first commit

.env.example

@@ -0,0 +1,5 @@
+consumer_key = "t5bxx#######"
+consumer_secret = "8CoU8####"
+access_token = "735#########"
+access_secret = "vKP########"
+mongo_uri = "###############"

.gitignore

@@ -0,0 +1,46 @@
+__pycache__/
+*.py[cod]
+*$py.class
+
+*.so
+
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+pip-log.txt
+pip-delete-this-directory.txt
+
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+*.mo
+*.pot
+__pypackages__/
+.env

README.md

@@ -0,0 +1 @@
+# spamerio

cleaner/cleaner.py

@@ -0,0 +1,9 @@
+import re
+
+class Cleaner:
+    def clean_tweet(self, tweet):
+        r = tweet.lower()
+        r = re.sub("@[A-Za-z0-9_]+","", r)
+        r = re.sub("#[A-Za-z0-9_]+","", r)
+        r = re.sub(r'http\S+', '', r)
+        return r

main.py

@@ -0,0 +1,50 @@
+import os
+import sys
+from dotenv import load_dotenv
+from pymongo import MongoClient
+from twitter import twitter
+from cleaner import cleaner
+
+def main():
+    db = new_database()
+    client = twitter.Twitter(
+        os.environ.get("consumer_key"), 
+        os.environ.get("consumer_secret"), 
+        os.environ.get("access_token"), 
+        os.environ.get("access_secret")
+    )
+    clean = cleaner.Cleaner()
+
+    try:
+        query = str(input("Enter a search query: "))
+        tweets = client.get_tweets(query, 1, "4.570868,-74.297333,100km")
+    except:
+        print("Could not get tweets")
+        sys.exit(1)
+
+
+    for tweet in tweets:
+        tw = {
+            "text": clean.clean_tweet(tweet.text),
+            "user": tweet.user.screen_name,
+            "location": tweet.user.location,
+            "date": tweet.created_at
+        }
+
+        try:
+            tw_id = db.tweets.insert_one(tw).inserted_id
+            print("Tweet inserted with id: ", tw_id)
+        except:
+            print("Could not insert tweet")
+
+def new_database():
+    try:
+        client = MongoClient(os.environ.get("mongo_uri"))
+        db = client.test
+        return db
+    except:
+        print("Could not connect to MongoDB")
+
+if __name__ == "__main__":
+    load_dotenv()
+    main()

twitter/twitter.py

@@ -0,0 +1,24 @@
+import tweepy
+
+class Twitter: 
+    def __init__(self, consumer_key, consumer_secret, access_token, access_secret):
+        self.consumer_key = consumer_key
+        self.consumer_secret = consumer_secret
+        self.access_token = access_token
+        self.access_secret = access_secret
+        self.auth = tweepy.OAuthHandler(consumer_key, consumer_secret)
+        self.auth.set_access_token(access_token, access_secret)
+        self.api = tweepy.API(self.auth)
+
+    def get_tweets(self, query, count, geocode, lang="es", result_type="recent"):
+        filtered = query + "-filter:retweets"
+
+        tweets = tweepy.Cursor(
+            self.api.search_tweets, 
+            q=filtered, 
+            lang=lang,
+            result_type=result_type,
+            geocode=geocode
+        ).items(count)
+
+        return tweets