win11下 demo 代码运行报错

[yunchenran]

不论是win11的python环境还是wsl ubuntu均报错：

   ...: parser = EpubParser(engine="unstructured") # epub engines: unstructured
   ...: epub_data = parser.parse(epub_path)
   ...:
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[1], line 4
      2 epub_path = "xxx.epub"
      3 parser = EpubParser(engine="unstructured") # epub engines: unstructured
----> 4 epub_data = parser.parse(epub_path)

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/wisup_e2m/parsers/doc/epub_parser.py:165, in EpubParser.parse(self, file_name, file, extract_images, include_image_link_in_text, ignore_transparent_images, work_dir, image_dir, relative_path, **kwargs)
    163     if k in _epub_parser_params:
    164         kwargs[k] = v
--> 165 return self.get_parsed_data(**kwargs)

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/wisup_e2m/parsers/doc/epub_parser.py:132, in EpubParser.get_parsed_data(self, file_name, file, extract_images, include_image_link_in_text, ignore_transparent_images, work_dir, image_dir, relative_path, **kwargs)
    129     EpubParser._validate_input_file(file_name)
    131 if self.config.engine == "unstructured":
--> 132     return self._parse_by_unstructured(
    133         file_name=file_name,
    134         file=file,
    135         extract_images=extract_images,
    136         include_image_link_in_text=include_image_link_in_text,
    137         ignore_transparent_images=ignore_transparent_images,
    138         work_dir=work_dir,
    139         image_dir=image_dir,
    140         relative_path=relative_path,
    141     )
    142 else:
    143     raise NotImplementedError(f"Engine {self.config.engine} not supported")

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/wisup_e2m/parsers/doc/epub_parser.py:84, in EpubParser._parse_by_unstructured(self, file_name, file, extract_images, include_image_link_in_text, ignore_transparent_images, work_dir, image_dir, relative_path)
     79 logger.info(f"Parsing {file_name} using unstructured engine...")
     81 import unstructured
     83 unstructured_elements: List[unstructured.documents.elements.Element] = (
---> 84     self.unstructured_parse_func(
     85         filename=file_name,
     86         file=file,
     87         languages=self.config.langs,
     88     )
     89 )
     91 if extract_images:
     92     epub_images = get_epub_images(
     93         file_name=file_name,
     94         file=file,
     95         target_image_dir=image_dir,
     96         ignore_transparent_images=ignore_transparent_images,
     97     )

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/unstructured/partition/epub.py:48, in partition_epub(filename, file, metadata_filename, metadata_last_modified, languages, detect_language_per_element, **kwargs)
     44 exactly_one(filename=filename, file=file)
     46 last_modified = get_last_modified_date(filename) if filename else None
---> 48 html_text = convert_file_to_html_text_using_pandoc(
     49     source_format="epub", filename=filename, file=file
     50 )
     52 return partition_html(
     53     text=html_text,
     54     encoding="unicode",
   (...)
     61     **kwargs,
     62 )

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/unstructured/file_utils/file_conversion.py:67, in convert_file_to_html_text_using_pandoc(source_format, filename, file)
     62         return convert_file_to_text(
     63             filename=tmp_file_path, source_format=source_format, target_format="html"
     64         )
     66 assert filename is not None
---> 67 return convert_file_to_text(
     68     filename=filename, source_format=source_format, target_format="html"
     69 )

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/unstructured/utils.py:217, in requires_dependencies.<locals>.decorator.<locals>.wrapper(*args, **kwargs)
    214 @wraps(func)
    215 def wrapper(*args: _P.args, **kwargs: _P.kwargs):
    216     run_check()
--> 217     return func(*args, **kwargs)

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/unstructured/file_utils/file_conversion.py:17, in convert_file_to_text(filename, source_format, target_format)
     14 import pypandoc
     16 try:
---> 17     text = pypandoc.convert_file(filename, target_format, format=source_format)
     18 except FileNotFoundError as err:
     19     msg = (
     20         f"Error converting the file to text. Ensure you have the pandoc package installed on"
     21         f" your system. Installation instructions are available at"
     22         f" https://pandoc.org/installing.html. The original exception text was:\n{err}"
     23     )

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/pypandoc/__init__.py:206, in convert_file(source_file, to, format, extra_args, encoding, outputfile, filters, verify_format, sandbox, cworkdir, sort_files)
    203 if len(discovered_source_files) == 1:
    204     discovered_source_files = discovered_source_files[0]
--> 206 return _convert_input(discovered_source_files, format, 'path', to, extra_args=extra_args,
    207                   outputfile=outputfile, filters=filters,
    208                   verify_format=verify_format, sandbox=sandbox,
    209                   cworkdir=cworkdir, sort_files=sort_files)

File ~/miniconda3/envs/e2m/lib/python3.10/site-packages/pypandoc/__init__.py:387, in _convert_input(source, format, input_type, to, extra_args, outputfile, filters, verify_format, sandbox, cworkdir, sort_files)
    384     input_file = []
    386 if sort_files:
--> 387     input_file = sorted(input_file)
    389 args = [__pandoc_path, '--from=' + format]
    391 args.append('--to=' + to)

TypeError: 'PosixPath' object is not iterable