77dbbc7 · helix · git.iko.soy

git.iko.soy/helix

Detect filetype from shebang line (#1001)

77dbbc7ath35 years ago

book/src/guides/adding_languages.md+2−1

@@ -33,10 +33,11 @@ These are the available keys and descriptions for the file.

33| scope | A string like `source.js` that identifies the language. Currently, we strive to match the scope names used by popular TextMate grammars and by the Linguist library. Usually `source.<name>` or `text.<name>` in case of markup languages |

34| injection-regex | regex pattern that will be tested against a language name in order to determine whether this language should be used for a potential [language injection][treesitter-language-injection] site. |

35| file-types | The filetypes of the language, for example `["yml", "yaml"]` |

36+| shebangs | The interpreters from the shebang line, for example `["sh", "bash"]` |

36| roots | A set of marker files to look for when trying to find the workspace root. For example `Cargo.lock`, `yarn.lock` |

37| roots | A set of marker files to look for when trying to find the workspace root. For example `Cargo.lock`, `yarn.lock` |

37| auto-format | Whether to autoformat this language when saving |

38| auto-format | Whether to autoformat this language when saving |

38| comment-token | The token to use as a comment-token |

39| comment-token | The token to use as a comment-token |

39−| indent | The indent to use. Has sub keys `tab-width` and `unit` |

40+| indent | The indent to use. Has sub keys `tab-width` and `unit` |

40| config | Language server configuration |

41| config | Language server configuration |

42## Queries

43## Queries

helix-core/src/indent.rs+1

@@ -450,6 +450,7 @@ where

450 language: vec![LanguageConfiguration {

451 scope: "source.rust".to_string(),

452 file_types: vec!["rs".to_string()],

453+ shebangs: vec![],

453 language_id: "Rust".to_string(),

454 language_id: "Rust".to_string(),

454 highlight_config: OnceCell::new(),

455 highlight_config: OnceCell::new(),

455 config: None,

456 config: None,

helix-core/src/syntax.rs+24

@@ -14,6 +14,8 @@ use std::{

14 cell::RefCell,

15 collections::{HashMap, HashSet},

16 fmt,

17+ fs::File,

18+ io::Read,

17 path::Path,

19 path::Path,

18 sync::Arc,

20 sync::Arc,

19};

21};

@@ -52,6 +54,7 @@ pub struct LanguageConfiguration {

52 pub language_id: String,

54 pub language_id: String,

53 pub scope: String, // source.rust

55 pub scope: String, // source.rust

54 pub file_types: Vec<String>, // filename ends_with? <Gemfile, rb, etc>

56 pub file_types: Vec<String>, // filename ends_with? <Gemfile, rb, etc>

57+ pub shebangs: Vec<String>, // interpreter(s) associated with language

55 pub roots: Vec<String>, // these indicate project roots <.git, Cargo.toml>

58 pub roots: Vec<String>, // these indicate project roots <.git, Cargo.toml>

56 pub comment_token: Option<String>,

59 pub comment_token: Option<String>,

@@ -254,6 +257,7 @@ pub struct Loader {

254 // highlight_names ?

257 // highlight_names ?

255 language_configs: Vec<Arc<LanguageConfiguration>>,

258 language_configs: Vec<Arc<LanguageConfiguration>>,

256 language_config_ids_by_file_type: HashMap<String, usize>, // Vec<usize>

259 language_config_ids_by_file_type: HashMap<String, usize>, // Vec<usize>

260+ language_config_ids_by_shebang: HashMap<String, usize>,

257}

261}

258

262

259impl Loader {

263impl Loader {

@@ -261,6 +265,7 @@ impl Loader {

261 let mut loader = Self {

265 let mut loader = Self {

262 language_configs: Vec::new(),

266 language_configs: Vec::new(),

263 language_config_ids_by_file_type: HashMap::new(),

267 language_config_ids_by_file_type: HashMap::new(),

268+ language_config_ids_by_shebang: HashMap::new(),

264 };

269 };

265

270

266 for config in config.language {

271 for config in config.language {

@@ -273,6 +278,11 @@ impl Loader {

273 .language_config_ids_by_file_type

278 .language_config_ids_by_file_type

274 .insert(file_type.clone(), language_id);

279 .insert(file_type.clone(), language_id);

275 }

280 }

281+ for shebang in &config.shebangs {

282+ loader

283+ .language_config_ids_by_shebang

284+ .insert(shebang.clone(), language_id);

285+ }

276

286

277 loader.language_configs.push(Arc::new(config));

287 loader.language_configs.push(Arc::new(config));

278 }

288 }

@@ -298,6 +308,20 @@ impl Loader {

298 // TODO: content_regex handling conflict resolution

308 // TODO: content_regex handling conflict resolution

299 }

309 }

300

310

311+ pub fn language_config_for_shebang(&self, path: &Path) -> Option<Arc<LanguageConfiguration>> {

312+ // Read the first 128 bytes of the file. If its a shebang line, try to find the language

313+ let file = File::open(path).ok()?;

314+ let mut buf = String::with_capacity(128);

315+ file.take(128).read_to_string(&mut buf).ok()?;

316+ static SHEBANG_REGEX: Lazy<Regex> =

317+ Lazy::new(|| Regex::new(r"^#!\s*(?:\S*[/\\](?:env\s+)?)?([^\s\.\d]+)").unwrap());

318+ let configuration_id = SHEBANG_REGEX

319+ .captures(&buf)

320+ .and_then(|cap| self.language_config_ids_by_shebang.get(&cap[1]));

321+

322+ configuration_id.and_then(|&id| self.language_configs.get(id).cloned())

323+ }

324+

301 pub fn language_config_for_scope(&self, scope: &str) -> Option<Arc<LanguageConfiguration>> {

325 pub fn language_config_for_scope(&self, scope: &str) -> Option<Arc<LanguageConfiguration>> {

302 self.language_configs

326 self.language_configs

303 .iter()

327 .iter()

helix-view/src/document.rs+3−1

@@ -494,7 +494,9 @@ impl Document {

494 /// Detect the programming language based on the file type.

495 pub fn detect_language(&mut self, theme: Option<&Theme>, config_loader: &syntax::Loader) {

496 if let Some(path) = &self.path {

497+ let language_config = config_loader

497− let language_config = config_loader.language_config_for_file_name(path);

498+ .language_config_for_file_name(path)

499+ .or_else(|| config_loader.language_config_for_shebang(path));

498 self.set_language(theme, language_config);

500 self.set_language(theme, language_config);

499 }

501 }

500 }

502 }

languages.toml+35

@@ -3,6 +3,7 @@ name = "rust"

3scope = "source.rust"

4injection-regex = "rust"

5file-types = ["rs"]

6+shebangs = []

6roots = []

7roots = []

7auto-format = true

8auto-format = true

8comment-token = "//"

9comment-token = "//"

@@ -17,6 +18,7 @@ name = "toml"

17scope = "source.toml"

18scope = "source.toml"

18injection-regex = "toml"

19injection-regex = "toml"

19file-types = ["toml"]

20file-types = ["toml"]

21+shebangs = []

20roots = []

22roots = []

21comment-token = "#"

23comment-token = "#"

@@ -27,6 +29,7 @@ name = "protobuf"

27scope = "source.proto"

29scope = "source.proto"

28injection-regex = "protobuf"

30injection-regex = "protobuf"

29file-types = ["proto"]

31file-types = ["proto"]

32+shebangs = []

30roots = []

33roots = []

31comment-token = "//"

34comment-token = "//"

@@ -37,6 +40,7 @@ name = "elixir"

37scope = "source.elixir"

40scope = "source.elixir"

38injection-regex = "elixir"

41injection-regex = "elixir"

39file-types = ["ex", "exs"]

42file-types = ["ex", "exs"]

43+shebangs = []

40roots = []

44roots = []

41comment-token = "#"

45comment-token = "#"

@@ -48,6 +52,7 @@ name = "mint"

48scope = "source.mint"

52scope = "source.mint"

49injection-regex = "mint"

53injection-regex = "mint"

50file-types = ["mint"]

54file-types = ["mint"]

55+shebangs = []

51roots = []

56roots = []

52comment-token = "//"

57comment-token = "//"

@@ -59,6 +64,7 @@ name = "json"

59scope = "source.json"

64scope = "source.json"

60injection-regex = "json"

65injection-regex = "json"

61file-types = ["json"]

66file-types = ["json"]

67+shebangs = []

62roots = []

68roots = []

64indent = { tab-width = 2, unit = " " }

70indent = { tab-width = 2, unit = " " }

@@ -68,6 +74,7 @@ name = "c"

68scope = "source.c"

74scope = "source.c"

69injection-regex = "c"

75injection-regex = "c"

70file-types = ["c"] # TODO: ["h"]

76file-types = ["c"] # TODO: ["h"]

77+shebangs = []

71roots = []

78roots = []

72comment-token = "//"

79comment-token = "//"

@@ -79,6 +86,7 @@ name = "cpp"

79scope = "source.cpp"

86scope = "source.cpp"

80injection-regex = "cpp"

87injection-regex = "cpp"

81file-types = ["cc", "hh", "cpp", "hpp", "h", "ipp", "tpp", "cxx", "hxx", "ixx", "txx", "ino"]

88file-types = ["cc", "hh", "cpp", "hpp", "h", "ipp", "tpp", "cxx", "hxx", "ixx", "txx", "ino"]

89+shebangs = []

82roots = []

90roots = []

83comment-token = "//"

91comment-token = "//"

@@ -90,6 +98,7 @@ name = "c-sharp"

90scope = "source.csharp"

98scope = "source.csharp"

91injection-regex = "c-?sharp"

99injection-regex = "c-?sharp"

92file-types = ["cs"]

100file-types = ["cs"]

101+shebangs = []

93roots = []

102roots = []

94comment-token = "//"

103comment-token = "//"

104

@@ -100,6 +109,7 @@ name = "go"

100scope = "source.go"

109scope = "source.go"

101injection-regex = "go"

110injection-regex = "go"

102file-types = ["go"]

111file-types = ["go"]

112+shebangs = []

103roots = ["Gopkg.toml", "go.mod"]

113roots = ["Gopkg.toml", "go.mod"]

104auto-format = true

114auto-format = true

105comment-token = "//"

115comment-token = "//"

@@ -113,6 +123,7 @@ name = "javascript"

113scope = "source.js"

123scope = "source.js"

114injection-regex = "^(js|javascript)$"

124injection-regex = "^(js|javascript)$"

115file-types = ["js", "mjs"]

125file-types = ["js", "mjs"]

126+shebangs = []

116roots = []

127roots = []

117comment-token = "//"

128comment-token = "//"

118# TODO: highlights-jsx, highlights-params

129# TODO: highlights-jsx, highlights-params

@@ -124,6 +135,7 @@ name = "typescript"

124scope = "source.ts"

135scope = "source.ts"

125injection-regex = "^(ts|typescript)$"

136injection-regex = "^(ts|typescript)$"

126file-types = ["ts"]

137file-types = ["ts"]

138+shebangs = []

127roots = []

139roots = []

128# TODO: highlights-jsx, highlights-params

140# TODO: highlights-jsx, highlights-params

129

141

@@ -135,6 +147,7 @@ name = "tsx"

135scope = "source.tsx"

147scope = "source.tsx"

136injection-regex = "^(tsx)$" # |typescript

148injection-regex = "^(tsx)$" # |typescript

137file-types = ["tsx"]

149file-types = ["tsx"]

150+shebangs = []

138roots = []

151roots = []

139# TODO: highlights-jsx, highlights-params

152# TODO: highlights-jsx, highlights-params

140

153

@@ -146,6 +159,7 @@ name = "css"

146scope = "source.css"

159scope = "source.css"

147injection-regex = "css"

160injection-regex = "css"

148file-types = ["css"]

161file-types = ["css"]

162+shebangs = []

149roots = []

163roots = []

150

164

151indent = { tab-width = 2, unit = " " }

165indent = { tab-width = 2, unit = " " }

@@ -155,6 +169,7 @@ name = "html"

155scope = "text.html.basic"

169scope = "text.html.basic"

156injection-regex = "html"

170injection-regex = "html"

157file-types = ["html"]

171file-types = ["html"]

172+shebangs = []

158roots = []

173roots = []

159

174

160indent = { tab-width = 2, unit = " " }

175indent = { tab-width = 2, unit = " " }

@@ -164,6 +179,7 @@ name = "python"

164scope = "source.python"

179scope = "source.python"

165injection-regex = "python"

180injection-regex = "python"

166file-types = ["py"]

181file-types = ["py"]

182+shebangs = ["python"]

167roots = []

183roots = []

168comment-token = "#"

184comment-token = "#"

169

185

@@ -176,6 +192,7 @@ name = "nix"

176scope = "source.nix"

192scope = "source.nix"

177injection-regex = "nix"

193injection-regex = "nix"

178file-types = ["nix"]

194file-types = ["nix"]

195+shebangs = []

179roots = []

196roots = []

180comment-token = "#"

197comment-token = "#"

181

198

@@ -187,6 +204,7 @@ name = "ruby"

187scope = "source.ruby"

204scope = "source.ruby"

188injection-regex = "ruby"

205injection-regex = "ruby"

189file-types = ["rb"]

206file-types = ["rb"]

207+shebangs = ["ruby"]

190roots = []

208roots = []

191comment-token = "#"

209comment-token = "#"

192

210

@@ -198,6 +216,7 @@ name = "bash"

198scope = "source.bash"

216scope = "source.bash"

199injection-regex = "bash"

217injection-regex = "bash"

200file-types = ["sh", "bash"]

218file-types = ["sh", "bash"]

219+shebangs = ["sh", "bash", "dash"]

201roots = []

220roots = []

202comment-token = "#"

221comment-token = "#"

203

222

@@ -209,6 +228,7 @@ name = "php"

209scope = "source.php"

228scope = "source.php"

210injection-regex = "php"

229injection-regex = "php"

211file-types = ["php"]

230file-types = ["php"]

231+shebangs = ["php"]

212roots = []

232roots = []

213

233

214indent = { tab-width = 4, unit = " " }

234indent = { tab-width = 4, unit = " " }

@@ -218,6 +238,7 @@ name = "latex"

218scope = "source.tex"

238scope = "source.tex"

219injection-regex = "tex"

239injection-regex = "tex"

220file-types = ["tex"]

240file-types = ["tex"]

241+shebangs = []

221roots = []

242roots = []

222comment-token = "%"

243comment-token = "%"

223

244

@@ -228,6 +249,7 @@ name = "julia"

228scope = "source.julia"

249scope = "source.julia"

229injection-regex = "julia"

250injection-regex = "julia"

230file-types = ["jl"]

251file-types = ["jl"]

252+shebangs = []

231roots = []

253roots = []

232comment-token = "#"

254comment-token = "#"

233language-server = { command = "julia", args = [

255language-server = { command = "julia", args = [

@@ -253,6 +275,7 @@ name = "java"

253scope = "source.java"

275scope = "source.java"

254injection-regex = "java"

276injection-regex = "java"

255file-types = ["java"]

277file-types = ["java"]

278+shebangs = []

256roots = []

279roots = []

257indent = { tab-width = 4, unit = " " }

280indent = { tab-width = 4, unit = " " }

258

281

@@ -261,6 +284,7 @@ name = "ledger"

261scope = "source.ledger"

284scope = "source.ledger"

262injection-regex = "ledger"

285injection-regex = "ledger"

263file-types = ["ldg", "ledger", "journal"]

286file-types = ["ldg", "ledger", "journal"]

287+shebangs = []

264roots = []

288roots = []

265comment-token = ";"

289comment-token = ";"

266indent = { tab-width = 4, unit = " " }

290indent = { tab-width = 4, unit = " " }

@@ -270,6 +294,7 @@ name = "ocaml"

270scope = "source.ocaml"

294scope = "source.ocaml"

271injection-regex = "ocaml"

295injection-regex = "ocaml"

272file-types = ["ml"]

296file-types = ["ml"]

297+shebangs = []

273roots = []

298roots = []

274comment-token = "(**)"

299comment-token = "(**)"

275indent = { tab-width = 2, unit = " " }

300indent = { tab-width = 2, unit = " " }

@@ -278,6 +303,7 @@ indent = { tab-width = 2, unit = " " }

278name = "ocaml-interface"

303name = "ocaml-interface"

279scope = "source.ocaml.interface"

304scope = "source.ocaml.interface"

280file-types = ["mli"]

305file-types = ["mli"]

306+shebangs = []

281roots = []

307roots = []

282comment-token = "(**)"

308comment-token = "(**)"

283indent = { tab-width = 2, unit = " "}

309indent = { tab-width = 2, unit = " "}

@@ -286,6 +312,7 @@ indent = { tab-width = 2, unit = " "}

286name = "lua"

312name = "lua"

287scope = "source.lua"

313scope = "source.lua"

288file-types = ["lua"]

314file-types = ["lua"]

315+shebangs = []

289roots = []

316roots = []

290comment-token = "--"

317comment-token = "--"

291indent = { tab-width = 2, unit = " " }

318indent = { tab-width = 2, unit = " " }

@@ -295,6 +322,7 @@ name = "svelte"

295scope = "source.svelte"

322scope = "source.svelte"

296injection-regex = "svelte"

323injection-regex = "svelte"

297file-types = ["svelte"]

324file-types = ["svelte"]

325+shebangs = []

298roots = []

326roots = []

299indent = { tab-width = 2, unit = " " }

327indent = { tab-width = 2, unit = " " }

300language-server = { command = "svelteserver", args = ["--stdio"] }

328language-server = { command = "svelteserver", args = ["--stdio"] }

@@ -305,6 +333,7 @@ name = "vue"

305scope = "source.vue"

333scope = "source.vue"

306injection-regex = "vue"

334injection-regex = "vue"

307file-types = ["vue"]

335file-types = ["vue"]

336+shebangs = []

308roots = []

337roots = []

309indent = { tab-width = 2, unit = " " }

338indent = { tab-width = 2, unit = " " }

310

339

@@ -312,6 +341,7 @@ indent = { tab-width = 2, unit = " " }

312name = "yaml"

341name = "yaml"

313scope = "source.yaml"

342scope = "source.yaml"

314file-types = ["yml", "yaml"]

343file-types = ["yml", "yaml"]

344+shebangs = []

315roots = []

345roots = []

316comment-token = "#"

346comment-token = "#"

317indent = { tab-width = 2, unit = " " }

347indent = { tab-width = 2, unit = " " }

@@ -331,6 +361,7 @@ name = "zig"

331scope = "source.zig"

361scope = "source.zig"

332injection-regex = "zig"

362injection-regex = "zig"

333file-types = ["zig"]

363file-types = ["zig"]

364+shebangs = []

334roots = ["build.zig"]

365roots = ["build.zig"]

335auto-format = true

366auto-format = true

336comment-token = "//"

367comment-token = "//"

@@ -343,6 +374,7 @@ name = "prolog"

343scope = "source.prolog"

374scope = "source.prolog"

344roots = []

375roots = []

345file-types = ["pl", "prolog"]

376file-types = ["pl", "prolog"]

377+shebangs = ["swipl"]

346comment-token = "%"

378comment-token = "%"

347

379

348language-server = { command = "swipl", args = [

380language-server = { command = "swipl", args = [

@@ -354,6 +386,7 @@ language-server = { command = "swipl", args = [

354name = "tsq"

386name = "tsq"

355scope = "source.tsq"

387scope = "source.tsq"

356file-types = ["scm"]

388file-types = ["scm"]

389+shebangs = []

357roots = []

390roots = []

358comment-token = ";"

391comment-token = ";"

359indent = { tab-width = 2, unit = " " }

392indent = { tab-width = 2, unit = " " }

@@ -362,6 +395,7 @@ indent = { tab-width = 2, unit = " " }

362name = "cmake"

395name = "cmake"

363scope = "source.cmake"

396scope = "source.cmake"

364file-types = ["cmake", "CMakeLists.txt"]

397file-types = ["cmake", "CMakeLists.txt"]

398+shebangs = []

365roots = []

399roots = []

366comment-token = "#"

400comment-token = "#"

367indent = { tab-width = 2, unit = " " }

401indent = { tab-width = 2, unit = " " }

@@ -371,6 +405,7 @@ language-server = { command = "cmake-language-server" }

371name = "perl"

405name = "perl"

372scope = "source.perl"

406scope = "source.perl"

373file-types = ["pl", "pm"]

407file-types = ["pl", "pm"]

408+shebangs = ["perl"]

374roots = []

409roots = []

375comment-token = "#"

410comment-token = "#"

376indent = { tab-width = 2, unit = " " }

411indent = { tab-width = 2, unit = " " }