Youtube playlist stats scraper - get number of videos, total views and last time updated of a youtube playlist

ytplstats.py

#!/usr/bin/env python

import sys
import time
import json
import argparse
import locale
import datetime
import urllib.request

parser = argparse.ArgumentParser()
parser.add_argument('ytpl_url', type=str)
args = parser.parse_args()
ytpl_url = args.ytpl_url

# test first with a html file, to avoid hitting URL so frequently
# from pathlib import Path
# f = Path("/home/eevmanu/test.html")
# content = f.read_text()

with urllib.request.urlopen(ytpl_url) as u:
    content = str(u.read(), 'utf-8')

pattern = '\"stats\"'
pos_start = content.find(pattern)
if pos_start == -1:
    sys.exit()
if content[pos_start + len(pattern)] != ':':
    sys.exit()
open_operators = ['[', '{']
close_operators = [']', '}']
if content[pos_start + len(pattern) + 1] not in open_operators:
    sys.exit()
pos_start = pos_start + len(pattern) + 1

stack = []
stack.append(content[pos_start])
pos = pos_start
fail = False
wait_time = 0
while len(stack) > 0:
    pos += 1
    # print(f"{content[pos]=!r}")
    if content[pos] in open_operators:
        stack.append(content[pos])
        # print(f"{stack=!r}")
        time.sleep(wait_time)
        continue
    elif content[pos] in close_operators:
        last_operator = stack.pop()
        if last_operator == '[' and content[pos] == ']':
            time.sleep(wait_time)
            # print(f"{stack=!r}")
            continue
        elif last_operator == '{' and content[pos] == '}':
            time.sleep(wait_time)
            # print(f"{stack=!r}")
            continue
        else:
            fail = True
            # print(f"Error not compatibility between {last_operator=!r} and {content[pos]=!r}")
            time.sleep(wait_time)
            break
    else:
        # FIX ME
        # if content on keys or values strings contains '[', ']', '{', '}', this logic will breaks
        time.sleep(wait_time)
        continue
if fail:
    sys.exit()
pos_end = pos
content = content[pos_start : pos_end + 1]
print(content)
d = json.loads(content)

pl_videos_count = int(d[0]['runs'][0]['text'])
pl_views = int(d[1]['simpleText'].split(' ')[0].replace(',', ''))
pl_last_update = d[2]['runs'][1]['text']

# not recommended to change whole locale
# https://stackoverflow.com/a/32785195/3889948
# couldn't get it work with babel
# https://github.com/python-babel/babel/blob/cc36c84a83dd447bf48a6af3eb03c97bf299e8cb/babel/dates.py#L1162-L1163
# couldn't get it work with arrow get() method
# https://stackoverflow.com/q/26502470
# save initial locale to rollback later
# https://docs.python.org/3/library/locale.html?highlight=locale#:~:text=returned%20by%20localeconv().-,Example,-%3A

# $ locale -a
# es_ES.utf8

initial_locale = locale.getlocale(category=locale.LC_TIME)
locale.setlocale(locale.LC_TIME, "es_ES.utf8")

# match final dot with actual month name
# ene. | feb. | mar. | abr. | may. | jun. | jul. | ago. | set. | oct. | nov. | dic.
pl_last_update = datetime.datetime.strptime(pl_last_update, '%d %b. %Y')
pl_last_update = pl_last_update.date().isoformat()
locale.setlocale(locale.LC_TIME, initial_locale)

print(f"{pl_videos_count=!r}")
print(f"{pl_views=!r}")
print(f"{pl_last_update=!r}")

if __name__ == '__main__':
    pass

# example

# minify
# [{"runs":[{"text":"7"},{"text":" videos"}]},{"simpleText":"14,340 vistas"},{"runs":[{"text":"Se actualizó por última vez el "},{"text":"11 mar. 2020"}]}]

# beautify
# [
#     {
#         "runs": [
#             {
#                 "text":"7"
#             },
#             {
#                 "text":" videos"
#             }
#         ]
#     },
#     {
#         "simpleText":"14,340 vistas"
#     },
#     {
#         "runs":[
#             {
#                 "text":"Se actualizó por última vez el "
#             },
#             {
#                 "text":"11 mar. 2020"
#             }
#         ]
#     }
# ]