import gradio as gr
from newsdataapi import NewsDataApiClient
import os
import json
import pandas as pd



def creating_data_dir(directory_path):
  # Use the os.makedirs() function to create the directory
  # The 'exist_ok=True' argument allows it to run without errors if the directory already exists
  os.makedirs(directory_path, exist_ok=True)

  # Check if the directory was created successfully
  if os.path.exists(directory_path):
      print(f"Directory '{directory_path}' created successfully.")
  else:
      print(f"Failed to create directory '{directory_path}'.")

def retrieve_news_per_keyword(api,keywords):
  # %time
  for keyword in keywords:
    # print(f"{api} \n {keyword}")
    response = api.news_api( q= keyword , country = "us", language = 'en', full_content = True)
    # writing to a file
    file_path = os.path.join(directory_path, f"response_{keyword}.json")
    with open(file_path, "w") as outfile:
        json.dump(response, outfile)

    print(f"News Response for keyword {keyword} is retrieved")
    keywords.remove(keyword)

def combine_responses_into_one(directory_path):

  # Use a list comprehension to get all file names in the directory
  file_list = [f for f in os.listdir(directory_path) if os.path.isfile(os.path.join(directory_path, f))]

  #retrieve the file_keyword by extracting the string after "_"
  # Extract the file_keyword from each filename
  file_keywords = [filename.split('_')[1].split('.')[0] for filename in file_list]


  # Initialize an empty list to store the combined JSON data
  combined_json = []

  # Loop through each file name
  for filename in file_list:
      # Read the current JSON file
      with open(directory_path+'/'+filename, 'r') as file:
          current_json = json.load(file)

      # Extract the file_keyword from the filename
      file_keyword = filename.split('_')[1].split('.')[0]

      # Add the file_keyword to each result in the current JSON
      for result in current_json['results']:
          result['file_keyword'] = file_keyword

      # Extend the combined JSON list with the results from the current JSON
      combined_json.extend(current_json['results'])
      print(f'{filename} is added to the combined json object')
      # break # using the break to check the loop code always

  # Save the combined_json object as a JSON file
  with open('combined_news_response.json', 'w') as combined_file:
      json.dump(combined_json, combined_file, indent=4)

def convert_json_to_csv(file_name):
  json_data_df = pd.read_json(file_name)
  # json_data_df.head()

  columns = [ 'title', 'keywords', 'creator', 'description', 'content', 'pubDate', 'country', 'category', 'language', 'file_keyword' ]
  csv_file_name = 'combined_news_response.csv'
  json_data_df[columns].to_csv(csv_file_name)
  print(f'{csv_file_name} is created')


# API key authorization, Initialize the client with your API key
NEWSDATA_API_KEY = "pub_2915202f68e543f70bb9aba9611735142c1fd"
keywords = [  "GDP",  "CPI",  "PPI",  "Unemployment Rate",  "Interest Rates",  "Inflation",  "Trade Balance",  "Retail Sales",  "Manufacturing Index",  "Earnings Reports",  "Revenue Growth",  "Profit Margins",  "Earnings Surprises",  "Geopolitical Events",  "Trade Tensions",  "Elections",  "Natural Disasters",  "Global Health Crises",  "Oil Prices",  "Gold Prices",  "Precious Metals",  "Agricultural Commodities",  "Federal Reserve",  "ECB",  "Forex Market",  "Exchange Rates",  "Currency Pairs",  "Tech Company Earnings",  "Tech Innovations",  "Retail Trends",  "Consumer Sentiment",  "Financial Regulations",  "Government Policies",  "Technical Analysis",  "Fundamental Analysis",  "Cryptocurrency News",  "Bitcoin",  "Altcoins",  "Cryptocurrency Regulations",  "S&P 500",  "Dow Jones",  "NASDAQ",  "Market Analysis",  "Stock Market Indices" ]

# creating a data directory
# Define the directory path you want to create
directory_path = './data'

def call_functions(directory_path='./data'):
  creating_data_dir(directory_path)
  items = os.listdir(directory_path)
  
  file_name = './combined_news_response.json'

  if len(items) == 0:
    print(f"Directory '{directory_path}' is empty.")
    api = NewsDataApiClient(apikey=NEWSDATA_API_KEY)
    retrieve_news_per_keyword(api,keywords)
    combine_responses_into_one(directory_path)
    convert_json_to_csv(file_name)
  elif len(items) >= 2:
    print(f"Directory '{directory_path}' contains at least two files.")
    combine_responses_into_one(directory_path)
    convert_json_to_csv(file_name)
  else:
      print(f"Directory '{directory_path}' contains only one file.")

  # Read the combined CSV file and display the first few rows
  csv_file_name = "combined_news_response.csv"
  if os.path.exists(csv_file_name):
      df = pd.read_csv(csv_file_name)
      # Assuming df is your DataFrame
      if 'Unnamed: 0' in df.columns:
          df.drop('Unnamed: 0', axis=1, inplace=True)
      first_few_rows = df.head(10)  # Adjust the number of rows as needed
      return first_few_rows
  else:
      return f"CSV file '{csv_file_name}' not found."


# Create a Gradio interface
iface = gr.Interface(
    fn=call_functions,
    inputs=gr.components.Textbox(label="Directory Path"),
    outputs=gr.components.Dataframe(type="pandas")
)


# Launch the Gradio app
iface.launch(debug=True)