import requests
import json
import pprint
import tabulate
import collections

sessions = []

for i in range(1, 95):
  url = f'https://agu.confex.com/agu/fm22/meetingapi.cgi/Search/0?date=2022-08-10T15%3A56%3A36&configtype=meetingapp_prelim&sort=Relevance&size=10&page={i}&ModelType=Session'
  response = requests.get(url)
  data = json.loads(response.text)
  lists = data['ChildList_Hits']
  for _list in lists:
    session_data = {
        'id': _list['id'],
        'title': _list['Title']
    }
    sessions.append(session_data)
  if i % 10 == 0:
    print(f'Retrieved {i*10} of 938 sessions.')

len(sessions)

abstracts = []
for i, session in enumerate(sessions):
  url = f"https://agu.confex.com/agu/fm22/meetingapi.cgi/Session/{session['id']}/ChildList_Paper"
  response = requests.get(url)
  abstract = json.loads(response.text)
  abstract_data = {
      'session_id': session['id'],
      'session_title': session['title'],
      'num_abstract': len(abstract)
  }
  abstracts.append(abstract_data)
  i += 1
  if i % 50 == 0:
    print(f'Retrieved {i} of {len(sessions)} sessions')


sorted_abstract = sorted(abstracts, key=lambda d: d['num_abstract'], reverse=True)

seen = set()
sorted_unique_abstract = []
rank = 1
last_count = 0
for i, d in enumerate(sorted_abstract):
  t = tuple(d.items())
  if t not in seen:
    num_abstract = d['num_abstract']
    if i == 0:
      last_count = num_abstract

    if num_abstract != last_count:
      last_count = num_abstract
      rank += 1

    seen.add(t)
    _d = collections.OrderedDict()
    _d['session_id'] = d['session_id']
    _d['num_abstract'] = d['num_abstract']
    _d['rank'] = rank
    _d['session_title'] = d['session_title']
    sorted_unique_abstract.append(_d)

header = sorted_unique_abstract[0].keys()
rows =  [x.values() for x in sorted_unique_abstract]
print(tabulate.tabulate(rows, header, tablefmt='grid'))