Spaces:
Runtime error
Runtime error
Upload 2 files
Browse files- app.py +58 -0
- requirements.txt +6 -0
app.py
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
import fitz # PyMuPDF
|
| 3 |
+
import arabic_reshaper
|
| 4 |
+
import easyocr
|
| 5 |
+
from PIL import Image
|
| 6 |
+
import io
|
| 7 |
+
|
| 8 |
+
reader = easyocr.Reader(['fa','ar','en'], gpu=False)
|
| 9 |
+
|
| 10 |
+
def extract_text_from_pdf(pdf_file):
|
| 11 |
+
if pdf_file is None:
|
| 12 |
+
return "لطفاً یک فایل PDF آپلود کنید.", None
|
| 13 |
+
try:
|
| 14 |
+
pdf_document = fitz.open(pdf_file)
|
| 15 |
+
all_text = []
|
| 16 |
+
for page_num in range(len(pdf_document)):
|
| 17 |
+
page = pdf_document[page_num]
|
| 18 |
+
text = page.get_text("text")
|
| 19 |
+
if not text.strip() or len(set(text)) < 10:
|
| 20 |
+
pix = page.get_pixmap(dpi=150)
|
| 21 |
+
img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
|
| 22 |
+
text_lines = reader.readtext(img, detail=0)
|
| 23 |
+
text = "\n".join(text_lines)
|
| 24 |
+
if any('\u0600' <= char <= '\u06FF' or '\u0750' <= char <= '\u077F' for char in text):
|
| 25 |
+
text = arabic_reshaper.reshape(text)
|
| 26 |
+
all_text.append(f"--- صفحه {page_num + 1} ---\n{text}\n")
|
| 27 |
+
pdf_document.close()
|
| 28 |
+
extracted_text = "\n".join(all_text)
|
| 29 |
+
output_file = "extracted_text.txt"
|
| 30 |
+
with open(output_file, "w", encoding="utf-8") as f:
|
| 31 |
+
f.write(extracted_text)
|
| 32 |
+
return extracted_text, output_file
|
| 33 |
+
except Exception as e:
|
| 34 |
+
return f"خطا در پردازش فایل: {str(e)}", None
|
| 35 |
+
|
| 36 |
+
def create_interface():
|
| 37 |
+
with gr.Blocks(theme=gr.themes.Soft()) as interface:
|
| 38 |
+
gr.Markdown("""# 📄 استخراج متن از PDF با OCR بهینه
|
| 39 |
+
این برنامه متن PDF را استخراج میکند.
|
| 40 |
+
- اگر PDF متن واقعی داشته باشد مستقیم میخواند.
|
| 41 |
+
- اگر اسکن شده باشد فقط صفحات خالی یا غیرقابلخواندن OCR میشوند.
|
| 42 |
+
### نحوه استفاده:
|
| 43 |
+
1. فایل PDF را آپلود کنید
|
| 44 |
+
2. روی "استخراج متن" کلیک کنید
|
| 45 |
+
3. متن استخراجشده را مشاهده و دانلود کنید""")
|
| 46 |
+
with gr.Row():
|
| 47 |
+
with gr.Column(scale=1):
|
| 48 |
+
pdf_input = gr.File(label="📂 آپلود PDF", file_types=[".pdf"], type="filepath")
|
| 49 |
+
extract_btn = gr.Button("🔍 استخراج متن", variant="primary")
|
| 50 |
+
with gr.Column(scale=2):
|
| 51 |
+
text_output = gr.Textbox(label="📝 متن استخراج شده", placeholder="متن PDF اینجا نمایش داده میشود...", lines=20, max_lines=30)
|
| 52 |
+
download_output = gr.File(label="⬇️ دانلود txt")
|
| 53 |
+
extract_btn.click(fn=extract_text_from_pdf, inputs=pdf_input, outputs=[text_output, download_output])
|
| 54 |
+
return interface
|
| 55 |
+
|
| 56 |
+
if __name__ == "__main__":
|
| 57 |
+
interface = create_interface()
|
| 58 |
+
interface.launch()
|
requirements.txt
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
gradio
|
| 2 |
+
PyMuPDF
|
| 3 |
+
arabic-reshaper
|
| 4 |
+
python-bidi
|
| 5 |
+
pillow
|
| 6 |
+
easyocr
|