refactor: 统一 HTML Reader 的 parse 签名,使用文件路径参数

将所有 HTML Parser 的函数签名从接收 HTML 字符串改为接收文件路径,
与其他 Reader(PDF、DOCX 等)保持一致。

主要变更:
- 修改 PARSERS 列表,移除 lambda 表达式,直接传递函数引用
- 在 HtmlReader.parse() 中统一管理临时文件(UTF-8 编码)
- 每个 Parser 使用独立的临时文件副本,用完即清理
- 移除 download_and_parse() 方法,逻辑合并到 parse() 中
- 更新相关测试,改为直接传递文件路径

受影响的 Parser:
- trafilatura.parse(html_content) -> parse(file_path)
- domscribe.parse(html_content) -> parse(file_path)
- markitdown.parse(html_content, temp_file_path) -> parse(file_path)
- html2text.parse(html_content) -> parse(file_path)
This commit is contained in:
2026-03-09 00:05:23 +08:00
parent 09904aefdc
commit 2b81dd49fe
10 changed files with 132 additions and 146 deletions

View File

@@ -25,20 +25,16 @@ class TestHtmlReadersConsistency:
</html>
""")
# 读取 HTML 内容
with open(file_path, 'r', encoding='utf-8') as f:
html_content = f.read()
parsers = [
("html2text", lambda c: html2text.parse(c)),
("markitdown", lambda c: markitdown.parse(c, file_path)),
("trafilatura", lambda c: trafilatura.parse(c)),
("domscribe", lambda c: domscribe.parse(c)),
("html2text", html2text.parse),
("markitdown", markitdown.parse),
("trafilatura", trafilatura.parse),
("domscribe", domscribe.parse),
]
successful_results = []
for name, parser in parsers:
content, error = parser(html_content)
content, error = parser(file_path)
if content is not None and content.strip():
successful_results.append((name, content))