@postalsys/email-text-tools 2.3.3 → 2.3.5

package/.ncurc.js

@@ -2,6 +2,7 @@ module.exports = {
     upgrade: true,
     reject: [
         // Block package upgrades that moved to ESM
-        'nanoid'
+        'nanoid',
+        'jsdom' // parse5 dependency in 27.0.1 is ESM only
     ]
 };

package/.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-    ".": "2.3.3"
+    ".": "2.3.5"
 }

package/CHANGELOG.md

@@ -1,5 +1,19 @@
 # Changelog
 
+## [2.3.5](https://github.com/postalsys/email-text-tools/compare/email-text-tools-v2.3.4...email-text-tools-v2.3.5) (2025-10-29)
+
+
+### Bug Fixes
+
+* **htmlToText:** Improve HTML-to-text conversion for better newline preservation ([6606d5c](https://github.com/postalsys/email-text-tools/commit/6606d5c5b9d3b05f41d54c434f2b68760141dacb))
+
+## [2.3.4](https://github.com/postalsys/email-text-tools/compare/email-text-tools-v2.3.3...email-text-tools-v2.3.4) (2025-10-23)
+
+
+### Bug Fixes
+
+* Downgraded jsdom to 27.0.0 ([a2969e9](https://github.com/postalsys/email-text-tools/commit/a2969e9399ed6c9d15a85d7d8c11342bc5bf00ce))
+
 ## [2.3.3](https://github.com/postalsys/email-text-tools/compare/email-text-tools-v2.3.2...email-text-tools-v2.3.3) (2025-10-23)
 
 

package/lib/html-to-text.js

@@ -1,16 +1,54 @@
 'use strict';
 const { compile } = require('html-to-text');
 
-const htmlToText = compile({
+// Post-process function to clean up excessive newlines
+function cleanupNewlines(text) {
+    // Replace 3 or more consecutive newlines with just 2
+    // This handles cases where HTML has many nested empty divs
+    return text.replace(/\n{3,}/g, '\n\n');
+}
+
+const htmlToTextCompiled = compile({
     wordwrap: 160,
     selectors: [
+        // Block elements that should preserve line breaks
         { selector: 'p', options: { leadingLineBreaks: 1, trailingLineBreaks: 1 } },
+        // Divs get minimal breaks - many HTML emails wrap each line in a div
+        { selector: 'div', options: { leadingLineBreaks: 1, trailingLineBreaks: 0 } },
         { selector: 'pre', options: { leadingLineBreaks: 1, trailingLineBreaks: 1 } },
-        { selector: 'table.class#id', format: 'dataTable' },
-        // ignore images
+        { selector: 'blockquote', options: { leadingLineBreaks: 1, trailingLineBreaks: 1, trimEmptyLines: true } },
+
+        // Headers should have spacing
+        { selector: 'h1', options: { leadingLineBreaks: 2, trailingLineBreaks: 1, uppercase: false } },
+        { selector: 'h2', options: { leadingLineBreaks: 2, trailingLineBreaks: 1, uppercase: false } },
+        { selector: 'h3', options: { leadingLineBreaks: 2, trailingLineBreaks: 1, uppercase: false } },
+        { selector: 'h4', options: { leadingLineBreaks: 2, trailingLineBreaks: 1, uppercase: false } },
+        { selector: 'h5', options: { leadingLineBreaks: 2, trailingLineBreaks: 1, uppercase: false } },
+        { selector: 'h6', options: { leadingLineBreaks: 2, trailingLineBreaks: 1, uppercase: false } },
+
+        // Line breaks
+        { selector: 'br', options: { leadingLineBreaks: 1, trailingLineBreaks: 0 } },
+        { selector: 'hr', options: { leadingLineBreaks: 1, trailingLineBreaks: 1 } },
+
+        // Lists should preserve structure
+        { selector: 'ul', options: { leadingLineBreaks: 1, trailingLineBreaks: 1 } },
+        { selector: 'ol', options: { leadingLineBreaks: 1, trailingLineBreaks: 1 } },
+
+        // Tables
+        { selector: 'table', format: 'dataTable' },
+
+        // Ignore images
         { selector: 'img', format: 'skip' }
     ],
-    hideLinkHrefIfSameAsText: true
+    hideLinkHrefIfSameAsText: true,
+    // Preserve whitespace better for content
+    preserveNewlines: true
 });
 
+// Wrapper function that applies post-processing
+function htmlToText(html) {
+    const text = htmlToTextCompiled(html);
+    return cleanupNewlines(text);
+}
+
 module.exports = htmlToText;

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@postalsys/email-text-tools",
-    "version": "2.3.3",
+    "version": "2.3.5",
     "description": "Email text tools",
     "main": "index.js",
     "scripts": {
@@ -34,7 +34,7 @@
         "dompurify": "3.3.0",
         "he": "1.2.0",
         "html-to-text": "9.0.5",
-        "jsdom": "27.0.1",
+        "jsdom": "27.0.0",
         "juice": "11.0.3",
         "linkify-it": "5.0.0",
         "moment-timezone": "0.6.0",

package/test/html-to-text.test.js

@@ -0,0 +1,226 @@
+'use strict';
+
+const { describe, it } = require('node:test');
+const assert = require('node:assert/strict');
+const htmlToText = require('../lib/html-to-text');
+
+describe('html-to-text conversion', () => {
+    it('should preserve newlines from br tags', () => {
+        const html = 'Line 1<br>Line 2<br>Line 3';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Line 1'));
+        assert.ok(text.includes('Line 2'));
+        assert.ok(text.includes('Line 3'));
+
+        // Check that lines are separated
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.equal(lines.length, 3);
+    });
+
+    it('should preserve newlines from div tags', () => {
+        const html = '<div>Line 1</div><div>Line 2</div><div>Line 3</div>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Line 1'));
+        assert.ok(text.includes('Line 2'));
+        assert.ok(text.includes('Line 3'));
+
+        // Check that lines are separated
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.equal(lines.length, 3);
+    });
+
+    it('should preserve newlines from paragraph tags', () => {
+        const html = '<p>Paragraph 1</p><p>Paragraph 2</p><p>Paragraph 3</p>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Paragraph 1'));
+        assert.ok(text.includes('Paragraph 2'));
+        assert.ok(text.includes('Paragraph 3'));
+
+        // Check that paragraphs are separated
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.equal(lines.length, 3);
+    });
+
+    it('should handle MS Graph style HTML with mixed formatting', () => {
+        // Typical MS Graph HTML structure
+        const html = `
+            <div>Hello there,</div>
+            <div><br></div>
+            <div>This is a test email with multiple lines.</div>
+            <div>Each line should be preserved.</div>
+            <div><br></div>
+            <div>Thanks,</div>
+            <div>User</div>
+        `;
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Hello there'));
+        assert.ok(text.includes('This is a test email'));
+        assert.ok(text.includes('Each line should be preserved'));
+        assert.ok(text.includes('Thanks'));
+        assert.ok(text.includes('User'));
+
+        // Verify structure is maintained
+        const lines = text.split('\n').map(l => l.trim()).filter(l => l);
+        assert.ok(lines.length >= 5, `Expected at least 5 lines, got ${lines.length}`);
+    });
+
+    it('should handle lists with proper line breaks', () => {
+        const html = '<ul><li>Item 1</li><li>Item 2</li><li>Item 3</li></ul>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Item 1'));
+        assert.ok(text.includes('Item 2'));
+        assert.ok(text.includes('Item 3'));
+
+        // Each list item should be on a new line
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.ok(lines.length >= 3);
+    });
+
+    it('should handle headers with appropriate spacing', () => {
+        const html = '<h1>Main Title</h1><p>Some content</p><h2>Subtitle</h2><p>More content</p>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Main Title'));
+        assert.ok(text.includes('Some content'));
+        assert.ok(text.includes('Subtitle'));
+        assert.ok(text.includes('More content'));
+
+        // Headers should create separation
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.ok(lines.length >= 4);
+    });
+
+    it('should handle blockquotes with line breaks', () => {
+        const html = '<p>Normal text</p><blockquote>Quoted text line 1<br>Quoted text line 2</blockquote><p>More normal text</p>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Normal text'));
+        assert.ok(text.includes('Quoted text line 1'));
+        assert.ok(text.includes('Quoted text line 2'));
+        assert.ok(text.includes('More normal text'));
+    });
+
+    it('should not create excessive blank lines', () => {
+        const html = '<p>Text 1</p><p>Text 2</p><p>Text 3</p>';
+        const text = htmlToText(html);
+
+        // Should not have more than 3 consecutive newlines
+        assert.ok(!text.includes('\n\n\n\n'), 'Should not have 4 or more consecutive newlines');
+    });
+
+    it('should handle tables with line breaks between rows', () => {
+        const html = '<table><tr><td>Cell 1</td><td>Cell 2</td></tr><tr><td>Cell 3</td><td>Cell 4</td></tr></table>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Cell 1'));
+        assert.ok(text.includes('Cell 2'));
+        assert.ok(text.includes('Cell 3'));
+        assert.ok(text.includes('Cell 4'));
+
+        // Rows should be separated
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.ok(lines.length >= 2);
+    });
+
+    it('should handle empty elements gracefully', () => {
+        const html = '<div>Text</div><div></div><div>More text</div>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Text'));
+        assert.ok(text.includes('More text'));
+    });
+
+    it('should preserve newlines in pre tags', () => {
+        const html = '<pre>Line 1\nLine 2\nLine 3</pre>';
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('Line 1'));
+        assert.ok(text.includes('Line 2'));
+        assert.ok(text.includes('Line 3'));
+    });
+
+    it('should handle complex nested structure', () => {
+        const html = `
+            <div>
+                <p>First paragraph</p>
+                <div>
+                    <div>Nested content</div>
+                </div>
+                <p>Last paragraph</p>
+            </div>
+        `;
+        const text = htmlToText(html);
+
+        assert.ok(text.includes('First paragraph'));
+        assert.ok(text.includes('Nested content'));
+        assert.ok(text.includes('Last paragraph'));
+
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.equal(lines.length, 3);
+    });
+
+    it('should handle real-world MS Graph HTML example', () => {
+        // Real example of how MS Graph might return HTML
+        const html = `
+            <html>
+            <head></head>
+            <body>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            Hi Team,</div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            <br>
+            </div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            I wanted to update you on the following:</div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            <br>
+            </div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            - Item 1</div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            - Item 2</div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            - Item 3</div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            <br>
+            </div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            Best regards,</div>
+            <div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
+            John</div>
+            </body>
+            </html>
+        `;
+        const text = htmlToText(html);
+
+        // Check all content is present
+        assert.ok(text.includes('Hi Team'));
+        assert.ok(text.includes('I wanted to update you'));
+        assert.ok(text.includes('- Item 1'));
+        assert.ok(text.includes('- Item 2'));
+        assert.ok(text.includes('- Item 3'));
+        assert.ok(text.includes('Best regards'));
+        assert.ok(text.includes('John'));
+
+        // Check structure is preserved
+        const lines = text.trim().split('\n').filter(l => l.trim());
+        assert.ok(lines.length >= 7, `Expected at least 7 content lines, got ${lines.length}`);
+
+        // Items should be on separate lines
+        const itemIndex1 = text.indexOf('- Item 1');
+        const itemIndex2 = text.indexOf('- Item 2');
+        const itemIndex3 = text.indexOf('- Item 3');
+
+        assert.ok(itemIndex2 > itemIndex1);
+        assert.ok(itemIndex3 > itemIndex2);
+
+        // There should be newlines between items
+        assert.ok(text.substring(itemIndex1, itemIndex2).includes('\n'));
+        assert.ok(text.substring(itemIndex2, itemIndex3).includes('\n'));
+    });
+});