initial commit

main.py

+import snscrape.modules.twitter as sntwitter
+import pandas as pd
+
+from src.functions import construct_query, convert_date
+
+table = pd.DataFrame(columns=['date', 'tweet'])
+
+terms = ["Metaverse"]
+negativeTerms = ["#metaverse"]
+
+filters = []
+negativeFilters = ["replies", "links", "retweets", "nativeretweets"]
+
+languages = ["en"]
+
+tweetNumber = 1000
+
+for i, tweet in enumerate(sntwitter.TwitterSearchScraper(query=construct_query(terms, negativeTerms, filters, negativeFilters)).get_items()):
+    if i > tweetNumber:
+        break
+    if (tweet.lang in languages):
+        dict = {'date': convert_date(tweet.date), 'tweet': tweet.rawContent}
+
+        if ("metaverse" in tweet.rawContent.lower()):
+            table = pd.concat([table, pd.DataFrame.from_records([dict])])
+
+    print(i/tweetNumber*100, " %")
+
+
+table = table.drop_duplicates(subset=["tweet"]).sort_values(by=["tweet"])
+
+table.to_json("output/twitter_data.json", orient="split")

src/functions.py

+import datetime as dt
+
+
+def construct_query(terms, negativeTerms, filters, negativeFilters):
+    query = ""
+
+    for t in terms:
+        query += t+" "
+
+    for nt in negativeTerms:
+        query += "-"+nt+" "
+
+    for f in filters:
+        query += "filter:"+f+" "
+
+    for nf in negativeFilters:
+        query += "-filter:"+nf+" "
+
+    return (query)
+
+
+def convert_date(date):
+    return (date.strftime(
+            "%m/%d/%Y, %H:%M:%S"))