加入pandoc解析docx

2026-02-15 21:54:54 +08:00
parent f167aa2111
commit 4324699a3d
3 changed files with 54 additions and 6 deletions
--- a/temp/scripts/parser.py
+++ b/temp/scripts/parser.py
@@ -64,6 +64,7 @@ def main() -> None:

    if file_type == "docx":
        parsers = [
+            ("pypandoc-binary", docx_parser.parse_docx_with_pypandoc),
            ("MarkItDown", docx_parser.parse_docx_with_markitdown),
            ("python-docx", docx_parser.parse_docx_with_python_docx),
            ("XML 原生解析", docx_parser.parse_docx_with_xml),