import requests
import time
from google.cloud import firestore
import firebase_admin
from firebase_admin import credentials
from google.cloud.firestore_v1.base_query import FieldFilter

import traceback
# Initialize Firestore client
from collections import Counter
import re

def get_keywords(text, top_n=100):
    # Define a simple set of stopwords
    stopwords = {
        "a", "an", "the", "we", "were", "you", "your", "yours", "he", "she", 
        "it", "they", "them", "his", "her", "our", "ours", "their", "was", 
        "is", "are", "am", "be", "been", "to", "in", "of", "and", "or", "on", 
        "for", "with", "as", "by", "at", "from", "this", "that", "these", "those"
    }
    
    # Convert text to lowercase and extract words
    words = re.findall(r"\b\w+\b", text.lower())
    
    # Filter out stopwords
    filtered_words = [word for word in words if word not in stopwords]
    
    # Count word frequencies
    word_counts = Counter(filtered_words)
    
    # Return top N keywords as a list
    return [word for word, _ in word_counts.most_common(top_n)]



cred = credentials.Certificate("servicekey.json")
try:
    firebase_admin.initialize_app(cred)
except Exception as e:
    print(e)
db = firestore.Client()
# Endpoint URL
url = "https://us-central1-farmbase-b2f7e.cloudfunctions.net/create_advisory_blog_predata2"

def get_jeevnai_advisories():
    to_time = int(time.time())
    from_time = int((to_time - 60 * 60 * 24*10))
    # JSON payload
    payload = {
    "isTesting": True,
    "RequiredBlogsCount": 10,
    "FromTime": from_time,
    "ToTime": to_time,
    "AvoidFromDays": 30
    }
   # print(payload)
    # Send POST request with JSON data
    response = requests.post(url, json=payload)
    all_content = []
    all_keywords = []
    try:
        #print(response.text)
        response_json = response.json()
        #print("Response JSON:", response_json)
        crop_blogs = response_json.get("Crops", None)
        #print(crop_blogs)
        if crop_blogs is not None:
            for crop_name, crop_blog_keys in crop_blogs.items():
                #print(crop_name, crop_blog_keys)
                final_advisories_arr = []
            # print("query_started", crop_blog_keys)
                for single_key in crop_blog_keys:
                    single_key = str(single_key)
                    advisory_ref = db.collection("JEEVN_AI_advisories").where(filter=FieldFilter("timestamp", "==", single_key)).get()
                # print("query_complete")
                # Print results
                    for doc in advisory_ref:
                       # print(f"{doc.id} => {doc.to_dict()}")
                        advisory_obj = doc.to_dict()
                        advisory_obj["UID"] = "None"
                        advisory_obj["fieldID"] = "None"
                        advisory_obj["fieldName"] = "None"
                        final_advisories_arr.append(str(advisory_obj))
                    #keywords = get_keywords(str(final_advisories_arr), top_n=100)
                content_obj = {
                'body': final_advisories_arr
            }
                all_content.append(content_obj)
                #all_keywords.append(keywords)
            # print(final_advisories_obj)
                #time.sleep(100000)
        #print(all_content)
        return all_content, ''
    except:
        print(traceback.format_exc())