| 1 |
function wordRegExp(words, end) { |
| 2 |
return new RegExp((end ? "" : "^") + "(?:" + words.join("|") + ")" + (end ? "$" : "\\b")); |
| 3 |
} |
| 4 |
|
| 5 |
function chain(tokenize, stream, state) { |
| 6 |
state.tokenize.push(tokenize); |
| 7 |
return tokenize(stream, state); |
| 8 |
} |
| 9 |
|
| 10 |
var operators = /^(?:[-+/%|&^]|\*\*?|[<>]{2})/; |
| 11 |
var conditionalOperators = /^(?:[=!]~|===|<=>|[<>=!]=?|[|&]{2}|~)/; |
| 12 |
var indexingOperators = /^(?:\[\][?=]?)/; |
| 13 |
var anotherOperators = /^(?:\.(?:\.{2})?|->|[?:])/; |
| 14 |
var idents = /^[a-z_\u009F-\uFFFF][a-zA-Z0-9_\u009F-\uFFFF]*/; |
| 15 |
var types = /^[A-Z_\u009F-\uFFFF][a-zA-Z0-9_\u009F-\uFFFF]*/; |
| 16 |
var keywords = wordRegExp([ |
| 17 |
"abstract", "alias", "as", "asm", "begin", "break", "case", "class", "def", "do", |
| 18 |
"else", "elsif", "end", "ensure", "enum", "extend", "for", "fun", "if", |
| 19 |
"include", "instance_sizeof", "lib", "macro", "module", "next", "of", "out", "pointerof", |
| 20 |
"private", "protected", "rescue", "return", "require", "select", "sizeof", "struct", |
| 21 |
"super", "then", "type", "typeof", "uninitialized", "union", "unless", "until", "when", "while", "with", |
| 22 |
"yield", "__DIR__", "__END_LINE__", "__FILE__", "__LINE__" |
| 23 |
]); |
| 24 |
var atomWords = wordRegExp(["true", "false", "nil", "self"]); |
| 25 |
var indentKeywordsArray = [ |
| 26 |
"def", "fun", "macro", |
| 27 |
"class", "module", "struct", "lib", "enum", "union", |
| 28 |
"do", "for" |
| 29 |
]; |
| 30 |
var indentKeywords = wordRegExp(indentKeywordsArray); |
| 31 |
var indentExpressionKeywordsArray = ["if", "unless", "case", "while", "until", "begin", "then"]; |
| 32 |
var indentExpressionKeywords = wordRegExp(indentExpressionKeywordsArray); |
| 33 |
var dedentKeywordsArray = ["end", "else", "elsif", "rescue", "ensure"]; |
| 34 |
var dedentKeywords = wordRegExp(dedentKeywordsArray); |
| 35 |
var dedentPunctualsArray = ["\\)", "\\}", "\\]"]; |
| 36 |
var dedentPunctuals = new RegExp("^(?:" + dedentPunctualsArray.join("|") + ")$"); |
| 37 |
var nextTokenizer = { |
| 38 |
"def": tokenFollowIdent, "fun": tokenFollowIdent, "macro": tokenMacroDef, |
| 39 |
"class": tokenFollowType, "module": tokenFollowType, "struct": tokenFollowType, |
| 40 |
"lib": tokenFollowType, "enum": tokenFollowType, "union": tokenFollowType |
| 41 |
}; |
| 42 |
var matching = {"[": "]", "{": "}", "(": ")", "<": ">"}; |
| 43 |
|
| 44 |
function tokenBase(stream, state) { |
| 45 |
if (stream.eatSpace()) { |
| 46 |
return null; |
| 47 |
} |
| 48 |
|
| 49 |
// Macros |
| 50 |
if (state.lastToken != "\\" && stream.match("{%", false)) { |
| 51 |
return chain(tokenMacro("%", "%"), stream, state); |
| 52 |
} |
| 53 |
|
| 54 |
if (state.lastToken != "\\" && stream.match("{{", false)) { |
| 55 |
return chain(tokenMacro("{", "}"), stream, state); |
| 56 |
} |
| 57 |
|
| 58 |
// Comments |
| 59 |
if (stream.peek() == "#") { |
| 60 |
stream.skipToEnd(); |
| 61 |
return "comment"; |
| 62 |
} |
| 63 |
|
| 64 |
// Variables and keywords |
| 65 |
var matched; |
| 66 |
if (stream.match(idents)) { |
| 67 |
stream.eat(/[?!]/); |
| 68 |
|
| 69 |
matched = stream.current(); |
| 70 |
if (stream.eat(":")) { |
| 71 |
return "atom"; |
| 72 |
} else if (state.lastToken == ".") { |
| 73 |
return "property"; |
| 74 |
} else if (keywords.test(matched)) { |
| 75 |
if (indentKeywords.test(matched)) { |
| 76 |
if (!(matched == "fun" && state.blocks.indexOf("lib") >= 0) && !(matched == "def" && state.lastToken == "abstract")) { |
| 77 |
state.blocks.push(matched); |
| 78 |
state.currentIndent += 1; |
| 79 |
} |
| 80 |
} else if ((state.lastStyle == "operator" || !state.lastStyle) && indentExpressionKeywords.test(matched)) { |
| 81 |
state.blocks.push(matched); |
| 82 |
state.currentIndent += 1; |
| 83 |
} else if (matched == "end") { |
| 84 |
state.blocks.pop(); |
| 85 |
state.currentIndent -= 1; |
| 86 |
} |
| 87 |
|
| 88 |
if (nextTokenizer.hasOwnProperty(matched)) { |
| 89 |
state.tokenize.push(nextTokenizer[matched]); |
| 90 |
} |
| 91 |
|
| 92 |
return "keyword"; |
| 93 |
} else if (atomWords.test(matched)) { |
| 94 |
return "atom"; |
| 95 |
} |
| 96 |
|
| 97 |
return "variable"; |
| 98 |
} |
| 99 |
|
| 100 |
// Class variables and instance variables |
| 101 |
// or attributes |
| 102 |
if (stream.eat("@")) { |
| 103 |
if (stream.peek() == "[") { |
| 104 |
return chain(tokenNest("[", "]", "meta"), stream, state); |
| 105 |
} |
| 106 |
|
| 107 |
stream.eat("@"); |
| 108 |
stream.match(idents) || stream.match(types); |
| 109 |
return "propertyName"; |
| 110 |
} |
| 111 |
|
| 112 |
// Constants and types |
| 113 |
if (stream.match(types)) { |
| 114 |
return "tag"; |
| 115 |
} |
| 116 |
|
| 117 |
// Symbols or ':' operator |
| 118 |
if (stream.eat(":")) { |
| 119 |
if (stream.eat("\"")) { |
| 120 |
return chain(tokenQuote("\"", "atom", false), stream, state); |
| 121 |
} else if (stream.match(idents) || stream.match(types) || |
| 122 |
stream.match(operators) || stream.match(conditionalOperators) || stream.match(indexingOperators)) { |
| 123 |
return "atom"; |
| 124 |
} |
| 125 |
stream.eat(":"); |
| 126 |
return "operator"; |
| 127 |
} |
| 128 |
|
| 129 |
// Strings |
| 130 |
if (stream.eat("\"")) { |
| 131 |
return chain(tokenQuote("\"", "string", true), stream, state); |
| 132 |
} |
| 133 |
|
| 134 |
// Strings or regexps or macro variables or '%' operator |
| 135 |
if (stream.peek() == "%") { |
| 136 |
var style = "string"; |
| 137 |
var embed = true; |
| 138 |
var delim; |
| 139 |
|
| 140 |
if (stream.match("%r")) { |
| 141 |
// Regexps |
| 142 |
style = "string.special"; |
| 143 |
delim = stream.next(); |
| 144 |
} else if (stream.match("%w")) { |
| 145 |
embed = false; |
| 146 |
delim = stream.next(); |
| 147 |
} else if (stream.match("%q")) { |
| 148 |
embed = false; |
| 149 |
delim = stream.next(); |
| 150 |
} else { |
| 151 |
if(delim = stream.match(/^%([^\w\s=])/)) { |
| 152 |
delim = delim[1]; |
| 153 |
} else if (stream.match(/^%[a-zA-Z_\u009F-\uFFFF][\w\u009F-\uFFFF]*/)) { |
| 154 |
// Macro variables |
| 155 |
return "meta"; |
| 156 |
} else if (stream.eat('%')) { |
| 157 |
// '%' operator |
| 158 |
return "operator"; |
| 159 |
} |
| 160 |
} |
| 161 |
|
| 162 |
if (matching.hasOwnProperty(delim)) { |
| 163 |
delim = matching[delim]; |
| 164 |
} |
| 165 |
return chain(tokenQuote(delim, style, embed), stream, state); |
| 166 |
} |
| 167 |
|
| 168 |
// Here Docs |
| 169 |
if (matched = stream.match(/^<<-('?)([A-Z]\w*)\1/)) { |
| 170 |
return chain(tokenHereDoc(matched[2], !matched[1]), stream, state) |
| 171 |
} |
| 172 |
|
| 173 |
// Characters |
| 174 |
if (stream.eat("'")) { |
| 175 |
stream.match(/^(?:[^']|\\(?:[befnrtv0'"]|[0-7]{3}|u(?:[0-9a-fA-F]{4}|\{[0-9a-fA-F]{1,6}\})))/); |
| 176 |
stream.eat("'"); |
| 177 |
return "atom"; |
| 178 |
} |
| 179 |
|
| 180 |
// Numbers |
| 181 |
if (stream.eat("0")) { |
| 182 |
if (stream.eat("x")) { |
| 183 |
stream.match(/^[0-9a-fA-F_]+/); |
| 184 |
} else if (stream.eat("o")) { |
| 185 |
stream.match(/^[0-7_]+/); |
| 186 |
} else if (stream.eat("b")) { |
| 187 |
stream.match(/^[01_]+/); |
| 188 |
} |
| 189 |
return "number"; |
| 190 |
} |
| 191 |
|
| 192 |
if (stream.eat(/^\d/)) { |
| 193 |
stream.match(/^[\d_]*(?:\.[\d_]+)?(?:[eE][+-]?\d+)?/); |
| 194 |
return "number"; |
| 195 |
} |
| 196 |
|
| 197 |
// Operators |
| 198 |
if (stream.match(operators)) { |
| 199 |
stream.eat("="); // Operators can follow assign symbol. |
| 200 |
return "operator"; |
| 201 |
} |
| 202 |
|
| 203 |
if (stream.match(conditionalOperators) || stream.match(anotherOperators)) { |
| 204 |
return "operator"; |
| 205 |
} |
| 206 |
|
| 207 |
// Parens and braces |
| 208 |
if (matched = stream.match(/[({[]/, false)) { |
| 209 |
matched = matched[0]; |
| 210 |
return chain(tokenNest(matched, matching[matched], null), stream, state); |
| 211 |
} |
| 212 |
|
| 213 |
// Escapes |
| 214 |
if (stream.eat("\\")) { |
| 215 |
stream.next(); |
| 216 |
return "meta"; |
| 217 |
} |
| 218 |
|
| 219 |
stream.next(); |
| 220 |
return null; |
| 221 |
} |
| 222 |
|
| 223 |
function tokenNest(begin, end, style, started) { |
| 224 |
return function (stream, state) { |
| 225 |
if (!started && stream.match(begin)) { |
| 226 |
state.tokenize[state.tokenize.length - 1] = tokenNest(begin, end, style, true); |
| 227 |
state.currentIndent += 1; |
| 228 |
return style; |
| 229 |
} |
| 230 |
|
| 231 |
var nextStyle = tokenBase(stream, state); |
| 232 |
if (stream.current() === end) { |
| 233 |
state.tokenize.pop(); |
| 234 |
state.currentIndent -= 1; |
| 235 |
nextStyle = style; |
| 236 |
} |
| 237 |
|
| 238 |
return nextStyle; |
| 239 |
}; |
| 240 |
} |
| 241 |
|
| 242 |
function tokenMacro(begin, end, started) { |
| 243 |
return function (stream, state) { |
| 244 |
if (!started && stream.match("{" + begin)) { |
| 245 |
state.currentIndent += 1; |
| 246 |
state.tokenize[state.tokenize.length - 1] = tokenMacro(begin, end, true); |
| 247 |
return "meta"; |
| 248 |
} |
| 249 |
|
| 250 |
if (stream.match(end + "}")) { |
| 251 |
state.currentIndent -= 1; |
| 252 |
state.tokenize.pop(); |
| 253 |
return "meta"; |
| 254 |
} |
| 255 |
|
| 256 |
return tokenBase(stream, state); |
| 257 |
}; |
| 258 |
} |
| 259 |
|
| 260 |
function tokenMacroDef(stream, state) { |
| 261 |
if (stream.eatSpace()) { |
| 262 |
return null; |
| 263 |
} |
| 264 |
|
| 265 |
var matched; |
| 266 |
if (matched = stream.match(idents)) { |
| 267 |
if (matched == "def") { |
| 268 |
return "keyword"; |
| 269 |
} |
| 270 |
stream.eat(/[?!]/); |
| 271 |
} |
| 272 |
|
| 273 |
state.tokenize.pop(); |
| 274 |
return "def"; |
| 275 |
} |
| 276 |
|
| 277 |
function tokenFollowIdent(stream, state) { |
| 278 |
if (stream.eatSpace()) { |
| 279 |
return null; |
| 280 |
} |
| 281 |
|
| 282 |
if (stream.match(idents)) { |
| 283 |
stream.eat(/[!?]/); |
| 284 |
} else { |
| 285 |
stream.match(operators) || stream.match(conditionalOperators) || stream.match(indexingOperators); |
| 286 |
} |
| 287 |
state.tokenize.pop(); |
| 288 |
return "def"; |
| 289 |
} |
| 290 |
|
| 291 |
function tokenFollowType(stream, state) { |
| 292 |
if (stream.eatSpace()) { |
| 293 |
return null; |
| 294 |
} |
| 295 |
|
| 296 |
stream.match(types); |
| 297 |
state.tokenize.pop(); |
| 298 |
return "def"; |
| 299 |
} |
| 300 |
|
| 301 |
function tokenQuote(end, style, embed) { |
| 302 |
return function (stream, state) { |
| 303 |
var escaped = false; |
| 304 |
|
| 305 |
while (stream.peek()) { |
| 306 |
if (!escaped) { |
| 307 |
if (stream.match("{%", false)) { |
| 308 |
state.tokenize.push(tokenMacro("%", "%")); |
| 309 |
return style; |
| 310 |
} |
| 311 |
|
| 312 |
if (stream.match("{{", false)) { |
| 313 |
state.tokenize.push(tokenMacro("{", "}")); |
| 314 |
return style; |
| 315 |
} |
| 316 |
|
| 317 |
if (embed && stream.match("#{", false)) { |
| 318 |
state.tokenize.push(tokenNest("#{", "}", "meta")); |
| 319 |
return style; |
| 320 |
} |
| 321 |
|
| 322 |
var ch = stream.next(); |
| 323 |
|
| 324 |
if (ch == end) { |
| 325 |
state.tokenize.pop(); |
| 326 |
return style; |
| 327 |
} |
| 328 |
|
| 329 |
escaped = embed && ch == "\\"; |
| 330 |
} else { |
| 331 |
stream.next(); |
| 332 |
escaped = false; |
| 333 |
} |
| 334 |
} |
| 335 |
|
| 336 |
return style; |
| 337 |
}; |
| 338 |
} |
| 339 |
|
| 340 |
function tokenHereDoc(phrase, embed) { |
| 341 |
return function (stream, state) { |
| 342 |
if (stream.sol()) { |
| 343 |
stream.eatSpace() |
| 344 |
if (stream.match(phrase)) { |
| 345 |
state.tokenize.pop(); |
| 346 |
return "string"; |
| 347 |
} |
| 348 |
} |
| 349 |
|
| 350 |
var escaped = false; |
| 351 |
while (stream.peek()) { |
| 352 |
if (!escaped) { |
| 353 |
if (stream.match("{%", false)) { |
| 354 |
state.tokenize.push(tokenMacro("%", "%")); |
| 355 |
return "string"; |
| 356 |
} |
| 357 |
|
| 358 |
if (stream.match("{{", false)) { |
| 359 |
state.tokenize.push(tokenMacro("{", "}")); |
| 360 |
return "string"; |
| 361 |
} |
| 362 |
|
| 363 |
if (embed && stream.match("#{", false)) { |
| 364 |
state.tokenize.push(tokenNest("#{", "}", "meta")); |
| 365 |
return "string"; |
| 366 |
} |
| 367 |
|
| 368 |
escaped = stream.next() == "\\" && embed; |
| 369 |
} else { |
| 370 |
stream.next(); |
| 371 |
escaped = false; |
| 372 |
} |
| 373 |
} |
| 374 |
|
| 375 |
return "string"; |
| 376 |
} |
| 377 |
} |
| 378 |
|
| 379 |
export const crystal = { |
| 380 |
name: "crystal", |
| 381 |
startState: function () { |
| 382 |
return { |
| 383 |
tokenize: [tokenBase], |
| 384 |
currentIndent: 0, |
| 385 |
lastToken: null, |
| 386 |
lastStyle: null, |
| 387 |
blocks: [] |
| 388 |
}; |
| 389 |
}, |
| 390 |
|
| 391 |
token: function (stream, state) { |
| 392 |
var style = state.tokenize[state.tokenize.length - 1](stream, state); |
| 393 |
var token = stream.current(); |
| 394 |
|
| 395 |
if (style && style != "comment") { |
| 396 |
state.lastToken = token; |
| 397 |
state.lastStyle = style; |
| 398 |
} |
| 399 |
|
| 400 |
return style; |
| 401 |
}, |
| 402 |
|
| 403 |
indent: function (state, textAfter, cx) { |
| 404 |
textAfter = textAfter.replace(/^\s*(?:\{%)?\s*|\s*(?:%\})?\s*$/g, ""); |
| 405 |
|
| 406 |
if (dedentKeywords.test(textAfter) || dedentPunctuals.test(textAfter)) { |
| 407 |
return cx.unit * (state.currentIndent - 1); |
| 408 |
} |
| 409 |
|
| 410 |
return cx.unit * state.currentIndent; |
| 411 |
}, |
| 412 |
|
| 413 |
languageData: { |
| 414 |
indentOnInput: wordRegExp(dedentPunctualsArray.concat(dedentKeywordsArray), true), |
| 415 |
commentTokens: {line: "#"} |
| 416 |
} |
| 417 |
}; |