Update app.py

app.py

@@ -8,17 +8,8 @@ from pathlib import Path
 import re
 
 # os.system('pip install -U magic-pdf==0.8.1')
-# os.system('pip install -U magic_pdf-0.9.0a9-py3-none-any.whl')
 os.system('pip install git+https://github.com/opendatalab/MinerU.git@dev')
 
-# from huggingface_hub import snapshot_download
-# model_dir = snapshot_download('opendatalab/PDF-Extract-Kit')
-# layoutreader_model_dir = snapshot_download('hantian/layoutreader')
-
-# os.system('cp magic-pdf.template.json ~/magic-pdf.json')
-# os.system(f"sed -i 's|/tmp/models|{model_dir}/models|g' /home/user/magic-pdf.json")
-# os.system(f"sed -i 's|/tmp/layoutreader|{layoutreader_model_dir}|g' /home/user/magic-pdf.json")
-
 os.system('wget https://github.com/opendatalab/MinerU/raw/master/docs/download_models_hf.py -O download_models_hf.py')
 os.system('python download_models_hf.py')
 os.system("sed -i 's|cpu|cuda|g' /home/user/magic-pdf.json")
@@ -36,8 +27,6 @@ from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
 from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
 from magic_pdf.tools.common import do_parse, prepare_env
 
-# import spaces
-
 
 def read_fn(path):
     disk_rw = DiskReaderWriter(os.path.dirname(path))
@@ -131,28 +120,13 @@ def to_markdown(file_path, end_pages, ocr):
     # 返回转换后的PDF路径
     new_pdf_path = os.path.join(local_md_dir, file_name + "_layout.pdf")
 
-    # return md_content, txt_content, archive_zip_path, show_pdf(new_pdf_path)
     return md_content, txt_content, archive_zip_path, new_pdf_path
 
 
-# def show_pdf(file_path):
-#     with open(file_path, "rb") as f:
-#         base64_pdf = base64.b64encode(f.read()).decode('utf-8')
-#     pdf_display = f'<embed src="data:application/pdf;base64,{base64_pdf}" ' \
-#                   f'width="100%" height="1000" type="application/pdf">'
-#     return pdf_display
-
-
-def show_pdf(file):
-    return file
-
-
 latex_delimiters = [{"left": "$$", "right": "$$", "display": True},
                     {"left": '$', "right": '$', "display": False}]
 
 
-from magic_pdf.model.doc_analyze_by_custom_model import ModelSingleton
-
 def init_model():
     from magic_pdf.model.doc_analyze_by_custom_model import ModelSingleton
     try:
@@ -205,7 +179,6 @@ if __name__ == "__main__":
                                          latex_delimiters=latex_delimiters, line_breaks=True)
                     with gr.Tab("Markdown text"):
                         md_text = gr.TextArea(lines=45, show_copy_button=True)
-        # file.upload(fn=show_pdf, inputs=file, outputs=pdf_show)
         change_bu.click(fn=to_markdown, inputs=[pdf_show, max_pages, is_ocr], outputs=[md, md_text, output_file, pdf_show])
         clear_bu.add([md, pdf_show, md_text, output_file, is_ocr])