Skip to main content

A language server for SPARQL

Project description

🦀 Qlue-ls 🦀

⚡Qlue-ls (pronounced "clueless") is a blazingly fast language server for SPARQL, written in Rust 🦀.

To learn more about the origin story of this project, read the blog post.

🚀 Getting Started

📦 Installation

Qlue-ls is available on crate.io:

cargo install qlue-ls

And on PyPI:

pipx install qlue-ls

You can also build it from source:

git clone https://github.com/IoannisNezis/Qlue-ls.git
cd Qlue-ls
cargo build --release --bin qlue-ls

CLI Usage

To run Qlue-ls as formatter run:

qlue-ls format <PATH>

To run Qlue-ls as language server run:

qlue-ls server

This will create a language server listening on stdio.

with Neovim

After you installed the language server, add this to your init.lua:

vim.api.nvim_create_autocmd({ 'FileType' }, {
  desc = 'Connect to Qlue-ls',
  pattern = { 'sparql' },
  callback = function()
    vim.lsp.start {
      name = 'qlue-ls',
      cmd = { 'qlue-ls', 'server' },
      root_dir = vim.fn.getcwd(),
      on_attach = function(client, bufnr)
        vim.keymap.set('n', '<leader>f', vim.lsp.buf.format, { buffer = bufnr, desc = 'LSP: ' .. '[F]ormat' })
      end,
    }
  end,
})

Open a .rq file and check that the buffer is attached to th server:

:checkhealth lsp

Configure keymaps in on_attach function.

🚀 Capabilities

📐 Formatting

Status: Full support

Formats SPARQL queries to ensure consistent and readable syntax. Customizable options to align with preferred query styles are also implemented.

🩺 Diagnostics

Status: Partial support

Diagnostics provide feadback on the query.
Diagnostics come in severity: error, warning and info

** provided diagnostics**:

Type Name Description
❌ error undefined prefix a used prefix is not declared
❌ error ungrouped select variable selected variable is not grouped
❌ error invalid projection variable projection variable is taken
⚠️ warning unused prefix a declared prefix is not used
ℹ️ info uncompacted uri a raw uncompacted uri is used

✨ Completion

Status: Full support

Completion provides suggestions how the query could continue.

For completion of subjects, predicates or objects the language server sends completion-queries to the backend and gets the completions from the knowledge-graph.

These completion queries have to be configured for each knowledge-graph.

🛠️ Code Actions

Status: Partial support

name description diagnostic
shorten uri shorten uri into compacted form uncompacted uri
declare prefix declares undeclared prefix (if known) undeclared prefix
shorten all uri's shorten all uri's into compacted form
add to result add variable to selected result
filter variable add filter for this variable

⚙️ Configuration

Qlue-ls can be configured through a qlue-ls.toml or qlue-ls.yml file.

Here is the full default configuration

[format]
align_predicates = true
align_prefixes = true
separate_prolouge = false
capitalize_keywords = true
insert_spaces = true
tab_size = 10
where_new_line = true
filter_same_line = true

[completion]
timeout_ms = 5000
result_size_limit = 42

🌐 use in web

If you want to connect from a web-based-editor, you can use this package as well.
For this purpose this can be compiled to wasm and is available on npm:

npm i qlue-ls

You will have to wrap this in a Web Worker and provide a language server client. There will be more documentation on this in the future...

Until then you can check out the demo in ./editor/

🏗 Development Setup

Here is a quick guide to set this project up for development.

Requirements

Initial Setup

You will only have to do this once.

In the justfile and Makefile you will find the target init_dev, run it:

just init_dev

or

make init_dev

It will:

  • install node dependencies
  • build wasm binaries
  • link against local packages
  • run the vite dev server

If you don't have just or make installed:

Install just

Automatically rebuild on change

When developping the cycle is:

  • Change the code
  • Compile to wasm (or run tests)
  • Evaluate

To avoid having to run a command each time to Compile I strongly recommend setting up a auto runner like watchexec.

watchexec --restart --exts rs --exts toml just build-wasm

or just:

just watch-and-run build-wasm

have fun!

🙏 Special Thanks

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

qlue_ls-0.9.0.tar.gz (712.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

qlue_ls-0.9.0-py3-none-manylinux_2_34_x86_64.whl (17.7 MB view details)

Uploaded Python 3manylinux: glibc 2.34+ x86-64

File details

Details for the file qlue_ls-0.9.0.tar.gz.

File metadata

  • Download URL: qlue_ls-0.9.0.tar.gz
  • Upload date:
  • Size: 712.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: maturin/1.8.6

File hashes

Hashes for qlue_ls-0.9.0.tar.gz
Algorithm Hash digest
SHA256 354b14a051fe341c5e0bce1cbb45b20c32808680389fdf3007ae18f1b76b0e04
MD5 495ced0dd212d1c65d1055aa217b269e
BLAKE2b-256 9dbc77e66f6d66e196f1eeffd28defc3251d6f4cf0b8cd1c6f5eb5d8b6631b56

See more details on using hashes here.

File details

Details for the file qlue_ls-0.9.0-py3-none-manylinux_2_34_x86_64.whl.

File metadata

File hashes

Hashes for qlue_ls-0.9.0-py3-none-manylinux_2_34_x86_64.whl
Algorithm Hash digest
SHA256 5dcfbeaabbe74f1d9fe6116592356e88afcbc4f73b7a6206a32eb576b8c0252b
MD5 e000f3731c6abbf216b35c7a2ca1571c
BLAKE2b-256 2334c64ad409a974ac3f5fe2828f6ed4e4f2a2164b7245679c9dc3237a60121d

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page