grzesiekp / thesaurus Goto Github PK

Cześć,

Aby pobrać dane dla TFI z Goldman Sachs można skorzystać z dwóch API, aby uniknąć przetwarzania plików HTML i mieć dostęp do danych archiwalnych równiez:

• Analizy.pl: Dostępne jest nie publiczne API, które można znaleźć przez devtools: https://www.analizy.pl/api/quotation/fio/ING87
• Export CSV z goldman sachs: Można wykorzystać eksport do pliku CSV, który jest dostępny na stronie:

Ja osobiście korzystam z ekspirtu . Na początku pobierałem dane e za pomocą CURL:

curl 'https://www.gstfi.pl/?action=quotes.getQuotesCsv&startDate=2023-03-28&endDate=2024-03-28&fundIds=1500088&unitCategoryIds=1' \
 -H 'Accept: */*' \
 -H 'User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36'

A teraz mam skrypt w Python, który możę cie zainspirować:

import pandas as pd

import requests

FUND_UNIT_URL = 'https://www.gstfi.pl/fundusze-inwestycyjne/fundusze-emerytura-ppk/goldman-sachs-emerytura-2055'
FUND_UNIT_ID_REGEXP = re.compile(r'data-fund-id="(\d+)"')
RATE_CSV_URL = 'https://www.gstfi.pl/'

def fetch_gstfi_rates():
    func_unit_url = FUND_UNIT_URL
    fund_unit_id = get_func_unit_id(func_unit_url)

    start_date = datetime.datetime(2023, 3, 28)
    end_date = datetime.datetime(2024, 3, 28)
    df = fetch_fund_rates(start_date, end_date, fund_unit_id)
    return df

def fetch_fund_rates(start_date, end_date, fund_unit_id):
    LOG.debug("Fetching fund rates: fund_unit_id=%s", fund_unit_id)

    rate_response = requests.get(RATE_CSV_URL, params={
        'action': 'quotes.getQuotesCsv',
        'startDate': start_date.strftime('%Y-%m-%d'),
        'endDate': end_date.strftime('%Y-%m-%d'),
        'fundIds': fund_unit_id,
        'unitCategoryIds': 1
    })
    rate_response.raise_for_status()
    df = pd.read_csv(StringIO(rate_response.text))
    return df 

def get_func_unit_id(func_unit_url):
    LOG.debug("Getting fund unit ID: func_unit_url=%s", func_unit_url)
    response = requests.get(func_unit_url)
    response.raise_for_status()
    text_content = response.text
    fund_unit_id_match = re.search(FUND_UNIT_ID_REGEXP, text_content)
    if not fund_unit_id_match:
        raise ValueError("Fund unit ID not found")
    fund_unit_id = fund_unit_id_match.group(1)
    return fund_unit_id

Poprosiłem ChatGPT o przepisanie na Google AppScript i dostałem taką odpowiedź:

Rewriting the provided Python code into Google Apps Script (GAS) involves a few steps and changes due to the different runtime environments and available libraries. GAS doesn't have a direct equivalent of Python's pandas or requests libraries, but you can achieve similar functionalities using Google Sheets for data manipulation and the UrlFetchApp class for HTTP requests.

Here's how the Python script can be translated into Google Apps Script:

// Global variables
const FUND_UNIT_URL = 'https://www.gstfi.pl/fundusze-inwestycyjne/fundusze-emerytura-ppk/goldman-sachs-emerytura-2055';
const RATE_CSV_URL = 'https://www.gstfi.pl/';

function fetchGstfiRates() {
  const fundUnitId = getFundUnitId(FUND_UNIT_URL);
  
  const startDate = new Date(2023, 2, 28); // Month is 0-indexed
  const endDate = new Date(2024, 2, 28); // Month is 0-indexed
  
  const data = fetchFundRates(startDate, endDate, fundUnitId);
  // Assuming you want to log or use the data here, for example, log to console.
  console.log(data);
}

function fetchFundRates(startDate, endDate, fundUnitId) {
  const formattedStartDate = Utilities.formatDate(startDate, Session.getScriptTimeZone(), 'yyyy-MM-dd');
  const formattedEndDate = Utilities.formatDate(endDate, Session.getScriptTimeZone(), 'yyyy-MM-dd');
  
  const params = {
    'action': 'quotes.getQuotesCsv',
    'startDate': formattedStartDate,
    'endDate': formattedEndDate,
    'fundIds': fundUnitId,
    'unitCategoryIds': 1
  };
  
  const queryString = Object.keys(params).map(key => `${key}=${encodeURIComponent(params[key])}`).join('&');
  const response = UrlFetchApp.fetch(`${RATE_CSV_URL}?${queryString}`);
  
  // Assuming CSV data is returned and you want to convert it to a 2D array
  const csvData = Utilities.parseCsv(response.getContentText());
  return csvData;
}

function getFundUnitId(funcUnitUrl) {
  const response = UrlFetchApp.fetch(funcUnitUrl);
  const textContent = response.getContentText();
  
  const fundUnitIdRegexp = /data-fund-id="(\d+)"/;
  const fundUnitIdMatch = textContent.match(fundUnitIdRegexp);
  
  if (!fundUnitIdMatch) {
    throw new Error("Fund unit ID not found");
  }
  
  const fundUnitId = fundUnitIdMatch[1];
  return fundUnitId;
}

Key Differences and Adjustments:

• Date handling: JavaScript and GAS use 0-indexed months, so the month numbers are adjusted accordingly.
• HTTP requests: GAS uses UrlFetchApp for making HTTP requests.
• Regular expressions: The syntax remains the same, but GAS uses JavaScript's match method.
• Data processing: Instead of pandas, this example directly manipulates the CSV data. If you need to work extensively with the data, consider using Google Sheets as a database or for further processing.
• Logging and output: The example uses console.log for logging, but you might want to replace this with appropriate GAS functionalities, such as appending to a Google Sheet.
• Error handling: GAS throws errors with throw new Error("message").

This script provides a foundation for fetching and processing data in Google Apps Script, mirroring the Python script's functionality within the GAS environment and its capabilities.

Mam nadzieje, że to komuś pomoże.