ids" which only makes sense in the
context of LSP. We replace "client ids" with standard API namespaces generated
from `nvim_create_namespace`.
This PR is *mostly* backward compatible (so long as plugins are only using the
publicly documented API): LSP diagnostics will continue to work as usual, as
will pseudo-LSP clients like null-ls and nvim-lint. However, the latter can now
use the new interface, which looks something like this:
```lua
-- The namespace *must* be given a name. Anonymous namespaces will not work with diagnostics
local ns = vim.api.nvim_create_namespace("foo")
-- Generate diagnostics
local diagnostics = generate_diagnostics()
-- Set diagnostics for the current buffer
vim.diagnostic.set(ns, diagnostics, bufnr)
```
Some public facing API utility methods were removed and internalized directly in `vim.diagnostic`:
* `vim.lsp.util.diagnostics_to_items`
## API Design
`vim.diagnostic` contains most of the same API as `vim.lsp.diagnostic` with
`client_id` simply replaced with `namespace`, with some differences:
* Generally speaking, functions that modify or add diagnostics require a namespace as their first argument, e.g.
```lua
vim.diagnostic.set({namespace}, {bufnr}, {diagnostics}[, {opts}])
```
while functions that read or query diagnostics do not (although in many cases one may be supplied optionally):
```lua
vim.diagnostic.get({bufnr}[, {namespace}])
```
* We use our own severity levels to decouple `vim.diagnostic` from LSP. These
are designed similarly to `vim.log.levels` and currently include:
```lua
vim.diagnostic.severity.ERROR
vim.diagnostic.severity.WARN
vim.diagnostic.severity.INFO
vim.diagnostic.severity.HINT
```
In practice, these match the LSP diagnostic severity levels exactly, but we
should treat this as an interface and not assume that they are the same. The
"translation" between the two severity types is handled transparently in
`vim.lsp.diagnostic`.
* The actual "diagnostic" data structure is: (**EDIT:** Updated 2021-09-09):
```lua
{
lnum = <number>,
col = <number>,
end_lnum = <number>,
end_col = <number>,
severity = <vim.diagnostic.severity>,
message = <string>
}
```
This differs from the LSP definition of a diagnostic, so we transform them in
the handler functions in vim.lsp.diagnostic.
## Configuration
The `vim.lsp.with` paradigm still works for configuring how LSP diagnostics are
displayed, but this is a specific use-case for the `publishDiagnostics` handler.
Configuration with `vim.diagnostic` is instead done with the
`vim.diagnostic.config` function:
```lua
vim.diagnostic.config({
virtual_text = true,
signs = false,
underline = true,
update_in_insert = true,
severity_sort = false,
}[, namespace])
```
(or alternatively passed directly to `set()` or `show()`.)
When the `namespace` argument is `nil`, settings are set globally (i.e. for
*all* diagnostic namespaces). This is what user's will typically use for their
local configuration. Diagnostic producers can also set configuration options for
their specific namespace, although this is generally discouraged in order to
respect the user's global settings. All of the values in the table passed to
`vim.diagnostic.config()` are resolved in the same way that they are in
`on_publish_diagnostics`; that is, the value can be a boolean, a table, or
a function:
```lua
vim.diagnostic.config({
virtual_text = function(namespace, bufnr)
-- Only enable virtual text in buffer 3
return bufnr == 3
end,
})
```
## Misc Notes
* `vim.diagnostic` currently depends on `vim.lsp.util` for floating window
previews. I think this is okay for now, although ideally we'd want to decouple
these completely.
This generalizes diagnostic handling outside of just the scope of LSP.
LSP clients are now a specific case of a diagnostic producer, but the
diagnostic subsystem is decoupled from the LSP subsystem (or will be,
eventually).
More discussion at [1].
[1]: https://github.com/neovim/neovim/pull/15585
* Simplify rpc encode/decode messages to rpc.send/rcp.receive
* Make missing handlers message throw a warning
* Clean up formatting style in log
* Move all non-RPC loop messages to trace instead of debug
* Add format func option to log to allow newlines in per log entry
Previously, the handler signature was:
function(err, method, params, client_id, bufnr, config)
In order to better support external plugins that wish to extend the
protocol, there is other information which would be advantageous to
forward to the client, such as the original params of the request that
generated the callback.
In order to do this, we would need to break symmetry of the handlers, to
add an additional "params" as the 7th argument.
Instead, this PR changes the signature of the handlers to:
function(err, result, ctx, config)
where ctx (the context) includes params, client_id, and bufnr. This also leaves
flexibility for future use-cases.
BREAKING_CHANGE: changes the signature of the built-in client handlers, requiring
updating handler calls
Resolve an issue with deferred clearing of highlight failing if the
buffer is deleted before the timeout by checking whether the
buffer is valid first.
Declaration, type-definition, and implementation capabilities were
previously disabled if the client received table output from the server
capabilities. The workDoneProgress capability is sent for many servers
for all supported capabilities as part of this table. Default to setting
capability to table instead of false.
The official developer documentation in in :h dev-lua-doc specifies to
use "--@" for special/magic tokens. However, this format is not
consistent with EmmyLua notation (used by some Lua language servers) nor
with the C version of the magic docstring tokens which use three comment
characters.
Further, the code base is currently split between usage of "--@",
"---@", and "--- @". In an effort to remain consistent, change all Lua
magic tokens to use "---@" and update the developer documentation
accordingly.
According to the protocol definition `rootPath`, `rootUri` and
`workspaceFolders` are allowed to be null.
Some language servers utilize this to provide "single file" support.
If all three are null, they don't attempt to index a directory but
instead only provide capabilities for a single file.
This changes the behavior of the hl_cache to the old one.
- when the capture exists as a hlgroup -> use it
- when hl_map contains a mapping -> use it
- else do nothing (before: map capture to non-existing capture)
Before also captures `@foo.bar` would intend to use the hlgroup `foo.bar`
which results in a confusing error since hlgroups can't contain dots.
Add a new function to redraw diagnostics from the current diagnostic
cache, without receiving a "publishDiagnostics" message from the server.
This is already being done in two places in the Lua stdlib, so this
function unifies that functionality in addition to providing it to third
party plugins.
An example use case for this could be a command or key-binding for
toggling diagnostics virtual text. The virtual text configuration option
can be toggled using `vim.lsp.with` followed by
`vim.lsp.diagnostic.redraw()` to immediately redraw the diagnostics
with the updated setting.
The handlers for textDocument/references, textDocument/documentSymbol,
and workspace/symbol open their results in the quickfix list by default
and are not configurable. They are also incompatible with `vim.lsp.with`
as they do not accept a configuration parameter.
Add a `config` parameter to the handler for these three messages which
allows them to be configured with `vim.lsp.with`. Additionally, add a
new configuration option 'loclist' that, when true, causes these
handlers to open their results in the location list rather than the
quickfix list.
Some language servers *cough*rust-analyzer*cough* need an empty/custom
workspaceFolders for certain usecases. For example, rust-analyzer
needs an empty workspaceFolders table for standalone file support
(See https://github.com/rust-analyzer/rust-analyzer/pull/8955).
This can also be useful for other languages that need to commonly
open a certain directory (like flutter or lua), which would help
prevent spinning up a new language server altogether.
In case no workspaceFolders are passed, we fallback to what we had
before.
Passing `nil` is equivalent to passing 0, i.e. it simply uses the
current buffer number.
This fixes a bug when vim.lsp.diagnostic.disable() is called without
arguments.
Add two new methods to allow diagnostics to be disabled (and re-enabled)
in the current buffer. When diagnostics are disabled they are simply not
displayed to the user, but they are still sent by the server and
processed by the client.
Disabling diagnostics can be helpful in a number of scenarios. For
example, if one is working on a buffer with an overwhelming amount of
diagnostic warnings it can be helpful to simply disable diagnostics
without disabling the LSP client entirely. This also allows users more
flexibility on when and how they may want diagnostic information to be
displayed. For example, some users may not want to display diagnostic
information until after the buffer is first written.
An empty table was previously always treated as a list, which means that
while merging tables, whenever an empty table was encountered it would
always truncate any table on the left.
`vim.tbl_deep_extend("force", { b = { a = 1 } }, { b = {} })`
Before: `{ b = {} }`
After: `{ b = { a = 1 } }`
the `textDocument/rangeFormatting` nad `textDocument/formatting` did not
pass bufnr to apply_text_edits, meaning edits were applied to
the user's currently active buffer. This could result in text being
applied to the wrong buffer.
ipairs terminates on the first nil index when iterating over table keys:
for i,k in ipairs( {[1] = 'test', [3] = 'test'} ) do
print(i, k)
end
prints:
1 test
Instead, use pairs which continues iterating over the entire table:
for i,k in pairs( {[1] = 'test', [3] = 'test'} ) do
print(i, k)
end
prints:
1 test
3 test
`return err_message(tostring(err))` caused errors to be printed as
`table: 0x123456789` instead of showing the error code and error
message.
This also removes some `if err` blocks that never got called because at
the end of `handlers.lua` all the handlers are wrapped with logic that
adds generic error handling.
RFC 8089, which defines the file URI scheme, also allows URIs without a
hostname, i.e. of the form file:/path/to/file. These are returned by
some language servers and accepted by other LSP implementations, such as
VSCode's, so it is reasonable for us to accept them as well.
Adds indentation that matches the number prefix to ensure diagnostic
messages spawning multiple lines align.
Before:
Diagnostics:
1. • Variable not in scope: red :: t0 -> t
• Perhaps you meant one of these:
‘rem’ (imported from Prelude), ‘read’ (imported from Prelude),
‘pred’ (imported from Prelude)
2. • Variable not in scope: repeDoubleColon :: [Char] -> t0
• Perhaps you meant ‘replaceDoubleColon’ (line 32)
After:
Diagnostics:
1. • Variable not in scope: red :: t0 -> t
• Perhaps you meant one of these:
‘rem’ (imported from Prelude), ‘read’ (imported from Prelude),
‘pred’ (imported from Prelude)
2. • Variable not in scope: repeDoubleColon :: [Char] -> t0
• Perhaps you meant ‘replaceDoubleColon’ (line 32)
Options formatted as a list of comma-separated key-value pairs may have
values that contain leading and trailing whitespace characters. For
example, the `listchars` option has a default value of
`"tab:> ,trail:-,nbsp:+"`. When converting this value to a lua table,
leading and trailing whitespace should not be trimmed.
Co-authored-by: Robert Hrusecky <robert.hrusecky@utexas.edu>
This closes#14677, but I also am a little unsure if there are times
where this may not be correct. However, this just changes the behavior
that even if `was_set` was false, we still get for
`nvim_win_get_option`.
The `onexit` handler could set `message_callbacks` to `nil` within the
luv event loop while the mainloop runs a function that tries to access
`message_callbacks`.
This adds some checks to prevent errors in that case.
Fixes https://github.com/neovim/neovim/issues/14863
Add the buffer number to the `textDocument/formatting` request, so
that it is passed to the handler.
The built-in formatting handlers do not use the buffer number, but user
overrides should have access to it.
`lsp.diagnostic.get_all()` was returning diagnotics for `:bwipeout`-ed
buffers because the diagnostic cache is not cleared. The first argument
of on_detach callback is the string "detach", not the bufnr.
Currently, for large number of diagnostics, the delay in populating
loclist may be sufficient for a user to switch to another window,
resulting in the loclist being populated on the wrong window.
In the documentation for `vim.lsp.util.open_floating_preview`
the opts table keys were prefixed with `--` instead of `---`,
preventing capture by docgen.
This pr allows the user to specify whether `lsp.utils.open_floating_preview`
is focusable via the `opts` parameter. Defaults to true.
It can be configured by setting the focusable key inside opts parameter:
```lua
vim.lsp.util.open_floating_preview(contents, syntax, {focusable = false})
```
