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.
