feat: prune some deps

tests/parser/file/test_html_parser.py

@@ -1,16 +1,19 @@
 import pytest
-from pathlib import Path
-from unittest.mock import patch, MagicMock
-
-import sys
-import types
 
 from application.parser.file.html_parser import HTMLParser
 
 
+HTML = (
+    "<html><head><title>My Page</title></head>"
+    "<body><h1>Heading</h1><p>Hello world.</p></body></html>"
+)
+
+
 @pytest.fixture
-def html_parser():
-    return HTMLParser()
+def html_file(tmp_path):
+    path = tmp_path / "page.html"
+    path.write_text(HTML)
+    return path
 
 
 def test_html_init_parser():
@@ -21,23 +24,33 @@ def test_html_init_parser():
     assert parser.parser_config_set
 
 
-def test_html_parser_parse_file():
-    parser = HTMLParser()
-    mock_doc = MagicMock()
-    mock_doc.page_content = "Extracted HTML content"
-    mock_doc.metadata = {"source": "test.html"}
+def test_html_parser_extracts_text(html_file):
+    text = HTMLParser().parse_file(html_file)
+    assert "Heading" in text
+    assert "Hello world." in text
 
-    fake_lc = types.ModuleType("langchain_community")
-    fake_dl = types.ModuleType("langchain_community.document_loaders")
 
-    bshtml_mock = MagicMock(return_value=MagicMock(load=MagicMock(return_value=[mock_doc])))
-    fake_dl.BSHTMLLoader = bshtml_mock
-    fake_lc.document_loaders = fake_dl
+def test_html_parser_returns_plain_text(html_file):
+    """bulk.py stringifies parser output, so it must not be a repr.
 
-    with patch.dict(sys.modules, {
-        "langchain_community": fake_lc,
-        "langchain_community.document_loaders": fake_dl,
-    }):
-        result = parser.parse_file(Path("test.html"))
-        assert result == [mock_doc]
-        bshtml_mock.assert_called_once_with(Path("test.html"))
+    The previous langchain BSHTMLLoader returned a Document whose ``str()`` is
+    ``page_content='...' metadata={...}``, which leaked into indexed content.
+    """
+    text = HTMLParser().parse_file(html_file)
+    assert isinstance(text, str)
+    assert "page_content" not in str(text)
+    assert "metadata=" not in str(text)
+
+
+def test_html_parser_reports_title_metadata(html_file):
+    assert HTMLParser().get_file_metadata(html_file) == {"title": "My Page"}
+
+
+def test_html_parser_metadata_without_title(tmp_path):
+    path = tmp_path / "no_title.html"
+    path.write_text("<html><body><p>Just text.</p></body></html>")
+    assert HTMLParser().get_file_metadata(path) == {}
+
+
+def test_html_parser_metadata_unreadable_file(tmp_path):
+    assert HTMLParser().get_file_metadata(tmp_path / "missing.html") == {}