Merge pull request 'wip-python -> removed cron for python scheduling, and added pyyaml library.' (#4) from wip-python into main

Reviewed-on: #4

Dockerfile

@@ -1,30 +1,17 @@
 FROM python:3.11-slim
 
-# Install required utilities
-RUN apt-get update && apt-get install -y \
-    curl \
-    cron \
-    tzdata \
-    && rm -rf /var/lib/apt/lists/*
+RUN apt-get update && apt-get install -y --no-install-recommends curl tzdata && rm -rf /var/lib/apt/lists/*
 
-# Install python dependencies
-RUN pip install --no-cache-dir requests
+RUN pip install --no-cache-dir requests pyyaml schedule
 
-# Create crontabs directory (if needed)
-RUN mkdir -p /etc/crontabs
+ENV TZ=Europe/Paris
 
-# Copy scripts
-COPY update-blocklist.py /usr/local/bin/update-blocklist.py
-COPY entrypoint.py /usr/local/bin/entrypoint.py
-
-# Make scripts executable
-RUN chmod +x /usr/local/bin/update-blocklist.py /usr/local/bin/entrypoint.py
-
-# Set default timezone (can be overridden with TZ env var)
-ENV TZ=UTC
-
-# Configure timezone (tzdata)
 RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone
 
-# Set entrypoint
-ENTRYPOINT ["/usr/local/bin/entrypoint.py"]
+WORKDIR /app
+
+COPY blocklist_scheduler.py /app/blocklist_scheduler.py
+
+RUN chmod +x /app/blocklist_scheduler.py
+
+ENTRYPOINT ["python3", "/app/blocklist_scheduler.py"]

README.md

@@ -13,28 +13,37 @@
 
 - [Features](#features)
 - [Environment Variables](#environment-variables)
+- [Volumes](#volumes)
 - [File Structure](#file-structure)
-- [Installation and Usage](#nstallation-and-usage)
+- [Installation and Usage](#installation-and-usage)
 
 ## Features
 
-- Automatically downloads IP CIDR blocks for specified countries to block.
-- Supports additional manually blocked IPs from a configurable file.
-- Updates the disallowed_clients section in the AdGuard Home config.
-- Configurable update frequency via cron expression environment variable.
-- Automatically restarts the AdGuard Home container after updates via Docker socket proxy.
-- Backup `AdguardHome.yaml` at first startup, then create a second backup at each update.
+- Downloads CIDR lists by country from GitHub  
+- Adds manual IPs from a `manually_blocked_ips.conf` file  
+- Updates the `AdGuardHome.yaml` file by replacing the `disallowed_clients` list  
+- Creates a backup of the original config (`AdGuardHome.yaml.first-start.bak`) on first run  
+- Creates a backup before each update (`AdGuardHome.yaml.last-update.bak`)  
+- Restarts the AdGuard Home container via Docker API  
+- Built-in Python scheduler using the `schedule` library, configurable to run updates daily or weekly  
+
 
 ## Environment Variables
 
-| Variable            | Description                                                | Default                           |
-| ------------------- | ---------------------------------------------------------- | --------------------------------- |
-| `TZ`    | Your Time Zone    | (required) |
-| `BLOCK_COUNTRIES`   | Comma-separated country codes to block (e.g., `CN,RU,IR`)  | (required)                        |
-| `BLOCKLIST_CRON`    | Cron expression for update frequency (e.g., `0 6 * * *`) | `0 6 * * *` (at 6:00 everydays)          |
-| `DOCKER_API_URL`    | URL of Docker socket proxy to restart AdGuard container    | `http://socket-proxy-adguard:2375` |
-| `ADGUARD_CONTAINER_NAME`    | Name of your adguard container    | `adguardhome` |
 
+| Variable                 | Description                                                              | Example                     | Possible Values                             |
+|--------------------------|--------------------------------------------------------------------------|-----------------------------|---------------------------------------------|
+| `TZ`                      | Timezone of the container to correctly schedule updates                | `Europe/Paris`              | Any valid timezone (e.g., `UTC`, `America/New_York`, etc.) |
+| `BLOCK_COUNTRIES`         | List of country codes for CIDR lists, separated by commas                | `cn,ru,ir`                  | ISO 2-letter country codes                  |
+| `BLOCKLIST_CRON_TYPE`     | Scheduling type: `daily` or `weekly`                                    | `daily`                     | `daily`, `weekly`                           |
+| `BLOCKLIST_CRON_TIME`     | Time to run update in `HH:MM` 24-hour format                            | `06:00`                     | 24-hour time format                         |
+| `BLOCKLIST_CRON_DAY`      | Day of the week for weekly schedule (e.g., `mon`, `tue`, etc.)          | `mon`                       | `mon`, `tue`, `wed`, `thu`, `fri`, `sat`, `sun` |
+| `ADGUARD_CONTAINER_NAME`  | Name of the AdGuard Home container to restart                           | `adguardhome`               | Valid Docker container name                 |
+| `DOCKER_API_URL`          | Docker API URL (used to restart the container)                          | `http://socket-proxy-adguard:2375` | HTTP URL                                   |
+
+## Volumes
+
+- `/path/to/adguard/confdir` : configuration directory containing `AdGuardHome.yaml` from your adguard container, and optionally `manually_blocked_ips.conf`.
 
 ## File Structure
 
@@ -46,7 +55,7 @@
 
 ## Installation and Usage
 
-### With our docker image
+### With our provided docker image
 
 1. **Create `docker-compose.yml` in your `adguard-cidre` folder**
 
@@ -60,7 +69,10 @@
         environment:
         - TZ=Europe/Paris # change to your timezone
         - BLOCK_COUNTRIES=cn,ru # choose countries listed IP to block. Full lists here https://github.com/vulnebify/cidre/tree/main/output/cidr/ipv4
-        - BLOCKLIST_CRON=0 6 * * * # at 6:00 every days
+        - BLOCKLIST_CRON_TYPE=daily # daily or weekly
+      # if weekly, choose the day
+      # - BLOCKLIST_CRON_DAY=mon
+        - BLOCKLIST_CRON_TIME=06:00
         - DOCKER_API_URL=http://socket-proxy-adguard:2375 # docker socket proxy
         - ADGUARD_CONTAINER_NAME=adguardhome # adguard container name
         volumes:
@@ -128,7 +140,6 @@
 3. **Build and start the container**
 
     ```bash
-    docker compose build
     docker compose up -d
     ```
 4. **Check logs to verify updates**

blocklist_scheduler.py

@@ -0,0 +1,159 @@
+#!/usr/bin/env python3
+import os
+import sys
+import logging
+import requests
+import yaml
+import schedule
+import time
+from pathlib import Path
+
+logging.basicConfig(
+    level=logging.INFO,
+    format='[blocklist] %(levelname)s: %(message)s',
+    stream=sys.stdout,
+)
+
+ADGUARD_YAML = Path("/adguard/AdGuardHome.yaml")
+TMP_YAML = ADGUARD_YAML.parent / (ADGUARD_YAML.name + ".tmp")
+MANUAL_IPS_FILE = Path("/adguard/manually_blocked_ips.conf")
+CIDR_BASE_URL = "https://raw.githubusercontent.com/vulnebify/cidre/main/output/cidr/ipv4"
+
+FIRST_BACKUP = ADGUARD_YAML.parent / "AdGuardHome.yaml.first-start.bak"
+LAST_UPDATE_BACKUP = ADGUARD_YAML.parent / "AdGuardHome.yaml.last-update.bak"
+
+BLOCK_COUNTRIES = os.getenv("BLOCK_COUNTRIES", "")
+BLOCKLIST_CRON_TYPE = os.getenv("BLOCKLIST_CRON_TYPE", "daily").lower()  # daily or weekly
+BLOCKLIST_CRON_TIME = os.getenv("BLOCKLIST_CRON_TIME", "06:00")  # HH:MM format
+BLOCKLIST_CRON_DAY = os.getenv("BLOCKLIST_CRON_DAY", "mon").lower()  # only if weekly
+
+ADGUARD_CONTAINER_NAME = os.getenv("ADGUARD_CONTAINER_NAME", "adguardhome")
+DOCKER_API_URL = os.getenv("DOCKER_API_URL", "http://socket-proxy-adguard:2375")
+
+def backup_first_start():
+    if not FIRST_BACKUP.exists():
+        logging.info(f"Creating first start backup: {FIRST_BACKUP}")
+        FIRST_BACKUP.write_text(ADGUARD_YAML.read_text())
+    else:
+        logging.info("First start backup already exists, skipping.")
+
+def backup_last_update():
+    logging.info(f"Creating last update backup: {LAST_UPDATE_BACKUP}")
+    LAST_UPDATE_BACKUP.write_text(ADGUARD_YAML.read_text())
+
+def download_cidr_lists(countries):
+    combined_ips = []
+    for code in countries:
+        url = f"{CIDR_BASE_URL}/{code.lower()}.cidr"
+        logging.info(f"Downloading CIDR list for {code} from {url}")
+        try:
+            r = requests.get(url, timeout=30)
+            r.raise_for_status()
+            ips = r.text.strip().splitlines()
+            logging.info(f"Downloaded {len(ips)} CIDR entries for {code}")
+            combined_ips.extend(ips)
+        except Exception as e:
+            logging.warning(f"Failed to download {code}: {e}")
+    return combined_ips
+
+def read_manual_ips():
+    if MANUAL_IPS_FILE.exists():
+        logging.info(f"Reading manual IPs from {MANUAL_IPS_FILE}")
+        valid_ips = []
+        with MANUAL_IPS_FILE.open() as f:
+            for line in f:
+                line = line.strip()
+                if line and (line.count('.') == 3 or '/' in line):
+                    valid_ips.append(line)
+        logging.info(f"Added {len(valid_ips)} manual IP entries")
+        return valid_ips
+    else:
+        logging.info("Manual IPs file does not exist, skipping.")
+        return []
+
+def update_yaml_with_ips(ips):
+    if not ADGUARD_YAML.exists():
+        logging.error(f"{ADGUARD_YAML} does not exist. Cannot update.")
+        return False
+
+    data = None
+    with ADGUARD_YAML.open() as f:
+        data = yaml.safe_load(f)
+
+    if data is None:
+        logging.error(f"Failed to parse YAML file {ADGUARD_YAML}")
+        return False
+
+    data['dns']['disallowed_clients'] = ips
+
+    with TMP_YAML.open('w') as f:
+        yaml.safe_dump(data, f)
+
+    TMP_YAML.replace(ADGUARD_YAML)
+    logging.info(f"Updated {ADGUARD_YAML} with new disallowed clients list.")
+    return True
+
+def restart_adguard_container():
+    restart_url = f"{DOCKER_API_URL}/containers/{ADGUARD_CONTAINER_NAME}/restart"
+    logging.info(f"Restarting AdGuard container '{ADGUARD_CONTAINER_NAME}'...")
+    try:
+        resp = requests.post(restart_url, timeout=10)
+        if resp.status_code == 204:
+            logging.info("AdGuard container restarted successfully.")
+        else:
+            logging.error(f"Failed to restart container: {resp.status_code} {resp.text}")
+    except Exception as e:
+        logging.error(f"Error restarting container: {e}")
+
+def update_blocklist():
+    if not BLOCK_COUNTRIES:
+        logging.error("No countries specified in BLOCK_COUNTRIES environment variable. Skipping update.")
+        return
+
+    countries_list = [c.strip() for c in BLOCK_COUNTRIES.split(",") if c.strip()]
+    cidr_ips = download_cidr_lists(countries_list)
+    manual_ips = read_manual_ips()
+    combined_ips = cidr_ips + manual_ips
+
+    backup_last_update()
+
+    success = update_yaml_with_ips(combined_ips)
+    if success:
+        restart_adguard_container()
+
+def schedule_job():
+    try:
+        hour, minute = [int(x) for x in BLOCKLIST_CRON_TIME.split(":")]
+    except Exception:
+        logging.error(f"Invalid BLOCKLIST_CRON_TIME '{BLOCKLIST_CRON_TIME}', must be HH:MM. Defaulting to 06:00.")
+        hour, minute = 6, 0
+
+    if BLOCKLIST_CRON_TYPE == "daily":
+        schedule.every().day.at(f"{hour:02d}:{minute:02d}").do(update_blocklist)
+        logging.info(f"Scheduled daily update at {hour:02d}:{minute:02d}")
+    elif BLOCKLIST_CRON_TYPE == "weekly":
+        valid_days = ["mon","tue","wed","thu","fri","sat","sun"]
+        day = BLOCKLIST_CRON_DAY[:3]
+        if day not in valid_days:
+            logging.error(f"Invalid BLOCKLIST_CRON_DAY '{BLOCKLIST_CRON_DAY}', must be one of {valid_days}. Defaulting to Monday.")
+            day = "mon"
+        getattr(schedule.every(), day).at(f"{hour:02d}:{minute:02d}").do(update_blocklist)
+        logging.info(f"Scheduled weekly update on {day.capitalize()} at {hour:02d}:{minute:02d}")
+    else:
+        logging.error(f"Invalid BLOCKLIST_CRON_TYPE '{BLOCKLIST_CRON_TYPE}', must be 'daily' or 'weekly'. Defaulting to daily.")
+        schedule.every().day.at(f"{hour:02d}:{minute:02d}").do(update_blocklist)
+        logging.info(f"Scheduled daily update at {hour:02d}:{minute:02d}")
+
+def main():
+    logging.info("Starting blocklist scheduler...")
+
+    backup_first_start()
+
+    update_blocklist()
+    schedule_job()
+    while True:
+        schedule.run_pending()
+        time.sleep(10)
+
+if __name__ == "__main__":
+    main()

docker-compose.yml

@@ -7,7 +7,10 @@ services:
     environment:
       - TZ=Europe/Paris # change to your timezone
       - BLOCK_COUNTRIES=cn,ru # choose countries listed IP to block. Full lists here https://github.com/vulnebify/cidre/tree/main/output/cidr/ipv4
-      - BLOCKLIST_CRON=0 6 * * * # at 6:00 every days
+      - BLOCKLIST_CRON_TYPE=daily # daily or weekly
+      # if weekly, choose the day
+      # - BLOCKLIST_CRON_DAY=mon
+      - BLOCKLIST_CRON_TIME=06:00
       - DOCKER_API_URL=http://socket-proxy-adguard:2375 # docker socket proxy
       - ADGUARD_CONTAINER_NAME=adguardhome # adguard container name
     volumes:

entrypoint.py

@@ -1,67 +0,0 @@
-#!/usr/bin/env python3
-import os
-import sys
-import subprocess
-import logging
-from pathlib import Path
-
-logging.basicConfig(
-    level=logging.INFO,
-    format='[entrypoint] %(message)s',
-    stream=sys.stdout
-)
-
-ADGUARD_YAML = Path("/adguard/AdGuardHome.yaml")
-FIRST_BACKUP = Path("/adguard/AdGuardHome.yaml.first-start.bak")
-
-def setup_cron():
-    cron_expr = os.getenv("BLOCKLIST_CRON", "0 6 * * *")
-    cron_line = f"{cron_expr} root /usr/local/bin/update-blocklist.py\n"
-    cron_dir = "/etc/crontabs"
-    cron_file = f"{cron_dir}/root"
-
-    logging.info(f"Setting cron job: {cron_line.strip()}")
-
-    # Ensure cron directory exists
-    os.makedirs(cron_dir, exist_ok=True)
-
-    with open(cron_file, "w") as f:
-        f.write(cron_line)
-
-def backup_first_start():
-    if not FIRST_BACKUP.exists():
-        logging.info("Creating first start backup...")
-        FIRST_BACKUP.write_text(ADGUARD_YAML.read_text())
-    else:
-        logging.info("First start backup already exists.")
-
-def run_initial_update():
-    logging.info("Running initial update-blocklist.py script...")
-    try:
-        subprocess.run(
-            ["/usr/local/bin/update-blocklist.py"],
-            check=True,
-            stdout=sys.stdout,
-            stderr=sys.stderr,
-        )
-    except subprocess.CalledProcessError as e:
-        logging.error(f"Initial update script failed: {e}")
-        sys.exit(1)
-
-def start_cron_foreground():
-    logging.info("Starting cron in foreground...")
-    os.execvp("cron", ["cron", "-f"])
-
-def main():
-    # Check AdGuardHome.yaml exists
-    if not ADGUARD_YAML.exists():
-        logging.error(f"{ADGUARD_YAML} not found. Exiting.")
-        sys.exit(1)
-
-    backup_first_start()
-    run_initial_update()
-    setup_cron()
-    start_cron_foreground()
-
-if __name__ == "__main__":
-    main()

update-blocklist.py

@@ -1,148 +0,0 @@
-#!/usr/bin/env python3
-import os
-import sys
-import logging
-import requests
-from pathlib import Path
-
-logging.basicConfig(
-    level=logging.INFO,
-    format='[update-blocklist] %(levelname)s: %(message)s',
-    stream=sys.stdout,
-)
-
-ADGUARD_YAML = Path("/adguard/AdGuardHome.yaml")
-FIRST_BACKUP = Path("/adguard/AdGuardHome.yaml.first-start.bak")
-LAST_CRON_BACKUP = Path("/adguard/AdGuardHome.yaml.last-cron.bak")
-TMP_YAML = ADGUARD_YAML.parent / (ADGUARD_YAML.name + ".tmp")
-MANUAL_IPS_FILE = Path("/adguard/manually_blocked_ips.conf")
-CIDR_BASE_URL = "https://raw.githubusercontent.com/vulnebify/cidre/main/output/cidr/ipv4"
-COUNTRIES = os.getenv("BLOCK_COUNTRIES", "")
-
-def backup_files():
-    if not FIRST_BACKUP.exists():
-        logging.info(f"Creating first-start backup: {FIRST_BACKUP}")
-        FIRST_BACKUP.write_text(ADGUARD_YAML.read_text())
-    else:
-        logging.info("First-start backup already exists, skipping.")
-
-    logging.info(f"Creating last-cron backup: {LAST_CRON_BACKUP}")
-    LAST_CRON_BACKUP.write_text(ADGUARD_YAML.read_text())
-
-def download_cidr_lists(countries):
-    combined_ips = []
-    for code in countries:
-        url = f"{CIDR_BASE_URL}/{code.lower()}.cidr"
-        logging.info(f"Downloading CIDR list for {code} from {url}")
-        try:
-            r = requests.get(url, timeout=30)
-            r.raise_for_status()
-            ips = r.text.strip().splitlines()
-            logging.info(f"Downloaded {len(ips)} CIDR entries for {code}")
-            combined_ips.extend(ips)
-        except Exception as e:
-            logging.warning(f"Failed to download {code}: {e}")
-    return combined_ips
-
-def read_manual_ips():
-    if MANUAL_IPS_FILE.exists():
-        logging.info(f"Reading manual IPs from {MANUAL_IPS_FILE}")
-        valid_ips = []
-        with MANUAL_IPS_FILE.open() as f:
-            for line in f:
-                line = line.strip()
-                # Simple check for IPv4 or IPv4 CIDR format
-                if line and line.count('.') == 3:
-                    valid_ips.append(line)
-        logging.info(f"Added {len(valid_ips)} manual IP entries")
-        return valid_ips
-    else:
-        logging.info("Manual IPs file does not exist, skipping.")
-        return []
-
-def update_yaml_with_ips(ips):
-    output_lines = []
-    inside_disallowed = False
-    disallowed_indent = ""
-
-    with ADGUARD_YAML.open() as f:
-        lines = f.readlines()
-
-    for line in lines:
-        stripped = line.lstrip()
-        indent = line[:len(line) - len(stripped)]
-
-        if stripped.startswith("disallowed_clients:"):
-            # Capture the indentation of the disallowed_clients key
-            disallowed_indent = indent
-
-            # Replace entire line with just 'disallowed_clients:' (remove any [])
-            output_lines.append(f"{disallowed_indent}disallowed_clients:")
-
-            # Add all IPs indented 2 spaces more than disallowed_clients
-            formatted_ips = [f"{disallowed_indent}  - {ip}" for ip in ips]
-            output_lines.extend(formatted_ips)
-
-            inside_disallowed = True
-            continue
-
-        if inside_disallowed:
-            # We skip all old lines inside disallowed_clients block.
-            # The block ends when we find a line with indentation
-            # less than or equal to disallowed_indent but not the key line itself.
-            # To detect end of block, compare indent length:
-            if len(indent) <= len(disallowed_indent) and stripped != "":
-                inside_disallowed = False
-                output_lines.append(line.rstrip("\n"))
-            else:
-                # skip this line (old disallowed_clients content)
-                continue
-        else:
-            output_lines.append(line.rstrip("\n"))
-
-    # Write temp file in same directory to avoid cross-device rename errors
-    with TMP_YAML.open("w") as f:
-        f.write("\n".join(output_lines) + "\n")
-
-    TMP_YAML.replace(ADGUARD_YAML)
-    logging.info(f"Updated {ADGUARD_YAML} with new disallowed clients list.")
-
-
-def restart_adguard_container():
-    docker_api_url = os.getenv("DOCKER_API_URL", "http://socket-proxy-adguard:2375")
-    container_name = os.getenv("ADGUARD_CONTAINER_NAME", "adguardhome")
-    restart_url = f"{docker_api_url}/containers/{container_name}/restart"
-
-    logging.info(f"Restarting AdGuard container '{container_name}'...")
-    try:
-        resp = requests.post(restart_url, timeout=10)
-        if resp.status_code == 204:
-            logging.info("AdGuard container restarted successfully.")
-        else:
-            logging.error(f"Failed to restart container: {resp.status_code} {resp.text}")
-    except Exception as e:
-        logging.error(f"Error restarting container: {e}")
-
-def main():
-    if not ADGUARD_YAML.exists():
-        logging.error(f"{ADGUARD_YAML} not found, exiting.")
-        sys.exit(1)
-
-    if not COUNTRIES:
-        logging.error("No countries specified in BLOCK_COUNTRIES environment variable, exiting.")
-        sys.exit(1)
-
-    backup_files()
-
-    countries_list = [c.strip() for c in COUNTRIES.split(",") if c.strip()]
-    cidr_ips = download_cidr_lists(countries_list)
-    manual_ips = read_manual_ips()
-
-    combined_ips = cidr_ips + manual_ips
-
-    update_yaml_with_ips(combined_ips)
-
-    restart_adguard_container()
-
-if __name__ == "__main__":
-    main()