Timezone aware datetime (#1)

* Add timezone to all datetime objects

Issue was occurring where the timestamps in the database
were not compatible with python datetime objects because
they were timezone unaware. UTC was given as the default
timezone to any calls to datetime.now().

* Fix bug where collector skipped very first event

Subtracted the first datetime object by one second to
compensate.

* Update values in preparation for 0.1.1

README.md

@@ -1,6 +1,6 @@
 # ScryWarden
 
-ScryWarden is a framework for detecting anomalies in general datasets through the use of behavioral profiling. The library provides simple interfaces to design behavioral profiles, to pull messages from datasets, and to report on found anomalies. Its also engineered to horizontally scale as needed for larger datasets.
+ScryWarden is a framework for detecting anomalies in general datasets through the use of behavioral models. The library provides simple interfaces to design behavioral profiles, to pull messages from datasets, and to report on found anomalies. Its also engineered to horizontally scale as needed for larger datasets.
 
 It aims to be an accessible anomaly detection method that doesn't require any advanced knowledge on the topic while also allowing those with relevant knowledge to extend it for their needs.
 
@@ -20,7 +20,7 @@ This will install a CLI command in your path called `scrywarden`. Try it out by
 $ scrywarden --help
 Usage: scrywarden [OPTIONS] COMMAND [ARGS]...
 
-  Detects anomalies in datasets using behavioral profiling.
+  Detects anomalies in datasets using behavioral modeling.
 
 Options:
   -c, --config TEXT  Path to the config file to use.  [default:

scrywarden/cli.py

@@ -22,7 +22,7 @@
 )
 @click.pass_context
 def main(ctx: Context, **kwargs):
-    """Detects anomalies in datasets using behavioral profiling."""
+    """Detects anomalies in datasets using behavioral modeling."""
     ctx.ensure_object(dict)
     ctx.obj['config_file'] = kwargs['config']
 

scrywarden/investigator/base.py

@@ -2,7 +2,7 @@
 import threading
 import typing as t
 import uuid
-from datetime import datetime
+from datetime import datetime, timezone
 from queue import Queue, Full
 
 import pandas as pa
@@ -167,7 +167,7 @@ def _investigate(self) -> t.Optional[
         logger.debug("\n%s", malicious_anomalies)
         with self._session(expire_on_commit=False) as session:
             session.add(investigation)
-            investigation.completed_at = datetime.now()
+            investigation.completed_at = datetime.now(timezone.utc)
             session.flush()
         return investigation, malicious_anomalies
 

scrywarden/profile/collectors.py

@@ -1,7 +1,7 @@
 import logging
 import threading
 import typing as t
-from datetime import timedelta, datetime
+from datetime import timedelta, datetime, timezone
 
 import pandas as pa
 from sqlalchemy.orm import sessionmaker, Session
@@ -247,7 +247,7 @@ def _loop_until_anomalies(
     def _wait(self, target: datetime) -> bool:
         timeout: float = 0.0
         while not self.shutdown.wait(timeout):
-            now = datetime.now()
+            now = datetime.now(timezone.utc)
             logger.debug("Target start time %s current time %s", target, now)
             if target + timedelta(seconds=self.delay) <= now:
                 break
@@ -297,7 +297,9 @@ def _create_initial_investigation(
         first_event = self._get_first_event(profile)
         if not first_event:
             return pa.DataFrame()
-        return self._loop_until_anomalies(profile, first_event.created_at)
+        return self._loop_until_anomalies(
+            profile, first_event.created_at - timedelta(seconds=1),
+        )
 
 
 PARSER = parsers.Options({

scrywarden/transport/message.py

@@ -1,6 +1,6 @@
 import typing as t
 from collections import KeysView, ItemsView, ValuesView
-from datetime import datetime
+from datetime import datetime, timezone
 from uuid import uuid4, UUID
 
 from scrywarden.typing import JSONValue
@@ -133,7 +133,9 @@ def create(
         Message
             Created message object.
         """
-        return cls(id or uuid4(), timestamp or datetime.now(), data)
+        return cls(
+            id or uuid4(), timestamp or datetime.now(timezone.utc), data,
+        )
 
     def __getitem__(self, item: t.Sequence[str]) -> JSONValue:
         if isinstance(item, str):

setup.py

@@ -5,10 +5,10 @@
 
 setup(
     name='scrywarden',
-    version='0.1.0',
+    version='0.1.1',
     author='Chase Brewer',
     author_email='chasebrewsky@gmail.com',
-    description="Detect anomalies in datasets using behavioral profiling",
+    description="Detect anomalies in datasets using behavioral modeling",
     long_description=README,
     long_description_content_type='text/markdown',
     url='https://github.com/chasebrewsky/scrywarden',