app.py

from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
import cv2
import uuid
import os
import requests
import random
import string
import json
import shutil
import ast
import tempfile
import base64

app = FastAPI()

def check_nsfw(img_url):
    session_hash = ''.join(random.choice(string.ascii_lowercase + string.digits) for _ in range(7))
    data = {
        'data': [
            {'path': img_url},
            "chen-convnext",
            0.5,
            True,
            True
            ],
        'session_hash': session_hash,
        'fn_index': 0,
        'trigger_id': 12
        }
    r = requests.post('https://yoinked-da-nsfw-checker.hf.space/queue/join', json=data)

    r = requests.get(f'https://yoinked-da-nsfw-checker.hf.space/queue/data?session_hash={session_hash}', stream=True)
    buffer = ""  # Buffer to accumulate the chunks


    for content in r.iter_content(100):
        # Decode the byte content to a string
        buffer += content.decode('utf-8')

    return json.loads(buffer.split('data:')[len(buffer.split('data:'))-2])["output"]["data"][0]['label']
    

def check_nsfw2(img_url):
    session_hash = ''.join(random.choice(string.ascii_lowercase + string.digits) for _ in range(7))
    data = {
        'data': [
            {'path': img_url}
            ],
        'session_hash': session_hash,
        'fn_index': 0,
        'trigger_id': 9
        }
    r = requests.post('https://jamescookjr90-falconsai-nsfw-image-detection.hf.space/queue/join', json=data)

    r = requests.get(f'https://jamescookjr90-falconsai-nsfw-image-detection.hf.space/queue/data?session_hash={session_hash}', stream=True)
    buffer = ""  # Buffer to accumulate the chunks


    for content in r.iter_content(100):
        # Decode the byte content to a string
        buffer += content.decode('utf-8')
    return json.loads(buffer.split('data:')[len(buffer.split('data:'))-2])["output"]["data"][0]['label']

def check_nsfw3(img_url):
    data = {
        'data': [
            {'path': img_url}
            ]

        }
    r = requests.post('https://zanderlewis-xl-nsfw-detection.hf.space/call/predict',json=data)
    json_data = r.json()
    event_id = json_data['event_id']
    r = requests.get(f'https://zanderlewis-xl-nsfw-detection.hf.space/call/predict/{event_id}', stream=True)
    event_stream = ''
    for chunk in r.iter_content(100):
        event_stream += chunk.decode('utf-8')
    return ast.literal_eval(event_stream.split('data:')[-1])[0]['label']

def get_replica_code(url):
    try:
        r = requests.get(url)

        return r.text.split('replicas/')[1].split('"};')[0]
    except:
        return None
def check_nsfw4(img_url):
    code = get_replica_code('https://error466-falconsai-nsfw-image-detection.hf.space')
    session_hash = ''.join(random.choice(string.ascii_lowercase + string.digits) for _ in range(7))
    data = {
        'data': [
            {'path': img_url}
            ],
        'session_hash': session_hash,
        'fn_index': 0,
        'trigger_id': 58
        }
    r = requests.post(f'https://error466-falconsai-nsfw-image-detection.hf.space/--replicas/{code}/queue/join', json=data)

    r = requests.get(f'https://error466-falconsai-nsfw-image-detection.hf.space/--replicas/{code}/queue/data?session_hash={session_hash}', stream=True)
    buffer = ""  # Buffer to accumulate the chunks


    for content in r.iter_content(100):
        # Decode the byte content to a string
        buffer += content.decode('utf-8')
    return json.loads(buffer.split('data:')[-1])["output"]["data"][0]['label']

def check_nsfw5(img_url):
    session_hash = ''.join(random.choice(string.ascii_lowercase + string.digits) for _ in range(7))
    data = {
        'data': [
            {'path': img_url}
            ],
        'session_hash': session_hash,
        'fn_index': 0,
        'trigger_id': 9
        }
    r = requests.post('https://phelpsgg-falconsai-nsfw-image-detection.hf.space/queue/join', json=data)

    r = requests.get(f'https://phelpsgg-falconsai-nsfw-image-detection.hf.space/queue/data?session_hash={session_hash}', stream=True)
    buffer = ""  # Buffer to accumulate the chunks


    for content in r.iter_content(100):
        # Decode the byte content to a string
        buffer += content.decode('utf-8')
    return json.loads(buffer.split('data:')[-1])["output"]["data"][0]['label']
    
def check_nsfw_final(img_url):
    result = None
    print(img_url)
    try:
        check = check_nsfw2(img_url)
        if check == 'nsfw':
            result = True
        else:
            result = False
    except Exception as e:
        print(e)

    if result is None:
        try:
            check = check_nsfw3(img_url)
            if check == 'nsfw':
                result = True
            else:
                result = False
        except Exception as e:
            print(e)
    
    if result is None:
        try:
            check = check_nsfw4(img_url)
            if check == 'nsfw':
                result = True
            else:
                result = False
        except:
            pass

    if result is None:
        try:
            check = check_nsfw5(img_url)
            if check == 'nsfw':
                result = True
            else:
                result = False
        except Exception as e:
            print(e)

    return result

# Frame extraction
def extract_frames(video_path, num_frames, temp_dir):
    print(video_path)
    vidcap = cv2.VideoCapture(video_path)
    total_frames = int(vidcap.get(cv2.CAP_PROP_FRAME_COUNT))
    frames = []
    
    for i in range(num_frames):
        frame_number = int(i * total_frames / num_frames) 
        vidcap.set(cv2.CAP_PROP_POS_FRAMES, frame_number)
        success, image = vidcap.read()
        if success:
            frame_filename = os.path.join(temp_dir, f"{uuid.uuid4().hex}.jpg")
            cv2.imwrite(frame_filename, image)
            frames.append(frame_filename)
        else:
            break

    vidcap.release()
    return frames

# Video processing
def process_video(video_path, num_frames):
    temp_dir = os.path.join("/tmp", f"frames_{uuid.uuid4().hex}")
    os.makedirs(temp_dir, exist_ok=True)

    frames = extract_frames(video_path, num_frames, temp_dir)
    
    nsfw_count = 0
    total_frames = len(frames)
    frame_results = []

    for frame_path in frames:
        img_url = f"file://{frame_path}"
        print(img_url)
        nsfw_detected = check_nsfw_final(img_url)
        frame_results.append({
            "frame_path": frame_path,
            "nsfw_detected": nsfw_detected
        })
        if nsfw_detected:
            nsfw_count += 1
    
    # Cleanup
    shutil.rmtree(temp_dir, ignore_errors=True)

    result = {
        "nsfw_count": nsfw_count,
        "total_frames": total_frames,
        "frames": frame_results
    }
    
    return result

# Request/Response models
class VideoRequest(BaseModel):
    video_url: str
    num_frames: int = 10

class VideoResponse(BaseModel):
    nsfw_count: int
    total_frames: int
    frames: list

# API Endpoints

@app.post("/process_video", response_model=VideoResponse)
async def process_video_endpoint(request: VideoRequest):
    # Download video
    try:
        video_path = os.path.join("/tmp", f"{uuid.uuid4().hex}.mp4")
        with requests.get(request.video_url, stream=True) as r:
            r.raise_for_status()
            with open(video_path, "wb") as f:
                for chunk in r.iter_content(chunk_size=8192):
                    f.write(chunk)
    except Exception as e:
        raise HTTPException(status_code=400, detail=f"Error downloading video: {e}")
    
    try:
        # Process video
        result = process_video(video_path, request.num_frames)
    finally:
        # Cleanup video file
        if os.path.exists(video_path):
            os.remove(video_path)
    
    return result