2023-06-07 20:36:35 +02:00

205 lines
6.7 KiB
Python

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
'''
Created on Tue Jun 6 11:40:07 2023
@author: michael
'''
import os
import tweepy
import pandas as pd
import numpy as np
import glob
import time
## Setup directories
# WD Michael
wd = '/home/michael/Documents/PS/Data/collectTweets/'
# WD Server
# wd = '/home/yunohost.multimedia/polsoc/Politics & Society/TweetCollection'
# WD Josie
# wd = '/home/michael/Documents/PS/Data/'
# WD Sam
# wd = '/home/michael/Documents/PS/Data/'
# Tweet-datafile directory
td = 'data/tweets/'
os.chdir(wd)
## Setup Api-connection
bearer_token = 'AAAAAAAAAAAAAAAAAAAAAMVDlQEAAAAAal9f5uZrM12CVPA4f4jr4mGH5Oc%3DuTg1Vd0YKYMwraA7ibX6LiGyd337OXkm3JwudEX7vatruswmoc'
client = tweepy.Client(bearer_token, return_type = dict, wait_on_rate_limit = True)
# Define time period of interest
# Define time periods of interest
time_slices = [
{
'start_time': '2020-01-01T00:00:00Z',
'end_time': '2020-06-01T00:00:00Z',
'suffix': '-slice1'
},
{
'start_time': '2020-06-01T00:00:01Z',
'end_time': '2021-01-01T00:00:00Z',
'suffix': '-slice2'
},
{
'start_time': '2021-01-01T00:00:01Z',
'end_time': '2021-06-01T00:00:00Z',
'suffix': '-slice3'
},
{
'start_time': '2021-06-01T00:00:01Z',
'end_time': '2023-01-03T00:00:00Z',
'suffix': '-slice4'
}
]
# gather keywords @chenTrackingSocialMedia2020
# line80 ff: lamsalCoronavirusCOVID19Tweets2020
# Initialize the keywords list
keywords = []
# Read the keywords from a file
with open('data/keywords.txt', 'r') as file:
lines = file.readlines()
for line in lines:
keyword = line.strip() # Remove the newline character
keywords.append(keyword)
tweet_fields = [
'id',
'text',
'attachments',
'author_id',
'context_annotations',
'conversation_id',
'created_at',
'entities',
'geo',
'lang',
'possibly_sensitive',
'public_metrics',
'referenced_tweets',
'reply_settings',
'source',
'withheld',
]
# Get accounts & alt-accounts from Senators-Datafile
accounts = pd.read_csv('data/senators-raw.csv')['twitter_handle'].tolist()
alt_accounts = pd.read_csv('data/senators-raw.csv')['alt_handle'].tolist()
print(accounts)
print(alt_accounts)
# Iterate over each Twitter account
for handle in accounts:
for slice_data in time_slices:
# define slice data variables from time_slices
start_time = slice_data['start_time']
end_time = slice_data['end_time']
suffix = slice_data['suffix']
# define tweepy query with twitter handle of current sen
query = f'from:{handle} -is:retweet'
# create empty tweetlist that will be filled with tweets of current sen
tweetlist = []
# statusmsg
msg = f'trying to fetch tweets for {handle}{suffix}'
print(msg)
# Fetch tweets using tweepy Twitter API v2 pagination
tweets = tweepy.Paginator(client.search_all_tweets,
query=query,
tweet_fields=tweet_fields,
start_time=start_time,
end_time=end_time,
max_results=20).flatten(20)
# for each tweet returned...
for tweet in tweets:
# ... add that tweet to tweetlist
tweetlist.append(tweet)
# Check if no tweets fetched for the current time slice. If there are no tweets, skip to next time_slices loop iteration
if len(tweetlist) == 0:
msg = f'return empty in {handle}{suffix} - from {start_time} to {end_time}'
print(msg)
continue
# convert to dataframe
tweet_df = pd.DataFrame(tweetlist)
# add handle column as api only provides user-ids
tweet_df['handle'] = handle
## Extract referenced_tweet info from column
tweet_df['referenced_tweet_type'] = None
tweet_df['referenced_tweet_id'] = None
# if cond. because in some cases column doesn't exist
if 'referenced_tweets' in tweet_df.columns:
for index, row in tweet_df.iterrows():
referenced_tweets = row['referenced_tweets']
if isinstance(referenced_tweets, list) and len(referenced_tweets) > 0:
referenced_tweet = referenced_tweets[0]
referenced_tweet_type = referenced_tweet['type']
referenced_tweet_id = referenced_tweet['id']
tweet_df.at[index, 'referenced_tweet_type'] = referenced_tweet_type
tweet_df.at[index, 'referenced_tweet_id'] = referenced_tweet_id
## Check if tweet-text contains keyword
# if cond. because in some cases column doesn't exist
if 'text' in tweet_df.columns:
tweet_df['contains_keyword'] = (tweet_df['text'].str.findall('|'.join(keywords))
.str.join(',')
.replace('', 'none'))
## Save two versions of the dataset, one with all fields and one without dict fields
# define filepaths
csv_path = f'data/tweets/{handle}{suffix}.csv'
csv_path2 = f'data/tweets/{handle}{suffix}-LONG.csv'
# save LONG csv
tweet_df.to_csv(csv_path2)
# Remove 'context_annotations', 'entities' and 'referenced_tweets' columns for short csv files
# if cond. because in some cases column doesn't exist
if all(k in tweet_df for k in ('context_annotations', 'entities', 'referenced_tweets')):
tweet_df = tweet_df.drop(['context_annotations', 'entities', 'referenced_tweets'], axis=1)
# save short csv
tweet_df.to_csv(csv_path)
# sleep 1 second to not get over 1sec api limit
time.sleep(1)
# Merge CSV-Files
# (it would also have been a possibility to build a dataframe with all senators' tweets but i found the other way around more useful)
path_to_tweetdfs = wd + td
os.chdir(path_to_tweetdfs)
tweetfiles = glob.glob('*.{}'.format('csv'))
print(tweetfiles)
# save merged csv as two files
df_all_senators = pd.DataFrame()
df_all_senators_long = pd.DataFrame()
for file in tweetfiles:
if 'LONG' in file:
df = pd.read_csv(file)
df_all_senators_long = pd.concat([df, df_all_senators_long])
else:
df = pd.read_csv(file)
df_all_senators = pd.concat([df, df_all_senators])
csv_path = td + 'ALL-SENATORS.csv'
csv_path2 = td + 'ALL-SENATORS-LONG-LONG.csv'
df_all_senators.to_csv(csv_path)
df_all_senators_long.to_csv(csv_path2)
os.chdir(wd)