Update gdpr.py

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 18 months 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 30 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
-python-dotenv==0.15.0
-python-dateutil==2.8.1
-git+git://github.com/Klemek/miniscord.git
-numpy
-matplotlib
+discord.py>=1.7.0
+python-dotenv>=0.15.0
+python-dateutil>=2.8.1
+matplotlib>=3.4.2
+cryptography>=2.8
+git+git://github.com/Klemek/miniscord.git

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
@@ -16,18 +15,6 @@ from utils import (
     mention,
 )
 
-CB_color_cycle = [
-    "#e41a1c",
-    "#984ea3",
-    "#377eb8",
-    "#4daf4a",
-    "#dede00",
-    "#ff7f00",
-    "#a65628",
-    "#f781bf",
-    "#999999",
-]
-
 
 class Frequency:
     def __init__(self):
@@ -71,12 +58,7 @@ class Frequency:
                 self.hours[i][0] * 7 / n_hours
             ]
             ax.plot(
-                times,
-                hours,
-                label=calendar.day_name[i],
-                linestyle="--",
-                linewidth=0.8,
-                c=CB_color_cycle[i],
+                times, hours, label=calendar.day_name[i], linestyle="--", linewidth=0.8
             )
 
         hours = [day[hour] / n_hours for hour in range(24)] + [day[0] / n_hours]

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
-MIN_MODIFICATION_TIME = 5 * 60
-# ~1 year, remove log file
-MAX_MODIFICATION_TIME = 365 * 24 * 60 * 60
+load_dotenv()
 
+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))
+
+# 30 days, remove log file
+MAX_MODIFICATION_TIME = int(os.getenv("MAX_MODIFICATION_TIME", 30 * 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
-            await code_message(progress, "Reading saved history (1/4)...")
+            encrypted_data = None
+            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",
+    "1.17.0",
     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.mention_count[message.author] += len(message.mentions)
+            pres.channel_total[channel.id] += 1
+            pres.mention_count[message.author] += len(message.mentions)
         if len(raw_members) > 0:
             for mention in message.mentions:
                 if mention in raw_members:

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

@@ -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 30 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**