fix: handle CR in rules to follow CommonMark spec (#493)

* fix: add CR to the line ending pattern

* wip: add test

* wip: add more tests

* wip: mark error prone patterns

* wip

* wip

* wip

* wip: `no-html`

* wip

* wip: simplify regex

* wip

* wip

* wip

* wip

Author: lumirlumir

src/rules/no-html.js

@@ -3,6 +3,12 @@
  * @author Nicholas C. Zakas
  */
 
+//-----------------------------------------------------------------------------
+// Imports
+//-----------------------------------------------------------------------------
+
+import { lineEndingPattern } from "../util.js";
+
 //-----------------------------------------------------------------------------
 // Type Definitions
 //-----------------------------------------------------------------------------
@@ -20,7 +26,6 @@
 
 const htmlTagPattern =
 	/<(?<tagName>[a-z0-9]+(?:-[a-z0-9]+)*)(?:\s(?:[^>"']|"[^"]*"|'[^']*')*)?>/giu;
-const lineEndingPattern = /\r\n?|\n/u;
 
 //-----------------------------------------------------------------------------
 // Rule Definition

src/rules/no-reference-like-urls.js

@@ -27,7 +27,7 @@ import { normalizeIdentifier } from "micromark-util-normalize-identifier";
 
 /** Pattern to match both inline links: `[text](url)` and images: `![alt](url)`, with optional title */
 const linkOrImagePattern =
-	/(?<imageBang>!)?\[(?<label>(?:\\.|[^()\\]|\([\s\S]*\))*?)\]\((?<destination>[ \t]*(?:\r\n?|\n)?(?<![ \t])[ \t]*(?:<[^>]*>|[^ \t()]+))(?:[ \t]*(?:\r\n?|\n)?(?<![ \t])[ \t]*(?<title>"[^"]*"|'[^']*'|\([^)]*\)))?[ \t]*(?:\r\n?|\n)?(?<![ \t])[ \t]*\)$/u;
+	/\[(?<label>(?:\\.|[^()\\]|\([\s\S]*\))*?)\]\((?<destination>[ \t]*\r?\n?(?<![ \t])[ \t]*(?:<[^>]*>|[^ \t()]+))(?:[ \t]*\r?\n?(?<![ \t])[ \t]*(?:"[^"]*"|'[^']*'|\([^)]*\)))?[ \t]*\r?\n?(?<![ \t])[ \t]*\)$/u;
 
 //-----------------------------------------------------------------------------
 // Rule Definition
@@ -75,17 +75,9 @@ export default {
 
 					const match = linkOrImagePattern.exec(text);
 					if (match !== null) {
-						const {
-							imageBang,
-							label,
-							destination,
-							title: titleRaw,
-						} = match.groups;
-						const title = titleRaw?.slice(1, -1);
-
-						const isImage = !!imageBang;
-						const type = isImage ? "image" : "link";
-						const prefix = isImage ? "!" : "";
+						const { label, destination } = match.groups;
+						const { type, title } = node;
+						const prefix = type === "image" ? "!" : "";
 						const url =
 							normalizeIdentifier(destination).toLowerCase();
 

src/rules/no-reversed-media-syntax.js

@@ -22,7 +22,7 @@
 
 /** Matches reversed link/image syntax like `(text)[url]`, ignoring escaped characters like `\(text\)[url]`. */
 const reversedPattern =
-	/(?<=(?<!\\)(?:\\{2})*)\((?<label>(?:\\.|[^()\\]|\([\s\S]*\))*)\)\[(?<url>(?:\\.|[^\]\\\n])*)\](?!\()/gu;
+	/(?<=(?<!\\)(?:\\{2})*)\((?<label>(?:\\.|[^()\\]|\([\s\S]*\))*)\)\[(?<url>(?:\\.|[^\]\\\r\n])*)\](?!\()/gu;
 
 /**
  * Checks if a match is within any skip range

tests/rules/no-reversed-media-syntax.test.js

@@ -31,6 +31,9 @@ ruleTester.run("no-reversed-media-syntax", rule, {
 		"![foo](bar)",
 		"![foo](#bar)",
 		"![foo](http://bar.com)",
+		"(foo)[bar\r\n]",
+		"(foo)[bar\r]",
+		"(foo)[bar\n]",
 		"(foo)[bar](http://bar.com)",
 		"		myObj.getFiles(test)[0]",
 		dedent`