Fix Windows Unicode encoding errors when writing HTML files (#7)

* Fix Windows Unicode encoding errors when writing HTML files

On Windows, the default file encoding is cp1252, which cannot encode
Unicode characters like emojis. This caused UnicodeEncodeError when
writing HTML files containing emoji characters.

Fixed by explicitly specifying encoding="utf-8" for all read_text()
and write_text() calls on HTML files.

* Fix Windows test encoding issues

Add encoding="utf-8" to all read_text() and write_text() calls in tests
to ensure proper handling of UTF-8 encoded HTML files on Windows.

* Set PYTHONUTF8 environment variable for tests

For windows, refs:
- #6

---------

Co-authored-by: Claude <noreply@anthropic.com>

Author: simonw

.github/workflows/test.yml

@@ -24,5 +24,7 @@ jobs:
       run: |
         pip install -e . --group dev
     - name: Run tests
+      env:
+        PYTHONUTF8: "1"
       run: |
         python -m pytest

src/claude_code_transcripts/__init__.py

@@ -755,13 +755,13 @@ def inject_gist_preview_js(output_dir):
     """Inject gist preview JavaScript into all HTML files in the output directory."""
     output_dir = Path(output_dir)
     for html_file in output_dir.glob("*.html"):
-        content = html_file.read_text()
+        content = html_file.read_text(encoding="utf-8")
         # Insert the gist preview JS before the closing </body> tag
         if "</body>" in content:
             content = content.replace(
                 "</body>", f"<script>{GIST_PREVIEW_JS}</script>\n</body>"
             )
-            html_file.write_text(content)
+            html_file.write_text(content, encoding="utf-8")
 
 
 def create_gist(output_dir, public=False):
@@ -894,7 +894,9 @@ def generate_html(json_path, output_dir, github_repo=None):
             pagination_html=pagination_html,
             messages_html="".join(messages_html),
         )
-        (output_dir / f"page-{page_num:03d}.html").write_text(page_content)
+        (output_dir / f"page-{page_num:03d}.html").write_text(
+            page_content, encoding="utf-8"
+        )
         print(f"Generated page-{page_num:03d}.html")
 
     # Calculate overall stats and collect all commits for timeline
@@ -977,7 +979,7 @@ def generate_html(json_path, output_dir, github_repo=None):
         index_items_html="".join(index_items),
     )
     index_path = output_dir / "index.html"
-    index_path.write_text(index_content)
+    index_path.write_text(index_content, encoding="utf-8")
     print(
         f"Generated {index_path.resolve()} ({total_convs} prompts, {total_pages} pages)"
     )
@@ -1308,7 +1310,9 @@ def generate_html_from_session_data(session_data, output_dir, github_repo=None):
             pagination_html=pagination_html,
             messages_html="".join(messages_html),
         )
-        (output_dir / f"page-{page_num:03d}.html").write_text(page_content)
+        (output_dir / f"page-{page_num:03d}.html").write_text(
+            page_content, encoding="utf-8"
+        )
         click.echo(f"Generated page-{page_num:03d}.html")
 
     # Calculate overall stats and collect all commits for timeline
@@ -1391,7 +1395,7 @@ def generate_html_from_session_data(session_data, output_dir, github_repo=None):
         index_items_html="".join(index_items),
     )
     index_path = output_dir / "index.html"
-    index_path.write_text(index_content)
+    index_path.write_text(index_content, encoding="utf-8")
     click.echo(
         f"Generated {index_path.resolve()} ({total_convs} prompts, {total_pages} pages)"
     )

tests/test_generate_html.py

@@ -66,23 +66,23 @@ def test_generates_index_html(self, output_dir, snapshot_html):
         fixture_path = Path(__file__).parent / "sample_session.json"
         generate_html(fixture_path, output_dir, github_repo="example/project")
 
-        index_html = (output_dir / "index.html").read_text()
+        index_html = (output_dir / "index.html").read_text(encoding="utf-8")
         assert index_html == snapshot_html
 
     def test_generates_page_001_html(self, output_dir, snapshot_html):
         """Test page-001.html generation."""
         fixture_path = Path(__file__).parent / "sample_session.json"
         generate_html(fixture_path, output_dir, github_repo="example/project")
 
-        page_html = (output_dir / "page-001.html").read_text()
+        page_html = (output_dir / "page-001.html").read_text(encoding="utf-8")
         assert page_html == snapshot_html
 
     def test_generates_page_002_html(self, output_dir, snapshot_html):
         """Test page-002.html generation (continuation page)."""
         fixture_path = Path(__file__).parent / "sample_session.json"
         generate_html(fixture_path, output_dir, github_repo="example/project")
 
-        page_html = (output_dir / "page-002.html").read_text()
+        page_html = (output_dir / "page-002.html").read_text(encoding="utf-8")
         assert page_html == snapshot_html
 
     def test_github_repo_autodetect(self, sample_session):
@@ -346,16 +346,16 @@ def test_injects_js_into_html_files(self, output_dir):
         """Test that JS is injected before </body> tag."""
         # Create test HTML files
         (output_dir / "index.html").write_text(
-            "<html><body><h1>Test</h1></body></html>"
+            "<html><body><h1>Test</h1></body></html>", encoding="utf-8"
         )
         (output_dir / "page-001.html").write_text(
-            "<html><body><p>Page 1</p></body></html>"
+            "<html><body><p>Page 1</p></body></html>", encoding="utf-8"
         )
 
         inject_gist_preview_js(output_dir)
 
-        index_content = (output_dir / "index.html").read_text()
-        page_content = (output_dir / "page-001.html").read_text()
+        index_content = (output_dir / "index.html").read_text(encoding="utf-8")
+        page_content = (output_dir / "page-001.html").read_text(encoding="utf-8")
 
         # Check JS was injected
         assert GIST_PREVIEW_JS in index_content
@@ -368,11 +368,13 @@ def test_injects_js_into_html_files(self, output_dir):
     def test_skips_files_without_body(self, output_dir):
         """Test that files without </body> are not modified."""
         original_content = "<html><head><title>Test</title></head></html>"
-        (output_dir / "fragment.html").write_text(original_content)
+        (output_dir / "fragment.html").write_text(original_content, encoding="utf-8")
 
         inject_gist_preview_js(output_dir)
 
-        assert (output_dir / "fragment.html").read_text() == original_content
+        assert (output_dir / "fragment.html").read_text(
+            encoding="utf-8"
+        ) == original_content
 
     def test_handles_empty_directory(self, output_dir):
         """Test that empty directories don't cause errors."""
@@ -389,8 +391,12 @@ def test_creates_gist_successfully(self, output_dir, monkeypatch):
         import click
 
         # Create test HTML files
-        (output_dir / "index.html").write_text("<html><body>Index</body></html>")
-        (output_dir / "page-001.html").write_text("<html><body>Page</body></html>")
+        (output_dir / "index.html").write_text(
+            "<html><body>Index</body></html>", encoding="utf-8"
+        )
+        (output_dir / "page-001.html").write_text(
+            "<html><body>Page</body></html>", encoding="utf-8"
+        )
 
         # Mock subprocess.run to simulate successful gh gist create
         mock_result = subprocess.CompletedProcess(
@@ -425,7 +431,9 @@ def test_raises_on_gh_cli_error(self, output_dir, monkeypatch):
         import click
 
         # Create test HTML file
-        (output_dir / "index.html").write_text("<html><body>Test</body></html>")
+        (output_dir / "index.html").write_text(
+            "<html><body>Test</body></html>", encoding="utf-8"
+        )
 
         # Mock subprocess.run to simulate gh error
         def mock_run(*args, **kwargs):
@@ -448,7 +456,9 @@ def test_raises_on_gh_not_found(self, output_dir, monkeypatch):
         import click
 
         # Create test HTML file
-        (output_dir / "index.html").write_text("<html><body>Test</body></html>")
+        (output_dir / "index.html").write_text(
+            "<html><body>Test</body></html>", encoding="utf-8"
+        )
 
         # Mock subprocess.run to simulate gh not found
         def mock_run(*args, **kwargs):
@@ -533,7 +543,7 @@ def mock_run(*args, **kwargs):
         assert result.exit_code == 0
         assert (output_dir / "index.html").exists()
         # Verify JS was injected
-        index_content = (output_dir / "index.html").read_text()
+        index_content = (output_dir / "index.html").read_text(encoding="utf-8")
         assert "gistpreview.github.io" in index_content
 
 
@@ -621,13 +631,13 @@ def test_long_text_in_continuation_appears_in_index(self, output_dir):
 
         # Write the session to a temp file
         session_file = output_dir / "test_session.json"
-        session_file.write_text(json.dumps(session_data))
+        session_file.write_text(json.dumps(session_data), encoding="utf-8")
 
         # Generate HTML
         generate_html(session_file, output_dir)
 
         # Read the index.html
-        index_html = (output_dir / "index.html").read_text()
+        index_html = (output_dir / "index.html").read_text(encoding="utf-8")
 
         # The long text summary should appear in the index
         # This is the bug: currently it doesn't because the continuation
@@ -942,7 +952,7 @@ def test_jsonl_generates_html(self, output_dir, snapshot_html):
         fixture_path = Path(__file__).parent / "sample_session.jsonl"
         generate_html(fixture_path, output_dir)
 
-        index_html = (output_dir / "index.html").read_text()
+        index_html = (output_dir / "index.html").read_text(encoding="utf-8")
         assert "hello world" in index_html.lower()
         assert index_html == snapshot_html
 
@@ -968,7 +978,7 @@ def test_gets_first_user_message_if_no_summary(self, tmp_path):
     def test_returns_no_summary_for_empty_file(self, tmp_path):
         """Test handling empty or invalid files."""
         jsonl_file = tmp_path / "empty.jsonl"
-        jsonl_file.write_text("")
+        jsonl_file.write_text("", encoding="utf-8")
         summary = get_session_summary(jsonl_file)
         assert summary == "(no summary)"