yeast-markdown-parser - npm Package Compare versions

Comparing version 1.5.1-devengage-2714-link-fixes.543 to 1.5.1-devengage-2714-link-fixes.544

dist/index.js

@@ -69,6 +69,6 @@ import { YeastBlockNodeTypes, YeastNodeFactory, YeastInlineNodeTypes, isYeastNodeType, scrapeText, ContentGroupType, isYeastTextNode, isYeastNode, YeastParser } from 'yeast-core';
 
-const ITALICS_REGEX_UNDERSCORES = /(\\?)(_)([^\s_]|\S.*?\S)(\\?)(_)/gi;
-const ITALICS_REGEX_ASTERISKS = /(\\?)(\*)([^\s_]|\S.*?\S)(\\?)(\*)/gi;
-const BOLD_REGEX_UNDERSCORES = /(\\?)(__)([^\s_]|\S.*?\S)(\\?)(__)/gi;
-const BOLD_REGEX_ASTERISKS = /(\\?)(\*\*)([^\s_]|\S.*?\S)(\\?)(\*\*)/gi;
+const ITALICS_REGEX_UNDERSCORES = /(?:^|([^\\]))_(?:(\\_|[^\\\s][^_\s]?)_|((?:\\_|\S)(?:\\_|[^_])*?[^\s\\])_)/gi;
+const ITALICS_REGEX_ASTERISKS = /(?:^|([^\\]))\*(?:(\\\*|[^\\\s][^\*\s]?)\*|((?:\\\*|\S)(?:\\\*|[^\*])*?[^\s\\])\*)/gi;
+const BOLD_REGEX_UNDERSCORES = /(?:^|([^\\]))__(?:(\\_|[^\\\s][^_\s]?)__|((?:\\_|\S)(?:\\_|[^_])*?[^\s\\])__)/gi;
+const BOLD_REGEX_ASTERISKS = /(?:^|([^\\]))\*\*(?:(\\\*|[^\\\s][^\*\s]?)\*\*|((?:\\\*|\S)(?:\\\*|[^\*])*?[^\s\\])\*\*)/gi;
 class InlineEmphasisPlugin {
@@ -78,13 +78,9 @@ tokenize(text, parser) {
         const parseMatch = (match, nodeType) => {
-            let node;
-            if (match.length == 6 && match[1] && match[4]) {
-                node = YeastNodeFactory.CreateText();
-                node.text = `${match[2]}${match[3]}${match[5]}`;
-            }
-            else {
-                node = YeastNodeFactory.Create(nodeType);
-                node.children = parser.parseInline(match[3]);
-            }
+            if (match.length < 4 || (!match[2] && !match[3]))
+                return;
+            const startOffset = (match[1] || '').length;
+            let node = YeastNodeFactory.Create(nodeType);
+            node.children = parser.parseInline(match[2] || match[3]);
             tokens.push({
-                start: match.index,
+                start: match.index + startOffset,
                 end: match.index + match[0].length,
@@ -107,36 +103,2 @@ from: 'InlineEmphasisPlugin',
         }
-        const textArr = text.split('');
-        let index = 0;
-        while (index < textArr.length) {
-            if (textArr[index] === '*' && textArr[index + 1] !== '*' && textArr[index + 1] !== ' ') {
-                if (index >= 0 && textArr[index - 1] !== '*') {
-                    let italizedText = '';
-                    let startIndex = index;
-                    let isInvalidSyntax = false;
-                    do {
-                        if (textArr[index + 1]) {
-                            italizedText += textArr[++index];
-                        }
-                        else {
-                            isInvalidSyntax = true;
-                            break;
-                        }
-                    } while (textArr[index + 1] !== '*' && index < textArr.length);
-                    if (isInvalidSyntax) {
-                        index++;
-                        continue;
-                    }
-                    const node = YeastNodeFactory.CreateItalicNode();
-                    node.children = parser.parseInline(italizedText);
-                    tokens.push({
-                        start: startIndex,
-                        end: startIndex + italizedText.length + 2,
-                        from: 'ItalicsInlinePlugin',
-                        nodes: [node],
-                    });
-                    index++;
-                }
-            }
-            index++;
-        }
         return tokens;
@@ -350,3 +312,3 @@ }
 
-const LINK_REGEX = /\[([^\[\]]*(?:\\.[^\[\]]*)*)\]\((.+?)(?:\s["'](.*?)["'])?\)/gi;
+const LINK_REGEX = /\[\s*([^\[\]]*?(?:\\.[^\[\]]*?)*?)\s*\]\(\s*(\S+)(?:\s+["']\s*(.*?)\s*["'])?\s*\)/gi;
 class InlineLinkPlugin {
@@ -356,3 +318,4 @@ tokenize(text, parser) {
         for (const match of text.matchAll(LINK_REGEX)) {
-            if (text.charAt(match.index - 1) === '!') {
+            const charBefore = text.charAt(match.index - 1);
+            if (charBefore === '!' || charBefore === '\\') {
                 continue;
@@ -359,0 +322,0 @@ }

package.json

 {
     "name": "yeast-markdown-parser",
-    "version": "1.5.1-devengage-2714-link-fixes.543",
+    "version": "1.5.1-devengage-2714-link-fixes.544",
     "description": "Parses markdown to yeAST (Yuri's Empathetic Arbitrary Syntax Tree) documents",
@@ -5,0 +5,0 @@ "exports": {

src/__tests__/resources/everythinginline.md

 Lorem ipsum dolor sit amet, consectetur adipiscing elit. ~Nulla viverra~ tortor non diam , non `func()` fringilla nunc vestibulum. _Donec faucibus_, odio a congue mollis, arcu nunc viverra leo, vitae viverra * turpis erat at quam. Pellentesque *foobar* hendrerit ligula neque, et tempus leo bibendum a.pulvinar, nunc a malesuada dignissim, nisi **vestibulum sapien a rutrum** diam maximus est,``function() ` `` a pharetra lorem tortor ac odio. Phasellus consectetur vestibulum sapien a rutrum. ![Alt Text](path/to/image.png 'Single quote text') Mauris vel ![Alt Text](path/to/image.png "Double quote text")pharetra libero, sollicitudin volutpat erat. ![](path/to/image.png "image without alt text")Nulla pulvinar libero sed vehicula sagittis. Mauris at quam fringilla, feugiat ante eget, dictu`m nibh. Vivamus aliquet, [/Genesys/developercenter](https://developer.genesys.cloud/ 'single quote text') tellus et pretium tempus, odio enim dictum sem, ut ultricies [Api Central](https://apicentral.dev-genesys.cloud/index/ "double quote text") nibh leo a dolor. [](https://apicentral.genesys.cloud/index/ "link without alt text") Phasellus suscipit libero rhoncus, euismod nulla ac, venenatis eros*. https://developer.genesys.cloud/
 
-This paragraph tests _non_standard touching use cases. Sin_gle_ cha*racters* _i_n *wo*rds are f_o_r i*talic*s. Characters * surrounded * by _ whitespace _ are ignored. Es\_ca\_ped ch\*ar\*acters a\__re\__ n\**ot\** \**parsed out\** \_but\_ \*are\* \__rendered without\__ escape characters. Dou__ble__ cha**racters** __i__n **wo**rds are f__o__r b**ol**d.
+This paragraph tests _non_standard touching use cases. Sin_gle_ cha*racters* _i_n *wo*rds are f_o_r i*talic*s. Characters * surrounded * by _ whitespace _ are ignored. Es\_ca\_ped ch\*ar\*acters a\_\_re\_\_ n\*\*ot\*\* \*\*parsed out\*\* \_but\_ \*are\* \_\_rendered without\_\_ escape characters. Dou__ble__ cha**racters** __i__n **wo**rds are f__o__r b**ol**d.
 
 Check **that \_escaped text\_ inside bold text** is reassembed into one text node.
 
-Dangling escaped markers, like \* or \__ or \|, should get unescaped.
+Dangling escaped markers, like \* or \_ or \|, should get unescaped.
+
+__bold underscore__ escape use cases 1) __bold text__ 2) __bold\_\_text__ 3) __bold\_text__ 4) \_\_notbold__isbold__ 5) __isbold__notbold\_\_ 6) __b__ __bb__ __bbb__ 7) __\___ 
+
+**bold asterisk** escape use cases 1) **bold text** 2) **bold\*\*text** 3) **bold\*text** 4) \*\*notbold**isbold** 5) **isbold**notbold\*\* 6) **b** **bb** **bbb** 7) **\*** 
+
+_italic underscore_ escape use cases 1) _italic text_ 2) _italic\_\_text_ 3) _italic\_text_ 4) \_notitalic_isitalic_ 5) _isitalic_notitalic\_ 6) _b_ _bb_ _bbb_ 7) _\__ 8) _ notitalic_ and _notitalic _ and _  _
+
+*italic asterisk* escape use cases 1) *italic text* 2) *italic\*\*text* 3) *italic\*text* 4) \*notitalic*isitalic* 5) *isitalic*notitalic\* 6) *b* *bb* *bbb* 7) *\** 8) * notitalic* and *notitalic * and *  *

src/__tests__/resources/links.ts

@@ -10,2 +10,8 @@ export const LINK_MARKDOWN = `[/Genesys/developercenter](https://developer.genesys.cloud/ 'single quote text')tellus et pretium tempus, https://developer.genesys.cloud/
 Not parsing escaped [brackets \\[inside\\] links](#asdf).
+
+This is not a link because it's not escaped: \\[brackets \\[inside\\] NOT links](#asdf). why \] ok.
+
+Links with whitespace [ why do you do this   ](    #somewhere  ).
+
+and alt text [     lots more poor spacing   ](    #somewhere   "  goes to somewhere's special. '    ).
 `;
@@ -164,3 +170,56 @@
 		},
+		{
+			type: 'paragraph',
+			children: [
+				{
+					text: "This is not a link because it's not escaped: [brackets [inside] NOT links](#asdf). why ] ok.",
+				},
+			],
+			indentation: 0,
+		},
+		{
+			type: 'paragraph',
+			children: [
+				{
+					text: 'Links with whitespace ',
+				},
+				{
+					type: 'link',
+					children: [
+						{
+							text: 'why do you do this',
+						},
+					],
+					href: '#somewhere',
+					title: 'Link',
+				},
+				{
+					text: '.',
+				},
+			],
+			indentation: 0,
+		},
+		{
+			type: 'paragraph',
+			children: [
+				{
+					text: 'and alt text ',
+				},
+				{
+					type: 'link',
+					children: [
+						{
+							text: 'lots more poor spacing',
+						},
+					],
+					href: '#somewhere',
+					title: "goes to somewhere's special.",
+				},
+				{
+					text: '.',
+				},
+			],
+			indentation: 0,
+		},
 	],
 };

src/__tests__/tests/MarkdownParser.test.ts

@@ -43,2 +43,3 @@ import * as fs from 'fs';
 import { TABLE_AST } from '../resources/table-data';
+import { EVERYTHING_INLINE_AST } from '../resources/everythinginline';
 
@@ -390,47 +391,6 @@ const standardBlockPluginCount = 10;
 	// Check document
-	checkAstStructureForDefaultDocument(ast, 4);
+	checkAstStructureForDefaultDocument(ast, 8);
 
-	// Paragraph 1
-	expect(ast.children[0].children.length).toBe(26);
-	expect((ast.children[0].children[1] as StrikethroughNode).type).toBe(YeastInlineNodeTypes.Strikethrough);
-	expect((ast.children[0].children[3] as InlineCodeNode).type).toBe(YeastInlineNodeTypes.Code);
-	expect((ast.children[0].children[5] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[0].children[7] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[0].children[9] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-	expect((ast.children[0].children[11] as InlineCodeNode).type).toBe(YeastInlineNodeTypes.Code);
-	expect((ast.children[0].children[13] as ImageNode).type).toBe(YeastInlineNodeTypes.Image);
-	expect((ast.children[0].children[15] as ImageNode).type).toBe(YeastInlineNodeTypes.Image);
-	expect((ast.children[0].children[17] as ImageNode).type).toBe(YeastInlineNodeTypes.Image);
-	expect((ast.children[0].children[19] as LinkNode).type).toBe(YeastInlineNodeTypes.Link);
-	expect((ast.children[0].children[21] as LinkNode).type).toBe(YeastInlineNodeTypes.Link);
-	expect((ast.children[0].children[23] as LinkNode).type).toBe(YeastInlineNodeTypes.Link);
-	expect((ast.children[0].children[25] as LinkNode).type).toBe(YeastInlineNodeTypes.Link);
-
-	// Paragraph 2 - test bold and italic formats and escaping
-	expect(ast.children[1].children.length).toBe(27);
-	expect((ast.children[1].children[1] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[1].children[3] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[1].children[5] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[1].children[7] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[1].children[9] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[1].children[11] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[1].children[13] as ItalicNode).type).toBe(YeastInlineNodeTypes.Italic);
-	expect((ast.children[1].children[15] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-	expect((ast.children[1].children[17] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-	expect((ast.children[1].children[19] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-	expect((ast.children[1].children[21] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-	expect((ast.children[1].children[23] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-	expect((ast.children[1].children[25] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-
-	// Paragraph 3 -- make sure text reassembly works for inline elements (paragraph above tests for blocks)
-	expect(ast.children[2].children.length).toBe(3);
-	expect((ast.children[2].children[0] as YeastText).text).toBeTruthy();
-	expect((ast.children[2].children[1] as BoldNode).type).toBe(YeastInlineNodeTypes.Bold);
-	expect((ast.children[2].children[1] as BoldNode).children.length).toBe(1);
-	expect(((ast.children[2].children[1] as BoldNode).children[0] as YeastText).text).toBeTruthy();
-	expect((ast.children[2].children[2] as YeastText).text).toBeTruthy();
-
-	// Paragraph 4 - dangling unescapes
-	expect(ast.children[3].children.length).toBe(1);
-	expect((ast.children[3].children[0] as YeastText).text).toBe('Dangling escaped markers, like * or __ or |, should get unescaped.');
+	// Validate AST
+	expect(JSON.stringify(ast)).toBe(JSON.stringify(EVERYTHING_INLINE_AST));
 });
@@ -437,0 +397,0 @@

src/plugins/inline/InlineEmphasisPlugin.ts

@@ -11,6 +11,13 @@ import {
 
-const ITALICS_REGEX_UNDERSCORES = /(\\?)(_)([^\s_]|\S.*?\S)(\\?)(_)/gi;
-const ITALICS_REGEX_ASTERISKS = /(\\?)(\*)([^\s_]|\S.*?\S)(\\?)(\*)/gi;
-const BOLD_REGEX_UNDERSCORES = /(\\?)(__)([^\s_]|\S.*?\S)(\\?)(__)/gi;
-const BOLD_REGEX_ASTERISKS = /(\\?)(\*\*)([^\s_]|\S.*?\S)(\\?)(\*\*)/gi;
+/**
+ * These expressions look for text between the various emphasis marker types with the following caveats:
+ * - the beginning of the match is the beginning of the string or not an escape slash
+ * - the next character is the opening marker
+ * - the first character after the opening marker is not whitespace
+ * - the last character before the closing marker is not whitespace
+ */
+const ITALICS_REGEX_UNDERSCORES = /(?:^|([^\\]))_(?:(\\_|[^\\\s][^_\s]?)_|((?:\\_|\S)(?:\\_|[^_])*?[^\s\\])_)/gi;
+const ITALICS_REGEX_ASTERISKS = /(?:^|([^\\]))\*(?:(\\\*|[^\\\s][^\*\s]?)\*|((?:\\\*|\S)(?:\\\*|[^\*])*?[^\s\\])\*)/gi;
+const BOLD_REGEX_UNDERSCORES = /(?:^|([^\\]))__(?:(\\_|[^\\\s][^_\s]?)__|((?:\\_|\S)(?:\\_|[^_])*?[^\s\\])__)/gi;
+const BOLD_REGEX_ASTERISKS = /(?:^|([^\\]))\*\*(?:(\\\*|[^\\\s][^\*\s]?)\*\*|((?:\\\*|\S)(?:\\\*|[^\*])*?[^\s\\])\*\*)/gi;
 
@@ -25,19 +32,13 @@ /**
 		/**
-		 * match[1] -> (optional) escape slash
-		 * match[2] -> opening characters
-		 * match[3] -> encased text
-		 * match[4] -> (optional) escape slash
-		 * match[5] -> closing characters
+		 * match[1] -> preceding character
+		 * match[2] -> encased text (1 char)
+		 * match[3] -> encased text (2+ chars)
 		 */
 		const parseMatch = (match: RegExpMatchArray, nodeType: YeastInlineNodeTypes) => {
-			let node: YeastInlineChild;
-			if (match.length == 6 && match[1] && match[4]) {
-				node = YeastNodeFactory.CreateText();
-				node.text = `${match[2]}${match[3]}${match[5]}`;
-			} else {
-				node = YeastNodeFactory.Create(nodeType) as YeastInlineNode;
-				node.children = parser.parseInline(match[3]);
-			}
+			if (match.length < 4 || (!match[2] && !match[3])) return;
+			const startOffset = (match[1] || '').length;
+			let node = YeastNodeFactory.Create(nodeType) as YeastInlineNode;
+			node.children = parser.parseInline(match[2] || match[3]);
 			tokens.push({
-				start: match.index,
+				start: match.index + startOffset,
 				end: match.index + match[0].length,
@@ -65,39 +66,4 @@ from: 'InlineEmphasisPlugin',
 
-		//Parse for asterisk syntax
-		const textArr = text.split('');
-		let index = 0;
-		while (index < textArr.length) {
-			if (textArr[index] === '*' && textArr[index + 1] !== '*' && textArr[index + 1] !== ' ') {
-				if (index >= 0 && textArr[index - 1] !== '*') {
-					let italizedText = '';
-					let startIndex = index;
-					let isInvalidSyntax = false;
-					do {
-						if (textArr[index + 1]) {
-							italizedText += textArr[++index];
-						} else {
-							//There was probably no closing tag
-							isInvalidSyntax = true;
-							break;
-						}
-					} while (textArr[index + 1] !== '*' && index < textArr.length);
-					if (isInvalidSyntax) {
-						index++;
-						continue;
-					}
-					const node = YeastNodeFactory.CreateItalicNode();
-					node.children = parser.parseInline(italizedText);
-					tokens.push({
-						start: startIndex,
-						end: startIndex + italizedText.length + 2,
-						from: 'ItalicsInlinePlugin',
-						nodes: [node],
-					});
-					index++;
-				}
-			}
-			index++;
-		}
 		return tokens;
 	}
 }

src/plugins/inline/InlineLinkPlugin.ts

 import { InlineTokenizerPlugin, Token, YeastParser, YeastNodeFactory, isYeastNodeType, YeastInlineNodeTypes, scrapeText } from 'yeast-core';
 
-const LINK_REGEX = /\[([^\[\]]*(?:\\.[^\[\]]*)*)\]\((.+?)(?:\s["'](.*?)["'])?\)/gi;
+const LINK_REGEX = /\[\s*([^\[\]]*?(?:\\.[^\[\]]*?)*?)\s*\]\(\s*(\S+)(?:\s+["']\s*(.*?)\s*["'])?\s*\)/gi;
 
@@ -9,4 +9,5 @@ export class InlineLinkPlugin implements InlineTokenizerPlugin {
 		for (const match of text.matchAll(LINK_REGEX)) {
-			// Don't process if it's an image
-			if (text.charAt(match.index - 1) === '!') {
+			// Don't process if it's an image or escaped
+			const charBefore = text.charAt(match.index - 1);
+			if (charBefore === '!' || charBefore === '\\') {
 				continue;
@@ -13,0 +14,0 @@ }

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

209923

increased by3.97%

Number of package files

103

increased by0.98%

Lines of code

5569

increased by15.42%

No dependency changes