WIP: Twint

modules/Neo4jDataAccess.py

@@ -322,7 +322,9 @@ def __save_df_to_graph(self, df, job_name, job_id=None):
         for index, row in df.iterrows():
             # determine the type of tweet
             tweet_type = 'TWEET'
-            if row["in_reply_to_status_id"] is not None and row["in_reply_to_status_id"] > 0:
+            if row['tweet_type_twint']:
+                tweet_type = row['tweet_type_twint']
+            elif row["in_reply_to_status_id"] is not None and row["in_reply_to_status_id"] > 0:
                 tweet_type = "REPLY"
             elif "quoted_status_id" in row and row["quoted_status_id"] is not None and row["quoted_status_id"] > 0:
                 tweet_type = "QUOTE_RETWEET"
@@ -347,8 +349,11 @@ def __save_df_to_graph(self, df, job_name, job_id=None):
                                'user_created_at': pd.Timestamp(row['user_created_at'], unit='s').to_pydatetime(),
                                'user_profile_image_url': row['user_profile_image_url'],
                                'reply_tweet_id': row['in_reply_to_status_id'],
+                               'conversation_id': row['conversation_id'] if 'conversation_id' in row else None,
                                'quoted_status_id': row['quoted_status_id'],
                                'retweet_id': row['retweet_id'] if 'retweet_id' in row else None,
+                               'geo': row['geo'] if 'geo' in row else None,
+                               'ingest_method': row['ingest_method']
                                })
             except Exception as e:
                 logging.error('params.append exn', e)
@@ -437,4 +442,4 @@ def __parse_urls(self, row, url_params, job_name, job_id=None):
                 logging.error(inst.args)     # arguments stored in .args
                 # __str__ allows args to be printed directly,
                 logging.error(inst)
-        return url_params
+        return url_params

modules/Twint.py

@@ -0,0 +1,117 @@
+import pyarrow as pa
+import twint
+from urlextract import URLExtractor
+from datetime import datetime, timedelta
+
+
+class TwintPool:
+    def __init__(self, fh_job=None, job_name="noname"):
+        self.fh = fh_job
+        self.config = twint.Config()
+        self.config.Limit = 100
+        self.config.Pandas = True
+        self.config.User_full = True
+        self.config.Hide_output = True
+
+    def twint_loop(self, since, until, stride_sec=600, limit=None):
+        def get_unix_time(time_str):
+            return datetime.strptime(time_str, "%Y-%m-%d %H:%M:%S")
+
+        since = get_unix_time(since)
+        until = get_unix_time(until)
+        t = since
+        tweets_returned = 0
+
+        while t < until and (not tweets_returned or tweets_returned < limit):
+            t0 = t
+            t1 = t + timedelta(seconds=stride_sec)
+            self.config.Since = str(t0)
+            self.config.Until = str(t1)
+            twint.run.Search(self.config)
+            tweets_returned += len(twint.storage.panda.Tweets_df)
+            yield (twint.storage.panda.Tweets_df, t0, t1)
+            t = t1
+
+    def _get_term(
+        self,
+        Search="IngSoc",
+        Since="1984-04-20 13:00:00",
+        Until="1984-04-20 13:30:00",
+        stride_sec=600,
+        **kwargs
+    ):
+        self.config.Search = Search
+        self.config.Retweets = True
+        for k, v in kwargs.items():
+            setattr(self.config, k, v)
+        # self.config.Search = term
+        for df, t0, t1 in self.twint_loop(Since, Until, stride_sec, self.config.Limit):
+            yield (df, t0, t1)
+
+
+    def _get_timeline(self, username="lmeyerov"):
+        self.config.Username = username
+        self.config.Retweets = True
+        #self.config.Search = term
+        twint.run.Search(self.config)
+        tweets_df = twint.storage.panda.Tweets_df
+        return tweets_df
+
+    def twint_df_to_neo4j_df(self, df):
+        neo4j_df = df.rename(
+            columns={
+                "id": "status_id",
+                "tweet": "full_text",
+                "created_at": "created_at",  # needs to be datetime
+                "nlikes": "favorite_count",
+                "nretweets": "retweet_count",
+                "user_id_str": "user_id",
+                "username": "user_name",
+                "name": "user_screen_name",
+            }
+        )
+
+        def row_to_tweet_type(row):
+            if row["quote_url"] is None or row["quote_url"] == "":
+                return "QUOTE_RETWEET"
+            elif row["retweet"]:
+                return "RETWEET"
+            elif row["id"] == row["conversation_id"]:
+                return "TWEET"
+            elif row["id"] != row["conversation_id"]:
+                return "REPLY"
+            else:
+                raise ("wat")
+
+        def row_to_quoted_status_id(row):
+            if row["quote_url"] and len(row["quote_url"]) > 0:
+                return row["quote_url"].split("/")[-1]
+            else:
+                return None
+
+        def row_tweet_to_urls(row):
+            extractor = URLExtract()
+            return list(extractor.gen_urls(row["tweet"]))
+
+        neo4j_df["user_location"] = None
+        neo4j_df["tweet_type_twint"] = df.apply(row_to_tweet_type, axis=1)
+        neo4j_df["hashtags"] = df["hashtags"].apply(
+            lambda x: [{"text": ht} for ht in x]
+        )
+        neo4j_df["user_followers_count"] = None
+        neo4j_df["user_friends_count"] = None
+        neo4j_df["user_created_at"] = None
+        neo4j_df["user_profile_image_url"] = None
+        neo4j_df["in_reply_to_status_id"] = None
+        neo4j_df["user_mentions"] = []  # Todo
+        # neo4j_df['retweet_id'] is suspiciously empty (always)
+
+        neo4j_df["quoted_status_id"] = df.apply(row_to_quoted_status_id, axis=1)
+        neo4j_df["urls"] = df.apply(row_tweet_to_urls, axis=1)
+
+        neo4j_df["ingest_method"] = 'twint'
+
+        return neo4j_df
+
+    def to_arrow(self, tweets_df):
+        pass