# -*- encoding: utf-8 -*-
# @Author: SWHL
# @Contact: liekkaskono@163.com
import gradio as gr
from gradio_pdf import PDF
from rapidocr_pdf import RapidOCRPDF
pdf_extracter = RapidOCRPDF()
def get_pdf_result(pdf_path: str, page_num: int = 0):
return pdf_extracter(pdf_path, page_num_list=[page_num])
with gr.Blocks(title="RapidOCR 📄 PDF", theme=gr.themes.Soft()) as demo:
gr.HTML(
"""
"""
)
pdf_path = PDF(label="Upload a PDF", interactive=True)
page_num = gr.Number(value=0, label="Select Extract Page", minimum=0, maximum=5)
btn = gr.Button("Run")
results = gr.Dataframe(
label="PDF Extract Results",
headers=["Index", "Txt", "Score"],
datatype=["number", "str", "number"],
show_copy_button=True,
show_fullscreen_button=True,
)
btn.click(get_pdf_result, inputs=[pdf_path, page_num], outputs=[results])
examples = gr.Examples(
examples=[
["./files/direct_and_image.pdf", 0],
["./files/direct_extract.pdf", 0],
],
examples_per_page=5,
inputs=[pdf_path, page_num],
fn=get_pdf_result,
outputs=[results],
cache_examples=False,
)
if __name__ == "__main__":
demo.launch()
