source-code.js 21.5 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585
/**
 * @fileoverview Abstraction of JavaScript source code.
 * @author Nicholas C. Zakas
 */
"use strict";

//------------------------------------------------------------------------------
// Requirements
//------------------------------------------------------------------------------

const
    { isCommentToken } = require("eslint-utils"),
    TokenStore = require("./token-store"),
    astUtils = require("../shared/ast-utils"),
    Traverser = require("../shared/traverser"),
    lodash = require("lodash");

//------------------------------------------------------------------------------
// Private
//------------------------------------------------------------------------------

/**
 * Validates that the given AST has the required information.
 * @param {ASTNode} ast The Program node of the AST to check.
 * @throws {Error} If the AST doesn't contain the correct information.
 * @returns {void}
 * @private
 */
function validate(ast) {
    if (!ast.tokens) {
        throw new Error("AST is missing the tokens array.");
    }

    if (!ast.comments) {
        throw new Error("AST is missing the comments array.");
    }

    if (!ast.loc) {
        throw new Error("AST is missing location information.");
    }

    if (!ast.range) {
        throw new Error("AST is missing range information");
    }
}

/**
 * Check to see if its a ES6 export declaration.
 * @param {ASTNode} astNode An AST node.
 * @returns {boolean} whether the given node represents an export declaration.
 * @private
 */
function looksLikeExport(astNode) {
    return astNode.type === "ExportDefaultDeclaration" || astNode.type === "ExportNamedDeclaration" ||
        astNode.type === "ExportAllDeclaration" || astNode.type === "ExportSpecifier";
}

/**
 * Merges two sorted lists into a larger sorted list in O(n) time.
 * @param {Token[]} tokens The list of tokens.
 * @param {Token[]} comments The list of comments.
 * @returns {Token[]} A sorted list of tokens and comments.
 * @private
 */
function sortedMerge(tokens, comments) {
    const result = [];
    let tokenIndex = 0;
    let commentIndex = 0;

    while (tokenIndex < tokens.length || commentIndex < comments.length) {
        if (commentIndex >= comments.length || tokenIndex < tokens.length && tokens[tokenIndex].range[0] < comments[commentIndex].range[0]) {
            result.push(tokens[tokenIndex++]);
        } else {
            result.push(comments[commentIndex++]);
        }
    }

    return result;
}

/**
 * Determines if two nodes or tokens overlap.
 * @param {ASTNode|Token} first The first node or token to check.
 * @param {ASTNode|Token} second The second node or token to check.
 * @returns {boolean} True if the two nodes or tokens overlap.
 * @private
 */
function nodesOrTokensOverlap(first, second) {
    return (first.range[0] <= second.range[0] && first.range[1] >= second.range[0]) ||
        (second.range[0] <= first.range[0] && second.range[1] >= first.range[0]);
}

/**
 * Determines if two nodes or tokens have at least one whitespace character
 * between them. Order does not matter. Returns false if the given nodes or
 * tokens overlap.
 * @param {SourceCode} sourceCode The source code object.
 * @param {ASTNode|Token} first The first node or token to check between.
 * @param {ASTNode|Token} second The second node or token to check between.
 * @param {boolean} checkInsideOfJSXText If `true` is present, check inside of JSXText tokens for backward compatibility.
 * @returns {boolean} True if there is a whitespace character between
 * any of the tokens found between the two given nodes or tokens.
 * @public
 */
function isSpaceBetween(sourceCode, first, second, checkInsideOfJSXText) {
    if (nodesOrTokensOverlap(first, second)) {
        return false;
    }

    const [startingNodeOrToken, endingNodeOrToken] = first.range[1] <= second.range[0]
        ? [first, second]
        : [second, first];
    const firstToken = sourceCode.getLastToken(startingNodeOrToken) || startingNodeOrToken;
    const finalToken = sourceCode.getFirstToken(endingNodeOrToken) || endingNodeOrToken;
    let currentToken = firstToken;

    while (currentToken !== finalToken) {
        const nextToken = sourceCode.getTokenAfter(currentToken, { includeComments: true });

        if (
            currentToken.range[1] !== nextToken.range[0] ||

                /*
                 * For backward compatibility, check speces in JSXText.
                 * https://github.com/eslint/eslint/issues/12614
                 */
                (
                    checkInsideOfJSXText &&
                    nextToken !== finalToken &&
                    nextToken.type === "JSXText" &&
                    /\s/u.test(nextToken.value)
                )
        ) {
            return true;
        }

        currentToken = nextToken;
    }

    return false;
}

//------------------------------------------------------------------------------
// Public Interface
//------------------------------------------------------------------------------

class SourceCode extends TokenStore {

    /**
     * Represents parsed source code.
     * @param {string|Object} textOrConfig The source code text or config object.
     * @param {string} textOrConfig.text The source code text.
     * @param {ASTNode} textOrConfig.ast The Program node of the AST representing the code. This AST should be created from the text that BOM was stripped.
     * @param {Object|null} textOrConfig.parserServices The parser services.
     * @param {ScopeManager|null} textOrConfig.scopeManager The scope of this source code.
     * @param {Object|null} textOrConfig.visitorKeys The visitor keys to traverse AST.
     * @param {ASTNode} [astIfNoConfig] The Program node of the AST representing the code. This AST should be created from the text that BOM was stripped.
     */
    constructor(textOrConfig, astIfNoConfig) {
        let text, ast, parserServices, scopeManager, visitorKeys;

        // Process overloading.
        if (typeof textOrConfig === "string") {
            text = textOrConfig;
            ast = astIfNoConfig;
        } else if (typeof textOrConfig === "object" && textOrConfig !== null) {
            text = textOrConfig.text;
            ast = textOrConfig.ast;
            parserServices = textOrConfig.parserServices;
            scopeManager = textOrConfig.scopeManager;
            visitorKeys = textOrConfig.visitorKeys;
        }

        validate(ast);
        super(ast.tokens, ast.comments);

        /**
         * The flag to indicate that the source code has Unicode BOM.
         * @type boolean
         */
        this.hasBOM = (text.charCodeAt(0) === 0xFEFF);

        /**
         * The original text source code.
         * BOM was stripped from this text.
         * @type string
         */
        this.text = (this.hasBOM ? text.slice(1) : text);

        /**
         * The parsed AST for the source code.
         * @type ASTNode
         */
        this.ast = ast;

        /**
         * The parser services of this source code.
         * @type {Object}
         */
        this.parserServices = parserServices || {};

        /**
         * The scope of this source code.
         * @type {ScopeManager|null}
         */
        this.scopeManager = scopeManager || null;

        /**
         * The visitor keys to traverse AST.
         * @type {Object}
         */
        this.visitorKeys = visitorKeys || Traverser.DEFAULT_VISITOR_KEYS;

        // Check the source text for the presence of a shebang since it is parsed as a standard line comment.
        const shebangMatched = this.text.match(astUtils.shebangPattern);
        const hasShebang = shebangMatched && ast.comments.length && ast.comments[0].value === shebangMatched[1];

        if (hasShebang) {
            ast.comments[0].type = "Shebang";
        }

        this.tokensAndComments = sortedMerge(ast.tokens, ast.comments);

        /**
         * The source code split into lines according to ECMA-262 specification.
         * This is done to avoid each rule needing to do so separately.
         * @type string[]
         */
        this.lines = [];
        this.lineStartIndices = [0];

        const lineEndingPattern = astUtils.createGlobalLinebreakMatcher();
        let match;

        /*
         * Previously, this was implemented using a regex that
         * matched a sequence of non-linebreak characters followed by a
         * linebreak, then adding the lengths of the matches. However,
         * this caused a catastrophic backtracking issue when the end
         * of a file contained a large number of non-newline characters.
         * To avoid this, the current implementation just matches newlines
         * and uses match.index to get the correct line start indices.
         */
        while ((match = lineEndingPattern.exec(this.text))) {
            this.lines.push(this.text.slice(this.lineStartIndices[this.lineStartIndices.length - 1], match.index));
            this.lineStartIndices.push(match.index + match[0].length);
        }
        this.lines.push(this.text.slice(this.lineStartIndices[this.lineStartIndices.length - 1]));

        // Cache for comments found using getComments().
        this._commentCache = new WeakMap();

        // don't allow modification of this object
        Object.freeze(this);
        Object.freeze(this.lines);
    }

    /**
     * Split the source code into multiple lines based on the line delimiters.
     * @param {string} text Source code as a string.
     * @returns {string[]} Array of source code lines.
     * @public
     */
    static splitLines(text) {
        return text.split(astUtils.createGlobalLinebreakMatcher());
    }

    /**
     * Gets the source code for the given node.
     * @param {ASTNode} [node] The AST node to get the text for.
     * @param {int} [beforeCount] The number of characters before the node to retrieve.
     * @param {int} [afterCount] The number of characters after the node to retrieve.
     * @returns {string} The text representing the AST node.
     * @public
     */
    getText(node, beforeCount, afterCount) {
        if (node) {
            return this.text.slice(Math.max(node.range[0] - (beforeCount || 0), 0),
                node.range[1] + (afterCount || 0));
        }
        return this.text;
    }

    /**
     * Gets the entire source text split into an array of lines.
     * @returns {Array} The source text as an array of lines.
     * @public
     */
    getLines() {
        return this.lines;
    }

    /**
     * Retrieves an array containing all comments in the source code.
     * @returns {ASTNode[]} An array of comment nodes.
     * @public
     */
    getAllComments() {
        return this.ast.comments;
    }

    /**
     * Gets all comments for the given node.
     * @param {ASTNode} node The AST node to get the comments for.
     * @returns {Object} An object containing a leading and trailing array
     *      of comments indexed by their position.
     * @public
     */
    getComments(node) {
        if (this._commentCache.has(node)) {
            return this._commentCache.get(node);
        }

        const comments = {
            leading: [],
            trailing: []
        };

        /*
         * Return all comments as leading comments of the Program node when
         * there is no executable code.
         */
        if (node.type === "Program") {
            if (node.body.length === 0) {
                comments.leading = node.comments;
            }
        } else {

            /*
             * Return comments as trailing comments of nodes that only contain
             * comments (to mimic the comment attachment behavior present in Espree).
             */
            if ((node.type === "BlockStatement" || node.type === "ClassBody") && node.body.length === 0 ||
                node.type === "ObjectExpression" && node.properties.length === 0 ||
                node.type === "ArrayExpression" && node.elements.length === 0 ||
                node.type === "SwitchStatement" && node.cases.length === 0
            ) {
                comments.trailing = this.getTokens(node, {
                    includeComments: true,
                    filter: isCommentToken
                });
            }

            /*
             * Iterate over tokens before and after node and collect comment tokens.
             * Do not include comments that exist outside of the parent node
             * to avoid duplication.
             */
            let currentToken = this.getTokenBefore(node, { includeComments: true });

            while (currentToken && isCommentToken(currentToken)) {
                if (node.parent && (currentToken.start < node.parent.start)) {
                    break;
                }
                comments.leading.push(currentToken);
                currentToken = this.getTokenBefore(currentToken, { includeComments: true });
            }

            comments.leading.reverse();

            currentToken = this.getTokenAfter(node, { includeComments: true });

            while (currentToken && isCommentToken(currentToken)) {
                if (node.parent && (currentToken.end > node.parent.end)) {
                    break;
                }
                comments.trailing.push(currentToken);
                currentToken = this.getTokenAfter(currentToken, { includeComments: true });
            }
        }

        this._commentCache.set(node, comments);
        return comments;
    }

    /**
     * Retrieves the JSDoc comment for a given node.
     * @param {ASTNode} node The AST node to get the comment for.
     * @returns {Token|null} The Block comment token containing the JSDoc comment
     *      for the given node or null if not found.
     * @public
     * @deprecated
     */
    getJSDocComment(node) {

        /**
         * Checks for the presence of a JSDoc comment for the given node and returns it.
         * @param {ASTNode} astNode The AST node to get the comment for.
         * @returns {Token|null} The Block comment token containing the JSDoc comment
         *      for the given node or null if not found.
         * @private
         */
        const findJSDocComment = astNode => {
            const tokenBefore = this.getTokenBefore(astNode, { includeComments: true });

            if (
                tokenBefore &&
                isCommentToken(tokenBefore) &&
                tokenBefore.type === "Block" &&
                tokenBefore.value.charAt(0) === "*" &&
                astNode.loc.start.line - tokenBefore.loc.end.line <= 1
            ) {
                return tokenBefore;
            }

            return null;
        };
        let parent = node.parent;

        switch (node.type) {
            case "ClassDeclaration":
            case "FunctionDeclaration":
                return findJSDocComment(looksLikeExport(parent) ? parent : node);

            case "ClassExpression":
                return findJSDocComment(parent.parent);

            case "ArrowFunctionExpression":
            case "FunctionExpression":
                if (parent.type !== "CallExpression" && parent.type !== "NewExpression") {
                    while (
                        !this.getCommentsBefore(parent).length &&
                        !/Function/u.test(parent.type) &&
                        parent.type !== "MethodDefinition" &&
                        parent.type !== "Property"
                    ) {
                        parent = parent.parent;

                        if (!parent) {
                            break;
                        }
                    }

                    if (parent && parent.type !== "FunctionDeclaration" && parent.type !== "Program") {
                        return findJSDocComment(parent);
                    }
                }

                return findJSDocComment(node);

            // falls through
            default:
                return null;
        }
    }

    /**
     * Gets the deepest node containing a range index.
     * @param {int} index Range index of the desired node.
     * @returns {ASTNode} The node if found or null if not found.
     * @public
     */
    getNodeByRangeIndex(index) {
        let result = null;

        Traverser.traverse(this.ast, {
            visitorKeys: this.visitorKeys,
            enter(node) {
                if (node.range[0] <= index && index < node.range[1]) {
                    result = node;
                } else {
                    this.skip();
                }
            },
            leave(node) {
                if (node === result) {
                    this.break();
                }
            }
        });

        return result;
    }

    /**
     * Determines if two nodes or tokens have at least one whitespace character
     * between them. Order does not matter. Returns false if the given nodes or
     * tokens overlap.
     * @param {ASTNode|Token} first The first node or token to check between.
     * @param {ASTNode|Token} second The second node or token to check between.
     * @returns {boolean} True if there is a whitespace character between
     * any of the tokens found between the two given nodes or tokens.
     * @public
     */
    isSpaceBetween(first, second) {
        return isSpaceBetween(this, first, second, false);
    }

    /**
     * Determines if two nodes or tokens have at least one whitespace character
     * between them. Order does not matter. Returns false if the given nodes or
     * tokens overlap.
     * For backward compatibility, this method returns true if there are
     * `JSXText` tokens that contain whitespaces between the two.
     * @param {ASTNode|Token} first The first node or token to check between.
     * @param {ASTNode|Token} second The second node or token to check between.
     * @returns {boolean} True if there is a whitespace character between
     * any of the tokens found between the two given nodes or tokens.
     * @deprecated in favor of isSpaceBetween().
     * @public
     */
    isSpaceBetweenTokens(first, second) {
        return isSpaceBetween(this, first, second, true);
    }

    /**
     * Converts a source text index into a (line, column) pair.
     * @param {number} index The index of a character in a file
     * @returns {Object} A {line, column} location object with a 0-indexed column
     * @public
     */
    getLocFromIndex(index) {
        if (typeof index !== "number") {
            throw new TypeError("Expected `index` to be a number.");
        }

        if (index < 0 || index > this.text.length) {
            throw new RangeError(`Index out of range (requested index ${index}, but source text has length ${this.text.length}).`);
        }

        /*
         * For an argument of this.text.length, return the location one "spot" past the last character
         * of the file. If the last character is a linebreak, the location will be column 0 of the next
         * line; otherwise, the location will be in the next column on the same line.
         *
         * See getIndexFromLoc for the motivation for this special case.
         */
        if (index === this.text.length) {
            return { line: this.lines.length, column: this.lines[this.lines.length - 1].length };
        }

        /*
         * To figure out which line rangeIndex is on, determine the last index at which rangeIndex could
         * be inserted into lineIndices to keep the list sorted.
         */
        const lineNumber = lodash.sortedLastIndex(this.lineStartIndices, index);

        return { line: lineNumber, column: index - this.lineStartIndices[lineNumber - 1] };
    }

    /**
     * Converts a (line, column) pair into a range index.
     * @param {Object} loc A line/column location
     * @param {number} loc.line The line number of the location (1-indexed)
     * @param {number} loc.column The column number of the location (0-indexed)
     * @returns {number} The range index of the location in the file.
     * @public
     */
    getIndexFromLoc(loc) {
        if (typeof loc !== "object" || typeof loc.line !== "number" || typeof loc.column !== "number") {
            throw new TypeError("Expected `loc` to be an object with numeric `line` and `column` properties.");
        }

        if (loc.line <= 0) {
            throw new RangeError(`Line number out of range (line ${loc.line} requested). Line numbers should be 1-based.`);
        }

        if (loc.line > this.lineStartIndices.length) {
            throw new RangeError(`Line number out of range (line ${loc.line} requested, but only ${this.lineStartIndices.length} lines present).`);
        }

        const lineStartIndex = this.lineStartIndices[loc.line - 1];
        const lineEndIndex = loc.line === this.lineStartIndices.length ? this.text.length : this.lineStartIndices[loc.line];
        const positionIndex = lineStartIndex + loc.column;

        /*
         * By design, getIndexFromLoc({ line: lineNum, column: 0 }) should return the start index of
         * the given line, provided that the line number is valid element of this.lines. Since the
         * last element of this.lines is an empty string for files with trailing newlines, add a
         * special case where getting the index for the first location after the end of the file
         * will return the length of the file, rather than throwing an error. This allows rules to
         * use getIndexFromLoc consistently without worrying about edge cases at the end of a file.
         */
        if (
            loc.line === this.lineStartIndices.length && positionIndex > lineEndIndex ||
            loc.line < this.lineStartIndices.length && positionIndex >= lineEndIndex
        ) {
            throw new RangeError(`Column number out of range (column ${loc.column} requested, but the length of line ${loc.line} is ${lineEndIndex - lineStartIndex}).`);
        }

        return positionIndex;
    }
}

module.exports = SourceCode;