links-notation 0.20.0 → 0.21.0

package/dist/index.js

@@ -275,7 +275,8 @@ var ELLIPSIS = "...";
 
 class ParseError extends Error {
   constructor(input, error) {
-    const start = error?.location?.start ?? { offset: 0, line: 1, column: 1 };
+    const start = positionAt(input, error?.location?.start?.offset ?? 0);
+    const end = positionAt(input, error?.location?.end?.offset ?? start.offset);
     const lineText = lineAt(input, start.offset);
     const summary = `line ${start.line}, column ${start.column}: ${error.message}`;
     const snippet = quote(start.line, lineText, start.column);
@@ -283,7 +284,7 @@ class ParseError extends Error {
 ${snippet}`);
     this.name = "ParseError";
     this.cause = error;
-    this.location = error.location;
+    this.location = { ...error.location, start, end };
     this.offset = start.offset;
     this.line = start.line;
     this.column = start.column;
@@ -294,12 +295,32 @@ ${snippet}`);
 }
 function lineAt(input, offset) {
   const at = Math.max(0, Math.min(offset, input.length));
-  const start = input.lastIndexOf(`
-`, at - 1) + 1;
-  const end = input.indexOf(`
-`, start);
-  const line = input.slice(start, end === -1 ? input.length : end);
-  return line.endsWith("\r") ? line.slice(0, -1) : line;
+  const before = input.slice(0, at);
+  const start = Math.max(before.lastIndexOf(`
+`), before.lastIndexOf("\r")) + 1;
+  const end = input.slice(start).search(/[\r\n]/);
+  return input.slice(start, end === -1 ? input.length : start + end);
+}
+function positionAt(input, offset) {
+  const at = Math.max(0, Math.min(offset, input.length));
+  let line = 1;
+  let column = 1;
+  for (let index = 0;index < at; index++) {
+    if (input[index] === "\r") {
+      line++;
+      column = 1;
+      if (input[index + 1] === `
+` && index + 1 < at)
+        index++;
+    } else if (input[index] === `
+`) {
+      line++;
+      column = 1;
+    } else {
+      column++;
+    }
+  }
+  return { offset: at, line, column };
 }
 function quote(number, lineText, column) {
   const [quoted, at] = windowAround(lineText, column);
@@ -400,7 +421,7 @@ function isSubstantiveBody(content) {
         return false;
       }
     }
-    if (!/\s/.test(character)) {
+    if (!/[ \t\n\r]/.test(character)) {
       hasVisible = true;
     }
   }
@@ -2220,7 +2241,7 @@ function peg$parse(input, options) {
           return false;
         }
       }
-      if (!/\s/.test(c)) {
+      if (!/[ \t\n\r]/.test(c)) {
         hasVisible = true;
       }
     }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "links-notation",
-  "version": "0.20.0",
+  "version": "0.21.0",
   "description": "Links Notation parser for JavaScript",
   "main": "dist/index.js",
   "types": "index.d.ts",

package/src/ParseError.js

@@ -19,7 +19,8 @@ export class ParseError extends Error {
    * @param {Error} error - The error the generated parser threw
    */
   constructor(input, error) {
-    const start = error?.location?.start ?? { offset: 0, line: 1, column: 1 };
+    const start = positionAt(input, error?.location?.start?.offset ?? 0);
+    const end = positionAt(input, error?.location?.end?.offset ?? start.offset);
     const lineText = lineAt(input, start.offset);
     const summary = `line ${start.line}, column ${start.column}: ${error.message}`;
     const snippet = quote(start.line, lineText, start.column);
@@ -29,7 +30,7 @@ export class ParseError extends Error {
     /** @type {Error} The error the generated parser threw */
     this.cause = error;
     /** @type {{start: {offset: number, line: number, column: number}}} */
-    this.location = error.location;
+    this.location = { ...error.location, start, end };
     /** @type {number} Offset of the offending position from the start of the document */
     this.offset = start.offset;
     /** @type {number} Line the offending position is on, counted from 1 */
@@ -53,10 +54,37 @@ export class ParseError extends Error {
  */
 function lineAt(input, offset) {
   const at = Math.max(0, Math.min(offset, input.length));
-  const start = input.lastIndexOf('\n', at - 1) + 1;
-  const end = input.indexOf('\n', start);
-  const line = input.slice(start, end === -1 ? input.length : end);
-  return line.endsWith('\r') ? line.slice(0, -1) : line;
+  const before = input.slice(0, at);
+  const start =
+    Math.max(before.lastIndexOf('\n'), before.lastIndexOf('\r')) + 1;
+  const end = input.slice(start).search(/[\r\n]/);
+  return input.slice(start, end === -1 ? input.length : start + end);
+}
+
+/**
+ * Count CR, LF and CRLF as line breaks, as the grammar does. Peggy's error
+ * location counts a lone CR as an ordinary character.
+ * @param {string} input - The document being parsed
+ * @param {number} offset - Offset of the position
+ * @returns {{offset: number, line: number, column: number}} The position
+ */
+function positionAt(input, offset) {
+  const at = Math.max(0, Math.min(offset, input.length));
+  let line = 1;
+  let column = 1;
+  for (let index = 0; index < at; index++) {
+    if (input[index] === '\r') {
+      line++;
+      column = 1;
+      if (input[index + 1] === '\n' && index + 1 < at) index++;
+    } else if (input[index] === '\n') {
+      line++;
+      column = 1;
+    } else {
+      column++;
+    }
+  }
+  return { offset: at, line, column };
 }
 
 /**

package/src/comments.js

@@ -155,7 +155,7 @@ function isSubstantiveBody(content) {
         return false;
       }
     }
-    if (!/\s/.test(character)) {
+    if (!/[ \t\n\r]/.test(character)) {
       hasVisible = true;
     }
   }

package/src/grammar.pegjs

@@ -83,7 +83,7 @@
           return false;
         }
       }
-      if (!/\s/.test(c)) {
+      if (!/[ \t\n\r]/.test(c)) {
         hasVisible = true;
       }
     }

package/src/parser-generated.js

@@ -1881,7 +1881,7 @@ function peg$parse(input, options) {
           return false;
         }
       }
-      if (!/\s/.test(c)) {
+      if (!/[ \t\n\r]/.test(c)) {
         hasVisible = true;
       }
     }

package/tests/Issue312Conformance.test.js

@@ -0,0 +1,46 @@
+import { test, expect } from 'bun:test';
+import { Parser } from '../src/Parser.js';
+
+const parser = new Parser();
+
+function render(node) {
+  if (!node.values || node.values.length === 0) {
+    return `<${node.id ?? ''}>`;
+  }
+  const head =
+    node.id === null || node.id === undefined ? '' : `<${node.id}>: `;
+  return `(${head}${node.values.map(render).join(' ')})`;
+}
+
+function parsesAs(source) {
+  return parser.parse(source).map(render).join('\n');
+}
+
+test('CR is a line break', () => {
+  expect(parsesAs('a\rb')).toBe('(<a>)\n(<b>)');
+  expect(parsesAs('a\r\nb')).toBe('(<a>)\n(<b>)');
+});
+
+test('only grammar whitespace makes an empty document', () => {
+  expect(parsesAs(' \t\n\r')).toBe('');
+  for (const character of ['\u0085', '\u00a0', '\u2028', '\u3000', '\ufeff']) {
+    expect(parsesAs(character)).toBe(`(<${character}>)`);
+  }
+});
+
+test('Unicode spaces are substantive even quote bodies', () => {
+  for (const character of ['\u0085', '\u00a0', '\u2028', '\u3000', '\ufeff']) {
+    expect(parsesAs(`''${character}''`)).toBe(`(<${character}>)`);
+  }
+});
+
+test('trailing indentation is document whitespace', () => {
+  expect(parsesAs('a\n  ')).toBe('(<a>)');
+  expect(parsesAs('a\n  \t')).toBe('(<a>)');
+});
+
+test('a single reference is a link value at every depth', () => {
+  expect(parsesAs('1')).toBe('(<1>)');
+  expect(parsesAs('(1)')).toBe('(<1>)');
+  expect(parsesAs('((1))')).toBe('((<1>))');
+});

package/tests/ParseErrorPosition.test.js

@@ -50,6 +50,14 @@ test('reports the line a late defect is on', () => {
   expect(error.lineText).toBe('e: f: g');
 });
 
+test('lone CR counts as one line break in diagnostics', () => {
+  const error = syntaxError('a\rb: c: d');
+
+  expect(error.line).toBe(2);
+  expect(error.column).toBe(5);
+  expect(error.lineText).toBe('b: c: d');
+});
+
 test('reports the end of the document when a group is never closed', () => {
   const error = syntaxError('a (b\n');