diff --git a/ConversationAnalyzer.py b/ConversationAnalyzer.py
index fe30520..cfe1a95 100644
--- a/ConversationAnalyzer.py
+++ b/ConversationAnalyzer.py
@@ -19,22 +19,22 @@ def __str__(self):
     def stats(self):
         return self.get_stats(self.df)
 
-    # TODO has to be tested
+    def get_stats(self, df, subject='all', start=None, end=None, period=None):
+        df = self.filter_by_input(df, subject=subject, start=start, end=end, period=period)
+        stats = ConversationStats(df)
+        return stats
+
     def get_time_series_data(self, subject='all', **kwargs):
         time_series = generate_time_series(**kwargs)
         return get_stats_for_intervals(self.get_stats, self.df, time_series, subject=subject)
 
-    def get_plotable_time_series_data(self, interval_stats, statistic):
+    @staticmethod
+    def get_plottable_time_series_data(interval_stats, statistic):
         for k, v in interval_stats.items():
             if isinstance(v, ConversationStats):
                 interval_stats[k] = getattr(v, statistic)
         return interval_stats
 
-    def get_stats(self, df, subject='all', start=None, end=None, period=None):
-        df = self.filter_by_input(df, subject=subject, start=start, end=end, period=period)
-        stats = ConversationStats(df)
-        return stats
-
     @staticmethod
     @subject_checker
     @date_checker
@@ -91,13 +91,13 @@ def unique_msg_count(self):
     # 3.
     @property
     def most_used_msgs(self):
-        # TODO first few (1-10) messages
+        # TODO LATER first few (1-10) messages
         return self.messages.value_counts()
 
     # 4.
     @property
     def msg_frequency(self):
-        # TODO this has been most likely depracated
+        # NOTE this has been most likely depracated OR?
         pass
 
     # 5.
@@ -132,12 +132,12 @@ def char_count(self):
     # 10.
     @property
     def most_used_chars(self):
-        return None  # TODO or not  https://stackoverflow.com/questions/4131123/finding-the-most-frequent-character-in-a-string
+        return None  # TODO LATER or not  https://stackoverflow.com/questions/4131123/finding-the-most-frequent-character-in-a-string
 
     # 11.
     @property
     def rate_of_media_messages(self):
-        pass  # TODO what?
+        pass  # NOTE what?
 
     def get_words(self):
         token_list = self.messages.str.lower().str.split()
diff --git a/Conversations.py b/Conversations.py
index 112f336..3fb1fbd 100644
--- a/Conversations.py
+++ b/Conversations.py
@@ -43,7 +43,7 @@ def extract_names_from_convos(jsons):
                 else:
                     name_data_map[key] = {
                         'title': msg.title,
-                        'compact_name': msg.compact_names,  # TODO is list ok for if length is  only  1??
+                        'compact_name': msg.compact_names,
                         # 'participants': msg.participants + ['Levente Csőke'],
                         'participants': msg.participants,
                         'messages': msg.df,
@@ -67,7 +67,7 @@ def to_df(self):
         self._df = pd.DataFrame(self.decoded.get('messages'))
 
     def set_date_as_index(self):
-        # TODO maybe not needed; could calculate real time
+        # NOTE maybe not needed; could calculate real time
         date_series = self._df.timestamp_ms.apply(self.ts_to_date)
         self._df = self._df.set_index(date_series).iloc[::-1]
 
diff --git a/Group.py b/Group.py
index 41b966f..2152d77 100644
--- a/Group.py
+++ b/Group.py
@@ -1,4 +1,4 @@
-# TODO groups should be searched by looking into jsons unfortunately :(
+# TODO LATER groups should be searched by looking into jsons unfortunately :(
 # because of directory says others
 # maybe we dont use groups right away?
 
diff --git a/MessagingAnalyzer.py b/MessagingAnalyzer.py
index 07965e5..0619505 100644
--- a/MessagingAnalyzer.py
+++ b/MessagingAnalyzer.py
@@ -1,26 +1,22 @@
-from utils import year_converter, month_converter, generate_time_series, get_stats_for_intervals
+from utils import year_converter, month_converter, generate_date_series, get_stats_for_intervals
 from datetime import datetime, date, timedelta
 import pandas as pd
 from ConversationAnalyzer import ConversationAnalyzer
 
-"""
-
-"""
-
 
 class MessagingAnalyzer:
     def __init__(self, names, people):
-        # TODO input people only. class ill know what to do
+        # TODO input people only. class will know what to do
         self.names = names
         self.people = people
 
     def time_series_analysis_for_all(self, subject=None, **kwargs):
-        time_series = generate_time_series(**kwargs)
+        time_series = generate_date_series(**kwargs)
         stacked_df = self.stack_dfs(self.people)
         interval_stats = get_stats_for_intervals(self.get_stats, stacked_df, time_series, subject=subject)
 
     def get_stats(self, df, subject='all', start=None, end=None, period=None):
-        # TODO
+        # TODO LATER
         # here you have to do something with it
         pass
 
@@ -29,14 +25,14 @@ def get_count(self, attribute, subject='all', start=None, end=None, period=None)
         # we have a list of names we want to iterate over
         for name in self.names:
             stats = self.get_conversation_stats(name=name, subject=subject, start=start, end=end, period=period)
-            if stats is not None:  # TODO too explicit; needed because it is possible that None will be returned, if t got an empty df
+            if stats is not None:
                 count += getattr(stats, attribute)
         return count
 
     def get_conversation_stats(self, name, subject='all', start=None, end=None, period=None):
         messages = self.people.get(name).messages
         analyzer = ConversationAnalyzer(name, messages)
-        if analyzer is None:  # TODO this is too explicit ?!
+        if analyzer is None:
             return None
         return analyzer.get_stats(messages, subject=subject, start=start, end=end, period=period)
 
@@ -80,7 +76,7 @@ def most_used_messages_by_me(self, **kwargs):
         >>> s2 = pd.Series([3, 2, 1, 1])
         >>> s1_vc = s1.value_counts()
         >>> s2_vc = s2.value_counts()
-        TODO (later) most used is already a problem:
+        TODO LATER most used is already a problem:
           - because its a series of all the unique messages/words ever used in a convo
           - it contains strings like ':d', ':p' and 'xd'
           - from all the convos the result of value_counts has to be cleared
@@ -101,14 +97,14 @@ def most_used_words_by_partners(self, **kwargs):
 
     # 5. Number of messages sent/got on busiest period (by year/month/day/hour)
     def days_when_most_messages_sent(self):
-        # TODO hard algorithmic problem
+        # TODO LATER hard algorithmic problem
         pass
 
     def days_when_most_messages_received(self):
         pass
 
     def hours_when_most_messages_sent(self):
-        # TODO
+        # TODO LATER
         # is this referring to the absolute hour most messages sent??
         # like: 2014.07.25. 15h-16h
         # OR
@@ -131,5 +127,4 @@ def stack_dfs(people):
         for data in people.values():
             if data.messages is not None:
                 dfs.append(data.messages)
-        # TODO do I need to sort by index (date)? yes!
-        return pd.concat(dfs).sort_index()  # TODO why ignore_index??
+        return pd.concat(dfs).sort_index()
diff --git a/Visualizer.py b/Visualizer.py
index b4a3779..052ecb3 100644
--- a/Visualizer.py
+++ b/Visualizer.py
@@ -4,33 +4,37 @@
 from People import People
 from ConversationAnalyzer import ConversationAnalyzer
 
-
 # plt.rcParams.update({'figure.figsize': (10, 7), 'figure.dpi': 120})
 
+TEST_DATA_PATH = '/home/levente/projects/facebook-data-miner/tests/test_data'
+
+
 class Visualizer:
     def __init__(self):
         pass
 
-    def plot_time_series(self, x, y, title="Time series", xlabel='Date', ylabel='Value', dpi=100):
+    def plot_convos(self, names):
+        people = People(path=TEST_DATA_PATH)
+        for name in names:
+            data = self.set_up_data(people, name, period='d')
+            df = pd.DataFrame(data.items(), columns=['date', 'value'])
+            v.plot_time_series(x=df.date, y=df.value, title=name)
+
+    @staticmethod
+    def set_up_data(people, name, period='y'):
+        analyzer = ConversationAnalyzer(name, people.individuals.get(name).messages)
+        interval_stats = analyzer.get_time_series_data(subject='all', start=None, end=None, period=period)
+        return analyzer.get_plottable_time_series_data(interval_stats, statistic='msg_count')
+
+    @staticmethod
+    def plot_time_series(x, y, title="Time series", xlabel='Date', ylabel='Value', dpi=100):
         plt.figure(figsize=(16, 5), dpi=dpi)
         plt.plot(x, y, color='tab:red')
         plt.gca().set(title=title, xlabel=xlabel, ylabel=ylabel)
         plt.show()
 
 
-def set_up(people, name, interval='y'):
-    analyzer = ConversationAnalyzer(name, people.individuals.get(name).messages)
-    interval_stats = analyzer.get_time_series_data()
-    stats = interval_stats.get(interval)
-    return analyzer.get_plotable_time_series_data(stats, statistic='msg_count')
-
-
 if __name__ == "__main__":
-    v = Visualizer()
-    TEST_DATA_PATH = '/home/levente/projects/facebook-data-miner/tests/test_data'
-    people = People(path=TEST_DATA_PATH)
     names = ['Teflon Musk', 'Tőke Hal']
-    for name in names:
-        data = set_up(people, name, interval='d')
-        df = pd.DataFrame(data.items(), columns=['date', 'value'])
-        v.plot_time_series(x=df.date, y=df.value, title=name)
+    v = Visualizer()
+    v.plot_convos(names)
diff --git a/tests/TestConversationAnalyzer.py b/tests/TestConversationAnalyzer.py
index 5fa87d8..015ac19 100644
--- a/tests/TestConversationAnalyzer.py
+++ b/tests/TestConversationAnalyzer.py
@@ -43,7 +43,7 @@ def _stats(name, **kwargs):
     return _stats
 
 
-# TODO extend all functions with all the data
+# TODO LATER or not extend all functions with all the data
 def test_stats_toke_hal_all(statistics):
     stats = statistics('Tőke Hal')
 
@@ -133,8 +133,8 @@ def test_stats_toke_hal_me_2014_12(statistics):
 def test_stats_teflon_musk(statistics):
     stats = statistics('Teflon Musk')
     assert stats.msg_count == 6
-    assert stats.unique_msg_count == 2  # TODO this does not count media messages
-    # assert stats.most_used_msgs == 0 # TODO should only return the most used or e.g. top10 most used
+    assert stats.unique_msg_count == 2
+    # assert stats.most_used_msgs == 0 # TODO LATER should only return the most used or e.g. top10 most used
     # assert stats.msg_frequency == 0
     assert stats.word_count == 14
     assert stats.unique_word_count == 7
@@ -224,6 +224,9 @@ def test_stats_teflon_musk_all_2014_12(statistics):
     assert stats.char_count == 0
     # assert stats.most_used_chars == 0
 
+class TestConversationAnalyzer: # Foo Bar
+    pass
+
 
 def test_time_series_analysis_for_user(analyze):
     analyzer = analyze('Teflon Musk')
diff --git a/tests/TestMessagingAnalyzer.py b/tests/TestMessagingAnalyzer.py
index d53af70..b803693 100644
--- a/tests/TestMessagingAnalyzer.py
+++ b/tests/TestMessagingAnalyzer.py
@@ -2,61 +2,94 @@
 from MessagingAnalyzer import MessagingAnalyzer
 from utils import dt
 
-
 @pytest.fixture(scope='session')
 def analyzer(people):
     return MessagingAnalyzer(people.names, people.individuals)
 
 
 def test_total_number_of_messages(analyzer):
-    assert analyzer.total_number_of_messages() == 14
+    assert analyzer.total_number_of_messages() == 29
+
+    assert analyzer.total_number_of_messages(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_messages(start=dt(year=2014), period='y') == 11
     assert analyzer.total_number_of_messages(start=dt(year=2018), period='y') == 3
+    assert analyzer.total_number_of_messages(start=dt(year=2020), period='y') == 15
 
+    assert analyzer.total_number_of_messages(start=dt(year=2011, month=11), period='m') == 0
     assert analyzer.total_number_of_messages(start=dt(year=2014, month=9), period='m') == 1
     assert analyzer.total_number_of_messages(start=dt(year=2014, month=11), period='m') == 8
     assert analyzer.total_number_of_messages(start=dt(year=2014, month=12), period='m') == 2
-    assert analyzer.total_number_of_messages(start=dt(year=2018, month=1), period='m') == 3
 
-    assert analyzer.total_number_of_messages(start=dt(year=2000), period='y') == 0
-    assert analyzer.total_number_of_messages(start=dt(year=2011, month=11), period='m') == 0
+    assert analyzer.total_number_of_messages(start=dt(year=2018, month=1), period='m') == 3
     assert analyzer.total_number_of_messages(start=dt(year=2018, month=5), period='m') == 0
 
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=2), period='m') == 10
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=3), period='m') == 1  # jpg
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=4), period='m') == 2
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=5), period='m') == 1
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=6), period='m') == 0
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=8), period='m') == 1
+
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=2, day=13), period='d') == 2
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=2, day=13, hour=6), period='h') == 2
+
+    assert analyzer.total_number_of_messages(start=dt(year=2020, month=2, day=13, hour=6), period='d') == 4
+
 
 def test_total_number_of_words(analyzer):
-    assert analyzer.total_number_of_words() == 24
+    assert analyzer.total_number_of_words() == 86
 
     assert analyzer.total_number_of_words(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_words(start=dt(year=2014), period='y') == 20
-    assert analyzer.total_number_of_words(start=dt(year=2018), period='y') == 4
+    assert analyzer.total_number_of_words(start=dt(year=2018), period='y') == 32
+    assert analyzer.total_number_of_words(start=dt(year=2020), period='y') == 34
 
     assert analyzer.total_number_of_words(start=dt(year=2014, month=9), period='m') == 6
     assert analyzer.total_number_of_words(start=dt(year=2014, month=11), period='m') == 13
     assert analyzer.total_number_of_words(start=dt(year=2014, month=12), period='m') == 1
 
-    assert analyzer.total_number_of_words(start=dt(year=2018, month=1), period='m') == 4
+    assert analyzer.total_number_of_words(start=dt(year=2018, month=1), period='m') == 32
     assert analyzer.total_number_of_words(start=dt(year=2018, month=2), period='m') == 0
 
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=2), period='m') == 27
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=3), period='m') == 0
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=4), period='m') == 4
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=5), period='m') == 1
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=6), period='m') == 0
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=8), period='m') == 2
+
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=2, day=13), period='d') == 14
+    assert analyzer.total_number_of_words(start=dt(year=2020, month=2, day=13, hour=5), period='d') == 14
+
 
 def test_total_number_of_characters(analyzer):
-    assert analyzer.total_number_of_characters() == 81
+    assert analyzer.total_number_of_characters() == 379
 
     assert analyzer.total_number_of_characters(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_characters(start=dt(year=2014), period='y') == 69
-    assert analyzer.total_number_of_characters(start=dt(year=2018), period='y') == 12
+    assert analyzer.total_number_of_characters(start=dt(year=2018), period='y') == 170
+    assert analyzer.total_number_of_characters(start=dt(year=2020), period='y') == 140
 
     assert analyzer.total_number_of_characters(start=dt(year=2014, month=9), period='m') == 24
     assert analyzer.total_number_of_characters(start=dt(year=2014, month=11), period='m') == 42
     assert analyzer.total_number_of_characters(start=dt(year=2014, month=12), period='m') == 3
 
-    assert analyzer.total_number_of_characters(start=dt(year=2018, month=1), period='m') == 12
+    assert analyzer.total_number_of_characters(start=dt(year=2018, month=1), period='m') == 170
     assert analyzer.total_number_of_characters(start=dt(year=2018, month=2), period='m') == 0
 
+    assert analyzer.total_number_of_characters(start=dt(year=2020, month=2), period='m') == 114
+    assert analyzer.total_number_of_characters(start=dt(year=2020, month=3), period='m') == 0
+    assert analyzer.total_number_of_characters(start=dt(year=2020, month=4), period='m') == 17
+    assert analyzer.total_number_of_characters(start=dt(year=2020, month=5), period='m') == 4
+    assert analyzer.total_number_of_characters(start=dt(year=2020, month=6), period='m') == 0
+    assert analyzer.total_number_of_characters(start=dt(year=2020, month=8), period='m') == 5
+
 
 def test_total_number_of_messages_sent(analyzer):
-    assert analyzer.total_number_of_messages_sent() == 8
+    assert analyzer.total_number_of_messages_sent() == 17
     assert analyzer.total_number_of_messages_sent(start=dt(year=2014), period='y') == 6
     assert analyzer.total_number_of_messages_sent(start=dt(year=2018), period='y') == 2
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020), period='y') == 9
 
     assert analyzer.total_number_of_messages_sent(start=dt(year=2014, month=9), period='m') == 1
     assert analyzer.total_number_of_messages_sent(start=dt(year=2014, month=11), period='m') == 4
@@ -67,58 +100,108 @@ def test_total_number_of_messages_sent(analyzer):
     assert analyzer.total_number_of_messages_sent(start=dt(year=2011, month=11), period='m') == 0
     assert analyzer.total_number_of_messages_sent(start=dt(year=2018, month=5), period='m') == 0
 
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=2), period='m') == 6
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=3), period='m') == 0
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=4), period='m') == 2
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=5), period='m') == 0
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=6), period='m') == 0
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=8), period='m') == 1
+
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=2, day=13), period='d') == 1
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=2, day=13, hour=6), period='h') == 1
+    assert analyzer.total_number_of_messages_sent(start=dt(year=2020, month=2, day=13, hour=18), period='h') == 0
+
 
 def test_total_number_of_words_sent(analyzer):
-    assert analyzer.total_number_of_words_sent() == 19
+    assert analyzer.total_number_of_words_sent() == 69
 
     assert analyzer.total_number_of_words_sent(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_words_sent(start=dt(year=2014), period='y') == 16
-    assert analyzer.total_number_of_words_sent(start=dt(year=2018), period='y') == 3
+    assert analyzer.total_number_of_words_sent(start=dt(year=2018), period='y') == 31
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020), period='y') == 22
 
     assert analyzer.total_number_of_words_sent(start=dt(year=2014, month=9), period='m') == 6
     assert analyzer.total_number_of_words_sent(start=dt(year=2014, month=11), period='m') == 9
     assert analyzer.total_number_of_words_sent(start=dt(year=2014, month=12), period='m') == 1
 
-    assert analyzer.total_number_of_words_sent(start=dt(year=2018, month=1), period='m') == 3
+    assert analyzer.total_number_of_words_sent(start=dt(year=2018, month=1), period='m') == 31
     assert analyzer.total_number_of_words_sent(start=dt(year=2018, month=2), period='m') == 0
 
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=2), period='m') == 16
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=3), period='m') == 0
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=4), period='m') == 4
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=5), period='m') == 0
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=6), period='m') == 0
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=8), period='m') == 2
+
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=2, day=13), period='d') == 5
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=2, day=13, hour=6), period='h') == 5
+    assert analyzer.total_number_of_words_sent(start=dt(year=2020, month=2, day=13, hour=7), period='h') == 0
+
 
 def test_total_number_of_characters_sent(analyzer):
-    assert analyzer.total_number_of_characters_sent() == 69
+    assert analyzer.total_number_of_characters_sent() == 311
 
     assert analyzer.total_number_of_characters_sent(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_characters_sent(start=dt(year=2014), period='y') == 60
-    assert analyzer.total_number_of_characters_sent(start=dt(year=2018), period='y') == 9
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2018), period='y') == 167
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020), period='y') == 84
 
     assert analyzer.total_number_of_characters_sent(start=dt(year=2014, month=9), period='m') == 24
     assert analyzer.total_number_of_characters_sent(start=dt(year=2014, month=11), period='m') == 33
     assert analyzer.total_number_of_characters_sent(start=dt(year=2014, month=12), period='m') == 3
 
-    assert analyzer.total_number_of_characters_sent(start=dt(year=2018, month=1), period='m') == 9
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2018, month=1), period='m') == 167
     assert analyzer.total_number_of_characters_sent(start=dt(year=2018, month=2), period='m') == 0
 
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=2), period='m') == 62
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=3), period='m') == 0
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=4), period='m') == 17
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=5), period='m') == 0
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=6), period='m') == 0
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=8), period='m') == 5
+
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=2, day=13, hour=6), period='d') == 21
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=2, day=13, hour=7), period='d') == 0
+
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=2, day=13, hour=6), period='h') == 21
+    assert analyzer.total_number_of_characters_sent(start=dt(year=2020, month=2, day=13, hour=7), period='h') == 0
+
 
 def test_total_number_of_messages_received(analyzer):
-    assert analyzer.total_number_of_messages_received() == 6
+    assert analyzer.total_number_of_messages_received() == 12
+    assert analyzer.total_number_of_messages_received(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_messages_received(start=dt(year=2014), period='y') == 5
     assert analyzer.total_number_of_messages_received(start=dt(year=2018), period='y') == 1
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020), period='y') == 6
+
+    assert analyzer.total_number_of_messages_received(start=dt(year=2011, month=11), period='m') == 0
 
     assert analyzer.total_number_of_messages_received(start=dt(year=2014, month=9), period='m') == 0
     assert analyzer.total_number_of_messages_received(start=dt(year=2014, month=11), period='m') == 4
     assert analyzer.total_number_of_messages_received(start=dt(year=2014, month=12), period='m') == 1
-    assert analyzer.total_number_of_messages_received(start=dt(year=2018, month=1), period='m') == 1
 
-    assert analyzer.total_number_of_messages_received(start=dt(year=2000), period='y') == 0
-    assert analyzer.total_number_of_messages_received(start=dt(year=2011, month=11), period='m') == 0
+    assert analyzer.total_number_of_messages_received(start=dt(year=2018, month=1), period='m') == 1
     assert analyzer.total_number_of_messages_received(start=dt(year=2018, month=5), period='m') == 0
 
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=2), period='m') == 4
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=3), period='m') == 1
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=4), period='m') == 0
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=5), period='m') == 1
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=8), period='m') == 0
+
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=2, day=13), period='d') == 1
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=2, day=14), period='d') == 2
+    assert analyzer.total_number_of_messages_received(start=dt(year=2020, month=2, day=18), period='d') == 1
+
 
 def test_total_number_of_words_received(analyzer):
-    assert analyzer.total_number_of_words_received() == 5
+    assert analyzer.total_number_of_words_received() == 17
 
     assert analyzer.total_number_of_words_received(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_words_received(start=dt(year=2014), period='y') == 4
     assert analyzer.total_number_of_words_received(start=dt(year=2018), period='y') == 1
+    assert analyzer.total_number_of_words_received(start=dt(year=2020), period='y') == 12
 
     assert analyzer.total_number_of_words_received(start=dt(year=2014, month=9), period='m') == 0
     assert analyzer.total_number_of_words_received(start=dt(year=2014, month=11), period='m') == 4
@@ -127,13 +210,22 @@ def test_total_number_of_words_received(analyzer):
     assert analyzer.total_number_of_words_received(start=dt(year=2018, month=1), period='m') == 1
     assert analyzer.total_number_of_words_received(start=dt(year=2018, month=2), period='m') == 0
 
+    assert analyzer.total_number_of_words_received(start=dt(year=2020, month=2), period='m') == 11
+    assert analyzer.total_number_of_words_received(start=dt(year=2020, month=3), period='m') == 0
+    assert analyzer.total_number_of_words_received(start=dt(year=2020, month=5), period='m') == 1
+
+    assert analyzer.total_number_of_words_received(start=dt(year=2020, month=2, day=13), period='d') == 9
+    assert analyzer.total_number_of_words_received(start=dt(year=2020, month=2, day=14), period='d') == 2
+    assert analyzer.total_number_of_words_received(start=dt(year=2020, month=2, day=18), period='d') == 0
+
 
 def test_total_number_of_characters_received(analyzer):
-    assert analyzer.total_number_of_characters_received() == 12
+    assert analyzer.total_number_of_characters_received() == 68
 
     assert analyzer.total_number_of_characters_received(start=dt(year=2000), period='y') == 0
     assert analyzer.total_number_of_characters_received(start=dt(year=2014), period='y') == 9
     assert analyzer.total_number_of_characters_received(start=dt(year=2018), period='y') == 3
+    assert analyzer.total_number_of_characters_received(start=dt(year=2020), period='y') == 56
 
     assert analyzer.total_number_of_characters_received(start=dt(year=2014, month=9), period='m') == 0
     assert analyzer.total_number_of_characters_received(start=dt(year=2014, month=11), period='m') == 9
@@ -141,3 +233,11 @@ def test_total_number_of_characters_received(analyzer):
 
     assert analyzer.total_number_of_characters_received(start=dt(year=2018, month=1), period='m') == 3
     assert analyzer.total_number_of_characters_received(start=dt(year=2018, month=2), period='m') == 0
+
+    assert analyzer.total_number_of_characters_received(start=dt(year=2020, month=2), period='m') == 52
+    assert analyzer.total_number_of_characters_received(start=dt(year=2020, month=3), period='m') == 0
+    assert analyzer.total_number_of_characters_received(start=dt(year=2020, month=5), period='m') == 4
+
+    assert analyzer.total_number_of_characters_received(start=dt(year=2020, month=2, day=13), period='d') == 30
+    assert analyzer.total_number_of_characters_received(start=dt(year=2020, month=2, day=14), period='d') == 22
+    assert analyzer.total_number_of_characters_received(start=dt(year=2020, month=2, day=18), period='d') == 0
diff --git a/tests/TestPeople.py b/tests/TestPeople.py
index e29448b..61295d8 100644
--- a/tests/TestPeople.py
+++ b/tests/TestPeople.py
@@ -9,7 +9,7 @@ def people_names():
 
 
 def test_specific_people_has_or_has_not_got_messages(people):
-    # TODO parametrize
+    # TODO LATER parametrize
     import pandas as pd
     assert isinstance(people.data.get('Benedek Elek').get('messages'), pd.DataFrame)
     assert isinstance(people.data.get('Teflon Musk').get('messages'), pd.DataFrame)
@@ -33,4 +33,4 @@ def test_some_convos_are_with_friends(people):
 def test_specific_people_has_or_has_not_got_media(people):
     assert people.data.get('Teflon Musk').get('media_dir')
 
-#TODO test individuals too
\ No newline at end of file
+#TODO LATER test individuals too
\ No newline at end of file
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/audio/audioclip15905232600004598_2621787141481389.mp4 b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/audio/audioclip15905232600004598_2621787141481389.mp4
new file mode 100644
index 0000000..7c1e48d
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/audio/audioclip15905232600004598_2621787141481389.mp4 differ
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/files/1810.04805.pdf b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/files/1810.04805.pdf
new file mode 100644
index 0000000..2394716
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/files/1810.04805.pdf differ
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/gifs/19349964_1624604560892442_7457726181358436352_n_487109582171361.gif b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/gifs/19349964_1624604560892442_7457726181358436352_n_487109582171361.gif
new file mode 100644
index 0000000..54f7837
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/gifs/19349964_1624604560892442_7457726181358436352_n_487109582171361.gif differ
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/gifs/97999627_1419172538270405_8596479473619042304_n_2963870430335255.gif b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/gifs/97999627_1419172538270405_8596479473619042304_n_2963870430335255.gif
new file mode 100644
index 0000000..079f486
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/gifs/97999627_1419172538270405_8596479473619042304_n_2963870430335255.gif differ
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/photos/apple-5391076_960_720.jpg b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/photos/apple-5391076_960_720.jpg
new file mode 100644
index 0000000..6c1d5ff
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/photos/apple-5391076_960_720.jpg differ
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/photos/blueberry-5417154_960_720.jpg b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/photos/blueberry-5417154_960_720.jpg
new file mode 100644
index 0000000..15530bd
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/photos/blueberry-5417154_960_720.jpg differ
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/videos/thumbnails/20562500_1414664788613193_5052712665724834582_n.jpg b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/videos/thumbnails/20562500_1414664788613193_5052712665724834582_n.jpg
new file mode 100644
index 0000000..0fca343
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/videos/thumbnails/20562500_1414664788613193_5052712665724834582_n.jpg differ
diff --git a/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/videos/video1501528035_1573509826004448.mp4 b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/videos/video1501528035_1573509826004448.mp4
new file mode 100644
index 0000000..147852b
Binary files /dev/null and b/tests/test_data/messages/inbox/FooBar_n5fd6gG50h/videos/video1501528035_1573509826004448.mp4 differ
diff --git a/tests/test_data/messages/inbox/benedekelek_s4f65sdg/message_1.json b/tests/test_data/messages/inbox/benedekelek_s4f65sdg/message_1.json
index 61c06a1..ecdb68d 100644
--- a/tests/test_data/messages/inbox/benedekelek_s4f65sdg/message_1.json
+++ b/tests/test_data/messages/inbox/benedekelek_s4f65sdg/message_1.json
@@ -23,7 +23,7 @@
     {
       "sender_name": "Levente Cs\u00c5\u0091ke",
       "timestamp_ms": 1515571228715,
-      "content": "yo",
+      "content": "yo Legyen az, hogy most megprobalok ekezet nelkul irni. Seems pretty easy. I need some english words in here. Right? A magyar szavak felismereset probalom tesztelni ezzekkel a mondatokkal.",
       "type": "Generic"
     }
   ],
diff --git a/tests/test_data/messages/inbox/foobar_n5fd6gG50h/message_1.json b/tests/test_data/messages/inbox/foobar_n5fd6gG50h/message_1.json
new file mode 100644
index 0000000..26953da
--- /dev/null
+++ b/tests/test_data/messages/inbox/foobar_n5fd6gG50h/message_1.json
@@ -0,0 +1,157 @@
+{
+  "participants": [
+    {
+      "name": "Foo Bar"
+    },
+    {
+      "name": "Levente Cs\u00c5\u0091ke"
+    }
+  ],
+  "messages": [
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1596910928321,
+      "content": "OUT! \u00e2\u009d\u00a4",
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Foo Bar",
+      "timestamp_ms": 1588500928123,
+      "content": "OUT!",
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1587850928047,
+      "videos": [
+        {
+          "uri": "messages/inbox/FooBar_n5fd6gG50h/videos/video1501528035_1573509826004448.mp4",
+          "creation_timestamp": 1587850940,
+          "thumbnail": {
+            "uri": "messages/inbox/FooBar_n5fd6gG50h/videos/thumbnails/20562500_1414664788613193_5052712665724834582_n.jpg"
+          }
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1585850928047,
+      "content": "Whet? Check this! :P",
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Foo Bar",
+      "timestamp_ms": 1583750928047,
+      "audio_files": [
+        {
+          "uri": "messages/inbox/FooBar_n5fd6gG50h/audio/audioclip15905232600004598_2621787141481389.mp4",
+          "creation_timestamp": 1583750927
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1582720928145,
+      "photos": [
+        {
+          "uri": "messages/inbox/FooBar_n5fd6gG50h/photos/apple-5391076_960_720.jpg"
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Foo Bar",
+      "timestamp_ms": 1582010928145,
+      "gifs": [
+        {
+          "uri": "messages/inbox/FooBar_n5fd6gG50h/gifs/19349964_1624604560892442_7457726181358436352_n_487109582171361.gif"
+        }
+      ],
+      "reactions": [
+        {
+          "reaction": "\u00e2\u009d\u00a4",
+          "actor": "Levente Cs\u00c5\u0091ke"
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1581980928047,
+      "content": "What the hack? xdddddd :D",
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1581700928145,
+      "gifs": [
+        {
+          "uri": "messages/inbox/FooBar_n5fd6gG50h/gifs/97999627_1419172538270405_8596479473619042304_n_2963870430335255.gif"
+        }
+      ],
+      "reactions": [
+        {
+          "reaction": "\u00f0\u009f\u0098\u00ae",
+          "actor": "Foo Bar"
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Foo Bar",
+      "timestamp_ms": 1581690928047,
+      "content": "Excepteur...laborum. :D",
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1581680928047,
+      "content": "Duis duia .. ! xdddddd :D",
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Foo Bar",
+      "timestamp_ms": 1581650928047,
+      "files": [
+        {
+          "uri": "messages/inbox/FooBar_n5fd6gG50h/files/1810.04805.pdf"
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1581640928145,
+      "photos": [
+        {
+          "uri": "messages/inbox/FooBar_n5fd6gG50h/photos/blueberry-5417154_960_720.jpg"
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Foo Bar",
+      "timestamp_ms": 1581570938715,
+      "content": "Ut akar ... consequat. oO wow :P xd :D",
+      "reactions": [
+        {
+          "reaction": "\u00e2\u009d\u00a4",
+          "actor": "Levente Cs\u00c5\u0091ke"
+        }
+      ],
+      "type": "Generic"
+    },
+    {
+      "sender_name": "Levente Cs\u00c5\u0091ke",
+      "timestamp_ms": 1581570928715,
+      "content": "Lorem lorim.. foo bar \u00f0\u009f\u0098\u00a1\u00f0\u009f\u0098\u00a1\u00f0\u009f\u0098\u00a1",
+      "type": "Generic"
+    }
+  ],
+  "title": "Foo Bar",
+  "is_still_participant": true,
+  "thread_type": "Regular",
+  "thread_path": "inbox/foobar_n5fd6gG50h"
+}
\ No newline at end of file
diff --git a/tests/test_utils.py b/tests/test_utils.py
index 76352c0..579569c 100644
--- a/tests/test_utils.py
+++ b/tests/test_utils.py
@@ -3,56 +3,85 @@
 from pathlib import Path
 import reusables
 from reusables.cli import *
+from datetime import datetime
 
+import pytest
 
-#TODO convert this to pytest
-
-class TestUtils(unittest.TestCase):
-    def setUp(self):
-        self.test_data_path = Path(f'{pwd()}/test_data')
-
-    def test_read_json(self):
-        dummy1 = {'0': 'a', '2': 'c'}
-        dummy1_path = Path(self.test_data_path / 'dummy1.json')
-        dump_to_json(file=dummy1_path, data=dummy1)
-
-        dummy2 = {'1': 'b', '3': 'd'}
-        dummy2_path = Path(self.test_data_path / 'dummy2.json')
-        dump_to_json(file=dummy2_path, data=dummy2)
-
-        dummy_unified = get_messages(dummy1_path, dummy2_path)
-        expected = {'0': 'a', '1': 'b', '2': 'c', '3': 'd'}
-
-        self.assertDictEqual(expected, dummy_unified)
-
-        dummy1_path.unlink()
-        dummy2_path.unlink()
-
-    def test_decode_text(self):
-        dummy = {'0': '\u00c5\u0091', '1': '\u00c3\u00a1', '2': ['\u00c5\u0091', {'0': '\u00c3\u00a1'}],
-                 '3': {'0': '\u00c5\u0091'}}
-        dummy_path = Path(self.test_data_path / 'dummy.json')
-        dump_to_json(file=dummy_path, data=dummy)
-        read = get_messages(dummy_path, decode=True)
-        expected = {'0': 'ő', '1': 'á', '2': ['ő', {'0': 'á'}],
-                    '3': {'0': 'ő'}}
-        self.assertEqual(expected, read)
-
-        dummy_path.unlink()
-
-    def later_tests(self):
-        data = {'participants': [{'name': 'Csőke Boglárka'}, {'name': 'Levente Csőke'}],
-                'title': 'Csőke Boglárka',
-                'is_still_participant': True,
-                'thread_type': 'Regular',
-                'thread_path': 'inbox/CsokeBoglarka_5A48Zi9P1w',
-                'messages': [...]}
-        msg_element = {'sender_name': 'Levente Csőke',
-                       'timestamp_ms': 1440948801592,
-                       'content': 'ahaa',
-                       'type': 'Generic'}
-
-    def test_read_real_data(self):
-        bogi_msg = Path('/home/levente/projects/facebook-data-miner/data/messages/inbox/csokeboglarka_5a48zi9p1w')
-        data = get_messages(bogi_msg/'message_1.json', bogi_msg/'message_2.json')
-        msg = data['messages']
\ No newline at end of file
+
+# class TestUtils(unittest.TestCase):
+#     def setUp(self):
+#         self.test_data_path = Path(f'{pwd()}/test_data')
+#
+#     def test_read_json(self):
+#         dummy1 = {'0': 'a', '2': 'c'}
+#         dummy1_path = Path(self.test_data_path / 'dummy1.json')
+#         dump_to_json(file=dummy1_path, data=dummy1)
+#
+#         dummy2 = {'1': 'b', '3': 'd'}
+#         dummy2_path = Path(self.test_data_path / 'dummy2.json')
+#         dump_to_json(file=dummy2_path, data=dummy2)
+#
+#         dummy_unified = get_messages(dummy1_path, dummy2_path)
+#         expected = {'0': 'a', '1': 'b', '2': 'c', '3': 'd'}
+#
+#         self.assertDictEqual(expected, dummy_unified)
+#
+#         dummy1_path.unlink()
+#         dummy2_path.unlink()
+#
+#     def test_decode_text(self):
+#         dummy = {'0': '\u00c5\u0091', '1': '\u00c3\u00a1', '2': ['\u00c5\u0091', {'0': '\u00c3\u00a1'}],
+#                  '3': {'0': '\u00c5\u0091'}}
+#         dummy_path = Path(self.test_data_path / 'dummy.json')
+#         dump_to_json(file=dummy_path, data=dummy)
+#         read = get_messages(dummy_path, decode=True)
+#         expected = {'0': 'ő', '1': 'á', '2': ['ő', {'0': 'á'}],
+#                     '3': {'0': 'ő'}}
+#         self.assertEqual(expected, read)
+#
+#         dummy_path.unlink()
+#
+#     def later_tests(self):
+#         data = {'participants': [{'name': 'Csőke Boglárka'}, {'name': 'Levente Csőke'}],
+#                 'title': 'Csőke Boglárka',
+#                 'is_still_participant': True,
+#                 'thread_type': 'Regular',
+#                 'thread_path': 'inbox/CsokeBoglarka_5A48Zi9P1w',
+#                 'messages': [...]}
+#         msg_element = {'sender_name': 'Levente Csőke',
+#                        'timestamp_ms': 1440948801592,
+#                        'content': 'ahaa',
+#                        'type': 'Generic'}
+#
+#     def test_read_real_data(self):
+#         bogi_msg = Path('/home/levente/projects/facebook-data-miner/data/messages/inbox/csokeboglarka_5a48zi9p1w')
+#         data = get_messages(bogi_msg/'message_1.json', bogi_msg/'message_2.json')
+#         msg = data['messages']
+#
+#
+# decode_text('Minden jot Levii \u00e2\u009d\u00a4\u00ef\u00b8\u008f')
+# decode_text('”Hat \u00f0\u009f\u008e\u00a9\u00f0\u009f\u00a4\u0094')
+# decode_text('\u00f0\u009f\u0098\u00a1\u00f0\u009f\u0098\u00a1\u00f0\u009f\u0098\u00a1')
+
+
+def test_generate_date_series():
+    start = datetime(2020, 1, 1, 0, 0)
+    end = datetime(2021, 1, 1, 0, 0)
+
+    date_range_year = generate_date_series(start, end, 'y')
+    assert len(date_range_year) == 1 + 1
+
+    date_range_month = generate_date_series(start, end, 'm')
+    assert len(date_range_month) == 12 + 1
+
+    date_range_day = generate_date_series(start, end, 'd')
+    assert len(date_range_day) == 366 + 1
+
+    date_range_hour = generate_date_series(start, end, 'h')
+    assert len(date_range_hour) == (366 * 24) + 1
+
+    for day in date_range_day:
+        assert isinstance(day, datetime)
+
+    with pytest.raises(ValueError):
+        faulty_date_range = generate_date_series(start, end, )
diff --git a/utils.py b/utils.py
index 8594258..2a48624 100644
--- a/utils.py
+++ b/utils.py
@@ -8,37 +8,6 @@
           'november', 'december']
 
 
-def get_stats_for_intervals(func, df, time_series, subject='all'):
-    data = {}
-    for offset, series in time_series.items():
-        data[offset] = {}
-        for i in range(len(series) - 1):  # only looping len - 1 times
-            start = series[i]
-            # TODO will we miss the last entry? I dont think so (99%), but check and correct hand in hand with the timeseries bug
-            # IT DOES NOT! HOWEVER test it with new data injected/modified at runtime <- this is hard
-            end = series[i + 1]
-            data[offset][start] = func(df, subject=subject, start=start, end=end)
-    return data
-
-
-# @date_checker
-def dt(year: int = 2004, month: int = 1, day: int = 1, hour: int = 0):
-    return datetime(year=year, month=month, day=day, hour=hour)
-
-
-def get_messages(*files, decode=True):
-    data = {}
-    for file in files:
-        temp = decode_text(read_json(file)) if decode else read_json(file)
-        if not data:
-            data = temp
-        elif data.get('messages') and temp.get('messages'):
-            data['messages'] += temp.get('messages')
-            if sorted(temp.keys()) != sorted(data.keys()):
-                data = {**temp, **data}
-    return data
-
-
 def read_json(file):
     with open(file) as f:
         return json.load(f)
@@ -49,19 +18,6 @@ def dump_to_json(data=None, file=None):
         json.dump(data, f)
 
 
-def decode_text(obj):
-    if isinstance(obj, str):
-        return obj.encode('latin_1').decode('utf-8')
-
-    if isinstance(obj, list):
-        return [decode_text(o) for o in obj]
-
-    if isinstance(obj, dict):
-        return {key: decode_text(item) for key, item in obj.items()}
-
-    return obj
-
-
 def order_list_of_dicts(lst, key='timestamp_ms'):
     return sorted(lst, key=lambda k: k[key])
 
@@ -128,25 +84,6 @@ def wrapper(*args, **kwargs):
     return wrapper
 
 
-# TODO period can refer to 2 things
-# 1. I used it for y/m/d/h
-# 2. return only with one date_range series
-# decide wtf
-def generate_time_series(start=None, end=None, period=None):
-    start = start or datetime(year=2009, month=10, day=2, hour=0)
-    end = end or datetime.now()
-    time_series = {
-        'y': pd.date_range(start=start, end=end, freq='YS'),  # TODO does not include 2009
-        'm': pd.date_range(start=start, end=end, freq='1MS'),  # TODO does not include october
-        'd': pd.date_range(start=start, end=end, freq='1D'),  # TODO does not include 2. ?! not sure if it is true
-        # TODO put this back after dev phase is over
-        # 'h': pd.date_range(start=start, end=end, freq='1H'), # TODO hour should only be run ONCE
-    }
-    if period and period in ('y', 'm', 'd', 'h'):
-        return time_series[period]
-    return time_series
-
-
 def subject_checker(func):
     def wrapper(*args, **kwargs):
         if not kwargs.get('subject') or kwargs.get('subject') not in ('all', 'me', 'partner'):
@@ -173,19 +110,79 @@ def wrapper(*args, **kwargs):
     return wrapper
 
 
+DELTA_MAP = {
+    'y': relativedelta(years=+1),
+    'm': relativedelta(months=+1),
+    'd': timedelta(days=1),
+    'h': timedelta(hours=1)
+}
+
+
 def period_checker(func):
     def wrapper(*args, **kwargs):
         if kwargs.get('start') is not None and kwargs.get('end') is not None:
             return func(*args, **kwargs)
-        delta_map = {
-            'y': relativedelta(years=+1),
-            'm': relativedelta(months=+1),
-            'd': timedelta(days=1),
-            'h': timedelta(hours=1)
-        }
-        if not kwargs.get('period') or delta_map[kwargs.get('period')] is None:
+
+        if not kwargs.get('period') or DELTA_MAP[kwargs.get('period')] is None:
             raise ValueError('Parameter `period` should be one of {y, m, d, h}')
-        kwargs['period'] = delta_map[kwargs.get('period')]
+        kwargs['period'] = DELTA_MAP[kwargs.get('period')]
         return func(*args, **kwargs)
 
     return wrapper
+
+
+def generate_date_series(start=None, end=None, period=None):
+    if period is None or DELTA_MAP.get(period) is None:
+        raise ValueError('Parameter `period` should be one of {y, m, d, h}')
+    start = start or datetime(year=2009, month=10, day=2, hour=0) # TODO change this to date when user joined FB
+    end = end or datetime.now()
+
+    dates = []
+    intermediate = start
+    while intermediate <= end:
+        dates.append(intermediate)
+        intermediate = intermediate + DELTA_MAP.get(period)
+    return dates
+
+
+def get_stats_for_intervals(func, df, time_series, subject='all'):
+    data = {}
+    for offset, series in time_series.items():
+        data[offset] = {}
+        for i in range(len(series) - 1):  # only looping len - 1 times
+            start = series[i]
+            # TODO LATER will we miss the last entry? I dont think so (99%), but check and correct hand in hand with the timeseries bug
+            # IT DOES NOT! HOWEVER test it with new data injected/modified at runtime <- this is hard
+            end = series[i + 1]
+            data[offset][start] = func(df, subject=subject, start=start, end=end)
+    return data
+
+
+def dt(year: int = 2004, month: int = 1, day: int = 1, hour: int = 0):
+    return datetime(year=year, month=month, day=day, hour=hour)
+
+
+def get_messages(*files, decode=True):
+    data = {}
+    for file in files:
+        temp = decode_text(read_json(file)) if decode else read_json(file)
+        if not data:
+            data = temp
+        elif data.get('messages') and temp.get('messages'):
+            data['messages'] += temp.get('messages')
+            if sorted(temp.keys()) != sorted(data.keys()):
+                data = {**temp, **data}
+    return data
+
+
+def decode_text(obj):
+    if isinstance(obj, str):
+        return obj.encode('latin_1').decode('utf-8')
+
+    if isinstance(obj, list):
+        return [decode_text(o) for o in obj]
+
+    if isinstance(obj, dict):
+        return {key: decode_text(item) for key, item in obj.items()}
+
+    return obj