import re
import os
import joblib
from fastapi import FastAPI, File, UploadFile
from pydantic import BaseModel
from fastapi.middleware.cors import CORSMiddleware
from sentence_transformers import SentenceTransformer, util
import cv2
import numpy as np

app = FastAPI()

app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],  # Or restrict to your domain
    allow_methods=["*"],
    allow_headers=["*"],
)

os.environ["HF_HOME"] = "/tmp"
os.environ["TRANSFORMERS_CACHE"] = "/tmp"
os.environ["SENTENCE_TRANSFORMERS_HOME"] = "/tmp"

# Load model and vectorizer
model = joblib.load("team_classifier_model.joblib") 
vectorizer = joblib.load("tfidf_vectorizer.joblib")
sbert_model = SentenceTransformer("sentence-transformers/paraphrase-MiniLM-L6-v2")
gender_list = ['Male', 'Female']
model = cv2.dnn.readNetFromCaffe("gender_deploy.prototxt", "gender_net.caffemodel")


def clean_text(text):
    text = re.sub(r"\s+", " ", str(text))
    text = re.sub(r"[^\w\s]", "", text)
    return text.lower().strip()


class InputText(BaseModel):
    subject: str
    message: str


class SimilarityRequest(BaseModel):
    text1: str
    text2: str


@app.get("/")
def root():
    return {"status": "running", "message": "Use POST /classify"}


@app.post("/classify")
async def classify_ticket(data: InputText):
    combined = clean_text(f"{data.subject} {data.message}")
    vec = vectorizer.transform([combined])
    prediction = model.predict(vec)[0]
    return {"team": prediction}


@app.post("/similarity")
async def compute_similarity(data: SimilarityRequest):
    emb1 = sbert_model.encode(data.text1, convert_to_tensor=True)
    emb2 = sbert_model.encode(data.text2, convert_to_tensor=True)
    score = util.pytorch_cos_sim(emb1, emb2).item()
    return {"similarity": score}


@app.post("/gender")
async def predict_gender(file: UploadFile = File(...)):
    try:
        contents = await file.read()
        npimg = np.frombuffer(contents, np.uint8)
        img = cv2.imdecode(npimg, cv2.IMREAD_COLOR)

        blob = cv2.dnn.blobFromImage(img, 1.0, (227, 227), (78.426337, 87.768914, 114.895847), swapRB=False)

        model.setInput(blob)
        gender_preds = model.forward()
        gender = gender_list[gender_preds[0].argmax()]
        return {"gender": gender}
    except Exception as e:
        return JSONResponse(content={"error": str(e)}, status_code=500)