diff --git a/main.py b/main.py new file mode 100644 index 0000000000000000000000000000000000000000..87e276bab85bafe393fe05d458c83bc9c915bba4 --- /dev/null +++ b/main.py @@ -0,0 +1,23 @@ +import snscrape.modules.twitter as sntwitter +import pandas as pd +import datetime as dt + +# cool +df = pd.DataFrame(columns=['date', 'tweet']) +limit = 100 + +for i, tweet in enumerate(sntwitter.TwitterSearchScraper(query=" Metaverse -filter:replies -filter:links -filter:retweets -filter:nativeretweets").get_items()): + if i > limit: + break + if (tweet.lang == "en"): + df2 = {'date': tweet.date.strftime( + "%m/%d/%Y, %H:%M:%S"), 'tweet': tweet.rawContent} + + df = pd.concat([df, pd.DataFrame.from_records([df2])]) + + print(i/limit*100) + + +df = df.drop_duplicates(subset=["tweet"]) +df = df.sort_values(by=["tweet"]) +df.to_json("twitter_data.json", orient="split")