* Some abs paths are not recognized for win32
Paths like "c:/abc/def.cpp" were not recognized as absolute on Windows.
* Fix test for mypy_handler
mypy handler not just simplifies path, but tries to convert path to
absolute.
This breaks detecting the project root when editing tests with a nested
pytest.ini file where other project files are available. The other files
are so common we can just removes this entirely and test that we
ignore it in one case.
https://github.com/facebook/pyrefly
The pyre project has evolved to pyrefly. This replaces the pyre linter
with a pyrefly one and removes the test files that were added for
finding the project root in the old pyre world.
Co-authored-by: Oliver Ruben Albertini <oliverruben@gmail.com>
Some linters will expand the paths of symlinked tempfiles when
reporting them back to ALE. This ensures that if they do, the filename
is still flagged appropriately as being temporary.
File local variables in Emacs are used in a way similar to Vim
modelines. For example, at the end of the file you might find something
like the following:
%% Local Variables:
%% erlang-indent-level: 2
%% something-weird: t
%% End:
The `erlang-indent-level' variable in this list instructs the Erlang
mode to use two columns per indentation level. But since the
`something-weird' variable is likely unknown, both may be ignored.
By default, Emacs in batch mode ignores all variable/value pairs if it
encounters at least one that is not known to be safe. Setting
`enable-local-variables' to `:safe' tells Emacs to use only safe values
and ignore the rest.
Version 2 of golangci-lint introduces several breaking changes to the
command line arguments and the configuration file.
Thi PR updates ale integration to support both version 1.6x.x and 2.x.x
of golangci-lint.
The code is already passed with the code key of location list entry.
Before this change, when using the default message format, it appeared
twice in the description of each location list entry and twice in each
echo message.
Implement the diagnostics pull model with the LSP Neovim client. We
must handle messages a little different and tweak client capabilities
for pull diagnostics to work through the Neovim client.
Implement pull diagnostics in the VimL implementation so ALE is able
to track when servers are busy checking files. Only servers that
support this feature will return diagnostics these ways.
Ensure that basic ALE functions `ale.var`, `ale.escape`, and `ale.env`
are available in Lua. Cover all Lua code so far with busted tests,
fixing bugs where ALE variables can be set with Boolean values instead
of numbers. Document all functionality so far.
1. Add ale.setup and ale.setup.buffer for pure Lua configuration.
2. Update many global settings to use Booleans instead of numbers to
make types easiert to work with in Lua.
3. Radically reformat documentation and fix errors to make
documentation more usable for Neovim users.
Support TCP connections to language servers through Neovim's built in
client. In all but what is currently the nightly builds of Neovim
connections via a hostname will fail, but connections via an IP address
should function. We will still enable the built in Neovim client by
default anyway, as LSP clients very rarely connect over TCP.
Change logic so ALE's LSP implementation and the Neovim LSP client
retrieve the language_id for language clients at roughly the same time
via the same means. This makes ALE inform the language server what the
language for the language is for clients.
Update documentation to advertise ALE's integration with Neovim's native
LSP client, and explain how functionality is integrated with ALE,
Neovim's native tools, and other plugins.
Now we ought to be able to handle any kind of response for any request
we send, clean up message handling so there are fewer changes to make
to LSP code to adjust for Neovim integration.
Save capabilities from language servers ALE connects to via Neovim's LSP
client and handle responses to requests ALE sends to language servers.
This enables ALE to work with Neovim's language client for its commands
while also letting users directly use the connected clients for native
Neovim keybinds and so on.
Get language servers starting and displaying diagnostics with Neovim's
API in Neovim 0.8 and up. With this set up, now ALE needs to take over
handling diagnostics returned by the language servers.
* feat: Add Zig zlint linter and handler for ALE
* docs: Add zlint documentation to ALE Zig integration guide
* docs: Updating docs for zlint support
* tests: Adding tests for checking zlint executable and command
* refactor: Move zlint configuration test to separate test file
* Refactor djlint linter code
This patch moves the code to the `autoload` directory, so it's available
when it's needed by a specific linter. This avoids redundant code when
another format supported by djlint is added.
* Add linting support for all formats supported by djlint
So far, the `djlint` linter in ALE only supported `html`, which is only
one of several file types supported by `djlint`.
This patch adds support for the following file types:
* gohtmltmpl
* handlebars
* htmlangular
* htmldjango
* jinja
* nunjucks
* Add djlint fixer for various HTML template formats
* Supported formats:
- html
- htmlangular
- htmldjango
- jinja
- handlebars
- nunjucks
- gohtmltmpl
* Add doc entries
* Add vader tests
---------
Co-authored-by: Adrian Vollmer <computerfluesterer@protonmail.com>
This adds Yara support to ALE using Avast's language server for Yara:
https://avast.github.io/yls/
A ".git" folder is used to determine the project_root so this means the
Yara rules must be in a git repo for the integration to work.
The server only have 1 optional argument (-v, --verbose). Since this is
the case, no additional configuration options are available.
---------
Co-authored-by: w0rp <w0rp@users.noreply.github.com>
`ModeChanged` looks like a more reliable way to detect an "exit insert mode" event and is a lot simpler (doesn't need a timer). Also, it can detect some other transitions like `\<C-o\>` in insert mode.
The `ModeChanged` event is available in:
* [Vim 8.2.3430](f1e8876fa2)
* [NeoVim 0.7.0](69bd1e4e36)
---------
Co-authored-by: Dmitry Zolotukhin <zlogic@gmail.com>
* Add support for the [djlinter](https://www.djlint.com/)
* Add documentation and tests.
* Fix the name of the variable for the executable name.
* Correct the name of the handler in the test.
* Correct the test adding the value of vcol.
* Format djlint.vim according to formatting rules.
Sometimes `s:HandleExit` can execute a deferred linter callback, which
ends up setting the `l:loclist` that's passed into
`ale#engine#HandleLoclist` at the end of `s:HandleExit` to a dictionary.
This dictionary cannot be iterated over, and thus errors out.
Guard against trying to iterate over values that don't make sense.
Co-authored-by: Alexander Huynh <git@e.sc>
* Fixed the issue with Black ignoring files is being processed. (#3406)
Add test for stdin-filename on test/fixers/test_ruff_format_fixer_callback.vader
* Fixed the issue with Black ignoring files is being processed. (#3406)
Fixed the problem on Windows's tests.
* Fixed the issue with Black ignoring config file to tell it which file is being processed. Trailing whitespace removed
Resolves#4314.
Add a fixer that's built into python for json formatting. Include a
couple arguments in docs to make these features more discoverable.
Uses stdin-based fixing so you don't need to save the file to fix.
In the vein of commit ea72d66b "Verilator current file search path (#3500)"
This includes the directory of the current file in the library
search path. From `man iverilog`:
-ylibdir
Append the directory to the library module search
path. When the compiler finds an undefined module, it
looks in these directories for files with the right name.
This might only be a problem for newer phpstan versions (2.1.1 here).
If you try to run `phpstan` the way ale will when it builds the option, you will get something like:
```
The "--memory-limit" option requires a value.
```
It wants you to use `--memory-limit=-1` instead.
The current xmllint fixer reads and formats the file that a buffer is
associated with from disk instead of accepting input from stdin. This
has the side effect that if the filename is changed in the buffer, but
not saved yet, the fixer discards all the pending changes and replaces
the buffer contents with the formatted text from the file contents on
disk.
* Add support for c3-lsp linter
Add support for c3-lang with the c3-lsp language server.
Link: http://github.com/pherrymason/c3-lsp
Link: http://c3-lang.org
* fix linter error
* fix: consistent use of the executable name
Consistently use the executable name 'c3lsp' instead of the project name
'c3-lsp'.
* c3lsp: add command line arguments to executable
* Add erlfmt fixer to the registry
Without this, the fixer will not appear in the list of suggested tools
and cannot be used without additional configuration.
* Handle stdin in the erlfmt fixer command
Previously, the full path to the file being edited was used, which
resulted in the loss of unsaved changes.
* Add executable selection tests for erlfmt fixer
Users can set the DOCKER environnement variable to select Docker or
Podman to run the tests.
Co-authored-by: L'HOSPITAL Logan <lhospitallogan@gmail.com>
* Added jq support
Cleaned up yq.vim file
* Updated docs
* Updated supported-tools.md
* Added yq tests
* Fix python linting/formatting error when in virtual environment (#4865)
Python fixers and linters were failing when vim is running in a virtual
environment that's located in a path containing text `poetry`. The cause
of this was the regular expression `poetry\|pipenv\|uv$` which matches
`poetry` and `pipenv` if they appear anywhere in the virtualenv path.
* Add cljfmt fixer for clojure files (#4860)
* When using `actionlint` look for & use a config file (#4858)
Actionlint supports a config file and it lives in a very searchable
path, as the only files it acts on are in the `.github` directory
already.
Look for an `actionlint.yml` and `.yaml` in that path, and use the
config if its there.
* Fix linting with jq (#4765) (#4862)
With the 1.6 version of jq the error message start with "parse error".
With the last version of jq the error message start with "jq: parse error".
Fix it by using a regular expression that works in both cases.
* Properly handle optional end_line_no/end_line_pos in sqlfluff (#4867)
end_line_no/end_line_pos are optional. Example SQL:
`SELECT NULL FROM {{ a_jinja_templated_table }};`
`sqlfluff lint --dialect ansi --format json` gives the following error
among others:
```
{"start_line_no": 1, "start_line_pos": 21, "code": "TMP", "description":
"Undefined jinja template variable: 'a_jinja_templated_table'", "name":
"", "warning": false}
```
As one can see there is no end_line_no/end_line_pos.
* Add golangci-lint fixer (#4853)
Closes#4616
* Fixed copy-paste misstakes and added filter to docs
* Added test vader file for yq
* Fixed and updated the test case
---------
Co-authored-by: Walter Kaunda <14844142+kwalter94@users.noreply.github.com>
Co-authored-by: rudolf ordoyne <49649789+casens5@users.noreply.github.com>
Co-authored-by: Bea Hughes <108035665+beahues@users.noreply.github.com>
Co-authored-by: benjos1234 <legrimlvl24@gmail.com>
Co-authored-by: Coacher <Coacher@users.noreply.github.com>
Co-authored-by: Ian Stapleton Cordasco <graffatcolmingov@gmail.com>
* fix ale_python_auto_virtualenv to correctly set virtualenv env vars
According to the documentation, `ale_python_auto_virtualenv` should automatically set environment variables for commands, but previously the variables were not set completely or correctly.
Before:
`PATH` variable was expanded to include `/path/to/venv`
After:
`PATH` variable is expanded to include `/path/to/venv/bin`
`VIRTUAL_ENV` variable is set to `path/to/venv`
This mimics exactly what the `activate` scripts do, and allows the configuration knob to work as expected.
For example, after this change, `jedi-language-server` can be installed globally (instead of inside every venv), and it will "just work" (e.g. find references to dependencies in the venv) when editing a file in a project that uses a venv, because the correct variables are set.
* fix test_python_virtualenv.vader test to expect output with both virtualenv vars
* remove unnecessary non-escape in test_python_virtualenv.vader
* fix accidentally removed space in windows test_python_virtualenv.vader
In addition to errors Zeek's parsing can also expose warning messages,
e.g., for the following code
```zeek
event http_stats(c: connection, stats: http_stats_rec) {
c$removal_hooks;
}
```
a warning is emitted
```
warning in /tmp/foo.zeek, line 2: expression value ignored (c$removal_hooks)
```
This patch adds parsing and propagation of these warning messages.
end_line_no/end_line_pos are optional. Example SQL:
`SELECT NULL FROM {{ a_jinja_templated_table }};`
`sqlfluff lint --dialect ansi --format json` gives the following error
among others:
```
{"start_line_no": 1, "start_line_pos": 21, "code": "TMP", "description":
"Undefined jinja template variable: 'a_jinja_templated_table'", "name":
"", "warning": false}
```
As one can see there is no end_line_no/end_line_pos.
With the 1.6 version of jq the error message start with "parse error".
With the last version of jq the error message start with "jq: parse error".
Fix it by using a regular expression that works in both cases.
Actionlint supports a config file and it lives in a very searchable
path, as the only files it acts on are in the `.github` directory
already.
Look for an `actionlint.yml` and `.yaml` in that path, and use the
config if its there.
Python fixers and linters were failing when vim is running in a virtual
environment that's located in a path containing text `poetry`. The cause
of this was the regular expression `poetry\|pipenv\|uv$` which matches
`poetry` and `pipenv` if they appear anywhere in the virtualenv path.
This fixer performs indentation with the Erlang mode for Emacs.
The Erlang mode is maintained in the Erlang/OTP source tree. It indents
some things differently than the Vim indent plugin, and provides more
customization options.
* Prefix user-defined commands with colons
This is consistent with Vim's own :help pages.
* Remove dot hack
Now that we have `:ALEInfo` and `ALEInfo`, we don't need `ALEInfo.` any
more to disambiguate them.
* Use colons in references
* Use angle brackets for command arguments
* Use `:Command` for command references
* Use a non-command reference for tsserver
* Prefix highlight references with hl-
* Fix some references into Vim's own :help
E.g. location-list or +features
* Misc hotlink improvements
* Undo previous changes to tsserver
Just leave it in backticks - even though I don't like it.
* Use bars for a command for consistency
* Append hotlinks to hl-groups
Remove minuses to make tables look more like in Vim's own :help
* Prefix features with +
* Provide full hotlink to ale.txt
* Fix double pipe typo
* Capitalize Error highlight
There seems to be no hotlink in Vim's own documentation for this.
I would have expected *hl-Error* - no such thing :-(
* Right align tags to col 79
Make ale#floating_preview#Show more similar to popup_create and return
the id of the window so it's easy to set the filetype of the resulting
buffer.
Update test stub version of Show() to return a win id (the current window
since it's not actually creating a window).
Test
* both tests still pass
The only option available to biome's `lsp-proxy` command used for
linting is `--config-path`. However, we are using ALE to find and set
the project root, and have a way to manually override, so that is no
longer necessary.
The LSP proxy also used the `g:ale_biome_options` config, which is
shared with the fixer's `check` command, but `lsp-proxy` will throw an
error if unknown options are included, making it so that option is only
useful to set the project root.
BREAKING CHANGE: We are no longer passing options to the biome LSP
proxy, but we can still set the project root with
`g:ale_biome_lsp_project_root`.
Since biome supports either `biome.json` or `biome.jsonc` config files,
we need to look for both when searching for the LSP project root. We can
also look for a package.json or .git folder to use. This uses mostly the
same logic as deno.
* Add Ruby linter with Steep
Fixes#3254
* Run steep instead of using language server
LSP presents a few issues and this works around those.
* Work around Steep path issue
See https://github.com/soutaro/steep/pull/975
* Add simple tests for steep
* Add steep to supported tools
* Pass linter
* Add a comment regarding Steep's column counting
* Make lnum an integer
* Add Steep handler test
* Fix separator for Windows
* Escape Windows path separators for substitute()
* Use ALEInfo (I) group
* Use fnameescape instead of quotes
* Skip linting for files not under steep root
* Add and pass tests covering proper steep root lookup
* Fix separator discrepancy
* Use strict operators (match case)
* Fix ordering
* Use `is#` instead of `==#`
Since Biome understands `typescriptreact` and `javascriptreact` as
languages, we can send the `filetype` to the LSP, rather than only
sending `typescript` for both `ts` and `tsx` files, or `javascript` for
`js` and `jsx` files.
fixes: #4752
biome handles utf8 characters differently between files and stdin, and
in some cases can replace emojis with ascii characters when using stdin
refs: biomejs/biome#2604
* Update dart analysis_server command
In 2021 the dart team added a new sub-command `language-server` to
replace the original `./snapshots/analysis_server.dart.snapshot --lsp`
convention for starting the language server.
c224cc2e0d
* Add ale_dart_analysis_server_enable_language_server option
This allows users to opt-in to the new `dart language-server` command.
* Enable ale_dart_analysis_server_enable_language_server option by default
* Update doc/ale-dart.txt
Include the dart version number where the `dart language-server` command
was added.
this commit is to fix#4756 which suggests to force disable applying
fixes when linting, particularly when `fix = true` was set in project
`pyproject.toml` file.
The flag `--no-fix` was added without checking the version of `ruff` at
this moment as it seems to be available in a quite early version.
php-cs-fixer command line options are ordered. Options that appear after the
main command are applied to the main command. Options that appear after the
subcommands are applied to the subcommands. This change enables a user to
specific fix options (like --config). This change also sets the plugin to
find the the configuraiton file in the current project tree. This matches
the default behavior of other linters like eslint.
As stated in the changelog:
"the original fields of line_pos and line_no have been renamed to start_line_pos and start_line_no, to distinguish them from the new fields starting end_*"
This commit appends `check` to the ruff executable if the version of
specified ruff executable is `>= 0.3.0`, as ruff version `0.3.0`
deprecates `ruff <path>` in favor of `ruff check <path>`:
https://github.com/astral-sh/ruff/releases/tag/v0.3.0
Checking whole packages instead of individual files is more sensible
default for golang projects. Without this we get many cryptic
`typecheck` errors in ALE that do not show when running in terminal or
CI.
* Fix '-s' to be '-S' when setting 'TabSize=1' for chktex
Fixes#4712Closes#4725
* Check if chktex's -S option is available
* Check chktex version instead of trying -S option
Most of the time it works to assume that the current working
directory is the root of the project. However, this is not the case
for Rebar3 checked out dependencies, for example.
It's also worth noting that because of the way Elvis handles file
patterns, and because directories in configuration are relative to the
project root, the path supplied to command must be also relative.
* super hacky way to get ember template lint to work on gjs files
* Clean up code so we use a handler which means we reuse all the config
also moves handler to the glimmer directory so it only fires
for gjs files
* fix tests
* PHPStan is now working with filename-mapping
See help ale-lint-other-machines for more info about filename-mapping.
* Add two tests to show and test what is expected
* Missed this update while creating previous commit
* Simplified the update
We only needed to refactor the processing loop.
No extra test are needed.
I have an LSP that is returning markdown code blocks on Hover with no
language specified, e.g.
````
```
Foobar
```
````
As a result, you get "```" in the message line which is not that useful.
I made the regex to catch the first code fence accept empty language as
well, and if it's empty, we set it to "text".
This makes it so that LSPs that return no language still produce legible
restuls on the message line.
Co-authored-by: Oliver Ruben Albertini <ora@fb.com>
* Fix chktex highlighting wrong column when using tabs instead of spaces
Fixes#723
chktex implemented feature request [1] for allowing setting options from
the command line. Thanks to that we can tell it to treat tab character
as of one space width, i.e. one char. That means, after we translate the
output back to Vim columns, we get correct numbers.
[1]: https://savannah.nongnu.org/bugs/?56486
* Add test_tex_chktex.vader
* Use functions to set g: variables in ale_linters/tex/chktex.vim
* Update ale_linters#tex#chktex#GetCommand() to use '%e'
[rubyfmt](https://github.com/fables-tales/rubyfmt) is a formatter for
`ruby` code.
This commit adds support for `rubyfmt` as a `ruby` fixer (#2991),
together with some tests and documentation.
* Add end_col and end_lnum to ShellCheck
ShellCheck supports a JSON format mode which includes an 'endLine' and
'endColumn' field.
We must use the newer 'json1' format as it properly treats tabs as a
single character. 'json1' was not supported until v0.7.0 in 2019[1], so
we maintain support for the older GCC based format.
[1] https://github.com/koalaman/shellcheck/blob/v0.9.0/CHANGELOG.md?plain=1#L121
* Add wiki link to ShellCheck json output
Since Neovim commit c4afb9788c4f139eb2e3b7aa4d6a6a20b67ba156, the sign
API uses extmarks internally. Virtual text is already rendered using
extmarks. ALE uses the same group name for both signs and virtual text
and as a result, both are placed in the same extmark group. Since ALE
deletes all extmarks in the virtual text group after all signs have been
placed, no signs are ever shown. This commit fixes this by renaming the
sign group from `ale` to `ale_signs`.
* Ruff use json-lines output format
* Fix Ruff: add -q to prevent non json output
Using the json-lines output format allows for setting of the end_line,
end_col and code field of the handle output.
Additionally, the first letter of the code is used to determine the type
field.
Co-authored-by: w0rp <w0rp@users.noreply.github.com>
Nickel(https://nickel-lang.org/) is a configuration language, like
Jsonnet, Cue, Dhall.
`nickel`(https://github.com/tweag/nickel) is the main command to run,
export and also format Nickel code.
this commit adds `nickel format` as a Nickel fixer, together with some
tests and documentation.
Fix solhint for versions >= 3.4.0, while still supporting older
versions.
The solhint linter code has been moved out of the `handlers` directory
as it does not need to be shared between different filetypes. Code has
been simplified.
Co-authored-by: Henrique Barcelos <16565602+hbarcelos@users.noreply.github.com>
We weren't joining and returning paths correctly for detecting project
roots for Haskell projects, and now we are.
Co-authored-by: Rodrigo Mesquita <rodrigo.m.mesquita@gmail.com>
* Remove some tests we no longer need
* Delete blocks of redundant code
* Compress some tests together to simplify them
* Remove a little code for ancient linter versions
* Escape more executables we didn't escape before
* Rename a deno option that didn't match our conventions
Add an ALEStopLSP command to stop all language servers that match a
given name. Completions are available for the command. This makes it
possible to keep other language servers running other than the one
you're interested in stopping.
Default `g:ale_disable_lsp` to a new mode `'auto'` by default. With this
setting applied, ALE will now check for the presence of nvim-lspconfig
and automatically turn off particular LSP linters if already configured
via nvim-lspconfig.
For users that do not use `nvim-lspconfig`, everything should work as
before.
Fix the ordering of virtualtext so we print the most severe problem on a
line. If two problems are the most severe, we will print the left-most
problem.
Show only a single virtualtext message per line by default. The setting
can be configured to whatever the user wants. This default prevents
several linters from spamming the editor with messages that run off into
the right margin.
Documentation now clarifies that problems have a predictable order, and
which message will come first.
.venv was going to be the officially recommended default virtualenv
directory name in PEP 704, which was not accepted. Still, poetry uses
this name by default, as do other projects. We can deem it the first
name we should try to search for.
ve-py3 was a directory name I can't find mentions of online, and was
used in my own projects during the days of migrating from Python 2 to 3.
We can just drop it, and people can update their settings if they still
need it.
Use Neovim's diagnostics API by default in recent enough Neovim
versions. This will make problems found by ALE play nicely with problems
found by other tools.
Use a repeating timer to emulate InsertLeave mode for users who have not
rebound <C-c> to <Esc>, like many experienced Vim users do. This allows
ALE to start linting when you finish typing by default without having
to know about this quirk in Vim or Neovim.
Make a series of sweeping changes to make :ALEInfo more useful.
1. Deprecate :ALEInfoToClipboard and support :ALEInfo -clipboard
2. Permit :ALEInfo -clip as a shorthand for :ALEInfo -clipboard
3. Support :ALEInfo -preview to render in the preview window
4. Support :ALEInfo -echo for the classic :ALEInfo mode
5. Change the default mode to 'preview', and make it configurable
6. Add syntax highlighting for ALEInfo in preview mode
7. Add a convenience to look up documentatation that explains itself
8. Don't show an empty 'Linter Variables' section
When commands are run, it can be useful to just save the hidden buffers
so language servers immediately get updated with changes to files
without you having to manually save each file. You can now enable this
by setting `g:ale_save_hidden` to `1`.
* Avoid performance problems with setbufline() and Treesitter
Call nvim_buf_set_lines() instead.
Since this is a performance problem only in Neovim (Treesitter is only
available there), it doesn't matter that this API is unavailable in Vim.
Note: nvim_buf_set_lines() returns E5555, when set nomodifiable is on.
Fixes#3669
* Avoid sign flickering
The signs flickered because nvim_buf_set_lines() removes all signs from
lines that it touches, which will immediately be readded by Ale (causing
the brief flicker). This is intended behaviour in neovim [0].
Neovim itself faced this problem in their own LSP formatting sync,
although they had the problem with marks instead of signs [1].
Similar to how neovim fixed it by storing and restoring the marks [2],
we can do the same thing with signs.
In fact it is easier with signs, because sign_placelist() will just
ignore and skip invalid line numbers, so we don't need to filter signs
that are not valid anymore.
[0] https://github.com/neovim/neovim/issues/10880#issuecomment-526466042
[1] https://github.com/neovim/neovim/issues/14307
[2] https://github.com/neovim/neovim/pull/14630
rust-analyzer sometimes returns a hover result with language being
"text", but there's no syntax/text.vim, so this would fail with:
Error detected while processing function <SNR>150_VimOutputCallback[6]..<lambda>8[1]..ale#lsp#HandleMessage[30]..ale#hover#HandleLSPResponse[42]..ale#floating_preview#Show[13]..<SNR>161_VimShow:
line 13:
E484: Cannot open file syntax/text.vim
Only including the file when it actually exists fixes this.
In #2637, support for numhl highlights was added for nvim.
In the meantime, vim added support for numhl highlights in patch 8.2.3874.
This patch allows numhl highlights to be enabled in ALE for vim >= 8.2.3874 too.
* fixed parsing errors when certain options are used in glslang
* Update glslang.vim
set column number to 0 like it is always set by glslangValidator
* Added a test for the handler of glslangValidator
* Fix a Ruby deprecation warning in the ERB linter
Before, the ERB linter used positional arguments. Newer versions of Ruby
have deprecated this method signature. We fixed the linter to use
keyword arguments.
* fixup! Fix a Ruby deprecation warning in the ERB linter
* Add fourmolu fixer
Fourmolu is aversion of Ormolu that supports configuration. This fixer
was modeled after the Ormolu one, but using the "stack executable"
approach of the Brittany and Stylish Haskell fixers.
* Sort supported-tools.md
* Add support for Bicep when installed as a plugin to Azure CLI
The compiler for Microsoft's DSL Bicep can be installed both
independently and as a plugin to Azure CLI. The latter is probably how
most people install it.
The program output is the same but Azure CLI wraps the arguments and has
a slightly different interface, hence I opted to copy the old linter and
modify it to match the plugin arguments.
* Fix bicep/az_bicep tests, arguments and parsing
* Actually test the ale_linters#bicep#az_bicep#Handle function in the
test that should test that function, not
ale_linters#bicep#bicep#Handle.
* Use the same method as in bicep/bicep for discarding output file, i.e.
by specifying --outfile to a null file.
* Fix parsing of occasionally occurring leading error type (such as
'ERROR: ').
* Correct option defaults for bicep & az_bicep specified in documentation
* Fix error from ansible-lint versions >=6.11.0.
The JSON output format of ansible-lint has changed since
6.11.0. Issue locations can have either a 'positions' or
a 'lines' member, rather than just a 'lines' member as it
was before. This fix checks which member is present, and
passes that member name to subsequent dictionary lookups.
The error was caused by the following change:
https://github.com/ansible/ansible-lint/pull/2897
* Add ansible-lint test to check each type of ansible-lint issue json.
* Change long single-line JSON in ansible test into multiline JSON.
* Fix linting errors in ansible_lint.vim.
When running the tests on aarch64, the run-tests script tries to
download a pre-built image that is built for x86-64, and thus does not
run.
This change adds a check for the Docker daemon host platform and only
downloads the image if it will run.
Furthermore, the image dependency testbed/vim:24 is also built unless
the platform is x86_64, since it is also only provided for this
platform.
PHPStan will only detect a configuration file in the current working
directory, so set that to the directory in which ALE finds the
configuration file.
Support replacing ALE's display of problems with sending problems to the Neovim diagnostics API.
:help g:ale_use_neovim_diagnostics_api
Co-authored-by: David Balatero <dbalatero@users.noreply.github.com>
Co-authored-by: Georgi Angelchev <angelchev@live.co.uk>
Co-authored-by: w0rp <devw0rp@gmail.com>
I am the President and Treasurer of Dense Analysis, and through my sole authority I transfer the copyright of ALE to my own nonprofit organisation, so it can live on throughout history as a protected FOSS project after I am dead. Nuff said.
Default virtual-text to the Comment highlight group and prefix
virtual-text messages with comment text for each language by default.
Messages can now be formatted with `%type%` to print the error type.
The Vim 9.0 version has been updated in the Docker image to add test
coverage for virtual-text.
Add functions to compute the cwd to be the same as the project root for
pylsp and Pyright to work around issues in each language server when
they encounter modules that share the same name as first or third party
libraries.
* Handle empty answer of ansible-lint
The variable a:lines might be empty if ansible-lint exited early, in
that case json_decode would trow an error.
* Use ales JSON decode function
this commit adds refurb as a Python linter, together with some tests
and documentation. it should fix issue: #4362
refurb repo: https://github.com/dosisod/refurb
`checkmake` by default checks config file "in the same folder it's
executed in" unless `--config` option is set.
This commit allows setting the `--config` option with an option
variable (with documentation updated).
Consider a file like
```
#lang racket
(require racket/gui)
```
Type `Go(eventspace-`.
Pressing <C-x><C-o> to trigger omnicomplete should suggest
```
eventspace-handler-thread
eventspace-shutdown?
eventspace-event-evt
```
It does not (instead producing "top-level" completions, as if
`(eventspace-` wasn't even there).
Debugging, place the cursor on a space _after_. Now
`ale#completion#OmniFunc(1, '')` correctly returns `1`, but when given
`(0, 'eventspace-')` it returns either the empty list or generic
completion results as described above. I'm not entirely sure of the
mechanism, but it seems that `b:ale_completion_info.prefix` is the key,
and that this is set by `ale#completion#GetPrefix`. Calling
`ale#completion#GetPrefix('racket', line('.'), col('.'))` returned `''`!
Now, it returns `eventspace-` and the completions work correctly again.
Ref #4293, #4186, #3870
- Add this option so command line arguments can be supplied to hadolint
- This will be respected when running in docker and via the executable
- Preserve the --no-color and - flags, and add these to the list
- Add to docs and tests
* ale.txt: fix indentation
* ale.txt: fix Type and Default markup
* ale.txt: use `not set` instead of `undefined`
This matches the way the variables are referenced in prose about
existence state.
Gcc does not support `x c*-header` when using `-` as input filename,
which is what ALE does.
Rework the feature to only use `-x c*-header` flag when using Clang and
not GCC.
The feature is now also controlled with the variable
`g:ale_c_cc_use_header_lang_flag` and
`g:ale_cpp_cc_use_header_lang_flag`.
Add configuration files for pyright (JSON and TOML) to list of files
which identify a project root directory. Update documentation
accordingly.
Co-authored-by: Andreas Doll <andreas.doll@posteo.de>
When linting an header file in C or C++, `-x c-header` or
`-x c++-header` should be used instead of `-x c` or `-x c++`.
Using `-x c` or `-x c++` for headers files can lead to unused variables
and functions marked as static inlined as seen in #4096.
Using `-x c-header` or `-x c++-header` solve these issues.
The list of file extensions that are considered as header files can be
configured with the variables `g:ale_c_cc_header_exts` and
`g:ale_cpp_cc_header_exts`.
I discovered that references to other Bicep files (modules) will be
broken if running on a temporary file in a different location. I've
found no way of providing an alternate path when invoking the command.
* Add support for Microsoft's DSL Bicep
The compilation command 'bicep build' catches compilation errors as well
as providing some lint warnings.
Repository for Bicep: https://github.com/Azure/bicep
* Different null file on Windows & hardcode commands
Deno LSP automatically detects config files named deno.json or
deno.jsonc since version 1.18.
For Deno 1.18+ this means that ALE no longer needs to resolve the
project root. However, removing the project root logic from ALE means
breaking changes for people that are still using an older version.
Adding deno.json to the list of looking files to look for will keep the
behavior consistent and compatible with the Deno config file naming
convention.
See also:
https://deno.com/blog/v1.18#auto-discovery-of-the-config-file
* Remove virtual text via types-filter
This is more robust and has the additional sideeffect that it will make
it easier to implement showing virtual text for all warnings
simultaneously.
We definitely do not want to do a call to prop_remove() for every
virtual text as that will cause noticeable lag when many warnings are
present, thus we can use this to remove all virtual text lines with one
call in the future.
Fixes#4294
refs: https://github.com/vim/vim/pull/10945
* Allow virtual text to appear for all warnings of the buffer
This can be enabled with:
let g:ale_virtualtext_cursor = 2
It is implemented both for neovim and vim 9.0.0297.
Note that sometimes it may appear like some warnings are displayed
multiple times. This is not a bug in the virtual text implementation,
but a sideeffect of multiple linters returning similar results.
For example for Rust, the 'cargo' and 'rls' linters appear to be
activated at the same time, but they sometimes return identical errors.
This causes the virtual text to show the same warning twice.
In the future we can mitigate this problem by removing duplicate errors
from our internal location list.
However users can also achieve cleaner warnings simply by activating
only one linter for each language (or multiple unambiguous linters).
For example for Rust, the problem could be solved with:
let g:ale_linters = {'rust': ['analyzer']}
Fixes#2962Fixes#3666
Regression was introduced in d93bc2baf7
The problem was that we did not handle the edge case where there is no
last popup to close, which caused old vim versions to enter code by
accident that was only supposed to be run by vim 9.
We fix this by guarding the if condition for vim 9.
Fixes#4290
* Add extra config options for virtualtext
* Undo virtualtext changes and move to floating preview
* revert changes to pass hightlight group to floating preview
* rename var
* Document changes
* Add updates based on feedback
* Check for string type and attempt to call the function
* Fix lint errors
Co-authored-by: Shaun Duncan <shaun@speedscale.com>
Our current virtual text implementation for vim emulates it by abusing
the textprop and popupwin feature from vim 8.2 (for more details see
commit 708e810414).
This implementation sometimes is janky, for example the popups may leak
into other vim windows next to the current window.
Luckily, vim just got native virtual-text support as a proper subtype to
the prop_add() function. By using the 'text' option, the text property
automatically becomes virtual text that is appended to the current line
if col is zero.
Note that the prop_add() method now returns negative IDs for virtual
text properties.
This feature was added in vim 9.0.0067, but it got a lot of bugfixes
which is why we only use this new API if vim has at least version
9.0.0214.
However, there are still some minor bugs with vim's native virtual text,
so we might have to bump the version check again in the future.
Also see #3906.
Now with proper virtual text support for both vim and neovim available,
we can tackle #2962 in the future by simply tracking multiple virt-texts
instead of just the last one.
In the future we might also want to disable our virtual text emulation
support for vim, as it is a total hack, but for now we should keep it
for backwards compatibility.
When 'close_cb' is set for job_start(), but out_cb or err_cb isn't, vim
buffers data instead of dropping it (in case someone wanted to read and
process it in close_cb), and additionally polls for new data every 10
milliseconds, causing excessive wakeups and CPU usage. Since we don't
read the data anywhere outside of out_cb/err_cb, any LSP that prints an
error to stderr triggers this and vim keeps spinning until :ALEStopAllLSPs.
Fix this by always setting both callbacks, thus dropping any data we're
not interested in.
See https://github.com/vim/vim/issues/10758 for an upstream report of
the excessive polling. It's possible this is intentional, I dunno.
Fixes: b42153eb17 ("Fix #4098 - Clear LSP data when servers crash")
In #4231 some code was added to stop the completion menu if any when
opening a new one. This resulted in an issue in Vim that fills the
buffer with Ctrl-Z characters when deleting to the end of a line in a
position that triggers auto-completion.
Since auto-completion seems to work fine on all my tests I am reverting
this specific change.
* add support, docs, tests for Laravel Pint
* fix php-cs-fixer link
* add missing project-without-pint
* fix indentation
* fix pint executable in pint fixer test
* fix variables, docs related to pint support
* fix: added support for local solhint executable
* feat: added support for matching parse errors
* test: added test for solhint command callback and handler
* chore: removed command callback test
* refactor: made solhint handler structure closer to eslint
* refactor(shfmt-fixer): remove derivation of default CLI arguments
The default `omni_start_map` is too restrictive for Lisps and Schemes
like Racket, which permit hyphens (among other special characters).
As recorded in #3870, trying to complete `file-name-from-path` when
typing `file-name<C-x><C-o>` would give completions like `namespace`
because the hyphen is ignored to find the start of the word for
completion.
Now the racket filetype searches for the start using the keyword class
`\k`, which is more precise.
* Allow customization of all floating window borders
Users may not necessarily want the same border character for top+bottom
or left+right, so allow all eight border characters to be configured in
g:ale_floating_window_border.
For backwards compatibility, the old rules are still applied if only six
elements are given.
* Reorder popup border array for compatibility
* Add support for HashiCorp Packer
* Add test for packer fmt
* Add doc for HCL/Packer
* Add link to Packer doc
* Also suggest packer fix for packer ft
* Add more links to TOC
* vscode-json-languageserver-bin support
VSCode JSON languageserver has schema support for linting and
completions.
I have enabled snippets support (`snippetSupport`) even if it is not
fully supported. `label` that comes with completions response can be
used as well.
* Test fix.
* vscode-json-languageserver instead of vscode-json-languageserver-bin
vscode-json-languageserver is more up-to-date (about 1 year old),
vscode-json-languageserver-bin is 4 years old.
* Use git root.
* Documentation update.
* Trying to sort ordering issue.
* One more attempt
* One more attempt
* Uppercase seems to win.
* Clean-up
* Clean-up 2
* Test removed.
The previous linter rule about stray echo lines has been restored, and
now all problems for custom linting rules can be ignored by adding a
comment above problem lines.
* rust-analyzer in non-cargo projects
rust-analyzer can also be used in non-cargo projects. This requires a
rust-project.json file in the project root [1].
Make the rust-analyzer linter search for a rust-project.json file if no
Cargo.toml file could be found.
[1]: https://rust-analyzer.github.io/manual.html#non-cargo-based-projects
* Document rust-analyzer without cargo
* Test rust-analyzer with non-cargo projects
Change the other rust tests to match the new directory structure of the
test files.
Currently, it's not possible to override the awk `--lint` option with
```viml
let g:ale_awk_gawk_options = '--lint=no-ext'
```
although this could be useful for those who only use gawk and don't want to get these lint errors:
> FEATURE X is a gawk extension
The idea is to move the default `--lint` option before the `awk_gawk_options` in the gawk.vim code to give the custom `--lint=...` option a higher precedence.
Co-authored-by: Barnabás Ágoston <barna@agoston.dev>
This patch adds support for opening jdt:// links on "go to definition" requests returned by Java language servers.
Co-authored-by: w0rp <devw0rp@gmail.com>
* Dispatch textDocument/didChange after rename
Previously whenever we renamed a symbol that was referenced from other
files we'd just edit those files in the background, and the LSP wouldn't
know about these changes. If we tried to rename the same symbol again,
the renaming would fail. In some scenarios, the operation would just be
wrong. Here is an attempt to fix this issue.
I also noticed another bug when using Go with `gopls` LSP and the `gofmt`
fixer. Whenever the file was saved, the `gofmt` would run and reformat
the file. But it seems there was some kind of a race condition so I
disabled saving for now, and all of the modified files will be unsaved,
so the user should call `:wa` to save them. I personally like this even
better because I can inspect exactly what changes happened, and I
instantly see them in the other opened buffers, which was previously not
the case.
Fixes#3343, #3642, #3781.
* Address PR comments
* Remove mode tests in corner case tests
* Address PR comments
* Save after ALERename and ALEOrganizeImports
Also provide options to disable automatic saving, as well as instructions to
enable `set hidden` before doing that.
* Fix broken test
* Save only when !&hidden
* Update doc
* Update doc
* Add silent
Unimport (https://github.com/hakancelik96/unimport/) is a linter,
formatter for finding and removing unused import statements.
This introduces linting support, although fixer support could come
later.
* Allows to use quickfix for references.
E.g. following mapping could be used to find references for item under
cursor and put result into quickfix list:
```
nnoremap <leader>af :ALEFindReferences -quickfix<CR>
```
Fixes#1759
* Documentation update.
Only open list window if the number of warnings or errors equals to or
exceeds the value of ale_open_list. No change when set to `1`.
Co-authored-by: cos <cos>
Closes#1810
Add ALEPopulateQuickfix and ALEPopulateLocList. They're not very useful
with ale's default auto-populate behaviour, so their useful configuration
is described in help.
Previously, it would not generate any lint messages for nix 2.5.
Moreover, it would cause this error whenever the nix command is
invoked, when paired with a custom `g:ale_command_wrapper`:
Error detected while processing function <SNR>92_NeoVimCallback[29]..<lambda>27[
1]..<SNR>90_ExitCallback[28]..ale_linters#nix#nix#Command:
line 1:
E684: list index out of range: 0
* Fix 4004 - Disable eslint by default for json.
This PR disables, or more correctly, excludes eslint from the list of
default linters for json files.
Also fixes elixir, go, json5, and jsonc files documentation and default
linters to make them consistent.
* Fix and improve tests
* ALEFileRename command added.
This command renames file and uses tsserver `getEditsForFileRename` to
fix import paths in Typescript files.
* ale#util#Input fix
* Even more fixes.
* Linting error fix.
When `let g:ale_ruby_reek_show_wiki_link = 1`, Reek linter is crashed
because `wiki_link` attribute does not exist in result.
It seems that `wiki_link` is now replaced with `documentation_link` in
recent version of Reek
* add support for checkov for terraform
* add tests for checkov handler
* add basic linter config tests for checkov
* update supported tools and languages lists
* simplify ale_linters#terraform#checkov#Handle
* ensure "-o json --quiet" is always set for checkov
* add documentation for checkov including config options
* fix tests after changing handling of default options for checkov
* add checkov to list of tools in doc/ale.txt
If virtualtext.vim is autoloaded first, it will link
ALEVirtualTextWarning to ALEWarning. But ALEWarning is not initialized
yet, so it will create ALEWarning, but with no color definition set.
Shortly after, highlight.vim is autoloaded, which would usually link
ALEWarning to SpellCap, but only if ALEWarning is not already set.
However since ALEWarning is already initialized due to the previous
link, we skip this and never actually come around to properly
initializing it.
We fix this by initializing all highlight groups in highlight.vim, thus
satisfying the dependency of ALEVirtualTextWarning being initialized
after ALEWarning.
Fixes#3585
* Add cppcheck handler match on misra msg
* Fix cppcheck --file-filter setting
This time, the tests and actually usage both work.
Co-authored-by: Dan George <dgeorge@anduril.com>
* Add cspell linter
Add cspell linter, with the languages it supports.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Add cspell Global Variables Documentation
Add documentation to /doc/ale.txt with cspell configuration options.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Add cspell to docs, Minor Cleanup
Add cspell for each supported language, adding some spaces and removing
others when caught navigating the file.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Add cppcheck handler match on misra msg
* Use --file-filter cppcheck option
Cppcheck recently added --file-filter so that cppcheck only checks the
filtered files, even when using --project option, which checks all files
in the project, by default. The --ccpcheck-build-dir option didn't help
enough (at all?).
* Added C test cases
Also fixed and assumed typo: foo.c, instead of foo.cpp
* Replace hard-coded full path filenames
Attempt to fix the windows platform test execution.
* Fix typo - foo.c, instead of foo.cpp
* Reset buffer var between tests
* Handle header files in cppcheck
Cppcheck isn't designed to check header files, stand-alone. Daniel
Marjamäki suggested using --suppress options to avoid FPs.
* Fix Vint complaint in cppcheck handler.
* Fix file path in cppcheck handler
Co-authored-by: Dan George <dgeorge@anduril.com>
Since having been added, the `alex` tool has added support for linting
on stdin. Rewrite this integration to reduce the number of tools
requiring disk-write access.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Adds phpactor lsp linter
* Fixes missing comma
* Adds tests for phpactor lsp linter
* Adds note that this part is not my own work
* Removes unused variable
* Adds phpactor to supported tools list
* Fixes doc sorting
* Wraps phpactor in code tags
* Add support for AVRA linting
* Add tests for AVRA linting and improve code
* Fix test
* Fix warning detection
* Fix test
* Fix test
* Add AVRA as a supported language in docs
* Adds --memmory-limit option for PHPStan linter
* Updates docs for phpstan --memory-limit option.
* Adds Arizard to authors
* Adds test for phpstan memory limit parameter
* Fixes order of parameters in test
* Changes dash to underscore
* Add Statix for Linting
Add `statix check` as a linter. Provides a simple set of definition
tests additionally. Variable names specify "check" to allow for later
addition of `statix fix` as a formatter once stream support is added.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Fixup Supported Tools List
I didn't realise there were two separate lists of tools, so add statix
to the other list. Also, remembered "S" comes after "R", and so
re-ordered it.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Fix statix Test File
I refactored the variables for statix to allow for writing a fixer
later, and forgot to update them in the test, so update them now. Also
remove a stray "i", add missing space before checks
Signed-off-by: David Houston <houstdav000@gmail.com>
* Update Output Stream for v0.4.0
statix v0.4.0 provides a breaking change of output stream from stderr to
stdout.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Add statix fix Fixer
Implement statix fix as a fixer for simple Nix antipatterns.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Fix statix Fixer Tests
Fix the statix fixer tests by removing the unnecessary
'read_temporary_file' value from the command, since it simply uses the
default value.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Add statix Handler Test
Add a test for the statix handler per @hsanson's request.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Fix to run only on stdin for linting
Signed-off-by: David Houston <houstdav000@gmail.com>
* Implement gofumpt Fixer
Add an implementation with test and documentation for the gofumpt go
code formatter, a stricter formatter than your standard "go fmt".
Signed-off-by: David Houston <houstdav000@gmail.com>
* Add gofumpt to ale.txt TOC
Forgot to add gofumpt to the ALE vim help Table of Contents, so do so.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Fix Test Setup Method Capitalization
I had put "Setup" instead of "SetUp" for "ale#assert#SetUpFixerTests".
Fix such.
Signed-off-by: David Houston <houstdav000@gmail.com>
* Fix typos
Add a missing space, remove an extra bracket by actually running tests
locally first. Would've been smart to do that from the beginning...
Signed-off-by: David Houston <houstdav000@gmail.com>
There is no need to filter for references in such a complicated way.
docker images already works if you just pass the image and tag as
an argument.
This caused problems if one was using podman with its docker-compatible
interface.
Previously podman would return the following error:
Error: cannot specify an image and a filter(s)
With this new method podman does not return an error anymore, causing
the image to not be redownloaded every time and it still works with
normal Docker.
Before this patch multiline warnings would appear in a single line with
'^@' as separator.
Now we use whitespace as separator to improve the appearance.
Also strip trailing whitespace, newlines, etc...
Fixes#3939
The `-T` option (for "taint checking") was deprecated in ruby 2.7
and removed entirely in ruby 3.0. This causes the linter to fail
entirely for users of ruby 3.0.
This was reported in #3537, and then fixed in #3538 - but it seems as
though in 9fe7b1fe6a, it was accidentally
and entirely undone.
This commit is essentially identical to #3538, aside from a path change
for the tests.
According to
https://microsoft.github.io/language-server-protocol/specifications/specification-3-17/#textDocument_codeAction,
the response to textDocument/codeAction is:
(Command | CodeAction)[] | null
and the code only handled the case where it was a CodeAction that either
specified an edit or a command, but didn't handle a direct Command.
Note that the specification also says that both can be specified and
then the edit is applied first, then the command. Furthermore, there
seems to be some hacky code handling arguments directly, which I suspect
is non-standard and only works with a specific LSP server that happens
to pass the edits in the arguments unmodified.
It is easier to explain this fix with an example:
* tsserver and LSPs ask for error information when you want to fix
error. tsserver `ts@getCodeFixes` command needs tsserver error code.
* now let's imagine that user has eslint and tsserver in use. Sometimes
both can report same error in different way.
* Now there is no guarantee which error will come first and if eslint
error comes first then tsserver will not return code fixes as we are
passing wrong error code to it.
This fix will return proper error code based on linter.
Without this, we have one `pyre` process running across different pyre
projects. With this change, files in different projects can be linted
with pyre at the same time.
Co-authored-by: Oliver Ruben Albertini <ora@fb.com>
It's necessary to provide a `-l` option to pyre with the closest parent
directory containing a `.pyre_configuration.local` file, or simply
change directory (cwd) to the root of the pyre project. Thanks to Ken
Verbosky for the code that fixes this.
Error seen when not using such a solution:
```
1031.473923 on 6: Dropping message 'ƛ Background task unexpectedly quited: Invalid configuration: Cannot find any source files to analyze. Either `source_directories` or `targets` must be specified.
```
Issue with this approach is that if you are editing files under
different projects, the `pyre persistent` process is not re-created for
each file. We have to do `:ALEStopAlllsps` in order for the process to
start with the new working directory.
Co-authored-by: Oliver Ruben Albertini <ora@fb.com>
More recent versions of thriftcheck use a more compliant GCC-style
output format which includes a space before the "severity" group.
This matches similar tools, like shellcheck.
This change adjusts the handler's pattern to parse this format in a
backwards-compatible way (even though backwards compatibility isn't
critical long-term as thriftcheck itself is close to its 1.0 release).
* Fix truncated echo
In typescript, when putting the cursor on a `>` character of an arrow
function, the displayString body comes back as an empty string, and
means the split operation has 0 items, causing a failure when attempting
to call TruncatedEcho.
Even if there's a better fix, I'd assume this is a good safety since we
are injesting external data.
* Convert to use `empty()`
This User autocommand is trigged immediately after an LSP process is
successfully initialized. This provides a way to perform any additional
initialization work, such as setting up buffer-level mappings.
* Add eslint as linter for JSON, JSONC and JSON5
Use the same lint configuration as eslint for javascript.
* Add documentation for JSON* eslint support
* Fix spacing in documentation
* Update docs to be unopinionated about plugins
Remove any preference for eslint plugins, since there are more thant one
that would work
* Reorder languages and tools in alphabetic order
* Fix misalignment
* Change orders to pass the tests
Look for all linters that have "lint_file" set to 1 and verify tools
that have it have the :floopy_disk: icon set and those that don't do not
have it.
Correspondingly added/removed !! on
ale-supported-languages-and-tools.txt file.
Co-authored-by: Horacio Sanson <horacio@allm.inc>
- Add .vintrc.yaml configuration that disables the scriptencoding check
(ProhibitMissingScriptEncoding) that is raised randomly.
- Upgrade vint to 0.3.21. Project seems to have stopped here and 0.4.0
was never released.
- Ensure the run-test scripts use the correct docker image (e.g. add tag)
.
Co-authored-by: Horacio Sanson <horacio@allm.inc>
For some reason CI tests started failing with these errors:
> ale_linters/eruby/erb.vim:1:1: Use scriptencoding when multibyte char exists (see :help :scriptencoding)
> ale_linters/mail/languagetool.vim:1:1: Use scriptencoding when multibyte char exists (see :help :scriptencoding)
Not sure at which point or what changed for this to happen but this MR
fixes it by removing the multibyte chars present on the problem files.
Co-authored-by: Horacio Sanson <horacio@allm.inc>
md5sum isn't available by default on macOS. Instead, it ships the
BSD-style md5(1) command, which does the same thing but with different
arguments.
With this change, run-tests works out-of-the-box on macOS.
* feat(deno): move init options to handlers
* feat(deno): add deno lsp support for js files
* feat(deno): use default map option
* feat(docs): add deno import map option
* feat(deno): add tests for importMap option
* fix(deno): use full path in importMap
* feat(deno): remove deno as linter for js, separate PR
* fix(deno): test for executable
* fix(deno-test): include filename to simplify function
* Add poetry support to python linters and black fixer.
* Update python.vim to detect poetry project.
* Update ale.vim, add an option for poetry `g:ale_python_auto_poetry`.
* Update ale-python.txt, add poetry support.
* Add and update poetry related tests.
Co-authored-by: unc0 <unc0@users.noreply.github.com>
* implement vim popups for preview
Details on implementation
-------------------------
- we make use of the |popupwin| api
- we split implementations (Nvim* vs. Vim* prefix) and call the right
one based on has('nvim')
- we follow a similar structure in each function, using the relevant API
- popup_list, win_execute, popup_settext in VimShow
- popup_create in VimCreate
- popup_close in VimClose
Some differences
----------------
- we DON'T have VimPrepareWindowContent because we use arguments to
popup_create for borders, padding, etc., and it also takes care of
buffer creation.
- we follow the protocol of setting and using w:preview for information,
but we only need the ID
- InsertEnter is the only autocommand required, because of
popup_create's moved argument. Any cursor movement with 'any' will
close the popup. This in turns means VimClose is only called from
InsertMode, so no mode-restoration necessary
- we don't tweak too much in the buffer because vim's popup buffers
already have most relevant settings and aren't editable without
calling popup functions.
- I enabled scrollbars, close buttons, dragging, and resizing
- vim popups get as big as they need to by default, so no worrying about
truncating/hiding/size
Note: we might want to consider changing w:preview to w:ale_preview to
avoid clashes if someone else tries to use the same variable
* floating window: document that vim supports it
* lint: fix indent/cont. lines
* Fix 3801 - Add ALEDummySign some width.
Due to changes in NeoVim 0.5 the g:ale_sign_column_always configuration
stopped working.
This PR sets the ALEDummySign to a blank space so when g:
ale_sign_column_always is set we have a sign with 1 width allowing the
configuration to work as before.
https://github.com/neovim/neovim/issues/13635
* Fix visual artifact on dummy sign
* Fix visual artifact on dummy sign (attempt 2)
Co-authored-by: Horacio Sanson <horacio@allm.inc>
* Correct typo in a config file filename.
.tool_versions should be .tool-versions
* Correct typo in config file names.
.tool_versions should be .tool-versions
* racket: support racket-langserver lsp
* racket-langserver: find highest dir with init.rkt
* autoload/ale/racket: re-indent to 4 spaces
* racket: lint: sort supported tools
* racket: lint: function!
This is _not_ needed anymore, but the lint wants it. See :help E127
* racket-langserver: do not use new dict format
* racket: lint: use snake_case
* add tests for racket-langserver
* racket-langserver tests: correct result values
* Also check for asdf-vm's .tool_versions file
A minimal python project may only be specifying a python version using a version management tool like asdf-vm, without providing other common python project configuration files. asdf-vm creates a single .tool_versions file in the managed directory. By checking for .tool_versions in addition to other common python config files we ensure that python linters (whose behaviour typically depends on a particular python version) will run with the same version of python used by the project. This will also be the same python version used by vim itself when it is run from inside the project's directories.
* add .tool_versions to ale-python-root documentation
This reflects the corresponding change to autoload/ale/python.vim
* Add yosys for verilog files.
* Add handler test for yosys.
* fix typo in yosys handler test
* fix array order in yosys handler test
* add yosys linter to filetype defaults test
* fix duplicate tag
* add 'yosys' to 'ale-supported-languages-and-tools.txt'
This is achieved by switching to JSON, which makes it much easier to
avoid confusion between an error message and the next one. It also
spares us from having to deal with regular expressions, and eliminates
some edge cases that no longer need to be tested.
- Show hadolint rule number in vim gutter in addition to `ALEDetails`
- Capture and show error in case of syntax errors
- Add tests for error capture
- Adapt existing tests
fixes: #2333fixes: #958
Both '-include' and '-imacros' take a file as an argument that will then
be searched in the include path like a regular '#include "..."'
statement in a source file. As such, they should not have their path
converted to an absolute path.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
The -o/--option flag was removed in version 2.0.0 of rpmlint. Providing
this causes rpmlint to fail and provide no output. Only provide this
flag to rpmlint if the version is less than 2.0.0.
This allows the location list from one buffer to point to an issue in
another; previously, the error message would be shown but with no way to
jump to it.
The ocaml filetype is currently used for several, different file
formats. This causes problems as not all tools support all formats.
New filetypes are introduced to support this separation, this needs some
changes in ale that are fortunately backwards-compatible.
These change add ocamlinterface file support for ocp-indent, merlin,
ocamlformat and ocaml-lsp. For ocaml-lsp I took the liberty to
add all recognised language ids, even if they are not supported.
ols has not been changed as the project has been abandoned since 2019.
cmake-format added support for reading from/outputting to stdin/out as
of v0.3.6, released 2018-04-10 (commit 2e2aff2) [0].
Reading from stdin is preferable over reading from a temporary file
because when given a concrete file cmake-format will look for its config
file (.cmake-format.py or similar) in the parent directories of the
provided file. If the temporary file is off in a tmpdir somewhere (e.g.,
/tmp on *nix), cmake-format will almost certainly not come across the
user's intended format configuration file, making it appear that
cmake-format is ignoring the config file.
If cmake-format reads from stdin, though, it'll look for its config file
in its current working directory and its parent directories, in a
similar manner to clang-format. This has a much higher chance of running
across the intended config file.
[0]: https://github.com/cheshirekow/cmake_format/releases/tag/v0.3.6
Co-authored-by: Alex Wang <ts826848@gmail.com>
* fix: added support for local solhint executable
* feat: added support for matching parse errors
* test: added test for solhint command callback and handler
* chore: removed command callback test
* refactor: made solhint handler structure closer to eslint
* improve DMD handler
- ignore errors from other files
- catch 'Deprecation' as warning
- add tests
* adding filename key instead of filtering
* update dmd test
* fix test dmd windows
Buildifier offers a -path command line option:
> Buildifier's reformatting depends in part on the path to the file
> relative to the workspace directory. Normally buildifier deduces
> that path from the file names given, but the path can be given
> explicitly with the -path argument. This is especially useful when
> reformatting standard input, or in scripts that reformat a temporary
> copy of a file.
This lets us drop our `-type`-guessing logic.
For the moment, we rely on the buffer filename being "relative to the
workspace directory", which it generally is, but in a future change, we
should introduce logic that will locate the WORKSPACE file and adjusts
the filename relative to that directory. This could be complicated based
on the working directory in which `buildifier` is executed, so a little
more research is necessary to implement that logic correctly.
* add support for svelte via svelteserver language server
* svelte: fix Vint error and add a `svelteserver` simple test.
Co-authored-by: Joakim Repomaa <mail@j.repomaa.com>
Co-authored-by: Joakim Repomaa <mail@jreinert.com>
* Fix texlab GetProjectRoot
* Fix indents in texlab#GetProjectRoot
* Prevent texlab from starting on every tex file
* Update texlab Vader tests
* Fix GetProjectRoot to return parent of .git
Previously, the function returned `../.git/`. We want the function to return the parent directory above that as the project root. This should help pass Vader tests.
The `ale_lsp_root` setting is now deprecated, and `ale_root` should be
used instead. The setting will be used for both setting the root easily
for LSP linters, and for running other linters over whole projects.
Working directories are now set seperately from the commands so they
can later be swapped out when running linters over projects is
supported, and also better support filename mapping for running linters
on other machines in future.
* Fix ansible-lint linter definition.
Use ansible-lint's feature auto-detection instead of temporary file.
For auto-detection to work, ansible project has to be also a git repository.
Don't use yaml rules. These are checked by yamllint.
Refactor pattern to work with ansible-lint >=5.0 version.
Clean-up obsolete test cases.
* Pull Request changes
This fixer enables buildifier's formatting and "lint fix" modes.
Additional options can be provided via `bazel_buildifier_options`.
It also implements some basic logic for guessing the file's type.
buildifier itself usually does this based on the filenames provided on
the command line, but because we're piping our buffer via stdin, we need
to do this manually.
* Allow clangformat to use a local style file.
* Add tests.
* Fix Vint issue.
* Improve explanation of feature in documentation.
* Fix failing test.
The test was checking the wrong directory.
* Simplify verilator linter using ale command format strings
* Verilator Linter: Restructure linter command tests
* Verilator Linter: adds to the handler test the returned filename
* Verilator Linter: add the current file path to the search path
* Verilator Linter: Add the search path to the tests
Co-authored-by: TG <tarik.graba@telecom-paris.fr>
Makefiles using GNU-make features might be called "GNUmakefile" instead
of "Makefile". This commit teaches the `c_parse_makefile` feature to
look for a GNUmakefile file if a Makefile is not present.
* The build status badge is now for GitHub Actions.
* The documentation now mentions GitHub instead.
* Warnings in the YAML file have been fixed or ignored.
* Add vim82 and neovim04 to CI tests.
* Fix test_sign_column_hightlighting test.
In vim82 with verbose=1 the output of highlight command changes breaking
the ale#sign#SetUpDefaultColumnWithoutErrorsHighlight(). This commit
forces verbose=0 when the method starts and restores the previous value
before exiting.
* No return values in vim82 returns a numeric value instead of a empty string.
* Fix test_reek_handler test
The FuzzyJSONDecode() method catches E474 when it fails to parse the
input as JSON but Vim8.2 throws E491 instead. This commit modifies the
function to catch both E474 or E491.
* Fix perl6 handler test.
Perl6 handler catches json parse errors using the E474 error but in
Vim82 it changed to E491. This commit modifies the handler so both
errors are considered.
* Fix list opening tests.
In Vim 8.2 the call `range(1, bufnr('$'))` always returns quickfix
buffers no matter if they are closed or not. Using `ls` does not show
them but the above range will always include them.
This new behavior breaks the ale#list#IsQuickfixOpen() method that in
turn breaks many other things. This commit fixes this by using the
getqflist() and getloclist() methods instead.
* Fix test updates loclist test.
For some reason in Vim 8.2 the sign offset seems to not reset between
tests causing the sign_id to not match in the Assert. When the test is
run individually it passes but when run as part of the whole suite the
sign_id is off by one.
Forcing the offset in the test setup seems to fix the issue.
* Fix omnifunc completion test.
For unknown reasons the SetCompletionResponse tests fail in Neovim 0.2
and 0.4. Unfortunatelly the only solution I found is to disable them
for neovim.
* Fix linter warnings
* Fix smoker test.
Add vim 8.2 to the list of versions that need some retires due to
randomly failing tests.
* Add docker image build job.
Trying some clever trick to build the docker image if not available
locally or in Docker hub. It uses the Dockerfile md5 checksum as tag so
only when changes on that file occur will the image be downloaded or
build.
* Add labels to Docker image
* Remove tests for middle versions 8.1 and 0.3.5
* Use same vader commit as appveyor
* Implement image push to Docker Hub
Co-authored-by: Horacio Sanson <horacio@allm.inc>
In the (unreleased) Nix 2.4 the error-messages have been reformatted[1].
This patch aims to retain proper `.nix`-support in `ale`, for both
stable Nix (2.3 and older) and unstable Nix (2.4 and newer).
[1] https://github.com/NixOS/nix/pull/3590
* master: (133 commits)
Add rnix-lsp for Nix diagnostics and completion
add spectral support for json
add spectral handler
add spectral linter for yaml
doc: Fix linter issues
doc: Add documentation for Deno
feat: Add Deno lsp support
feat: Add Deno fmt fixer
Add document for apkbuild filetype
Add tests for atools handler, basic and dealing with Error and Warning
Test default linters for apkbuild
Document new default linters for apkbuild
Make apkbuild_lint and secfixes_check default for apkbuild filetype
document support for apkbuild-lint and secfixes-check for apkbuild
Add linters for apkbuild-lint and secfixes-check from atools
Add handler for the output of atools
Fix typos
Add command callback tests
Add support for standalone files
Fix linting errors
...
atools is a collection of tools written in ash shell and Lua that
provide linting for Alpine Linux's APKBUILD.
APKBUILDs are build recipes used by Alpine Linux's build system, abuild,
an equivalent would be Arch Linux's PKGBUILD and Gentoo's ebuild.
Seems standardrb fails to properly use the --config option when using
temporary files but works fine when reading from stdin. This commit
changes the fixer so it uses stdin instead of temporary files.
* Fix 354 - Migrate CI from travis to Github Actions
* Use matrix strategy for parallel tests
* Don't build image on each run
* Add push trigger on tags
Co-authored-by: Horacio Sanson <horacio@allm.inc>
* Add nvim floating window hover support
* Add configuration for float to replace preview
* preview#ShowFloating: qualify local variables
* Configure floating preview usecases individually
Also:
* Extract floating preview to its own file.
* Ignore 'stay_here' option. Moving into the floating preview window
seems confusing at best.
* Re-use existing floating preview window if it's still up.
* Flush out floating preview documentation.
* Watch cursor position changes per window
Floating previews open a new window, so when that window is written to,
it moves briefly there at a different position than the original window.
This makes repeated positions detected when positions are tracked at a
s: level. Instead, we change the variable to window scoped, which only
fires a message if the cursor has changed from the last position in
*that window*.
* g:ale_floating_preview cleanup
* floating_preview: add ALEDetail tests
* Fix fecs test missing runtime call
* Add ALEHover floating preview tests
Co-authored-by: Jan-Grimo Sobez <jan-grimo.sobez@phys.chem.ethz.ch>
Prior to #3448, several linters should have been failing the
custom-checks that look for non-snake-cased lint names. They weren't,
but now the bug that hid those is fixed. So to avoid breaking users, we
just exclude those from the check. Linters excluded:
* clojure/clj_kondo.vim
* elixir/elixir_ls.vim
* go/golangci_lint.vim
* swift/swiftformat.vim
This adds a linter for Inko (https://inko-lang.org/). The linter makes
use of Inko's own compiler, and a newly introduced --check flag to only
check for errors; instead of also compiling source code.
* origin/master: (40 commits)
fix: correct suggested filetype for yamlfix
feat: add yamlfix fixer
Use _config for LSP config options
Add support for R languageserver (#3370)
Fix 3103 - add shellcheck shell directive detection. (#3216)
Added the Vundle command in installation instructions (#3400)
Adds support for Tlint - A Tighten Opinionated PHP Linter (#3291)
Add php phpcbf options (#3383)
Use has('gui_running') instead of has('gui')
Close#2727 - Add a hover-only setting for balloons
Fix#3332 - Modify everything for rename/actions
Add a missing blank line in documentation
Add luafmt fixer (#3289)
#3442 Fix code fix clangd issue
Close#1466 - Add GVIM refactor menu support
Look for node packages in .yarn/sdks as well
Update documentation for code actions and rename
cmp forwards, and reverse the code actions
Support for LSP/tsserver Code Actions (#3437)
Move the test for buffer-local variables
...
Hadolint is in the process of adding the severity of a lint rule to the
commandline output: https://github.com/hadolint/hadolint/pull/501
This change utilizes that to show the severity in vim.
NOTE: The custom-linting-rules test fails due to the following (legit)
warnings:
ale_linters/clojure/clj_kondo.vim:29 Use snake_case names for linters
ale_linters/elixir/elixir_ls.vim:15 Use snake_case names for linters
ale_linters/go/golangci_lint.vim:54 Use snake_case names for linters
ale_linters/swift/swiftformat.vim:56 Use snake_case names for linters
The message wasn't getting printed because docker was explicitly only
being asked to connect stdout (ignoring stderr). Unclear yet why the
error code wasn't getting bubbled up.
sed wasn't using -E, so '|' wasn't being handled properly. Seems likely
that's sed-implementation specific, so now it runs through docker's sed
to support portability.
* Fix 3103 - add shellcheck shell directive detection.
Searches for shellcheck shell directive to detect dialects for scripts
that do not have shebang.
* Change order of detection of shellcheck dialect
In a situation where the filetype can be wrong (example: something.sh
which is written in bash dialect) and has no hash-bang (since it is
meant to be sourced) then the override specified within the script will
be ignored.
It probably is the most right thing to do if the script author has added
a specific directive; it should trump everything else.
Co-authored-by: Horacio Sanson <horacio@allm.inc>
Co-authored-by: Dino Korah <dino.korah@redmatter.com>
ALE now just modifies every open buffer for rename and actions, and sets
up a one-time use BufEnter event to reload buffers that are changed so
you don't have to think about what to do with changed buffers.
* Added tsserver and LSP code action support.
* tsserver refactors support added.
* Handling special case when new text is added after new line symbol.
* ale#code_action#ApplyChanges simplified.
* Initial attempt on LSP Code Actions.
* workspace/executeCommand added.
* Some null checks added.
* Add last column to LSP Code Action message.
* Pass diagnostics to LSP code action.
Previously ApplyChanges code was applied from top-to-bottom that required
extra parameters to track progress and there was bug. I have changed code
to bottom-to-top approach as that does not require those extra parameters
and solved the bug.
Tested with typescript-language-server and it is working.
The "ale#handlers#sh#GetShellType()" function currently falls back
to the file type without checking for buffer-local variables first.
This causes the function to return "sh" even when a script is known
by Vim to be a script of a more specific type (e.g., "bash").
The "ale#handlers#shellcheck#GetDialectArgument()" function then
erroneously uses this type even though a more fitting type should be
used instead. Files without a "#!" line will be of type "sh" even
though they may have a ".bash" suffix.
This commit fixes the problem by checking for buffer-local shell
type variables (set by Vim) before falling back to the file type.
* origin/master:
Add tests for maven.vim file
Fix grammatical error in doc
Add maven helper file; use maven wrapper if available instead of global 'mvn' executable
fix lint, fix variable semantics and update tests
bibclean: update matchlist reges for bibclean > v2.11.4
Update rubocop_auto_correct_all tag
A recent(?) update to swipl changed the error format from
Warning: some.pl:2:
Singleton variables: [Y]
to
Warning: some.pl:2:
Warning: Singleton variables: [Y]
The old error handler doesn't report the correct line numbers and
messages on the old format.
I've chosen to add a function that covers the second case and detect it,
rather than rewrite the current function. This way, both versions should
be able to live together.
---
Example file that demonstrates the issue (some.pl above):
```
% vim: ft=prolog
ii(X, Y) :- X.
```
---
Newer versions of pylint will now check your code as you type. Older
versions will still only check the file on disk.
Co-authored-by: Oliver Wiegers <oliver.wiegers@gmail.com>
Add an `ALECompletePost` event along with everything needed to make it
useful for its primary purpose: fixing code after inserting completions.
* `ALEFix` can now be called with a bang (`!`) to suppress errors.
* A new `ALELintStop` command lets you stop linting, and start it later.
A new command, `:ALEImport`, has been added, which lets you import words
at your cursor if a completion provider can provide a completion for
that word which includes some additional text changes.
* ember-template-lint: Lint from stdin
* This feature has recently been implemented in ember-template-lint.
* Refactor ember-template-lint executable
* Fallback on a temporary file for old template-lint
Co-authored-by: w0rp <w0rp@users.noreply.github.com>
Option `per-file-ignores` was introduced in flake8 version 3.7.0.
It allows to ignore specific errors in specific files using glob syntax.
For example `per-file-ignores = src/generated/*.py:F401` will
ignore `F401` error in all python files in `src/generated`.
Thus ale has to run flake8 from project root where .flake8 config
is placed otherwise glob won't match linted file.
`lint_file` can now be computed dynamically with a callback function,
which can return a deferred result, as per `ale#command#Run`. This
allows linters to dynamically switch between checking files on disk,
or checking code on the fly.
Some tests have been fixed on Windows.
Prettier does not use `.prettierignore` unless the current directory is the root where the `.prettierignore` file resides.
* Update Prettier tests
* Look for prettierignore to determine project root
ALE now converts paths from compile_commands.json files into absolute
paths and prefers matching against absolute file and directory names for
determining which flags to use for files. As a result, parsing
compile_commands.json to determine flags should work for a lot more C
and C++ projects.
ALE was incorrectly detecting completion results from servers such as
rust-analyzer as wanting to add import lines when additionalTextEdits
was present, but empty.
Now ALE only filters out completion results if the autoimport setting is
off, and one of the additionalTextEdits starts on some line other than
the current line. If any additionalTextEdits happen to be identical to
the change from completion anyway, ALE will skip them.
Users can easily be confused when they set some options for a C or C++
compiler, and another compiler is run with different options, which
still reports errors. To remedy this, the existing `gcc` and `clang`
linters have been replaced with a `cc` linter that will run either
compiler.
This is a breaking change for ALE v3.0.0.
Certain tests could break if you ran them separately from other tests.
They have been patched.
`run-tests` now has a `--fast` option which runs tests with only the
fastest Vim version ALE tests with, and the custom checks.
* Restore old behavior of ALEFix command for Rubocop
Since RuboCop 0.60 ALEFix command stopped to fix all found offenses. This change restores the
previous behavior by allowing rubocop to fix all detected offenses.
* Fix tests
* Allow to configure auto-correct option for Rubocop
The options for parsing `make -n` and `compile_commands.json` flags
are now enabled by default, so people can start getting better flags
for their files by default.
`compile_commands.json` flags are now preferred over `make -n` results,
to make the options work better by default.
* fix cppcheck for 1.89+, and add column support
In cppcheck 1.89 the output changed to be more like GCC. This commit
forces any version of cppcheck to output in that same format. This also
allows for ALE to pick up the linter's column information
* Add parameters to tests. Vader passes.
* Fix c cppcheck for v1.89
* Added hdl_checker support
* Added hdl_checker tests
HDL Checker searches for files when no config file is found, which could lead to very long searches when the user is not really on a project setting
* Split FindNearestExecutable from FindExecutable
The path searching in ale#node#FindExecutable() will be useful for
eslint. Refactor it into a separate function so it can be used without
regard for the state of the _use_global and _executable variables.
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
* eslint: Set project root from local executable
Using the nearest directory with node_modules does not work correctly
for nested projects where the eslint dependencies are in the outer
project. For example:
https://github.com/dense-analysis/ale/issues/3143#issuecomment-652452362
Adopt the behavior of SublimeLinter, which runs from project_root
determined by the presence of the eslint executable in node_modules/.bin
(or eslint in dependencies/devDependencies of package.json, which we can
add later as necessary). See [NodeLinter#find_local_executable].
[NodeLinter#find_local_executable]: https://github.com/SublimeLinter/SublimeLinter/blob/056e6f6/lint/base_linter/node_linter.py#L109
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
* Split eslint#GetCdString from eslint#GetCommand
Move the code for finding the project root and building the cd string
into a separate function so that it can be reused in the eslint fixer.
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
* Run ESLint fixer from project root dir
To match the ESLint linter, as changed in 9ee57d43 (which I forgot to
apply to the fixer, whoops).
Fixes: #3094Closes: #3095
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
So that I can find the relevant information in the vint
linting policy summary and policies can be easily configured
https://github.com/Vimjas/vint/wiki/Vint-linting-policy-summary
Before this change an example warning message appears as:
autocmd should execute in an augroup or execute with a group (see :help :autocmd)
After this change the same example appears as:
ProhibitAutocmdWithNoGroup - autocmd should execute in an augroup or execute with a group (see :help :autocmd)
1. The often longish `description` moved away from (supposedly short)
statusline `message` into the `detail` section.
2. dockerfile_lint sends `reference_url` pointing to issue explanations.
Use that.
* Add terraform-lsp integration
https://github.com/juliosueiras/terraform-lsp
* Add tests & docs for terraform-lsp integration
terraform_langserver_options setting added to send custom flags to
terraform-lsp.
Vader tests have been added to test custom executable, custom flags, and
finding the project root. All tests pass.
Initial documentation has been added for the above.
Resolvesdense-analysis/ale#2758, juliosueiras#57
* Fix tag alignment
Co-authored-by: = <Aubrey.S.Lavigne@gmail.com>
Co-authored-by: w0rp <w0rp@users.noreply.github.com>
It was returning 0 when it should be returning an empty string.
The 'AssertEqual' in the ale image is from an old version so it does not
check the types of the arguments.
This is already fixed in 427fe19104Closes#3120
* Swap substitution order for echoed message
This prevents 'code' string in liter_name to be substituted by accident.
Linters including pycodestyle have been affected by this problem.
* Add test for linter whose name contains 'code'
Test for c525db8cb4
Since version 4.032 (04/2020) verilator linter messages also contain the
column number, and look like:
%Error: /tmp/test.sv:3:1: syntax error, unexpected endmodule, expecting ';'
To stay compatible with old versions of the tool, the column number is
optional in the researched pattern regular expression.
See commit:
81c659957e
Default navigation for commands that jump to new locations has been
implemented with the `ale_default_navigation` variable, and all commands
that jump to locations now support `-tab`, `-split`, or `-vsplit`
arguments for overriding the default navigation behavior.
Windows may insert carriage return line endings, which ALE does not handle
well. These characters should not be displayed.
Adds a line to remove these characters for all messages.
ALE appends flags from {c,cpp}_{clang,gcc}_options after those found by
parsing compile_commands.json or Makefile output. If -std=* flags are
present in both the ALE flags and parsed flags, the last one present
(i.e., ALE's -std=* flag) will determine the mode the compiler works in.
This can result in errors showing up in vim but not in the actual build
or vice-versa.
For example, say you have foo.cpp:
#include <type_traits>
int main() {
return std::is_same_v<float, int>;
}
If cpp_clang_options contains -std=c++17 and -std=c++14 is parsed from
compile_commands.json, then ALE would end up running something like:
clang++ -S -x c++ -fsyntax-only -std=c++14 -std=c++17 - < foo.cpp
This would result in no errors showing up in Vim, but the actual build
would fail with:
<stdin>:3:14: error: no template named 'is_same_v' in namespace 'std'; did you mean 'is_same'?
return std::is_same_v<float, int>;
~~~~~^~~~~~~~~
is_same
/Library/Developer/CommandLineTools/usr/bin/../include/c++/v1/type_traits:872:61: note: 'is_same' declared here
template <class _Tp, class _Up> struct _LIBCPP_TEMPLATE_VIS is_same : public false_type {};
^
<stdin>:3:35: error: expected '(' for function-style cast or type construction
return std::is_same_v<float, int>;
~~~~~~~~~~~~~~~~~~~~~~~~~~^
2 errors generated.
as the actual build would not have the -std=c++17 flag added by ALE.
If cpp_clang_options contains -std=c++14 and -std=c++17 is parsed from
compile_commands.json, then the opposite problem would occur. ALE would
end up running something like:
clang++ -S -x c++ -fsyntax-only -std=c++17 -std=c++14 - < foo.cpp
and would show an error on line 3 of foo.cpp:
[clang] No template named 'is_same_v' in namespace 'std'; did you mean 'is_same'? (fix available)
The actual build, on the other hand, would succeed without any
complaints.
Removing -std=* from ALE's flags if it is already present in the parsed
flags ensures that the wrong -std=* flag is not used.
An alternative would have been to switch the order in which parsed flags
and ALE flags were concatenated when producing the command to execute,
but that could prevent a user from intentionally using ALE's flags to
override some other flags, e.g. -W* flags to enable/disable warnings in
a project whose flags are not under the developer's control.
-std=* flags are also present in cuda/nvcc.vim, objc/clang.vim,
objcpp/clang.vim, and vhdl/ghdl.vim, but none of those linters appear to
parse compile_commands.json or `make` output.
The standard linter --fix fails if the file being input is not relative
to the project root (https://github.com/standard/standard/issues/1384).
This MR attempts to fix this by changing the command so the input file
is relative to the project root and the output is to a temporary file.
Preliminary tests with toy javascript projects seem to indicate this
works fine.
* Refactor the "s:LoadArgCount()" function
Previously, this function would always set "v:errmsg" on the first
call with a given function. This is because autoloaded functions
are not defined on the first call.
A number of improvements have been made:
- a useless local function ("l:Function") is removed
- the "execute()" builtin captures the output, instead of ":redir"
- a ":try" block handles the case where a function is not defined
- a useless ":if" is removed since ":redir" always defines the var
- confusing quoting is re-written (remove double "'" chars)
Fixes: #3021
Rather than requiring users to alias ps1 to powershell themselves,
include it in s:default_ale_linter_aliases. Since [vim-ps1] is a
popular (the only?) PowerShell ftplugin and there do not appear to be
any other uses of ft=ps1 on vim.org, this seems like a safe and
reasonable default.
[vim-ps1]: http://www.vim.org/scripts/script.php?script_id=1327
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
This is kind of a peculiar reason for a PR, but I no longer control the email listed. I want to change it to avoid people getting the wrong email for me. Also, I still control the domain, but if at any point I don't, I want to put down in writing that if you get an email from this, it's not from me.
* Add autoimport support for deoplete
* Fix test_deoplete_source.py
* Use callback instead of is_async for deoplete
Shuogo, the author of Deoplete, does not recommend using the `is_async`
option:
> I think is_async is not recommended. It is not so useful and broken.
> You should use callback system instead.
Link: https://github.com/Shougo/deoplete.nvim/issues/1006#issuecomment-526797857
Incidentally, the same thread mentiones an issue started by w0rp:
https://github.com/Shougo/deoplete.nvim/issues/976
The deoplete docs also say is_async is deprecated:
> is_async (Bool)
> If the gather is asynchronous, the source must set
> it to "True". A typical strategy for an asynchronous
> gather_candidates method to use this flag is to
> set is_async flag to True while results are being
> produced in the background (optionally, returning them
> as they become ready). Once background processing
> has completed, is_async flag should be set to False
> indicating that this is the last portion of the
> candidates.
>
> Note: The feature is deprecated and not recommended.
> You should use callback system by
> |deoplete#auto_complete()| instead.
Link: https://github.com/Shougo/deoplete.nvim/blob/master/doc/deoplete.txt
Co-authored-by: w0rp <w0rp@users.noreply.github.com>
ESLint 6 loads all plugins/configs/parsers relative to the project root
which, by default, is the directory in which ESLint is invoked, as
described in [ESLint RFC 2018-simplified-package-loading].
Therefore, ALE should run ESLint from the project root, when possible,
so that dependencies will load. This commit does so.
[ESLint RFC 2018-simplified-package-loading]: https://github.com/eslint/rfcs/blob/master/designs/2018-simplified-package-loading/README.mdFixes: #2787
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
Before this change, prettier_standard would run and ignore any
.prettierrc, now it will respect the configuration of the file being
linted.
This change relies on prettier-standard 16.1.0 for the --stdin-filepath
flag, but is backward compatible: older versions of prettier-standard
will ignore the unknown flag and continue to run with no configuration
file.
If checkstyle is configured with custom options that contain "-c" then
the checkstyle config file option is ignored. This PR modifies the
regular expression when creating the checkstyle command to avoid this.
ESLint errors are contained in an array that can contain different
stuff other than JSON error messages. This patch iterates over the whole
array ignoring any non-json data.
Some files lack a hashbang line but still have an unambiguous filetype.
For example, the file `.zshrc` has the filetype `zsh`.
Augment ale#handlers#sh#GetShellType to fall back to the filetype if
no hashbang line can be found.
* Refactor stylelint fixer test
* Support additional stylelint fixer options
* Support changing working directory for stylelint fixer
* Force css syntax for stylelint fixer
* Added base handling for window/showMessage
* Ignoring severity log
* Code formatting
* Added user configurable severity
* Preferring ale#util#ShowMessage over echo'ing directly
* Using format similar to ale_echo_msg_format for consistency
* Updating docs
* Added LSP log config string; improved tests
* Use warning as fallback for incorrect user config
* Add support for nimlsp.vim
* Add test and docs for nimlsp
* Add nimlsp to supported-tools.md
* Add nimlsp to doc/ale-supported-languages-and-tools.txt
Some messages of the crystal compiler are not tied to a file.
This causes a 'Key not present in Dictionnary' error (E716).
For the record, the json output on ```require "./nonexistent.cr"```
is the following :
```json
[
{ "file":"/tmp/file.cr", "line":1, "column":1, "size":0,
"message":"while requiring \"./nonexistent.cr\"" },
{ "message":"can't find file './nonexistent.cr' relative to '/tmp'" }
]
```
The second message does not have line/column attributes.
* fix tflint handler for 0.11+
* fixup! fix tflint handler for 0.11+
* maintain compatibility with previous tflint output format
* fixup! maintain compatibility with previous tflint output format
* Add comment about tflint's output format accross versions
* Use sign-group only on supported vim versions.
The sign-group feature is only available in nvim 0.4.0 and vim 8.1.614.
* Add priority to ALE signs.
This allows users to set a priority to ALE signs to take precedence over
other plugin signs.
This commit adds support for renaming symbols in tsserver and with LSP tools, and for organising imports with tsserver. Completion results for symbols that can be imported are now suggested if enabled for tsserver completion done via ALE.
jscs.info appears to have nothing to do with the linter, and just contains a blogpost about student debt.
This appears to be the closest to canonical site for the project (although it's now merged with ESLint I suppose some still use it?)
* Trying to keep win view from bouncing
* Adjusting when views are saved and restored
* Also restore view when closing quickfix
* Don't restore view when opening list vertically
* Parse CFLAGS that can be passed using a whitelist
I went through GCC's man page and selected flags that can safely be
passed to GCC and that can be useful to syntax checking. These include:
- -I/-i* include flags
- preprocessor flags such as -D
- -W* warning flags
- -O* optimization flags
- most -f options
- -m arch dependent options
* Fix CFLAGS tests: -Idir is now parsed to -I dir
* Added two tests for flags we want or don't want to pass.
* Also check for / in addition to s:sep
This makes some of the run-test output less misleading.
Also fix a minor shellcheck issue: "\*" and "\\*" are equivalent, but
the second one makes clear that the literal backslash is intentional.
* added omitted global variables which was breaking this test when run standalone
* invert logic for s:GetLinterVariables excluding disabled linters, so that linter global options can appear in output
* additional tests for s:GetLinterVariables for linter global options
This commit add support for ink-language-server, which it does by
largely copying and pasting from the pure-language-server PR that was
merged recently.
The most interesting things to note are:
- ink-language-server is distributed upstream via npm, which is why we
search through node_modules
- With some coaxing, it can be installed globally - which is why we
search for a global binary.
- Ink is a funky language, and users will likely need to add
initialization options.
- I am not incredibly familiar with vimscript; and I may not have done
some of the buffer searching correctly.
On some systems, notably NixOS, there is no `/bin/ls` and thus this test
can fail unnecessarily on those systems. This commit uses
`/usr/bin/env ls` which resolves the issue.
Allows the user to override $GO111MODULE environment variable through
ale options. This gives control over the default behavior of Go module
resolution.
Golang documentation:
https://github.com/golang/go/wiki/Modules#how-to-use-modules
Add `ale#Go#EnvString()` function to make it easy to add similar Go
environment variables in the future.
Use the new `EnvString` function in all available Go tools callbacks
& update tests
Also add test of linter command callback for `gofmt`
The default for `g:ale_lint_on_insert_leave` was recently changed to 1,
so it now needs to be explicitly set to 0 to run linters only when files
are saved.
Deoplete needs `get_complete_position` method and it has a different
signature. It already fetches the input string and attempts to detect
the position with `\k*` regexp patterns.
This option is used to determine if `min_pattern_length` is ignored.
In usual, it does not start completion when the matched input string is
shorter than `min_pattern_length`. But when the string matches
`input_pattern`, it starts completion even when ths string is `''`.
This MR adds a new configuration variable `g:ale_java_javalsp_config`
that allows to configure external dependencies and class paths to the
language server.
The variable accepts a dictionary similar to the one supported by the
[vscode/settings.json](https://github.com/georgewfraser/java-language-server#settings)
file.
Deprecates: #2561
Checkstyle xml configuration is mandatory and not providing one causes
the tool to fail with the following error:
Must specify a config XML file.
Checkstyle itself contains a default configuration as part of its
assests named `/google_checks.xml`. Invoking checkstyle with this config
works even if such file does not exists in the file system:
checkstyle -c /google_checks.xml
This should be the default invocation to allow ALE to use checkstyle
with zero configuration.
Also when a user sets `g:ale_java_checkstyle_config` option, ALE should
use it to invoke checkstyle even such file does not exists in the
filesystem. This is because checkstyle is able to use configuration files
within JAR files defined in the CLASSPATH. The default `/google_checks.xml`
is an example of such configuration available within a JAR resource.
The default binary "launcher" is too generic and can get mixed with
other tools. To use this linter user must explicitly set the absolute
path of the launcher path.
isort is great, but I've come to prefer reorder-python-imports. The tool
has a focus on smaller diffs than isort. reorder-python-imports is also
a little smarter than isort which is nice.
The existing option setting handles setting additional compile flags to
pass to clang-tidy. The new option setting added here allows setting
additional clang-tidy specific flags to be passed as well.
Fixes#2324
The command used to invoke the language server is missing some options
to include additional java modules. Without these modules the server
was not working properly.
The correct command can be found in a `launcher` script on the same
directory the `java` executable for the language server is found.
This commit changes the docs to prefer the launcher script over the java
executable. For backward compatibility it also fixes the command
invocation in case the java executable is configured.
cppcheck is now run without the --project option and from the buffer's
directory instead when the buffer has been modified. Saving the buffer
will get results by linting the project instead.
The checkstyle handler is capable of parsing the new and old output
formats. Unfortunately there are some particular output messages that
matched both the new and old regular expressions:
[WARN] whatever:11:7: WhitespaceAround: ''if'' is not followed by whitespace. [WhitespaceAround]
This caused ALE to report extra errors since the message was being
matched twice, once as a warning and another (incorrect) old formatted
error.
This MR fixes this by stopping any parsing using the old format regexp
is any errors of the new format are correcly parsed. There is no reason
to expect checkstyle to output both styles in the same report.
This linter uses the check functionality built into terraform. ALE
already has a fixer using `terraform fmt` but this doesn't provide error
messages. ALE already has a linter using `tflint` but this requires an
extra application to be installed.
For example this linter will give a warning that ! is an illegal
character in the line below:
variable "example" !{}
This linter runs the buffer through the command below and parses the
output:
terraform fmt -no-color -check=true -
This commit includes a basic implementation, documentation and tests.
The only option is to control which executable is run.
Tested with:
$ terraform -version
Terraform v0.11.13
To find the buffer corresponding to URIs reported by LSP the
HandleLSPDiagnostics() method uses the built-in bufnr() function. From
the documentation we learn that the first parameter of bufnr() is
an expression, not a path.
EclipseLSP will report project wide errors (e.g. gradle errors) that are
not related to any actual source file with an URI that corresponds to the
project root folder, e.g:
file:///home/username/Projects/gradle-simple
This URI will match any open buffer of files within the project root
hiearchy, thus project-wide errors appear as part of every file within
the project, e.g:
file:///home/username/Projects/gradle-simple/src/main/java/Hello.java
To fix this, this MR adds '^' to the beginning and '$' at the end of the
URI path to force an exact match. This is how is recommended in vim
help (see :h bufname).
We were setting the -data parameter to the project root but this caused
the language server to fail initialization and synch of gradle
dependencies. As consequence ALE failed to work fully on gradle
projects.
This fix sets the workspace to the parent folder of the project root.
Normally this corresponds to the correct Eclipse workspace path.
When this is not the case, this fix also allows users to explicitly set
the absolute path to the workspace via configuration variable.
* Search eclipselsp jar and config files within system package path
* Allow setting an alternate eclipselsp configuration directory
* Add test for ale_java_eclipselsp_config_path
The official configuration files for `flake8` are `.flake8`, `tox.ini`,
and `setup.cfg`.
After investigation, it is safe to remove `flake8.cfg` as it appears to
only exist as a typo in other tooling documentation (e.g.,
`python-language-server`).
Even though no linters automatically read `.flake8rc`, it is kept in
case projects may be using it for detecting the projects root directory.
Make it very clear in every single place that the setting for ALE's own
completion implementation is mentioned that you should not enable it if
you want to use ALE as a completion source for other plugins like
Deoplete.
- Set default value to $HOME/eclipse.jdt.ls
- Make JAR search regexp more specific.
- Allow to set the VSCode extensions folder as ale_java_eclipselsp_path.
The recommended format for _vim's internal help files_ is "<tag> <for vim version> <last change>", (see `:help help-writing` but this format is not parsed the same way for plugins. For plugins the recommended format includes a description of the plugin such as "<tag> <description>". See `:help write-local-help` for the different template.
The `settagstack` and `gettagstack` functions don't exist prior to Vim
8.1.0519. And the function definition was unclear whether it intended
to grab the *old* or the *new* file/line/col.
* [doc] Add swift support documentation
* [doc] Add swift bullets in main help file
* [doc] Add to supported languages and tools txt file as well
* Ensure same name styling for help/readme files
* move php-langserver "test for .git dir" test-project to its own directory
* search for composer.json file in php-langserver first then .git dir
* add test for php-langserver composer.json
When using `gotype` without the `-e` option, it will only output the
first 10 errors. When working on a larger package that ofter means taht
those 10 errors are in other files then the one that you are currently
working on which then seems to indicate that there are no errors.
By adding the `-e` flag, all errors will be returned and shown properly
in the file that you are working on.
* Add credo --strict option
If a user sets 'let g:ale_elixir_credo_strict=1' it will run credo with
--strict instead of suggest. The default (0) is to run as suggest.
* Added credo docs
The Rust compiler returns the first column that is _not_ part of the
current span as `column_end`, while Ale expects `end_col` to signify
the last column of the span.
Bandit automatically [uses any .bandit file] within the directories on
which it is invoked. Since ALE invokes bandit on stdin, it does not
load a .bandit file automatically. Add support for automatically
finding a .bandit file and passing it to bandit via the --ini option
along with a variable to disable this behavior if desired.
Note: This is useful for the skips and tests configuration options, but
not exclude which would require invoking bandit using a file name, which
may or may not be a good trade-off.
[uses any .bandit file]: https://github.com/PyCQA/bandit/blob/1.5.1/bandit/cli/main.py#L70-L73
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
Pylint only [checks for pylintrc] (and .pylintrc) files in the packages
aboves its current directory before falling back to user and global
pylintrc. For projects with a src dir, running pylint from the
directory containing the file will not use the project pylintrc.
Adopt the convention used by many other Python linters of running from
the project root, which solves this issue. Add pylintrc and .pylintrc
to FindProjectRoot. Update docs.
[checks for pylintrc]: https://github.com/PyCQA/pylint/blob/pylint-2.2.2/pylint/config.py#L106
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
* The README now points to a valid helptag for linter options.
* The now very, very large part of the table of contents for linter and
fixer options has been moved into a section so the initial table is
smaller.
* Special linter or fixer options now lie beneath the general linter
or fixer options.
Although using %t to lint changes was desirable, many pylama checks use
surrounding paths and file contents (e.g. C0103 module name, E0402
relative import beyond top, etc.) The more such errors I find during
testing, the less %t seems like a good idea. Switch to %s.
Also set `lint_file` to 1 and mark Pylama as a file linter in the docs.
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
As discussed in w0rp/ale#1051, there are cases where it would be useful
to be able to specify the dialect explicitly. This commit allows users
to do so using the ale_sh_shellcheck_dialect variable.
Fixes: w0rp/ale#1051
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
The vulture linter already supports ale_python_vulture_options, but it
is not documented or tested. Since vulture only supports configuration
via options, it is an important use case. Add docs and test.
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
* Support filtered jump based on loclist item type (E or W for now)
* Use flags to customize the behavior of ALENext and ALEPrevious
* Update <plug> bindings with flags
* Update documentation about ALENext and ALEPrevious
* Use ale#args#Parse in JumpWrap
A new function is added here which will later be modified for public use
in linter and fixer callbacks. All linting and fixing now goes through
this new function, to prove that it works in all cases.
This little error caused that when parsing compile_commands json, the
filename was used to fetch entries in directory dictionary, hence, when
adding new json commands, it never found anything in dir_lookup and
instead rewrote the previous entry. Hence, the dir_lookup always
contained list of only one compile_command per directory instead of all
compile_commands for given directory.
The executable for the Alex linter is currently hard-coded as 'alex',
which is an issue given the fact that it conflicts with the Haskell
lexer generator, whose executable is also called 'alex', has been around
a dozen years before the linter, and is packaged in the official
repositories of the major Linux distributions.
This commit adds options to use a local executable for the alex linter
(which is a node package), and an option to set a custom executable.
As side changes:
* The pattern in the alex handler is made more readable by turnig it
into a very-magic regex.
* Alex handles plain text, markdown, and HTML. Specific flags for HTML
and markdown are provided when instantiating the linters for the
respective filetypes, while before those formats were treated as plain
text.
Similar to other linters/fixers, by default change to the directory of
the file being fixed before invoking `black`, which allows the tool to
read project-specific configuration (pyproject.toml)
Fixes#2218
* Add initial ameba (crystal linter) support
Note that this depends on saved file as `ameba` does not have STDIN
support
* Fix formatting of crystal linter documentation
* Add tests for ameba executable customization
* Extended statusline.vim to provide an efficient way to access the first errors,warnings,stylerrors,stylewarnings,etc from the loclist.
* Added documentation and help for the new API function.
Currently, we detect the linter root based on a variety of techniques.
However, these techniques are not foolproof. For example, clangd works
fine for many things without a compile_commands.json file, and Go
projects may be built outside of the GOPATH to take advantage of Go
1.11's automatic module support.
Add global and buffer-specific variables to allow the user to specify
the root, either as a string or a funcref. Make the funcrefs accept the
buffer number as an argument to make sure that they can function easily
in an asynchronous environment.
We define the global variable in the main plugin, since the LSP linter
code is not loaded unless required, and we want the variable to be able
to be read correctly by :ALEInfo regardless.
All linters should have a name variable set in their dictionary, and
code should be able to rely on that. Fix this test such that its example
linter contains a name entry.
* Mimic Prettier's default parser by setting it to `babylon`
* Add tests to check default Prettier `parser`
* Set Prettier default parser based on version
* Update the comment to explain the reason for an explicit default
* Add textDocument/typeDefinition for LSP
Doc to spec https://microsoft.github.io/language-server-protocol/specification#textDocument_typeDefinition
This works like textDocument/definition but resolves a location of a
type of an expression under the cursor.
I'm not sure what to do with tsserver though.
* Fix passing column to LSP
* test_go_to_definition: wording
* Add tests for textDocument/typeDefinition
* Add docs for textDocument/typeDefinition
From LSP spec:
> A range in a text document expressed as (zero-based) start and end
> positions. A range is comparable to a selection in an editor. Therefore
> the end position is exclusive.
ale#Escape function seems to prepend and append ' to the file name, which
are not present in the pydocstyle output. Having the parsing regexp match
the file name was overkill anyway, since there is an obvious 1:1
correspondence between the buffer number and the (potential) errors
reported by pydocstyle.
When using a compilation database (compile_commands.json) in very large
projects, significant delays would occur when changing files --
particularly those that happened to be far down the db. Rather than
iterating over the whole list every time, we now build up a lookup table
based on the tail of the filename (and tail of the directory for
widening searches) and iterate over the much smaller list of compile
commands for files with the given name.
Test metrics (from compile_database_perf/test.sh) show a 90% performance
improvement -- from 25 seconds to 2.5 seconds per run.
* Add support for https://github.com/saibing/bingo
* Add docs for ale-go-bingo
* Use go.mod when found
* Add test for bingo FindProjectRoot
* Simplify ale_linters#go#bingo#GetCommand
With earlier elm versions, a separate package file is maintained for
tests, which when properly configured enabled the compiler to find what
it needed to compile the tests. Under elm 0.19, test dependencies are
managed in the top-level package file, so `elm make` will fail on the
tests. `elm-test make` is required in this case.
See https://github.com/elm-explorations/test/issues/64
Split by space instead of dash.
This prevents incorrect parsing where space-separated arguments are
merged (in particular, .c or .o files were appended to -I or -D
arguments).
Handle shell escape: quotes and escaped quotes \" and shell
substitutions are recognised. This is done by verifying that no special
character (" ' ` ()) has not a matching character.
Fixes#2049
- added a cd into the direcotry containing the file in the buffer
in order to properly check for a config file
- added command_callback tests for graphql
In some situations, errors reported by `perl -c` can have multiple
listings of "at <file> line <number>". If the l:pattern is changed to
use non-greedy matching it will also match these.
For example:
```
use strict;
use DateTime;
$asdf=1;
```
Results in:
```
Global symbol "$asdf" requires explicit package name (did you forget to declare "my $asdf"?) at /Users/mgrimes/t.pl line 3, <DATA> line 1.
/Users/mgrimes/t.pl had compilation errors.
```
I am not 100% sure why `perl -c` generates errors with the extra "file
line <num>". It only happens in some versions of perl when certain
modules are used.
See: https://github.com/testdouble/standard
StandardRB is to RuboCop what StandardJS is to ESLint. This commit
naively copies the RuboCop linter and fixer to point at the standardrb
executable. Any other adjustments are very minor (the only I can think
of is that standardrb takes a `--fix` option instead of
`--auto-correct`).
This raises a confusing point to me as both developer and a user: since
ale enables all linters by default, won't this run both RuboCop and
StandardRB (the results of which will almost always be in conflict with
one another)? How does ale already solve for this for the similar case
of StandardJS and ESLint?
It's common to add SwiftLint as a CocoaPod dependency, instead of as a global
binary. In this case we should use that version of SwiftLint before looking
for any others. Note that I'm also adding support for SwiftLint in ReactNative
projects here as well, where the Pods directory would be nested inside an ios
directory.
The linter can correctly parse pydocstyle output with any of the following
command-line options enabled: --explain, --source, --debug, and/or
--verbose
The command used to invoke the LSP process was being escaped wrong.
Also added a new option to set a different java executable and fixed the
documentation.
This fixes performance problems in Neovim, where every character results
in spawning a new clipboard-tool process.
Behaviour is not similarly pathological in Vim, but it still results in
an unnecessary amount of register churn.
There is currently a check that tries to prevent c-flags that contain
'-' in them from being unintentionally split and included in the list of
commands. For example, we wouldn't want "-fno-exceptions " to appear as
"-fno" and "-exceptions ". The way this check was done was by making sure
the last character of the split string was a space.
This meant that the very last option to appear in the compile command
was ignored (as it doesn't end with a space). This fix explicitly skips
the ends-with-space check on the last option in the command-line.
This isn't the best fix. Really we should be using the same
argument-processing rules as a shell would rather than just splitting on
'-'. That's a much larger and more complicated change though.
The output format used by older checkstyle versions differs from the one
of new versions. This commit adds a second parsing iteration on the
output lines with a suitable pattern to support both versions in
parallel. Due to the differences in the order of matching groups this is
hard to achieve in a single pass through the output lines.
An appropriate test case is added.
Removed ale_virualtext_prefix from debugging since it's not requried for
the functionality to work.
Sorted debugging info to make the list easier to navigate/diff.
- Add g:ale_virtualtext_cursor boolean to enable/disable it
- Add g:ale_virtualtext_prefix to configure what prefix to use (default:
'> ')
- Requires neovim 0.3.2's unreleased API `nvim_buf_set_virtual_text`
Problem: ocamlformat is configured to format files in-place and thus go
via creating a temporary file for that. Because temporary file resides
in a different directory ocamlformat can't find `.ocamlformat`
configuration files in an original location of source files.
Solution: ocamlformat since version 0.8 can read sources on stdin and
spur result on stdout. We reconfigure ocamlformat to use a simpler
interface.
Previously, elixir-ls would treat each sub-project within an umbrella as
standalone, which isn't desirable from a language server perspective.
Added ale#handlers#elixir#FindMixUmbrellaRoot, which locates the current
project's root and then continues searching upwards for a potential
umbrella project root. This literally looks just two levels up to keep
things simple while keeping in line with Elixir project conventions.
Use this new function to determine elixir-ls's LSP project root.
* Allow configuration of hamllint executable
The hamllint executable was hard-coded, preventing it from being
overridden. Fix the executable to be dynamic to allow custom executable
paths.
This adds generic configuration dictionary support to the elixir-ls
linter. This is useful for disabling its built-in Dialyzer support, for
example, which can improve startup time.
The configuration dictionary is a little verbose. I considered reducing
the user configuration to only the nested settings dictionary (and
having the linter implementation wrap it in the top-level `elixirLS`
dictionary), but leaving it fully configurable simplifies the code and
removes any assumptions about current or future ElixirLS behavior.
Each LSP connection now stores its configuration dictionary. It is
initially empty (`{}`) and is updated each time the LSP connection is
started. When a change is detected, the workspace/didChangeConfiguration
message is sent to the LSP servers with the updated configuration.
This is the callback-based variant of the existing `lsp_config` linter
option. It serves the same purpose but can be used when more complicated
processing is needed.
`lsp_config` and `lsp_config_callback` are mutually exclusive options;
if both an given, a linter preprocessing error will be raised.
The runtime logic has been wrapped in `ale#lsp_linter#GetConfig` for
convenience, similar to `ale#lsp_linter#GetOptions`.
This also adds documentation and an `AssertLSPConfig` test function for
completeness.
* add prolog/swipl linter
* use load_files/2 instead of read_term/2
Because it also checks some semantic warnings / errors
not only syntactic warnings / errors.
e.g.:
* singleton warning
* discontiguous warning
* ...
cf. http://www.swi-prolog.org/pldoc/doc_for?object=style_check/1
* support error messages with no line number
:- module(module_name, [pred/0]).
causes
ERROR: Exported procedure module_name:pred/0 is not defined
* add test for prolog/swipl handler
* cosmetic fixes
* detect timeout using SIGALRM
* rename g:prolog_swipl_goals to g:prolog_swipl_load
* write doc for prolog/swipl linter
* update toc and README
* fix ignore patterns
* Only run stack if a stack.yaml config is found
It is necessary to check for a stack.yaml file to distinguish between
cabal-only projects or stack projects (which are also cabal projects
since stack is built on top of cabal).
* Test that stack is called if stack.yaml exists
ElixirLS (https://github.com/JakeBecker/elixir-ls) is an LSP server for
Elixir. It's distributed as a release package that can be downloaded
from https://github.com/JakeBecker/elixir-ls/releases or built locally.
The easiest way to start it is via Unix- and Win32-specific helper
scripts, so that's the basis of this command integration. Alternatively,
we could implement the contents of those platform-specific scripts in
the linter's command callback in a language-neutral way, but there isn't
any benefit to doing that aside from eliminating the platform check, and
that could prove to be too tight of a coupling going forward.
* FIX: use mix from the project root directory
* Move find root project function to autoloaded handlers
* add tests for #ale#handlers#elixr#FindMixProjectRoot
PMD is currently not working properly for Java classes that use [unnamed
packages](https://docs.oracle.com/javase/specs/jls/se11/html/jls-7.html#jls-7.4.2).
Consider the following Java class that does not contain a `package`
declaration:
```java
public class App {
String getGreeting() {
return "Hello world.";
}
static void main(String... args) {
System.out.println(new App().getGreeting());
}
}
```
Running PMD in the command line agaist the Java class above produces an
output with empty string `""` in the `"Package"` column:
```sh
$ pmd -R category/java/bestpractices.xml -f csv -d './src/main/java/App.java'
Oct 02, 2018 9:10:39 PM net.sourceforge.pmd.PMD processFiles
WARNING: This analysis could be faster, please consider using Incremental Analysis: https://pmd.github.io/pmd-6.7.0/pmd_userdocs_incremental_analysis.html
"Problem","Package","File","Priority","Line","Description","Rule set","Rule"
"1","","/Users/diego/Projects/github.com/dlresende/kata-fizz-buzz/src/main/java/App.java","2","7","System.out.println is used","Best Practices","SystemPrintln"
```
But the pmd.vim handler's current pattern refuses everything coming
from a Java class that does not have a package name (2nd column):
```vim
let l:pattern = '"\(\d\+\)",".\+","\(.\+\)","\(\d\+\)","\(\d\+\)","\(.\+\)","\(.\+\)","\(.\+\)"$'
```
The solution I am proposing is to also accept empty strings as package names.
These test vars were covering up a bug in the hlint linter
implementation. Without these vars we can see the behavior that is
exhibited in `vim` proper.
* Add better support for Haskell stack compiler tools
This commit adds support for `stack` as the executable of a tool. This
follows a pattern that has been implemented for `bundler`'s tool chain.
* Move hlint command to linter file
* Add vader test for stack exec handling
* Update ghc-mod to support stack execution
`ghc-mod` was previously broken into 2 linters.
1. ghc_mod
2. stack_ghc_mod
This additional linter is not necessary with proper support for
executable variables and `stack exec` handling.
* Support stack exec in hfmt
* Support stack in hdevtools
* Don't add newlines when not a control statement for Python
* Add test for accidental newline fix
* Add docstring detection to avoid adding unnecessarily newlines
* Add tests for docstring detection
In a lint context, it's useful to assume that included files sit next to
the current file by default. Users can still further customize this
configuration variable to add more include paths.
When set to true, and the buffer is currently inside a pipenv,
GetExecutable will return "pipenv", which will trigger the existing
functionality to append the correct pipenv arguments to run each linter.
Defaults to false.
I was going to implement ale#python#PipenvPresent by invoking
`pipenv --venv` or `pipenv --where`, but it seemed to be abominably
slow, even to the point where the test suite wasn't even finishing
("Tried to run tests 3 times"). The diff is:
diff --git a/autoload/ale/python.vim b/autoload/ale/python.vim
index 7baae079..8c100d41 100644
--- a/autoload/ale/python.vim
+++ b/autoload/ale/python.vim
@@ -106,5 +106,9 @@ endfunction
" Detects whether a pipenv environment is present.
function! ale#python#PipenvPresent(buffer) abort
- return findfile('Pipfile.lock', expand('#' . a:buffer . ':p:h') . ';') isnot# ''
+ let l:cd_string = ale#path#BufferCdString(a:buffer)
+ let l:output = systemlist(l:cd_string . 'pipenv --where')[0]
+ " `pipenv --where` returns the path to the dir containing the Pipfile
+ " if in a pipenv, or some error text otherwise.
+ return strpart(l:output, 0, 18) !=# "No Pipfile present"
endfunction
Using vim's `findfile` is much faster, behaves correctly in the majority
of situations, and also works reliably when the `pipenv` command doesn't
exist.
Solargraph allows to set configuration options by creating a
.solargraph.yml file at the root of the project using it. Therfore this
file is a good canditate for finding ruby projects root paths.
Initial discussion:
https://github.com/w0rp/ale/issues/1874#issuecomment-418316168
* The project style linter now runs while you type.
* Now the scripts for checking the project require blank lines.
* Many style issues have been found and fixed.
It can be necessary to pass options to the puppet parser validation. The
most glaring example of this is when using Puppet 3, with the
`parser = future` option enabled. This update allows adding
`--parser=future` to the options passed to Puppet.
* Add stylish-haskell as a fixer
`stylish-haskell` is a common formatting tool for the haskell toolchain.
It is not as advanced as `brittany` or `hindent`, but it is commonly
used for formatting of imports and data declarations. This adds it as a
fixer in ALE.
I see no reason to do this? It is just setting the environment to what
it already is?
It was originally added in #297, but that entire PR is not a great idea
in the first place; that PR (together with #270) tried to make the Go do
non-standard and non-supported stuff like compiling packages outside of
GOPATH.
That's not something that works well (I tried), so was eventually
removed in #465, but these "go env" calls remained, for no reason in
particular, as far as I can think of.
This will improve on #1834; you will now no longer get a confusing error
(but still won't get a meaningful error; need to think how to do that).
It wasn't immediately obvious that the `g:ale_fixers` cannot be a list,
and would allow the use of `*` to match all filetypes. I was hoping to
add a bit more detail to the README to make this clearer.
* Adding support for haskell-ide-engine
* Work with the current directory if no stack.yaml file is found
* Added Cabal file detection, updated documentation and added tests
* Updated help
fixes#1738
- Replace previous `hh_client` usage with LSP client
- Add `HHAST` linter
- Split Hack from PHP: Hack is increasingly diverging from PHP:
- Hack tools do not understand PHP
- Most PHP tools do not handle Hack code well (including vim's syntax
highightling files)
- http://github.com/hhvm/vim-hack now sets filetype to `hack`
When dializer isn't a dependency, mix dialyzer recompiles the whole
project because it's not possible to know if this command dialyzer exist
or not until recompilation is done. Then the timestamps of the project
is messed up which results in broken hot-loading. In this case, mix help
dialyzer would return zero which prevents compilation of the whole
project since dialyzer isn't installed, it's help manual doesn't exist.
When dialyzer is a dependency, mix dialyzer would just run the command.
In this case, mix help dialyzer would return 1 which allows mix dialyzer
to run.
* Add kotlin languageserver linter definition
* Added kotlin languageserver references in docs, fix missing !! on other linters
* Added Vader tests for root path detection in Kotlin Language Server
"Pipfile" and "Pipfile.lock" files are also often located in Python module or
package directories and their presence is an okay heuristic for finding project
roots.
Pipenv doesn't do local virtualenvs by default, it uses a special local
directory to store them all.
However, if you run Pipenv with the PIPENV_VENV_IN_PROJECT environment variable
set to 1, it creates the virtual environment in the root of the project, under
the name ".venv". This is why I've added this as a possible virtualenv dir
name.
* Rust Cargo linter: Improve workspace support
When using Cargo workspaces [1], there is a 'Cargo.toml' directory in a
top level directory, listing all the crates in the project. If we are
currently editing one of the crates, 'cargo build' should execute in
that directory for that crate's separate `Cargo.toml`, otherwise Cargo
may spend more time possibly rebuilding the entire workspace, and maybe
failing on one of the other crates, instead of succeeding on the current.
[1] https://doc.rust-lang.org/book/second-edition/ch14-03-cargo-workspaces.html
Like many other linters, use variables for the executable and options
used by the linter.
By default, the linter now report every warnings as errors with
`--warning-errors`.
Also add include directory and set working directory to file directory.
* Set `--parser` option in Prettier's fixer
* Add expected `--parser` option to tests
* Disable Prettier `--parser` detection if file extension exists
* Manually default Prettier `--parser` to "babylon"
* Add `--parser` test for TypeScript
* Add tests for Prettier `--parser`
* Add JSON5 to the suggested fixer for Prettier
* Guard the ballooneval settings
* Mark main objectives to do to get nice Hover
* Make tweaks to make the tooltip work - See " XXX: comments
* Guard balloon_show call
* Use return instead of finish for functions
* ale#hover#show : Add optional arguments to specify arbtirary position
This change is requested to be able to call the function with mouse
position to enable hover information in vim's balloon
* ale#ballon#Disable : Remove feature guards
* ale#balloon : Show 'ALEHover' output on balloon if no diagnostic found
* ale#hover#HandleLSPResponse : remove the check for cursor position
This check prevented the 'ALEHover in balloon' feature, since mouse
position is almost never cursor position.
* ale#balloon#MessageForPos : Change the return of balloonexpr
balloonexpr evaluation now works even without balloon_show for basic
diagnostics, leaving the balloon_show call to ale#hover#Show, which can
then feature guard the call to avoid errors
* ale#hover#Response : Feature guard balloon_show calls
* ale#hover : always display 'Hover' information in messages
Also add a small comment to warn readers the different outputs the
ale#hover#Show will write to
* {LSP,TS}Response : use only variables from the Response
It is clearer that we only rely on l:options to get the relevant data to
build the LSP Response string
* hover#ShowDetails : fix an issue where not having focus broke balloons
The issue was caused by not using a buffer-specific version of getline()
to cap the value of the column sent in the message to LSP. Therefore a
cursor on column 10 in an inactive window could send a message with
column=0, if the active window had a buffer with too few lines
* {LSP,TS}Response : Remove redundant checks for balloon_show call
With the upcoming change in ale_set_balloons default value (see Pull
Request w0rp/ale#1565), this check will be useless
* balloonexpr? : Add a flag to separate hover#Show() calls
The goal of this flag is to make `:ALEHover` calls not pop a balloon
under the cursor, since the user has probably no interest in their
cursor while typing the command
The flag is a default argument which is overridden only in ballonexpr
call of ale#hover#Show, and stays set in the hover_map until the
callback for the LSP handles it.
There are no automated tests for this feature right now, and the nature
of the addition (one optional argument in the API) should make it
transparent to existing tests.
Since the differentiation is now possible, the check for moved cursor
has been put back in ale#hover#HandleLSPResponse
* ale#hover#hover_map : Protect accesses to hover_map
Using get() is safer than trying to access directly with ., as the tests
show.
* Raise timeout to try to get Appveyor happy
* Review : Fix comments
* Review : pass the optional argument 'called_from_balloonexpr' in a Dict
This optional dictionary has documentation just before the function
using it, ale#hover#Show, and allows easier extension in the future.
* Update section 5.viii in the README with ALEJobStarted and re-format
the example.
* Add an extra line after documentation update to ensure consistency
with the rest of the doc.
Adding a couple of tests to demonstrate how IsCheckingBuffer behaves
during specific autocommand hooks:
* At ALELintPre, no linters have actually executed yet, hence
IsCheckingBuffer should be returning false.
* ALEJobStarted, fires as early as reasonably possible after a job has
successfully started, and hence hooking into IsCheckingBuffer here
should return true.
This distinction is important when using these two events during things
like statusline refreshes, namely for "linter running" indicators.
The ALELintPre and ALELintPost autocommand events are currently being
used by lightline-ale to refresh the status line and check the linter
status for a current buffer. One of the plugin's checks looks to see if
linters are currently running, via ale#engine#IsCheckingBuffer(). This
currently only works partially in certain situations. In my particular
case, working with Go files, this only seems to function properly when a
file is initially opened. Saving a file does not correctly update the
status.
This seems to be due to the fact that ALELintPre actually runs before
any jobs are carried out, making it plausible that hooking into
ALELintPre for the purpose of checking to see if there are any currently
running linters for a buffer is unreliable as it would be prone to
pretty obvious race conditions.
This adds a new User autocommand, ALEJobStarted, that gets fired at the
start of every new job that is successfully run. This allows a better
point to hook into checking the linter status of a buffer using
ale#engine#IsCheckingBuffer() by ensuring that at least one job has
started by the time IsCheckingBuffer is run.
For now, it only detects undefined steps. The nearest `features` dir
above the buffer file is loaded, so step definitions should be found
correctly.
Tested only with Cucumber for Ruby, but it should work for any cucumber
that follows a substantially similar directory structure.
This removes the argument if the specified toolchain is empty.
As far as I can tell there is no +nighly (or similar) option [1] leading to
the termination of the server. But since people needed this option and
have yet to complain about it it stays the default for now.
[1] https://github.com/rust-lang-nursery/rls/blob/master/src/main.rs#L87
* Add first qmlfmt support
* Add GetCommand() function
- pass --error/-e option
* Add handle unittest
- fix pattern regex
- store col as integer
* Update docs
* Add command callback unit test
* Add fsc as a Scala linter
* Pull reused code into `autoload/ale/` directory
* Include fsc into the README
* Add unit test for testing the scala handler
* Add unit test for scala's fsc linter
* Rename scala unit tests for clarity
* Fix typo in README
* Fix typos in doc/ale.txt
* Fix author headline
* Put methods for fsc commands back into fsc.vim
* Move command_callback tests to correct location
* Rewrite handler test so it actually tests handler
* Clarify description of test in test_scala_handler
* handle multibyte string when linting text with redpen
* fix error when no string is provided, fix test's expect value
* remove ambiguious `==` operator
* Fixed (g)awk linter
* Made it secure, albeit less useful.
* Added gawk handler; the cpplint one was not working?
* Added gawk handler test.
* added warning to gawk handler.
* added gawk command callback test
* added comment about --source
* added back optional commandline option
* Handle flawfinder severity level
* Reverted code allowing Flawfinder to piggyback off of gcc's format handler
* Gave Flawfinder its own format handler and made requested changes.
Support for elm-format as a fixer has existed since Sept 2017, but it's not
easy to discover because the fixer was named `format`. This breaks the
convention of the other fixers that use the full name in the registry.
I've gone ahead and fixed this discrepancy, but I left the existing registry
entry in place. We should move people towards using `elm-format` as the fixer
name, but I'd hate to break existing setups.
* If a Perl script compiles, there are only warnings and no errors
* Let the first Perl error or warning win.
Take the following example:
***
sub foo {
my $thing;
***
This might have the following messages when we compile it:
Missing right curly or square bracket at warning.pl line 7, at end of
line
syntax error at warning.pl line 7, at EOF
warning.pl had compilation errors.
With the current behaviour, we just get a "syntax error" message, which
isn't all that helpful. With this patch we get "Missing right curly or
square bracket".
* Fix variable scope and pattern matching syntax
* Use named variable to enhance clarity when matching Perl output
* Add more tests for Perl linter
* Remove unnecessary parens
* Simplify check for pattern match
* Add configuration option to open lists vertically
* Add tests, clean up vertical list config
* Vertical list option cleanup
* Use is# for tests
* Order properties in documentation alphabetically
* Flawfinder support added for C and C++
A minor modification to gcc handler was made to support flawfinder's
single-line output format that does not have a space following the
colon denoting the warning level. gcc handler still passes its
Vader tests after this modification.
* Documentation fixes
* Revert documentation regression
* Added Flawfinder to table of contents
* Removed trailing whitespace
* Follow ALE conventions better
Added additional documentation and Vader tests
* Add Elixir linter for dialyxir
* Update doc/ale.txt with dialyxir
* Keep elixir tools alphabetically ordered in README
* Add a missing entry for dialyxir to the main documentation file.
Erubi is yet another parser for eRuby. This is the default parser in
Rails as of version 5.1. It supports some additional syntax with similar
behavior to Rails' extensions to the language, though incompatible.
Rails currently still recommends their own syntax, so GetCommand still
has to do the translation introduced in
https://github.com/w0rp/ale/pull/1114 .
Erubi does not supply an executable—It is intended to be invoked only
from within a Ruby program. In this case, a one-liner on the command
line.
* When working on rust/cargo projects of varying sizes, it may be useful
to either build all possible features (i.e. lint all possible
conditionally compiled code), or even turn off other features for a
quicker edit-lint cycle (e.g. for large projects with large build times)
* Added a g:ale_rust_cargo_default_feature_behavior flag for instructing
cargo to not build any features at all (via `--no-default-features`),
building default features (via no extra flags), or building all possible
features (via `--all-features`)
* Also added a g:ale_rust_cargo_include_features flag for including
arbitrary features to be checked by cargo. When coupled with
g:ale_rust_cargo_default_feature_behavior this allows for full
customization of what features are checked and which ones are ignored
Typically proto files depend on and make use of proto definitions in
other files. When invoking protoc user can supply paths to inspect for
dependencies.
This patch makes it possible to configure flags passed to protoc. This
makes it e.g., possible to change include paths of the linter's protoc
invocation.
The test already handled arbitrary paths reasonably well, but setting
the directory interfered via leakage with others tests for some reason.
This patch removes the call to `SetDirectory` in the fixture setup and
the subsequent cleanup in the teardown as they are not required.
On macOS, Apple's command line toolchain installs very old `tidy`
command (It was released on 31 Oct 2006). It does not consider new specs
such as HTML5 so we should avoid it.
rustfmt normally acts on a file in place, and applies configuration
from rustfmt.toml files according to the path of the file.
Using a temporary file for rustfmt breaks this functionality, so
removing the '%t' from the rustfmt command.
Switches all vale instances to JSON output and provides an appropriate
handler for that. Without JSON, no end_col is provided and text
highlighting only catches the first character of every result.
In Go you can "vendor" packages by putting them in the `vendor/`
directory for a project. Adding the `-srcdir` argument makes `goimports`
pick up these packages, in addition to what you have in GOPATH.
Without this, `goimports` is not very useful, since most projects vendor
their packages.
This grew out of my work in #1193; to ensure the statusline was being
updated I had to add:
fun! s:redraw(timer)
redrawstatus
endfun
augroup ALEProgress
autocmd!
autocmd BufWritePost * call timer_start(100, function('s:redraw'))
autocmd User ALELint redrawstatus
augroup end
Which kind of works, but is ugly. With this, I can replace the
`BufWritePost` with:
autocmd User ALEStartLint redrawstatus
Which is much better, IMHO.
Actually, this patch actually replaces adding a function, since you can
do:
augroup ALEProgress
autocmd!
autocmd User ALEStartLint hi Statusline ctermfg=darkgrey
autocmd User ALELint hi Statusline ctermfg=NONE
augroup end
or:
let s:ale_running = 0
let l:stl .= '%{s:ale_running ? "[linting]" : ""}'
augroup ALEProgress
autocmd!
autocmd User ALEStartLint let s:ale_running = 1 | redrawstatus
autocmd User ALELint let s:ale_running = 0 | redrawstatus
augroup end
Both seem to work very well in my testing.
No need to `ale#Statusline#IsRunning()` anymore, I think?
* puppet: add test for puppet parser validate
* puppet: handle where parser validate doesn't supply the column
* puppet: add test for when parser validate doesn't supply column
* Fix puppet regex to handle Windows paths
- Re: f224ce8a37
- The issues that prompted the above commit which reverted changes made to `go build` and
`gometalinter` seemed to suggest that the main issue was with gometalinter and that
changes should be put into different commits so they are independent of each other
- This commit reinstates the changes to the `go build` linter which seem to be uncontested
and it also seems absolutely necessary to show errors from all files in the package which
may have caused a build failure.
The previous version relied on a zsh-specific behavior where
`<filename` after a pipe could redirect to the first command. This
is the standard way to do it.
* Added filename keys to gobuild and gometalinter
* Removed skipping files not in current package
* Removed `--include` for gometalinter
* Fixed the tests
GetCommand conditionally adds a filter (implemented as inline Ruby code
in the command line) to transform some of the problematic
Rails-specific eRuby syntax. Specifically, <%= tags are replaced with
<%.
This does not reduce the effectiveness of the linter, because the
transformed code is still evaluated.
This solution was suggested by @rgo at
https://github.com/w0rp/ale/issues/580#issuecomment-337676607.
GetCommand conditionally adds a filter (implemented as inline Ruby code
in the command line) to transform some of the problematic
Rails-specific eRuby syntax. Specifically, <%= tags are replaced with
<%.
This does not reduce the effectiveness of the linter, because the
transformed code is still evaluated.
This solution was suggested by @rgo at
https://github.com/w0rp/ale/issues/580#issuecomment-337676607.
Ale saves a temporary file (%t) which does not share the same path as
the original file, breaking import statements with relative URLs.
This fix sends content to `lessc` over stdin and adds
the current file (%s) as one of the included paths, so statements like
`@import '../utils' will correctly resolve based on the current file path.
There were a couple of issues
- `paste` requires a file argument
- `mktemp` requires a pattern argument
- `sort` doesn't support `-h`, but `-n` is enough for sorting on numbers, and `-s` was introduced to perform a stable sort instead.
The main issues were that BSD `sed` does not support:
- Alternation (`\|`) - solved by splitting to multiple patterns
- Bound shortcuts (`x\+`, `x\?`) - solved by replacing with `xx*` and `x\{0,1\}` respectively
- Lower-casing (`\L`) - solved by piping through `tr` instead (this will lowercase everything and not only the integration names, but I assumed that wasn't too much of an issue, as a portable alternative for the selective downcasing would be much more involved).
Looks like elm-make only respects /dev/null, even on Windows. The person
who wrote this linter maybe did not test it on Windows, and wrote the
code in the way you would expect to be solid by using NUL on Windows.
However it seems elm-make is not actually making use of /dev/null but
rather using it as a form of flag. Ironically this seems to be what is
already described in the comments; I added some clarification.
Implements suggestions and recommendations suggested by the first review
of the "Advance C# linter based on mcs -t:module (#952)" pull request.
- Clarifies and simplifies description of linters and options
- Added links to help file and marked the mcsc linter as to be run only
when file in buffer is saved or loaded.
- Added comments to the mcsc.vim file to clarify code
- removed type checks considered not necessary be reviewer.
- addresses findings by vader
- removed call to getcwd and cd in vim script
- handler expands file names relative to route of source tree into
absolute pathes. Fixes errors not being marked when vim is started
from subdirectory of source tree.
- implements tests for mcs.vim and mcsc.vim linter
The existing c-charp linter used the --syntax check mode of the mono mcs
compiler only. The new mcsc linter tries to compile the files located in
a directory tree located bejond the specified source directory or the
current one if no source is explicitly specified. The resulting module
target is placed in a temporary file managed by ale.
The existing c-charp linter used the --syntax check mode of the mono mcs
compiler only. The new mcsc linter tries to compile the files located in
a directory tree located bejond the specified source directory or the
current one if no source is explicitly specified. The resulting module
target is placed in a temporary file managed by ale.
This fixes slim-lint not honoring a `.rubocop.yml` in the file's or
parent directory. Due to the way slim-lint calls rubocop, it requires
the special `SLIM_LINT_RUBUCOP_CONF` env var to pick up the
`.rubocop.yml` if it is not run on the real file (which is the case
here).
See https://github.com/sds/slim-lint/blob/master/lib/slim_lint/linter/README.md#rubocop
* Add prettier fixer support for typescript
* Add prettier fixer support for css and scss
* Add prettier fixer support for json
* Use getbufvar() to get &filetype
* Detect and use CM files for smlnj
* Split into two checkers
- one for CM projects
- one for single SML files
* Fix some typos
* Fix error caught by writing tests
We want to actually use `glob` to search in paths upwards from us.
(Previously we were just searching in the current directory every time!)
* Fix errors from former test run
* Write tests for GetCmFile and GetExecutableSmlnj
* Typo in 'smlnj/' fixture filenames
This linter works by invoking the `thrift` compiler with the buffer
contents and reporting any parser and code generation issues.
The handler rolls its own output-matching loop because we have the
(unfortunate) requirement of handling error output that spans multiple
lines.
Unit tests cover both the command callback and handler, and there is
initial documentation for all of the option variables.
* Add support for prettier configuration file.
As of version 1.6.0, prettier allows passing a `--config` argument with
a path to a configuration file.
* Add test prettier configuration file.
* Add option to use local prettier configuration.
* Add description for new prettier option.
* Also check if the config is present before using it.
Right now ghc-mod linter check temp file instead of current buffer,
which cause the problem that it can't detect cabal file and raise
missing package error.
To fix that we need to run ghc-mod check with actual path of the current
file and with ghc-mod option `--map-file` to redirect temp file source
code to actual one
SwiftFormat is a tool that can be used to format Swift files. This commit adds
support for using SwiftFormat as a fixer from ALE. It looks for executables in
the Pods directory, then the Pods directory for a React Native project, then
finally falls back to the globally installed instance if neither of those were
found.
https://github.com/nicklockwood/SwiftFormat
A limited number of clang-tidy checks can be used with C, too. I pretty much
copied and refactored the C++ clang-tidy linter, and added some documentation
about C-compatible checks.
* Add support for scalastyle
* Add scalastyle docs
* scalastyle support for column numbers
* off by one column
* Add tests for scalastyle command and handler
* update readme for scalastyle
* allow full scalastyle options instead of just config file
* fix indentation
* allow scalastyle config file in parent directories by a couple names.
* check for missing match args with empty
* remove echo
* use a for loop
The handler previously assumed there would be at least one entry in the
'files' array in the output JSON. It looks like this in the normal case:
"files":[{"path":"app/models/image.rb","offenses":[]}]
But if RuboCop's config excludes the specified input files, causing no
files to be linted, the output is emptier:
"files":[]
This change causes the handler to treat that case correctly, and also
exit early if the reported offense_count is zero.
* Move FindRailsRoot() to more general location
* Add rails_best_practices handler (resolves#655)
* Update documentation for rails_best_practices
Also add brakeman to *ale* documentation.
* rails_best_practices: allow overriding the executable
* rails_best_practices: format help correctly
* rails_best_practices: capture tool output on Windows
The real fix was not using absolute paths anymore (so not expanding with the `:p` option). The regex was correct and should at least include the `^` character to make sure the string starts with the given path/filename and not references the path/filename in some error description.
* Look for ini file to spot project root
When looking for the project root folder it would be better
to check for some well-known init file instead of __init__.py.
Indeed, with python3 it is now possible to have namespace modules
where intermediate dirs are not required to include the __init__.py file.
* Break if statement conditions over several lines
* Add blank lines for the if block
* Add test for FindProjectRoot
* Typo: missing / for MANIFEST.in
* Fix test for non-namespace package
* Add more test cases
* Vim scripts shouldn't have hyphens
Especially not ones that will be autoloaded. You can't have a hyphen in
a function name, so autoloading functions based on filename will fail.
* Add g:haskell_stack_build_options, default: --fast
If we're going to use the --fast option, we may as well go the whole 9
yards and let the user configure the 'stack build' flags.
* Create documentation for stack-build options
* Add stack-build linter for Haskell
The stack-build linter works better than the other two linters when
you're working with an entire Haskell project. It builds the project
entirely and reports any errors.
The other two Haskell GHC linters only work on single files, which can
result in spurious errors (for example, not being able to find imports).
* Document all available Haskell linters
* Split GHC checkers into separate files
* Use rubocop's JSON output format (resolves#339)
Rubocop's emacs formatter seems to have changed format in some
not-so-ancient version. The JSON formatter should provide a more stable
interface than parsing lines with a regex.
The JSON formatter was introduced in mid-2013, so it should be safe to
assume available in any reasonably-modern environment. The oldest
currently-supported version of ruby (according to ruby-lang.org) was
not supported by rubocop until 2014.
* Rubocop: Use global function for GetType
* Rubocop: Use scope prefix in GetType
* Rubocop: Update command_callback test
* Rubocop: add end_col to Handle
On macOS, the `$TMPDIR` is in `/var`. However, `/var -> /private/var`.
This means that fully resolved temp filenames weren't always getting
checked against the proper prefix.
This was affecting some of the Haskell plugins, though I'm sure it could
have affected any program that resolved past the generated `$TMPDIR/foo`
and wound up at a different filename.
* Added missing statement about a `<Plug>` mapping.
* Fixed logical mistake related to "option calling".
* Rectified article usage according to a request.
* Use different reporter to support older versions of jscs
* Add test and make more consistent with other code
* Add documentation for jscs
* Add more test coverage
* Add documentation for hadolint (doc/ale-hadolint.txt)
* Allow `hadolint` linter to run via docker image
These changes enable the `hadolint` linter to run via the author's
docker image, if present. Three modes are supported:
* never use docker;
* always use docker; and
* use docker as a failback.
Just to prevent any confusion, the documentation now explicitly states
that setting `g:ale_perl_perlcritic_profile` to an empty string merely
disables passing an explicit profile to `perlcritic` and does not cause
`--no-profile` to be set.
* Adds an option to pass additional arguments to the verilog/verilator linter
The new otion is g:ale_verilog_verilator_options
+ doc
* Spell check verilog linter doc file
* Add entries to the verilog linters in the doc table of content
* Vader test for verilog/verilator linter args option verilog_verilator_options
* Improve elm linter
Some types of errors do not return nice JSON.
Show them on the first line instead of showing nothing.
* Remove unnecessary properties from elm linter
* Add a vader test for elm-make linter
* Test non-JSON elm-make errors are shown
* Add column number to perlcritic linting output
This returns the column number of the perlcritic error so that ale can
show the column in addition to the line where perlcritic found an error.
* Add perlcritic configuration for rule names
This adds a configuration setting so that the name of the perlcritic
rule is shown [Rule::Name] after the error message.
This is useful to lookup the rule failure.
* Add a vader test for perlcritic#GetCommand
* Add :ALEFirst and :ALELast commands
* Add documentation for ALEFirst and ALELast commands
* Add tests for ale#loclist_jumping#JumpToIndex()
* Fix the loclist jumping tests
* Add ktlint support (without formatting) for kotlin filetype
* Fix code style and refactor to use ALE utility functions (GetMatches)
* Remove options for configuration file
* Refactor: Rename exec variable and use ale#Set for variable configuration
* Update list.vim
Set qflist/loclist window title properly ...
* Update list.vim
1. Remove redundant code.
2. Get absolute path from 'a:buffer'.
* Set the list window titles appropriately for each version of Vim, and add tests
* Include span label in rust lints
This turns relatively unhelpful error messages like
mismatched types
into more expressive messages along the lines of
mismatched types: expected bool, found integral variable
Fixes#597.
* Exclude rust lint span label if empty
* Use single-quoted strings in vimscript
* Add test for detailed rust errors
* Prune Cargo JSON
* Use matching error file name
* Byte offsets not char offsets
"-X Disables all warnings regardless of use warnings or $^W". See
"perldoc perlrun" or http://perldoc.perl.org/perlrun.html
With the current defaults, warnings are squashed. For example:
$ perl -X -Mwarnings -c -e'BEGIN { 42 + undef }'
-e syntax OK
$ perl -Mwarnings -c -e'BEGIN { 42 + undef }'
Use of uninitialized value in addition (+) at -e line 1.
-e syntax OK
So, it's not clear from the current defaults whether Ale wants to remove
warnings or enable them. As it stands, it's trying to do both and the
disabling appears to win.
This commit enables warnings by default.
* Improve performance when using gometalinter
Before this change when I opened a big project that had 6000+ warnings/errors it took ages to get the actual warnings/errors and it caused my CPU to be busy for quite some time. The call to gometalinter alone took about 24 seconds, but after that vim was struggling as well.
After this change the gometalinter call just takes 2 seconds and nothing noticable happens with the CPU and/or vim.
* Removed obsolete test
This logic is no longer done by the `ale` plugin, but by `gometalinter` itself.
* Initial attempt at an rpmlint linter.
* Add some basic documentation.
* Play with indentation in the test file.
* Another attempt to fix the rpmlint test.
* Hopefully this does it.
Linter is disabled by default (see g:ale_go_gometalinter_enabled) as it
conflicts with a number of established ALE linters (golint, govet,
gosimple, staticcheck, etc).
* Added ruby mri linter
* Added to the list of supported linters
* Async and now with 4 spaces
* Vader tests for ruby
* Match style choices
* Vader test for the Ruby handler now works and passes
* Adds options to foodcritic linter
Adds a way to pass command line options to the foodcritic command and
documentation about it.
* Creates a simple test for foodcritic command callback
This test simply runs the GetCommand function for the foodcritic linter
and feeds it with some test variables to assert the command line is
being created/escaped correctly.
* Makes foodcritic linter use a command callback
Following review comments, changes the foodcritic linter to use a
`GetCommand` callback for the `command_callback` linter option.
Makes sure that `~` are escaped: flags on foodcritic command line are
negated by adding a `~` in front of the specific cop name:
```
foodcritic -t ~FC011
```
But the way the commands are executed cause foodcritic to fail (since
tilde is recognized as home directory).
* Fixes the doc to include new variables
* Remove 'col' from linters where it is hardcoded to 1
When 'col' is 1, the first column will get highlighted for no reason. It
should be 0 (which is the default).
In the scalac linter there was also a check about the outcome of
`stridx`. It would set l:col to 0 if it was -1, and then it uses
`'col': l:col + 1` to convert the outcome of `stridx` to the actual
column number. This will make 'col' equals 1 when there is no match. We
can remove the check because `-1 + 1 = 0`.
* Remove outdated comments about vcol
vcol was added as a default, and the loclists that follow these comments
do not contain 'vcol' anymore
* [#420] Add options to facilitate linting only in normal mode
ale_lint_on_text_changed:
Allow setting to 'insert' or 'normal' to lint when text is changed only in
insert or normal mode respectively.
ale_lint_on_insert_leave:
This flag can be set to 1 to enable linting when leaving insert mode.
* [#420] Test updated global options
Ale should
- bind to TextChanged events when g:ale_lint_on_text_changed = 1
- bind to TextChanged events when g:ale_lint_on_text_changed = 'always'
- bind to InsertLeave event when g:ale_lint_on_insert_leave = 1
* Fix problems with nim check
- Multi file errors are not shown in the same buffer
- Fixes parsing of error type that contain ':'
* Remove redundant fnameescape
In particular, if we're working with a leex (.xrl) or yecc (.yrl) source
file, erlc would otherwise generate the corresponding .erl file in the
current directory (often the project root), which is generally not what
we want.
Unconditionally writing erlc output to a temporary directory also
matches Flycheck's behavior.
I've noticed that signs weren't unplaced and, learning that this was an issue depending on locale and :sign place, I've fixed the regular expression used to match those messages in my locale (it_IT.UTF-8).
* PHP: Fix column matching for unexpected single quotes
Unexpected single quotes resulted in an empty match, because PHP
surrounds the errors with quotes, and we check for the next quote to be
the ending delimiter.
For example: an unexpected string 'foo' would be presented as
`unexpected ''foo''`, and then the match would be `''`. The inner part
of that match is an empty string.
This adds a check for the keyword "expecting". Any quote after
"expecting" won't be matched, so we can use greedy matching instead of
non-greedy.
* PHP: Use "very magic"
The pattern started to get unreadable
Also replaced non-greedy matching (`\{-}`) by greedy matching, because
we don't need to match non-greedily anymore and it reads a little nicer.
* PHP: Add tests for column matches
And with that, also a test for unexpected single quotes.
* Fix Credo's line-matching pattern
In d3e7d3d5, the line matching pattern was changed to handle filenames
other than `stdin`. Unfortunately, this broke the pattern's ability to
reliably extract both line and column numbers because the latter is an
optional match and the filename portion was very greedy. This resulted
in line numbers being discarded (treated as part of the filename) and
column numbers being interpreted as line numbers.
This change simplifies the pattern to only anchor on the line's suffix,
ignoring the filename portion entirely.
Alternatively, we could use vim's `\f` ("file name characters") class,
but that could still run into problems when `:`'s naturally appear in
the filename.
* Add a Vader test case for the Credo handler
This adds support for the hdevtools haskell linter
https://github.com/hdevtools/hdevtools
The output for hdevtools is near identical to the ghc output so this
also extracts the ghc handler into the handle file and adds tests
* Add testing for previous major release of ghc
This adds support for the hdevtools haskell linter
https://github.com/hdevtools/hdevtools
The output for hdevtools is near identical to the ghc output so this
also extracts the ghc handler into the handle file and adds tests
* A try at javac support for ALE
* Small cleanup: moved '/tmp/java_ale' string into script var
* Fixed Travis-CI build failing on autocmd not being in augroup and stupid omission
* One more fix for Travis-CI
* For some reason, expandtab was not set
* Indentation and removal of header guard.
Used examples from ale_linters/c/gcc.vim and
ale_linters/javascript/eslint.vim for the indentation of string concat blocks.
By default, Credo attributes input from STDIN as though it came from a
file named `stdin`. This change passes the buffer's filename, too, so
that Credo can use that information when applying its configuration.
This is a nice improvement because files like `mix.exs` are normally
excluded from Credo-based linting. Previously, ALE would show lint
warnings for those files as they were edited. Now, they are correctly
honor the Credo configuration and don't produce lint output.
* try fixing go build
* cache some system calls
* fix /dev/null
* use chained commands, use `go test -c` instead of `go tool compile`
* fix some unescaped shell commands
* fix a bug with explicitly setting GOPATH
* implement changes requested in code review. handle errors from multiple files. fix issue when starting a new package
* run `go env` as a job
* ensure all functions return the proper type
* fix loclist line numbers in some cases
* remove multibuffer support for now
In my previous change, I updated the Rubocop linter to pass the filename
to Rubocop. This change was tested on a file I expected Rubocop to
ignore and the experience in vim was as I expected. However, I soon
found that ALE wasn't finding errors in files that should not be
ignored. After investigation, I found a few issues that this commit
fixes:
1. We were not properly passing the current filename. We now use
`expand` to get the filename.
2. The regular expression used in the callback was expecting the static
value of `_` for the filename in output. We now use a looser regular
expression that begins matching on the first `:`.
3. The linter was defined statically. By using the current filename when
defining the command the linter would always use the filename of the
first Ruby file the user opened. We now use a `command_callback` to
inject the proper filename.
I tested these changes on a configuration with included and excluded
files and found it to work as I expected. Apologies for the earlier
incorrect change.
When using `--stdin`, Rubocop requires that you also pass the associated
file name. ALE was previously passing `_` as the filename. By passing
the actual relative path to the file and enabling the
`--force-exclusion` option, we can get Rubocop to respect excluded files
in the configuration.
Closes#197
* Add ALEInfo command to get list of available/enabled linters for current filetype
* Add Vader tests for ALEInfo command
* Fix ALEInfo tests breaking CI by echoing too much output to screen
* Speculative change to Makefile which seems to fix test hanging problem locally.
* Fix Vader tests to not require a TTY
* Add erlc lint for Erlang (#248)
* Ignore certain errors in Erlang .hrl files (#248)
A .hrl file does not need to have a -module definition. Additionally, it
is common to have unused elements in such a file, as the entities will
be used in a file including the header.
* Address change requests to Erlang linter
* Add option to open loclist/quicklist when there are errors
I copied PR #137, and tries to complete it by correcting some issues and
adding vader tests.
About tests, first time with vader, can you give some feedback if there
are what you expected in PR #137.
* Remove old code + fix indent issue
* add g:ale_keep_list_window_open option
* Correct bug with keep open option
* Add comment into vader file
* Fix errors for Travis CI build
* Support netcore project linting.
* Support check on the fly.
* Remove debug.
* Rename csc.vim to mcs.vim as it should be.
* Update README.
* Update doc.
* Using `=~#` instead of `=~`.
* Add rustc checker for rust files
* Add documentation for rustc
* Use a nice helper function
* Add cargo as linter
* Complete the doc for rust linters
* Put l: in front of every local variable
* Apply the requested stylistic changes
This makes php output more specific error messages. The format is the normal one ALE expects, but on some systems ALE does not work with PHP unless the display_errors=1 option is used. Without that option php will only output a generic message without a line number like "Errors parsing index.php"
* add go build for build errors
* Add go build to doc and README
* Improvement for Go build
Go build works on package level, so copy over the other files
that belong to the same package to the temp folder as well.
* revert back to simple go build
* change gobuild script var name
* Filters out unrelated errors in Elm linter
The function now filters out errors that are unrelated to the file,
those that were found in imported modules.
It does this by comparing the temp directory environment variable to the
file name in the elm output. If the file begins with the temp directory,
then it sould be included (it's from the buffer).
* Changing output to '/dev/null'
Turns out the compiler only accepts /dev/null as an ignorable name. It's
hard-coded here
https://github.com/elm-lang/elm-make/blob/master/src/Flags.hs
Changing this allows Windows linting to work. Otherwise the compiler
errors when using "nul"
* Fixes for Windows
Should now be able to successfully handle Windows.
Windows seemed to not handle the ";" properly, so I switched it to "&&",
which probably should've been done anyway to prevent false positives.
Oddly, matchend(l:error.file, l:temp_dir), and various other regex
solutions, couldn't properly match the two. Subsetting did though, hence
the new solution.
* Applying corrections
Made the file check case-insensitive for Windows, case-sensitive for
Unix/non-windows.
Added comment explaining hard coding of 'dev/null'
* Spelling correction
* Minor corrections
Actually uses the is_file_buffer variable now, added space between the
if statements, and added space between '-'
neovim/neovim#5529 merged support for Vim's partial functions, which
made nvim more strict about dictionary functions and callbacks, to
match Vim behavior.
For ghc, it seemed that the conditional
```
if l:corrected_lines[-1] =~# ': error:$'
let l:line = substitute(l:line, '\v^\s+', ' ', '')
endif
```
was never being reached. It's actually better to unconditionally
collapse whitespace anyway and so I simply removed the conditional
check.
For hlint I added more information about the error. This changes the
reported error from `Error:` to something like:
` Error: Avoid lambda. Found: \ x -> foo x Why not: foo`
* Add support for Elm linting
* Adding documentation for Elm
* Adjusting spacing
* Addressing concerns listed in pull request
Removed the s:FindRootDirectory function as it does not make much sense
in this context. Adjusted the rest of the code to handle the removal of
that function, including using the ale#util function to find the nearest
file.
Ensured that when an empty filepath is found, the code does not attempt
to change directories.
Ensured that the linter would take from stdin using the wrapper.
* Add chktex linter
* Alias plaintex to tex
* Add lacheck linter
Closes#179
* Add the chktex warning code
This very useful to have when you want to suppress lint warnings with LaTeX
comments. chktex tends to be a bit noisy so this often needed.
* lacheck: Make regex less specific
To be more robust future changes in `stdin-wrapper`
* Start adding Puppet linters
* Use the correct output stream for puppet parser
* Finish Puppet and puppet-lint linters
* Add Puppet information to documentation
* Fix flow linter to provide filename of the buffer
Related #173
* Fix flow linter not to fail on empty response
* Various improvement to message parsing
Adding support the foodcritic linter for Chef files.
Listing all issues as warnings for now
Doesn't get in the way of rubocop linting if ft=ruby.chef
Updated documentation
Closes#127
* Add `javascript/flow` linter
* Add documentation for flow
* Remove a line from the docs that was from eslint
* Only run if flow gives output; Correct link in doc
* Address PR feedback #157
Shellcheck is smart enough to check the shebang in a given file to
determine which dialect to use. Unfortunately this doesn't work for
files without shebangs, even if it might be apparent what dialect should
be used, such as "bashrc" or "foo.bash". Luckily `filetype.vim` defines
specific vars based on which shell dialect is being used based on a huge
list of conditions. With this change we take those into account for all
the types shellcheck supports, otherwise we fallback to letting it try
and decide.
* Use `cat` instead of `read -r` to stream stdin to file
* Cleanup dmd-wrapper
* Fix typo
* Make wrapper work on macOS
* Use fifo instead of temporary file
* Fix stdin-wrapper
* Use `awk` instead of `read` hackery
* Finish refactoring
* Fix `exec` issue
* Add myself as an coauthor of wrapper scripts (no shame at all :P)
* Fix dmd-wrapper
* Extract check_dubfile
This PR first and formost implements support for dot-seperate filetypes,
a very trivial change.
This closes#132
But more importantly, this PR vastly improves the test quality for
`ale#linter#Get`. It enables us to reset the state of ale's internal
linter cache, to facilitate better testing, as well as making use of
mocked linters instead of depending on linters on disk (which may
change). In addition, a dummy linter is defined to test the autoloading
behavior.
Header guards were removed from all linters as:
* A: ale won't try and load linters if they already exist in memory
* B: we can't reset state for testing if they can't be loaded again
This is the template for reporting ALE bugs. Make sure you try updating ALE
to a more recent version before reporting a bug. Look through existing bug
reports for similar issues before reporting a new one. Don't leave comments
about new bugs in the comment section for old issues.
Make sure to try disabling other plugins and trying to repeat your bug before
reporting it in ALE. Some times problems can arise when two plugins are used
together, but often your issues might be problems with other plugins.
-->
## Information
**VIM version**
<!-- Paste just the first two lines of :version here. -->
Operating System: <!-- Describe your operating system version. -->
## What went wrong
<!-- Describe what went wrong here. Be specific. -->
Something went wrong in specifically this place, and I also searched through both open and closed issues for the same problem before reporting a bug here.
Are you having trouble configuring ALE? Try asking for help on [Stack Exchange](https://vi.stackexchange.com/) or perhaps on [Reddit](https://www.reddit.com/r/vim/) instead. The GitHub issue tracker should be used for reporting bugs or asking for new features.
## Reproducing the bug
<!-- Write a list of steps below. -->
1. I did this.
2. Then this happened.
### :ALEInfo
<details>
<summary>Expand</summary>
<!-- Paste the output of :ALEInfo here. Try :ALEInfo -clipboard -->
<!-- Make sure to run :ALEInfo from the buffer where the bug occurred. -->
<!-- Read the output. You might figure out what went wrong yourself. -->
" warning: (3) unable to parse response signature, expected 'response [<HTTP status code>] [(<media type>)]'; line 4, column 3k - line 4, column 22
" warning: (10) message-body asset is expected to be a pre-formatted code block, separate it by a newline and indent every of its line by 12 spaces or 3 tabs; line 30, column 5 - line 30, column 9; line 31, column 9 - line 31, column 14; line 32, column 9 - line 32, column 14
letl:pattern='\(^.*\): (\d\+) \(.\{-\}\); line \(\d\+\), column \(\d\+\) - line \d\+, column \d\+\(.*; line \d\+, column \d\+ - line \(\d\+\), column \(\d\+\)\)\{-\}$'
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
