SHOW:
|
|
- or go back to the newest paste.
| 1 | import pandas as pd | |
| 2 | import pprint | |
| 3 | import requests | |
| 4 | ||
| 5 | def collapse_results_by_party(results_by_candidate, candidates): | |
| 6 | results_by_party = {}
| |
| 7 | for candidate, count in results_by_candidate.iteritems(): | |
| 8 | party = candidates[candidate]['party'] | |
| 9 | results_by_party[party] = results_by_party.get(party, 0) + count | |
| 10 | ||
| 11 | return results_by_party | |
| 12 | ||
| 13 | states = [ | |
| 14 | 'Alaska', 'Alabama', 'Arkansas', 'Arizona', 'California', 'Colorado', | |
| 15 | 'Connecticut', 'Delaware', 'Florida', 'Georgia', | |
| 16 | 'Hawaii', 'Iowa', 'Idaho', 'Illinois', 'Indiana', 'Kansas', 'Kentucky', | |
| 17 | 'Louisiana', 'Massachusetts', 'Maryland', 'Maine', 'Michigan', | |
| 18 | 'Minnesota', 'Missouri', 'Mississippi', 'Montana', 'North Carolina', | |
| 19 | 'North Dakota', 'Nebraska', 'New Hampshire', 'New Jersey', 'New Mexico', | |
| 20 | 'Nevada', 'New York', 'Ohio', 'Oklahoma', 'Oregon', 'Pennsylvania', | |
| 21 | 'Rhode Island', 'South Carolina', 'South Dakota', 'Tennessee', 'Texas', | |
| 22 | 'Utah', 'Virginia', 'Vermont', 'Washington', 'Wisconsin', | |
| 23 | 'West Virginia', 'Wyoming', | |
| 24 | ] | |
| 25 | ||
| 26 | all_results = {}
| |
| 27 | for state in states: | |
| 28 | print 'Downloading {}'.format(state)
| |
| 29 | formatted_state = state.lower().replace(' ', '-')
| |
| 30 | state_results = requests.get('https://static01.nyt.com/elections-assets/2020/data/api/2020-11-03/race-page/{}/president.json'.format(formatted_state)).json()
| |
| 31 | all_results[formatted_state] = state_results | |
| 32 | ||
| 33 | records = [] | |
| 34 | for state, state_results in all_results.iteritems(): | |
| 35 | race = state_results['data']['races'][0] | |
| 36 | ||
| 37 | for candidate in race['candidates']: | |
| 38 | if candidate['party_id'] == 'republican': | |
| 39 | candidate['party'] = 'rep' | |
| 40 | elif candidate['party_id'] == 'democrat': | |
| 41 | candidate['party'] = 'dem' | |
| 42 | else: | |
| 43 | candidate['party'] = 'trd' | |
| 44 | candidates = { candidate['candidate_key']: candidate for candidate in race['candidates'] }
| |
| 45 | ||
| 46 | for data_point in race['timeseries']: | |
| 47 | data_point['state'] = state | |
| 48 | data_point['expected_votes'] = race['tot_exp_vote'] | |
| 49 | data_point['trump2016'] = race['trump2016'] | |
| 50 | data_point['votes2012'] = race['votes2012'] | |
| 51 | data_point['votes2016'] = race['votes2016'] | |
| 52 | ||
| 53 | vote_shares = collapse_results_by_party(data_point['vote_shares'], candidates) | |
| 54 | for party in ['rep', 'dem', 'trd']: | |
| 55 | data_point['vote_share_{}'.format(party)] = vote_shares.get(party, 0)
| |
| 56 | ||
| 57 | data_point.pop('vote_shares')
| |
| 58 | records.append(data_point) | |
| 59 | ||
| 60 | time_series_df = pd.DataFrame.from_records(records) | |
| 61 | time_series_df.to_csv('data/nyt_ts.csv', encoding='utf-8') |
