new_feature.py

import os
from flask import Flask, request, jsonify
from pathlib import Path
from logging_utils import setup_logger
from textblob import TextBlob
import datetime

def analyze_sentiment(text: str) -> dict:
    """Analyze sentiment of the given text using TextBlob."""
    blob = TextBlob(text)
    polarity = blob.sentiment.polarity
    subjectivity = blob.sentiment.subjectivity
    sentiment = "positive" if polarity > 0.1 else "negative" if polarity < -0.1 else "neutral"
    return {
        "sentiment": sentiment,
        "polarity": polarity,
        "subjectivity": subjectivity
    }

def log_sentiment_analysis(text: str, result: dict, logger):
    timestamp = datetime.datetime.utcnow().isoformat()
    logger.info(f"[{timestamp}] Sentiment analysis for text: '{text}' | Result: {result}")

def new_feature():
    """
    Flask API endpoint for sentiment analysis.
    POST /api/sentiment-analysis
    Body: { "text": "..." }
    Response: { "sentiment": "...", "polarity": ..., "subjectivity": ... }
    """
    app = Flask(__name__)
    LOG_PATH = Path(os.getenv("TARGET_REPO_PATH", os.getcwd())) / "sentiment_analysis.log"
    logger = setup_logger("sentiment_api", str(LOG_PATH), level=os.getenv("API_LOG_LEVEL", "INFO"))

    @app.route("/api/sentiment-analysis", methods=["POST"])
    def sentiment_analysis():
        data = request.get_json()
        if not data or "text" not in data:
            logger.warning("No text provided for sentiment analysis.")
            return jsonify({"error": "Missing 'text' in request body"}), 400
        text = data["text"]
        result = analyze_sentiment(text)
        log_sentiment_analysis(text, result, logger)
        return jsonify(result), 200

    app.run(host="0.0.0.0", port=5050)

if __name__ == "__main__":
    new_feature()