fix counter

v1.16.1

.env.example

@@ -0,0 +1,4 @@
+DISCORD_TOKEN=
+PYTHONPATH=./src
+CRYPT_KEY=
+LOG_DIR=logs

GDPR.md

@@ -0,0 +1,46 @@
+# About Analyst-bot's data usage
+
+## TL;DR
+
+Analyst-bot collects text message information. It does not share collected data with any third-party and data is retained 90 days or until the bot is leaving the guild/server.
+
+## Data collection
+
+Analyst-bot collects a Discord guild/server's history when asked to.
+
+This includes:
+
+- Visible text channel names
+- Visible text messages: date and time of creation and edition,  author,  content,  reactions and other available metadata (pinned, tts, etc.)
+
+This does __not__ includes:
+
+- Voice channels and not visible channels
+- Not visible text messages
+- Visible text messages' embedded content, images and other attachments
+
+## Data processing
+
+Any data collected is only processed in order to produce a one-time report sent to the user immediately. No temporary data are retained.
+
+## Data storage and retain policy
+
+Analyst-bot stores the collected data in files that are accessible by the software and its administrator only.
+
+Any collected data are retained maximum 90 days until deletion or when the bot is leaving a guild/server.
+
+## Data sharing
+
+Analyst-bot does not share the data collected with any third-party.
+
+## Right to retract
+
+If you want to have your data removed, you can use the `%gdpr revoke` command or remove this bot from your guild/server.
+
+## Terms agreement
+
+By agreeing to these terms, you ensure having the legal age if you are in a country that does have one and you also ensure having the consent of every member involved.
+
+*If you want more information, please contact the creator of this bot: <https://github.com/Klemek/discord-analyst>.*
+
+Type `%gdpr agree` to agree to these terms, `%gdpr revoke` to remove this guild/server's collected data or `%gdpr` to see this message again.

README.md

@@ -125,11 +125,14 @@ python3 src/main.py
 
 ## Changelog
 
+* **v1.17**
+  * compliency with 30 days data keeping policy and data encryption
+  * improvements and bug fix
 * **v1.16**
   * `%freq graph` graph hours frequency along the week
   * uses discord new time format
   * `%freq` now shows quietest day of week and hour of day
-  * improvments and bug fix
+  * improvements and bug fix
 * **v1.15**
   * `nsfw:allow/only` filter nsfw channels
   * `%find` can use regexes

requirements.txt

@@ -1,6 +1,6 @@
-discord.py==1.7.0
+discord.py>=2.0.0
-python-dotenv==0.15.0
+python-dotenv>=0.15.0
-python-dateutil==2.8.1
+python-dateutil>=2.8.1
-git+git://github.com/Klemek/miniscord.git
+matplotlib>=3.4.2
-numpy
+cryptography>=2.8
-matplotlib
+git+https://github.com/Klemek/miniscord.git

src/data_types/counter.py

@@ -4,7 +4,7 @@ from collections import defaultdict
 
 # Custom libs
 
-from utils import plural, from_now, percent, val_sum, top_key
+from utils import plural, from_now, percent, val_sum, top_key, utc_today
 
 
 class Counter:
@@ -25,7 +25,7 @@ class Counter:
         if self.last_used is None:
             return 0
         return self.all_usages() + 1 / (
-            100000 * ((datetime.today() - self.last_used).days + 1)
+            100000 * (abs((utc_today() - self.last_used).days) + 1)
         )
 
     def all_usages(self) -> int:

src/data_types/emoji.py

@@ -5,7 +5,7 @@ import discord
 
 # Custom libs
 
-from utils import mention, plural, from_now, top_key, percent
+from utils import mention, plural, from_now, top_key, percent, utc_today
 
 
 class Emoji:
@@ -44,14 +44,14 @@ class Emoji:
         )
 
     def life_days(self) -> int:
-        return (datetime.today() - self.emoji.created_at).days
+        return (utc_today() - self.emoji.created_at).days
 
     def use_days(self) -> int:
         # If never used, use creation date instead
         if self.last_used is None:
             return self.life_days()
         else:
-            return (datetime.today() - self.last_used).days
+            return (utc_today() - self.last_used).days
 
     def get_top_member(self) -> int:
         return top_key(self.members)

src/data_types/frequency.py

@@ -2,7 +2,6 @@ from typing import List
 from datetime import timedelta
 import calendar
 import matplotlib.pyplot as plt
-import numpy as np
 from io import BytesIO
 import discord
 import time

src/data_types/presence.py

@@ -54,16 +54,16 @@ class Presence:
             f"- **most visited channel**: {channel_mention(top_channel)} ({self.channel_usage[top_channel]:,} msg, {percent(self.channel_usage[top_channel]/channel_sum)})"
             if show_top_channel
             else "",
-            f"- **most contributed channel**: {channel_mention(found_in)} ({self.channel_usage[found_in]:,} msg, {percent(self.channel_usage[found_in]/self.channel_total[found_in])} of {type})"
+            f"- **most contributed channel**: {channel_mention(found_in)} ({self.channel_usage[found_in]:,} msg,)"
             if show_top_channel and member_specific
             else "",
-            f"- **was mentioned**: {plural(mention_sum, 'time')} ({percent(mention_sum/val_sum(self.mention_count))} of {type})"
+            f"- **was mentioned**: {plural(mention_sum, 'time')}"
             if member_specific and len(self.mentions) > 0
             else "",
             f"- **mostly mentioned by**: {mention(top_mention)} ({plural(self.mentions[top_mention], 'time')}, {percent(self.mentions[top_mention]/mention_sum)})"
             if member_specific and len(self.mentions) > 0
             else "",
-            f"- **mentioned others**: {plural(mention_others_sum, 'time')} ({percent(mention_others_sum/val_sum(self.mention_count))} of {type})"
+            f"- **mentioned others**: {plural(mention_others_sum, 'time')}"
             if len(self.mention_others) > 0 and member_specific
             else "",
             f"- **mostly mentioned**: {mention(top_mention_others)} ({plural(self.mention_others[top_mention_others], 'time')}, {percent(self.mention_others[top_mention_others]/mention_others_sum)})"
@@ -81,7 +81,7 @@ class Presence:
             f"- **reactions**: {plural(total_reaction_used, 'time')}"
             if len(self.reactions) > 0 and not member_specific
             else "",
-            f"- **reactions**: {plural(total_reaction_used, 'time')} ({percent(total_reaction_used/val_sum(self.used_reaction))} of {type})"
+            f"- **reactions**: {plural(total_reaction_used, 'time')}"
             if len(self.reactions) > 0 and member_specific
             else "",
             f"- **top reactions**: {mention(top_reaction_member)} ({plural(self.used_reaction[top_reaction_member], 'time')}, {percent(self.used_reaction[top_reaction_member]/val_sum(self.used_reaction))})"

src/logs/channel_logs.py

@@ -66,8 +66,8 @@ class ChannelLogs:
         is_empty = self.last_message_id is None
         try:
             if is_empty:
-                sanity_check = len(await channel.history(limit=1).flatten())
+                sanity_check = len([message async for message in channel.history(limit=1)])
-                if sanity_check != 1:
+                if sanity_check < 1:
                     yield len(self.messages), True
                     return
             # load backward

src/logs/guild_logs.py

@@ -8,28 +8,30 @@ import time
 import logging
 import asyncio
 import threading
-
+from dotenv import load_dotenv
+from cryptography.fernet import Fernet
 
 from . import ChannelLogs
 from utils import code_message, delta, deltas
 
-
-LOG_DIR = "logs"
-LOG_EXT = ".logz"
-
 current_analysis = []
 current_analysis_lock = threading.Lock()
 
-
 ALREADY_RUNNING = -100
 CANCELLED = -200
 NO_FILE = -300
 
-# 5 minutes, assume 'fast' arg
+load_dotenv()
-MIN_MODIFICATION_TIME = 5 * 60
-# ~1 year, remove log file
-MAX_MODIFICATION_TIME = 365 * 24 * 60 * 60
 
+LOG_DIR = os.getenv("LOG_DIR", "logs")
+LOG_EXT = os.getenv("LOG_EXT", ".logz")
+CRYPT_KEY = os.getenv("CRYPT_KEY", "")
+
+# 5 minutes, assume 'fast' arg
+MIN_MODIFICATION_TIME = int(os.getenv("MAX_MODIFICATION_TIME", 5 * 60))
+
+# 90 days, remove log file
+MAX_MODIFICATION_TIME = int(os.getenv("MAX_MODIFICATION_TIME", 90 * 24 * 60 * 60))
 
 class Worker:
     def __init__(
@@ -129,29 +131,41 @@ class GuildLogs:
         channels = {}
         try:
             last_time = os.path.getmtime(self.log_file)
-            gziped_data = None
+            encrypted_data = None
-            await code_message(progress, "Reading saved history (1/4)...")
+            await code_message(progress, "Reading saved history (1/5)...")
             t0 = datetime.now()
             with open(self.log_file, mode="rb") as f:
-                gziped_data = f.read()
+                encrypted_data = f.read()
             logging.info(f"log {self.guild.id} > read in {delta(t0):,}ms")
             if self.check_cancelled():
                 return CANCELLED, 0
-            await code_message(progress, "Reading saved history (2/4)...")
+            await code_message(progress, "Reading saved history (2/5)...")
+            if CRYPT_KEY == "" or CRYPT_KEY is None:
+                gziped_data = encrypted_data
+            try:
+                t0 = datetime.now()
+                fernet = Fernet(CRYPT_KEY)
+                gziped_data = fernet.decrypt(encrypted_data)
+                logging.info(f"log {self.guild.id} > decrypted in {delta(t0):,}ms")
+            except:
+                gziped_data = encrypted_data
+            if self.check_cancelled():
+                return CANCELLED, 0                
+            await code_message(progress, "Reading saved history (3/5)...")
             t0 = datetime.now()
             json_data = gzip.decompress(gziped_data)
             del gziped_data
             logging.info(f"log {self.guild.id} > gzip decompress in {delta(t0):,}ms")
             if self.check_cancelled():
                 return CANCELLED, 0
-            await code_message(progress, "Reading saved history (3/4)...")
+            await code_message(progress, "Reading saved history (4/5)...")
             t0 = datetime.now()
             channels = json.loads(json_data)
             del json_data
             logging.info(f"log {self.guild.id} > json parse in {delta(t0):,}ms")
             if self.check_cancelled():
                 return CANCELLED, 0
-            await code_message(progress, "Reading saved history (4/4)...")
+            await code_message(progress, "Reading saved history (5/5)...")
             t0 = datetime.now()
             self.channels = {
                 int(id): ChannelLogs(channels[id], self) for id in channels
@@ -288,7 +302,7 @@ class GuildLogs:
                 return CANCELLED, 0
             await code_message(
                 progress,
-                f"Saving history (1/3)...\n{real_total_msg:,} messages in {real_total_chan:,} channels",
+                f"Saving history (1/4)...\n{real_total_msg:,} messages in {real_total_chan:,} channels",
             )
             t0 = datetime.now()
             json_data = bytes(json.dumps(self.dict()), "utf-8")
@@ -299,7 +313,7 @@ class GuildLogs:
                 return CANCELLED, 0
             await code_message(
                 progress,
-                f"Saving history (2/3)...\n{real_total_msg:,} messages in {real_total_chan:,} channels",
+                f"Saving history (2/4)...\n{real_total_msg:,} messages in {real_total_chan:,} channels",
             )
             t0 = datetime.now()
             gziped_data = gzip.compress(json_data)
@@ -311,12 +325,28 @@ class GuildLogs:
                 return CANCELLED, 0
             await code_message(
                 progress,
-                f"Saving history (3/3)...\n{real_total_msg:,} messages in {real_total_chan:,} channels",
+                f"Saving history (3/4)...\n{real_total_msg:,} messages in {real_total_chan:,} channels",
+            )
+            if CRYPT_KEY == "" or CRYPT_KEY is None:
+                encrypted_data = gziped_data
+            try:
+                t0 = datetime.now()
+                fernet = Fernet(CRYPT_KEY)
+                encrypted_data = fernet.encrypt(gziped_data)
+                logging.info(f"log {self.guild.id} > encrypted in {delta(t0):,}ms ->  {len(gziped_data) / deltas(t0):,.3f} b/s")
+            except:
+                encrypted_data = gziped_data
+            if self.check_cancelled():
+                return CANCELLED, 0
+            await code_message(
+                progress,
+                f"Saving history (4/4)...\n{real_total_msg:,} messages in {real_total_chan:,} channels",
             )
             t0 = datetime.now()
             with open(self.log_file, mode="wb") as f:
-                f.write(gziped_data)
+                f.write(encrypted_data)
             del gziped_data
+            del encrypted_data
             logging.info(
                 f"log {self.guild.id} > saved in {delta(t0):,}ms -> {real_total_msg / deltas(t0):,.3f} m/s"
             )

src/main.py

@@ -18,7 +18,7 @@ emojis.load_emojis()
 
 bot = Bot(
     "Discord Analyst",
-    "1.16.1",
+    "1.17.5",
     alias="%",
 )
 

src/scanners/presence_scanner.py

@@ -69,8 +69,8 @@ class PresenceScanner(Scanner):
             for mention in message.mentions:
                 pres.mention_others[mention] += 1
             pres.messages[message.author] += 1
-        pres.channel_total[channel.id] += 1
+            pres.channel_total[channel.id] += 1
-        pres.mention_count[message.author] += len(message.mentions)
+            pres.mention_count[message.author] += len(message.mentions)
         if len(raw_members) > 0:
             for mention in message.mentions:
                 if mention in raw_members:

src/scanners/scanner.py

@@ -18,6 +18,7 @@ from utils import (
     command_cache,
     FilterLevel,
     SPLIT_TOKEN,
+    utc_now
 )
 from logs import (
     GuildLogs,
@@ -131,7 +132,7 @@ class Scanner(ABC):
                 self.start_date = None if len(dates) < 1 else min(dates)
                 self.stop_date = None if len(dates) < 2 else max(dates)
 
-                if self.start_date is not None and self.start_date > datetime.now():
+                if self.start_date is not None and self.start_date > utc_now():
                     await message.channel.send(
                         f"Start date is after today", reference=message
                     )
@@ -228,7 +229,7 @@ class Scanner(ABC):
                                 ),
                             )
                             if self.stop_date is None:
-                                self.stop_date = datetime.utcnow()
+                                self.stop_date = utc_now()
 
                         self.msg_count = 0
                         self.total_msg = 0

src/tools/convert_logs.py

@@ -0,0 +1,26 @@
+import os
+import os.path
+from dotenv import load_dotenv
+from cryptography.fernet import Fernet
+
+load_dotenv()
+
+LOG_DIR = os.getenv("LOG_DIR", "logs")
+LOG_EXT = os.getenv("LOG_DIR", ".logz")
+CRYPT_KEY = os.getenv("CRYPT_KEY", "")
+
+fernet = Fernet(CRYPT_KEY)
+
+for item in os.listdir(LOG_DIR):
+    if item.endswith(LOG_EXT):
+        path = os.path.join(LOG_DIR, item)
+        data = None
+        with open(path, mode="rb") as f:
+            data = f.read()
+        try:
+            fernet.decrypt(data)
+            print(f"{item} already encrypted")
+        except:
+            with open(path, mode="wb") as f:
+                f.write(fernet.encrypt(data))
+            print(f"{item} was encrypted")

src/utils/gdpr.py

@@ -13,7 +13,7 @@ arguments:
 TEXT = """
 __**About Analyst-bot's data usage**__
 **TL;DR**
-Analyst-bot collects text message information. It does not share collected data with any third-party and data is retained 18 months or until the bot is leaving the guild/server.
+Analyst-bot collects text message information. It does not share collected data with any third-party and data is retained 90 days or until the bot is leaving the guild/server.
 **Data collection**
 Analyst-bot collects a Discord guild/server's history when asked to.
 This includes:
@@ -27,7 +27,7 @@ This does __not__ includes:
 Any data collected is only processed in order to produce a one-time report sent to the user immediately. No temporary data are retained.
 **Data storage and retain policy**
 Analyst-bot stores the collected data in files that are accessible by the software and its administrator only.
-Any collected data are retained maximum 18 months until deletion or when the bot is leaving a guild/server.
+Any collected data are retained maximum 90 days until deletion or when the bot is leaving a guild/server.
 **Data sharing**
 Analyst-bot does not share the data collected with any third-party.
 **Right to retract**

src/utils/utils.py

@@ -4,7 +4,7 @@ import os
 import logging
 import discord
 import math
-from datetime import datetime, timedelta
+from datetime import datetime, timedelta, timezone
 import re
 import time
 import dateutil.parser
@@ -116,7 +116,7 @@ def escape_text(text: str) -> str:
     return discord.utils.escape_markdown(discord.utils.escape_mentions(text))
 
 
-class FakeMessage:
+class FakeMessage(discord.abc.Snowflake):
     def __init__(self, id: int):
         self.id = id
 
@@ -258,13 +258,19 @@ def parse_iso_datetime(str_date: str) -> datetime:
 RELATIVE_REGEX = r"(yesterday|today|\d*hours?|\d+h(ours?)?|\d*days?|\d+d(ays?)?|\d*weeks?|\d+w(eeks?)?|\d*months?|\d+m(onths?)?|\d*years?|\d+y(ears?)?)"
 
 
+def utc_now() -> datetime:
+    return datetime.now(tz=timezone.utc)
+
+def utc_today() -> datetime:
+    today = utc_now().date()
+    return datetime(today.year, today.month, today.day, tzinfo=timezone.utc)
+
+
 def parse_relative_time(src: str) -> datetime:
-    today = datetime.utcnow().date()
-    today = datetime(today.year, today.month, today.day)
     if src == "today":
-        return today
+        return utc_today()
     elif src == "yesterday":
-        return today - relativedelta(days=1)
+        return utc_today() - relativedelta(days=1)
     else:
         m = re.match("(\d*)(\w+)", src)
         delta = None
@@ -280,7 +286,7 @@ def parse_relative_time(src: str) -> datetime:
             delta = relativedelta(months=value)
         elif unit == "y":
             delta = relativedelta(years=value)
-        return datetime.utcnow() - delta
+        return utc_now() - delta
 
 
 def parse_time(src: str) -> datetime:
@@ -344,7 +350,7 @@ def get_intro(
     """
     time_text = ""
     if start_datetime is not None:
-        stop_datetime = datetime.now() if stop_datetime is None else stop_datetime
+        stop_datetime = utc_now() if stop_datetime is None else stop_datetime
         time_text = f" (in {str_delta(stop_datetime - start_datetime)})"
     # Show all data (members, channels) when it's less than 5 units
     if len(members) == 0: