Merge branch 'main' into @Skalakid/remove-nbsp-from-inline-code-blocks

__tests__/ExpensiMark-HTML-test.js

@@ -472,29 +472,29 @@ test('Test inline code blocks', () => {
     expect(parser.replace(inlineCodeStartString)).toBe('My favorite language is <code>JavaScript</code>. How about you?');
 });
 
-test('Test inline code blocks with double backticks', () => {
-    const inlineCodeStartString = 'My favorite language is ``JavaScript``. How about you?';
-    expect(parser.replace(inlineCodeStartString)).toBe('My favorite language is <code>&#x60;JavaScript&#x60;</code>. How about you?');
+test('Test double backticks', () => {
+    const testString = 'My favorite language is ``JavaScript``. How about you?';
+    expect(parser.replace(testString)).toBe('My favorite language is &#x60;&#x60;JavaScript&#x60;&#x60;. How about you?');
 });
 
 test('Test inline code blocks with triple backticks', () => {
     const inlineCodeStartString = 'My favorite language is ```JavaScript```. How about you?';
-    expect(parser.replace(inlineCodeStartString)).toBe('My favorite language is <code>&#x60;&#x60;JavaScript&#x60;&#x60;</code>. How about you?');
+    expect(parser.replace(inlineCodeStartString)).toBe('My favorite language is &#x60;&#x60;<code>JavaScript</code>&#x60;&#x60;. How about you?');
 });
 
 test('Test multiple inline codes in one line', () => {
     const inlineCodeStartString = 'My favorite language is `JavaScript`. How about you? I also like `Python`.';
     expect(parser.replace(inlineCodeStartString)).toBe('My favorite language is <code>JavaScript</code>. How about you? I also like <code>Python</code>.');
 });
 
-test('Test inline code with one backtick as content', () => {
-    const inlineCodeStartString = '```';
-    expect(parser.replace(inlineCodeStartString)).toBe('&#x60;&#x60;&#x60;');
+test('Test triple backticks', () => {
+    const testString = '```';
+    expect(parser.replace(testString)).toBe('&#x60;&#x60;&#x60;');
 });
 
-test('Test inline code with multiple backtick symbols as content', () => {
-    const inlineCodeStartString = '``````';
-    expect(parser.replace(inlineCodeStartString)).toBe('&#x60;&#x60;&#x60;&#x60;&#x60;&#x60;');
+test('Test multiple backtick symbols', () => {
+    const testString = '``````';
+    expect(parser.replace(testString)).toBe('&#x60;&#x60;&#x60;&#x60;&#x60;&#x60;');
 });
 
 test('Test inline code blocks with ExpensiMark syntax inside', () => {
@@ -508,9 +508,9 @@ test('Test inline code blocks inside ExpensiMark', () => {
     expect(parser.replace(testString)).toBe(resultString);
 });
 
-test('Test inline code blocks with two backticks', () => {
+test('Test words enclosed in double backticks', () => {
     const testString = '``JavaScript``';
-    expect(parser.replace(testString)).toBe('<code>&#x60;JavaScript&#x60;</code>');
+    expect(parser.replace(testString)).toBe('&#x60;&#x60;JavaScript&#x60;&#x60;');
 });
 
 test('Test code fencing with ExpensiMark syntax inside', () => {
@@ -1240,10 +1240,9 @@ test('Test for backticks with complete escaped backtick characters inside it', (
     expect(parser.replace(testString)).toBe(resultString);
 });
 
-// Backticks with only tab characters inside it are not replaced with <code>
-test('Test for backticks only tab characters inside it', () => {
+test('Test for inline code blocks with only tab characters as content', () => {
     const testString = '`\u0009`';
-    const resultString = '&#x60;\u0009&#x60;';
+    const resultString = '<code>	</code>';
     expect(parser.replace(testString)).toBe(resultString);
 });
 
@@ -1254,10 +1253,15 @@ test('Test for backticks with only space characters as content', () => {
     expect(parser.replace(testString)).toBe(resultString);
 });
 
-// Code-fence with spaces as content
 test('Test for inline code block with triple backtick with spaces as content', () => {
     const testString = '```   ```';
-    const resultString = '<code>&#x60;&#x60;   &#x60;&#x60;</code>';
+    const resultString = '&#x60;&#x60;<code>&nbsp;&nbsp;&nbsp;</code>&#x60;&#x60;';
+    expect(parser.replace(testString)).toBe(resultString);
+});
+
+test('Test for a normal code block, an empty code block, followed by a word and a backtick', () => {
+    const testString = '`hello` `` hi`';
+    const resultString = '<code>hello</code> &#x60;&#x60; hi&#x60;';
     expect(parser.replace(testString)).toBe(resultString);
 });
 

__tests__/ExpensiMark-HTMLToText-test.js

@@ -195,7 +195,11 @@ test('Mention report html to text', () => {
     expect(parser.htmlToText(testString, extras)).toBe('#room-name');
 });
 
-test('Test replacement for <img> tags', () => {
+test('Test replacement for attachment tags', () => {
     const testString = '<img src="https://example.com/image.png" alt="Image description" />';
+    const testString2 = '<video data-expensify-source="https://www.expensify.com/chat-attachments/123/test.mp4" data-expensify-width=720 data-expensify-height=1640 data-expensify-duration=20>test.mp4</video>';
+    const testString3 = '<a href="https://www.expensify.com/chat-attachments/123/test.csv" data-expensify-source="https://www.expensify.com/chat-attachments/123/test.csv">test.csv</a>';
     expect(parser.htmlToText(testString)).toBe('[Attachment]');
+    expect(parser.htmlToText(testString2)).toBe('[Attachment]');
+    expect(parser.htmlToText(testString3)).toBe('[Attachment]');
 });

lib/ExpensiMark.ts

@@ -227,9 +227,12 @@ export default class ExpensiMark {
                 // but we should not replace backtick symbols if they include <pre> tags between them.
                 // At least one non-whitespace character or a specific whitespace character (" " and "\u00A0")
                 // must be present inside the backticks.
-                regex: /(\B|_|)&#x60;((?:&#x60;)*(?!&#x60;).*?[\S| |\u00A0]+?.*?(?<!&#x60;)(?:&#x60;)*)&#x60;(\B|_|)(?!&#x60;|(?!<pre>)[^<]*(?:<(?!pre>)[^<]*)*<\/pre>|[^<]*<\/video>)/gm,
+                regex: /(\B|_|)&#x60;(.*?)&#x60;(\B|_|)(?!(?!<pre>)[^<]*(?:<(?!pre>)[^<]*)*<\/pre>|[^<]*<\/video>)/gm,
                 replacement: (_extras, _match, g1, g2, g3) => {
                     const g2Value = g2.trim() === '' ? g2.replaceAll(' ', '&nbsp;') : g2;
+                    if (!g2Value) {
+                        return _match;
+                    }
                     return `${g1}<code>${g2Value}</code>${g3}`;
                 },
                 rawInputReplacement: (_extras, _match, g1, g2, g3) => `${g1}<code>${g2}</code>${g3}`,
@@ -802,6 +805,16 @@ export default class ExpensiMark {
                 regex: /<img[^><]*src\s*=\s*(['"])(.*?)\1(?:[^><]*alt\s*=\s*(['"])(.*?)\3)?[^><]*>*(?![^<][\s\S]*?(<\/pre>|<\/code>))/gi,
                 replacement: '[Attachment]',
             },
+            {
+                name: 'video',
+                regex: /<video[^><]*data-expensify-source\s*=\s*(['"])(\S*?)\1(.*?)>([^><]*)<\/video>*(?![^<][\s\S]*?(<\/pre>|<\/code>))/gi,
+                replacement: '[Attachment]',
+            },
+            {
+                name: 'otherAttachments',
+                regex: /<a[^><]*data-expensify-source\s*=\s*(['"])(\S*?)\1(.*?)>([^><]*)<\/a>*(?![^<][\s\S]*?(<\/pre>|<\/code>))/gi,
+                replacement: '[Attachment]',
+            },
             {
                 name: 'reportMentions',
                 regex: /<mention-report reportID="(\d+)" *\/>/gi,

package.json

@@ -1,6 +1,6 @@
 {
   "name": "expensify-common",
-  "version": "2.0.110",
+  "version": "2.0.114",
   "author": "Expensify, Inc.",
   "description": "Expensify libraries and components shared across different repos",
   "homepage": "https://expensify.com",