JasonData commited on
Commit
675ed84
·
verified ·
1 Parent(s): 1af945b

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +1 -55
app.py CHANGED
@@ -1,56 +1,2 @@
1
-
2
-
3
- import os
4
-
5
  import gradio as gr
6
- import csv
7
- import fitz # PyMuPDF
8
-
9
-
10
- def pdf_to_csv(pdf_file):
11
- # Open the uploaded PDF file (pdf_file is a TemporaryFile)
12
- # pdf_reader = PyPDF2.PdfReader(pdf_file.name)
13
- text_lines = []
14
-
15
- file_name = os.path.basename(pdf_file.name)
16
-
17
- text_lines.append(f"File Name: {file_name}")
18
- text_lines.append(' 地区 (Region): 2010\n* 收入/支出金额 (Income/Expense Amount): +10,000.00\n* ')
19
- csv_filename = "extracted_text.csv"
20
- # Write each line into the CSV file (each line in its own row)
21
- with open(csv_filename, "w", newline="", encoding="utf-8-sig") as csvfile:
22
- writer = csv.writer(csvfile)
23
- for line in text_lines:
24
- writer.writerow([line])
25
-
26
- # Return the CSV file path so Gradio can offer it as a download
27
- return csv_filename
28
-
29
-
30
- def pdf_to_pngs(pdf_file):
31
- # Open the PDF
32
- doc = fitz.open(pdf_file)
33
- pix = None
34
- outputs = []
35
- # Loop through each page and save as PNG
36
- for page_num in range(doc.page_count):
37
- page = doc.load_page(page_num) # Get the page
38
- pix = page.get_pixmap() # Get the image of the page
39
-
40
- output_path = f'page_{page_num + 1}.png'
41
- pix.save(output_path) # Save as PNG
42
- print(f'Saved {output_path}')
43
- outputs.append(output_path)
44
- return outputs
45
-
46
- # Create a simple single-page Gradio interface
47
- demo = gr.Interface(
48
- fn=pdf_to_pngs,
49
- inputs=gr.File(label="Upload PDF", file_types=[".pdf"]),
50
- outputs=gr.File(label="Download CSV"),
51
- title="PDF to CSV Converter",
52
- description="Upload a PDF file, extract its text line-by-line, and download a CSV."
53
- )
54
-
55
- demo.launch()
56
-
 
 
 
 
 
1
  import gradio as gr
2
+ gr.Interface.load("models/Hidden_GEMINI_extractor/melaura-v1-0").launch()