Spaces:
Running
Running
Upload 3 files
Browse files- Dockerfile +18 -0
- main.py +25 -0
- requirements.txt +4 -0
Dockerfile
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Use official lightweight Python image
|
| 2 |
+
FROM python:3.9-slim
|
| 3 |
+
|
| 4 |
+
# Set working directory
|
| 5 |
+
WORKDIR /app
|
| 6 |
+
|
| 7 |
+
# Copy requirements and install
|
| 8 |
+
COPY requirements.txt .
|
| 9 |
+
RUN pip install --no-cache-dir -r requirements.txt
|
| 10 |
+
|
| 11 |
+
# Copy application code
|
| 12 |
+
COPY . .
|
| 13 |
+
|
| 14 |
+
# Expose port 7860 (Hugging Face default)
|
| 15 |
+
EXPOSE 7860
|
| 16 |
+
|
| 17 |
+
# Command to run the app (Adjust 'main:app' to your file name, e.g., 'app:app')
|
| 18 |
+
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
|
main.py
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from fastapi import FastAPI, UploadFile, File
|
| 2 |
+
from pypdf import PdfReader
|
| 3 |
+
from io import BytesIO
|
| 4 |
+
import uvicorn
|
| 5 |
+
|
| 6 |
+
app = FastAPI()
|
| 7 |
+
|
| 8 |
+
@app.post("/extract-text")
|
| 9 |
+
async def extract_text(file: UploadFile = File(...)):
|
| 10 |
+
try:
|
| 11 |
+
print(f"📄 Parsing: {file.filename}")
|
| 12 |
+
content = await file.read()
|
| 13 |
+
pdf_reader = PdfReader(BytesIO(content))
|
| 14 |
+
|
| 15 |
+
text = ""
|
| 16 |
+
for page in pdf_reader.pages:
|
| 17 |
+
text += page.extract_text() + "\n"
|
| 18 |
+
|
| 19 |
+
return {"status": "success", "text": text}
|
| 20 |
+
except Exception as e:
|
| 21 |
+
print(f"❌ Error: {str(e)}")
|
| 22 |
+
return {"status": "error", "error": str(e)}
|
| 23 |
+
|
| 24 |
+
if __name__ == "__main__":
|
| 25 |
+
uvicorn.run(app, host="127.0.0.1", port=8000)
|
requirements.txt
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
fastapi
|
| 2 |
+
uvicorn
|
| 3 |
+
pypdf
|
| 4 |
+
python-multipart
|