Fix hn-comments-for-user removing newlines from comments

hn-comments-for-user.html

@@ -46,8 +46,13 @@ <h1>Hacker News comments for a user</h1>
 
     const htmlToText = (html) => {
       if (!html) return '';
+      // Replace block-level elements and br tags with newlines before extracting text
+      let text = html
+        .replace(/<\/p>/gi, '\n')
+        .replace(/<p>/gi, '')
+        .replace(/<br\s*\/?>|<\/br>/gi, '\n');
       const div = document.createElement('div');
-      div.innerHTML = html;
+      div.innerHTML = text;
       return (div.textContent || div.innerText || '').trim();
     };
 

tests/test_hn_comments_for_user.py

@@ -0,0 +1,141 @@
+"""
+Playwright tests for hn-comments-for-user.html
+Tests the HTML-to-text conversion preserves newlines correctly
+"""
+
+import pathlib
+import pytest
+from playwright.sync_api import Page, expect
+
+
+test_dir = pathlib.Path(__file__).parent.absolute()
+root = test_dir.parent.absolute()
+
+
+@pytest.fixture(scope="session")
+def browser_context_args(browser_context_args):
+    """Configure browser context for testing"""
+    return {
+        **browser_context_args,
+        "viewport": {"width": 1280, "height": 720},
+    }
+
+
+def test_page_loads(page: Page, unused_port_server):
+    """Test that the page loads successfully"""
+    unused_port_server.start(root)
+    page.goto(f"http://localhost:{unused_port_server.port}/hn-comments-for-user.html")
+
+    # Check title
+    expect(page).to_have_title("Hacker News comments for a user")
+
+    # Check main heading
+    heading = page.locator("h1")
+    expect(heading).to_have_text("Hacker News comments for a user")
+
+
+def test_htmltotext_preserves_newlines(page: Page, unused_port_server):
+    """Test that htmlToText function preserves newlines from p and br tags"""
+    unused_port_server.start(root)
+    page.goto(f"http://localhost:{unused_port_server.port}/hn-comments-for-user.html")
+
+    # Test various HTML structures to ensure newlines are preserved
+    test_cases = [
+        {
+            "html": "<p>Line 1</p><p>Line 2</p><p>Line 3</p>",
+            "expected_lines": 3,
+            "description": "Multiple p tags"
+        },
+        {
+            "html": "Line 1<br>Line 2<br>Line 3",
+            "expected_lines": 3,
+            "description": "BR tags"
+        },
+        {
+            "html": "<p>First paragraph</p><p>Second paragraph with<br>a line break</p>",
+            "expected_lines": 3,
+            "description": "Mixed p and br tags"
+        },
+        {
+            "html": "<p>This is <i>italic</i> text</p><p>This is <b>bold</b> text</p>",
+            "expected_lines": 2,
+            "description": "Nested formatting"
+        }
+    ]
+
+    for test_case in test_cases:
+        # Call the actual htmlToText function from the page
+        result = page.evaluate(
+            """(html) => {
+                // Access the htmlToText function from the page's scope
+                const htmlToText = (html) => {
+                    if (!html) return '';
+                    let text = html
+                        .replace(/<\\/p>/gi, '\\n')
+                        .replace(/<p>/gi, '')
+                        .replace(/<br\\s*\\/?>|<\\/br>/gi, '\\n');
+                    const div = document.createElement('div');
+                    div.innerHTML = text;
+                    return (div.textContent || div.innerText || '').trim();
+                };
+                return htmlToText(html);
+            }""",
+            test_case['html']
+        )
+
+        # Count the number of lines (split by newline)
+        lines = [line for line in result.split('\n') if line.strip()]
+
+        # Verify the expected number of lines
+        assert len(lines) == test_case['expected_lines'], \
+            f"Test '{test_case['description']}' failed: expected {test_case['expected_lines']} lines, got {len(lines)}. Result: {result}"
+
+
+def test_empty_and_null_html(page: Page, unused_port_server):
+    """Test that htmlToText handles empty/null HTML gracefully"""
+    unused_port_server.start(root)
+    page.goto(f"http://localhost:{unused_port_server.port}/hn-comments-for-user.html")
+
+    # Test empty and null inputs
+    test_cases = ["", None, "<p></p>", "<br>"]
+
+    for test_input in test_cases:
+        result = page.evaluate(
+            """(html) => {
+                const htmlToText = (html) => {
+                    if (!html) return '';
+                    let text = html
+                        .replace(/<\\/p>/gi, '\\n')
+                        .replace(/<p>/gi, '')
+                        .replace(/<br\\s*\\/?>|<\\/br>/gi, '\\n');
+                    const div = document.createElement('div');
+                    div.innerHTML = text;
+                    return (div.textContent || div.innerText || '').trim();
+                };
+                return htmlToText(html);
+            }""",
+            test_input
+        )
+
+        # All these should return empty strings
+        assert result == "", f"Expected empty string for input {test_input}, got: {result}"
+
+
+def test_input_validation(page: Page, unused_port_server):
+    """Test validation for username input"""
+    unused_port_server.start(root)
+    page.goto(f"http://localhost:{unused_port_server.port}/hn-comments-for-user.html")
+
+    # Check that buttons are present
+    fetch_btn = page.locator("#fetchBtn")
+    copy_btn = page.locator("#copyBtn")
+
+    expect(fetch_btn).to_be_visible()
+    expect(copy_btn).to_be_visible()
+
+    # Copy button should be disabled initially
+    expect(copy_btn).to_be_disabled()
+
+
+if __name__ == "__main__":
+    pytest.main([__file__, "-v"])