moved biggest_single and worst_offenses into RoboPage

data.py

@@ -93,6 +93,7 @@ class MsgData:
         if len(msg.reactions) > 0:
             new = reaction_df(msg)
             async with self.lock:
+                # self.reactions = self.reactions.join(new, how='outer')
                 self.reactions = pd.concat([self.reactions, new])
             LOGGER.info(str(new.droplevel(level=0, axis=0).loc[:, 'count']))
 
@@ -114,13 +115,17 @@ class MsgData:
 
         # If there were actually some message ids found
         if message_id_counts.shape[0] > 0:
+            # Select the relevant messages from the self.msgs DataFrame using the filtered message ids
             res: pd.DataFrame = self.msgs.loc[message_id_counts]
 
+            # Add the 'count' column
             res['count'] = counts
 
+            # If necessary, filter by days into the past
             if days is not None and days > 0:
                 res = res[res['created'] >= (datetime.today() - timedelta(days=days)).astimezone()]
 
+            # return the message DataFrame, sorted by created (sent) time
             return res.sort_values('created', ascending=False)
 
         else:
@@ -142,32 +147,19 @@ class MsgData:
 
     def emoji_totals(self, emoji_name: str, days: int = None) -> pd.Series:
         """Creates a Series indexed by user id and with the number of reactions with emoji_name as values"""
-        return (self
-                .emoji_messages(emoji_name, days)
-                .groupby('user id')
-                .apply(lambda gdf: gdf['count'].sum())
-                .sort_values(ascending=False))
+        messages = self.emoji_messages(emoji_name, days)
+        if messages.shape[0] > 0:
+            return (messages
+                    .groupby('user id')
+                    .apply(lambda gdf: gdf['count'].sum())
+                    .sort_values(ascending=False))
+        else:
+            raise ValueError(f'No messages found for' + \
+                             f' {type(emoji_name)}:{emoji_name}, {type(days)}:{days}')
+            # return pd.DataFrame()
 
-    async def emoji_user_counts(self, client: discord.Client, emoji_name: str, days: int):
+    async def emoji_user_counts(self, client: discord.Client, emoji_name: str, days: int = None):
+        """Creates a Series indexed by user display_name with the number of reactions with emoji_name as values"""
         counts: pd.Series = self.emoji_totals(emoji_name, days)
         counts.index = pd.Index([(await client.fetch_user(user_id=uid)).display_name for uid in counts.index])
         return counts
-
-    def worst_offsenses(self, user: str, days: int):
-        cdf = self.emoji_messages('cancelled', days=days)
-        cdf = cdf[cdf['display_name'].str.contains(user, case=False)]
-
-        if cdf.shape[0] > 0:
-            res = f'{user}\'s top 5 cancellations in the last {days} days:\n'
-            res += f'\n'.join(
-                f'`{row["count"]:<2.0f}cancellations`\n{row["link"]}' for idx, row in cdf.iloc[:5].iterrows())
-        else:
-            res = f'No cancellations for {user} in the past {days} days'
-
-        return res
-
-    async def biggest_single(self, client: discord.Client, emoji: str, days: int) -> str:
-        data: pd.Series = self.emoji_totals(emoji_name=emoji, days=days)
-        user: discord.User = await client.fetch_user(user_id=data.index[0])
-        LOGGER.info(f'User: {user.mention}')
-        return f'{user.mention} with {data.iloc[0]:.0f} over the past {int(days)} days'