mirror-checker/main.py

#!/usr/bin/env python3

"""
This mirror status checker determines whether CSC mirror is up-to-date with upstream
"""

import time
import requests
from arch import Arch
from ceph import Ceph
from debian import Debian
from eclipse import Eclipse
from gnu import GNU
from kernel import Kernel
from openbsd import OpenBSD
from dateparser.search import search_dates # this library seems to be super slow but the other library: dateutil.parser gets some errors
# http://theautomatic.net/2018/12/18/2-packages-for-extracting-dates-from-a-string-of-text-in-python/
import re # import regular expressions to remove stray numbers in string that might interfere with date finding
import json # import json to read distro info stored in json file

import datefinder # another date finding library

def checker(directory_URL, file_name):
    page = requests.get(directory_URL).text
    indexOfFile = page.find(file_name)
    # print(page)

    # remove stray numbers (file size numbers in particular) that might interfere with date finding
    segment_clean = re.sub(r'\s\d+\s', ' ', page[indexOfFile:]) # removes numbers for size
    segment_clean = re.sub(r'\s\d+\w*\s', ' ', page[indexOfFile:]) # removes numbers + size unit. e.x. 50kb
    # print(segment_clean)

    # implementation using dateparser.search.search_dates
    # notes: some dates don't parse correctly with this tool
    # print(search_dates(page[indexOfFile:], languages=['en']))
    # print(search_dates(page[indexOfFile:])[0])

    # finds the dates in the segment after the file name
    # notes: a generator will be returned by the datefinder module. I'm typecasting it to a list. Please read the note of caution provided at the bottom.
    matches = list(datefinder.find_dates(segment_clean))
    # print(matches)

    if len(matches) > 0:
        date = matches[0] # date is of type datetime.datetime
        return(date.strftime("%m/%d/%Y, %H:%M:%S"))
    else:
        return('No dates found')

if __name__ == "__main__":
    """for distro in [Arch, Ceph, Debian, Eclipse, GNU, Kernel, OpenBSD]:
        try:
            distro.print_output(distro.check())
        except requests.exceptions.RequestException as err:
            print(f"Error: {distro.name()}\n{err}")"""

    """distros = json.load(open('distros.json',))
    print(distros)

    for distro in distros:
        print(distro[0] + ":")
        print("CSC mirror: " + checker(distro[1], distro[3]))
        print("Official distro: " + checker(distro[2], distro[3]))"""

    with open("data.json", "r", encoding = "utf-8") as file:
        data = json.load(file)
        current_time = int(time.time())
        for distro in data:
            csc_url = "http://mirror.csclub.uwaterloo.ca/" + data[distro]["csc"] \
                + data[distro]["file"]
            upstream_url = data[distro]["upstream"] + data[distro]["file"]
            try:
                if requests.get(csc_url).text == requests.get(upstream_url).text:
                    data[distro]["out_of_date_since"] = None
                elif data[distro]["out_of_date_since"] is None:
                    data[distro]["out_of_date_since"] = current_time
                elif current_time - data[distro]["out_of_date_since"] \
                    > data[distro]["out_of_date_interval"]:
                    print(f"Failure: {distro} out-of-sync")
                    continue
                print(f"Success: {distro} up-to-date")
            except requests.exceptions.RequestException as err:
                print(f"Error: {distro}\n{err}")
    with open("data.json", "w", encoding = "utf-8") as file:
        json.dump(data, file, indent = 4)