app.py

from fastapi import FastAPI ,Request ,Form, UploadFile, File
from fastapi.responses import HTMLResponse, FileResponse,StreamingResponse,JSONResponse
import os
import io
from PIL import ImageOps,Image ,ImageFilter
#from transformers import pipeline
import matplotlib.pyplot as plt
import numpy as np
import ast
from server import *
import cv2
from typing import Optional
import base64
from fastapi import FastAPI, Request
from slowapi import Limiter
from slowapi.util import get_remote_address
from fastapi.responses import JSONResponse


#http://localhost:8000
app = FastAPI()

limiter = Limiter(key_func=get_remote_address)
app.state.limiter = limiter
# Register the Limiter with FastAPI
@app.middleware("http")
async def add_process_time_header(request: Request, call_next):
    response = await call_next(request)
    return response

@app.exception_handler(429)
async def rate_limit_exceeded(request: Request, exc):
    '''
    return JSONResponse(
        status_code=429,
        content={"detail": "Too Many Requests ,please just try 3 times per hour"},
    )'''
    return JSONResponse(
          status_code=429,
           content={
                "detail": "Too Many Requests ,please just try 3 times per hour",
                "status": 0
            })



# Root route
@app.get('/')
def main():
    return "Hello From Background remover !"

@app.post("/items")
@limiter.limit("3/hour")  # must have parameter request
async def read_items(request: Request, type_of_filters: str = Form(...)):
    return {"items": str(type_of_filters)}


@app.post('/imageStep1')
@limiter.limit("1/hour")  # must have parameter request
async def image_step1(request: Request,image_file: UploadFile = File(...),background_image: Optional [UploadFile] = File(None),type_of_filters: str = Form(...), blur_radius: str = Form(...)):
        #return {"type ":type_of_filters ,"radius":blur_radius,"image":image_file,"back":background_image}
        #type_of_filters : cam / ...remove / ...back
        input_to_type_of_filters = None

        if background_image and background_image.filename:
                contents__back = await background_image.read()
                image_back = Image.open(io.BytesIO(contents__back))
                input_to_type_of_filters = image_back
        else:
               input_to_type_of_filters = type_of_filters

        contents_img = await image_file.read()
        image = Image.open(io.BytesIO(contents_img))
        

        output_step1 =SegmenterBackground().Back_step1(image,input_to_type_of_filters,int(blur_radius))

        if (output_step1[-1] == 0):
            return {
                "detail": output_step1[0],
                "status": output_step1[-1]
            }
        
        produced_image = output_step1[0]
        
        '''
        # Save the processed image to a temporary file
        #output_file_path_tmp = "/tmp/tmp_processed_image.png"
        #produced_image.save(output_file_path_tmp)
        # return FileResponse(output_file_path_tmp, media_type='image/png', filename="/tmp/tmp_processed_image.png")
        '''
        
        # Convert the image to base64 to return it in the response
        buffered = io.BytesIO()
        produced_image.save(buffered, format="PNG")
        encoded_img = base64.b64encode(buffered.getvalue()).decode("utf-8")
            
        # Returning both text and the base64 image
        return {
                "message": output_step1[1],
                "image_base64": encoded_img,
                "status": output_step1[-1]
        }
        

@app.post('/imageStep2')
async def image_step2(image_file: UploadFile = File(...),background_image: Optional [UploadFile] = File(None),type_of_filters: str = Form(...),
                      things_replace: str = Form(...), blur_radius: str = Form(...)):
            #things_replace : from what detected.
        things_replace=ast.literal_eval(things_replace)
        blur_radius=int(blur_radius)


        input_to_type_of_filters=None

        if background_image and background_image.filename:
                contents__back = await background_image.read()
                image_back = Image.open(io.BytesIO(contents__back))
                input_to_type_of_filters = image_back
        else:
               input_to_type_of_filters = type_of_filters


        contents = await image_file.read()
        image = Image.open(io.BytesIO(contents))
        

        produced_image=SegmenterBackground().Back_step2(image,input_to_type_of_filters,things_replace,int(blur_radius))

        # Save the processed image to a temporary file
        output_file_path_tmp = "/tmp/tmp_processed_image.png"
        produced_image.save(output_file_path_tmp)

        # Return the processed image for download
        return FileResponse(output_file_path_tmp, media_type='image/png', filename="/tmp/tmp_processed_image.png")
    



@app.post('/Video')
async def Video(video_file: UploadFile = File(...),background_image: Optional [UploadFile] = File(None),kind_back: str = Form(...) 
                ,type_of_filters: str = Form(...),blur_radius: str = Form(...)):#--->,background_image: UploadFile = File(...)):
        
        #video_data = await video_file.read()
        #nparr = np.frombuffer(video_data, np.uint8)
        #video_path=cv2.imdecode(nparr, cv2.IMREAD_COLOR) #named this as just passed as it's path
        blur_radius=int(blur_radius)
        kind_back=ast.literal_eval(kind_back)

        input_to_type_of_filters=None

        if background_image and background_image.filename:
                contents__back = await background_image.read()
                image_back = Image.open(io.BytesIO(contents__back))
                input_to_type_of_filters = image_back
        else:
               input_to_type_of_filters = type_of_filters


        input_path_toWrite = f'/tmp/tmp_imput.avi'#{video_file.filename}
        output_path = '/tmp/tmp_output.avi'
        with open(input_path_toWrite, 'wb') as f:
            f.write(await video_file.read())

        #--------> mp4? (when tried ,worked on it although)   sound??

        SegmenterBackground().Back_video(input_path_toWrite, output_path,input_to_type_of_filters,kind_back,blur_radius)#video,background_image,what_remove,blur_radius=23)
        
        return StreamingResponse(open(output_path, "rb"), media_type="video/avi")


if __name__ == "__main__":
    import uvicorn
    uvicorn.run(app, host="0.0.0.0", port=8000)