Jouons avec l'API des membres

#!/bin/env python

import sys
import json
import requests
import time
import csv
import lxml
import lxml.html
from datetime import datetime
from rich.progress import Progress


ID_MEMBRE = 502
NB_PAGES = 157
URL_LIST_MESSAGES_FORMAT = "https://zestedesavoir.com/forums/messages/{id_membre}/?page={page}"
URL_KARMAT_FORMAT = "https://zestedesavoir.com/api/forums/message/{id_message}/karma/"


"""
To get a token:

    $ ZDS_CRED=(echo -n '$CLIENT_ID:$CLIENT_SECRET' | base64)
    $ curl -X POST \
           -H "Authorization: Basic $ZDS_CRED" \
           -H "Cache-Control: no-cache" \
           -H "Content-Type: application/x-www-form-urlencoded" \
           "https://zestedesavoir.com/oauth2/token/" \
           -d '{"grant_type": "client_credentials"}'

But the API is down so we use an already authenticated session.
"""
# AUTH_TOKEN = 'SCnoKsAcXIFXboAtiqzGJQanCHWEwb' # Faux TOKEN
# AUTH_DATA = {'Authorization': f'Bearer {AUTH_TOKEN}'}
AUTH_DATA = {"csrftoken": "CpPrWfeRJheLyQTojPLtbOSFHxNngHJAkUwWDFaGjuhrAFLDAQRblLQOIRtfxMIe","sessionid": "oyOXhkXnTSCXpfwNoaSoJVqDSIpUzCwO"} # Faux TOKEN

def getKarma(s, post_id):
    link = URL_KARMAT_FORMAT.format(id_message=post_id)
    resp = s.get(link, headers={'accept': 'application/json'}, cookies=AUTH_DATA)
    data = []

    if resp.ok:
        karma_data = json.loads(resp.text)

        # Counts every member who likes
        for user in karma_data['like']['users']:
            data.append({"id": post_id, 'type': 'like', 'user_id': user['id'], 'username': user['username']})
        # Add a row for every anonymous likes
        for _ in range(karma_data['like']['count'] - len(karma_data['like']['users'])):
            data.append({"id": post_id, 'type': 'like', 'user_id': '', 'username': ''})

        # Counts every member who dislikes
        for user in karma_data['dislike']['users']:
            data.append({"id": post_id, 'type': 'dislike', 'user_id': user['id'], 'username': user['username']})
        # Add a row for every anonymous dislikes
        for _ in range(karma_data['dislike']['count'] - len(karma_data['dislike']['users'])):
            data.append({"id": post_id, 'type': 'dislike', 'user_id': '', 'username': ''})
    else:
        print(resp, file=sys.stderr)

    return resp.status_code, data


if __name__ == "__main__":
    s = requests.Session()

    list_post_id = []
    page = 0

    data = []

    while True:
        page += 1

        link = URL_LIST_MESSAGES_FORMAT.format(page=page, id_membre=ID_MEMBRE)
        resp = s.get(link, cookies=AUTH_DATA)
        print(f"Processing page n°{page}")

        if resp.ok:
            html = lxml.html.fromstring(resp.content)
            msg_linkgs = [a.attrib['href'] for a in html.cssselect(".forum-entry-title a")]
            post_id = [a[a.rfind('#p') + 2:] for a in msg_linkgs]
            list_post_id.extend(post_id)
        elif resp.status_code == 429:
            print("Too many requests waiting 10s\n")
            time.sleep(60)
            page -= 1
        else:
            break

    with Progress() as progress:
        karmaCount = progress.add_task("[red]Counting...", total=len(list_post_id))

        for i, post_id in enumerate(list_post_id):
            ret, data_post = getKarma(s, post_id)

            while ret == 429:
                # Retry later
                time.sleep(60)
                ret, data_post = getKarma(s, post_id)
            if 200 <= ret < 300:
                data.extend(data_post)

            progress.update(karmaCount, completed=i)

    with open(f'karma_{ID_MEMBRE}_{datetime.today().strftime("%Y-%m-%d")}.csv', 'w', newline='') as csvfile:
        writer = csv.DictWriter(csvfile, fieldnames=['id', 'type', 'user_id', 'username'])
        writer.writeheader()

        for row in data:
            writer.writerow(row)