Merge pull request 'wip-python -> Python rewriting + better logs + backup + update at startup' (#3) from wip-python into main

Reviewed-on: #3

Dockerfile

@@ -1,10 +1,30 @@
-FROM alpine:latest
+FROM python:3.11-slim
 
-RUN apk add --no-cache curl bash busybox tzdata
+# Install required utilities
+RUN apt-get update && apt-get install -y \
+    curl \
+    cron \
+    tzdata \
+    && rm -rf /var/lib/apt/lists/*
 
-COPY update-blocklist.sh /usr/local/bin/update-blocklist.sh
-COPY entrypoint.sh /entrypoint.sh
+# Install python dependencies
+RUN pip install --no-cache-dir requests
 
-RUN chmod +x /usr/local/bin/update-blocklist.sh /entrypoint.sh
+# Create crontabs directory (if needed)
+RUN mkdir -p /etc/crontabs
 
-ENTRYPOINT ["/entrypoint.sh"]
+# 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"]

README.md

@@ -23,6 +23,7 @@
 - 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.
 
 ## Environment Variables
 
@@ -73,6 +74,6 @@
 5. **Check logs to verify updates**
 
    ```bash
-   docker-compose logs -f
+   docker compose logs -f
    ```
 

docker-compose.yml

@@ -3,6 +3,7 @@ services:
   adguard-cidre:
     build: .
     container_name: adguard-cidre
+    restart: unless-stopped
     environment:
       - TZ=Europe/Paris # change to your timezone
       - BLOCK_COUNTRIES=cn,ru,ir # choose countries listed IP to block. Full lists here https://github.com/vulnebify/cidre/tree/main/output/cidr/ipv4

entrypoint.py

@@ -0,0 +1,67 @@
+#!/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()

entrypoint.sh

@@ -1,14 +0,0 @@
-#!/bin/sh
-set -e
-
-if [ -n "$TZ" ]; then
-  if [ -f "/usr/share/zoneinfo/$TZ" ]; then
-    cp "/usr/share/zoneinfo/$TZ" /etc/localtime
-    echo "$TZ" > /etc/timezone
-  fi
-fi
-
-CRON_EXPR="${BLOCKLIST_CRON:-"0 6 * * *"}"
-echo "$CRON_EXPR /usr/local/bin/update-blocklist.sh" > /etc/crontabs/root
-
-exec crond -f -c /etc/crontabs

update-blocklist.py

@@ -0,0 +1,148 @@
+#!/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()

update-blocklist.sh

@@ -1,75 +0,0 @@
-#!/bin/bash
-
-set -e
-
-ADGUARD_YAML="/adguard/AdGuardHome.yaml"
-TMP_YAML="/tmp/AdGuardHome.yaml"
-MANUAL_IPS_FILE="/adguard/manually_blocked_ips.conf"
-CIDR_BASE_URL="https://raw.githubusercontent.com/vulnebify/cidre/main/output/cidr/ipv4"
-COUNTRIES=${BLOCK_COUNTRIES:-""}
-DOCKER_API_URL=${DOCKER_API_URL:-"http://socket-proxy-adguard:2375"}
-CONTAINER_NAME=${ADGUARD_CONTAINER_NAME:-"adguard-home"}
-
-if [ -z "$COUNTRIES" ]; then
-  echo "No countries specified in BLOCK_COUNTRIES."
-  exit 1
-fi
-
-mkdir -p /tmp/cidr
-> /tmp/cidr/all.txt
-
-IFS=',' read -ra CODES <<< "$COUNTRIES"
-for CODE in "${CODES[@]}"; do
-  echo "Downloading CIDR list for $CODE..."
-  curl -sf "$CIDR_BASE_URL/${CODE,,}.cidr" -o "/tmp/cidr/${CODE}.cidr" || continue
-  cat "/tmp/cidr/${CODE}.cidr" >> /tmp/cidr/all.txt
-done
-
-if [ -f "$MANUAL_IPS_FILE" ]; then
-  echo "Validating and adding manually blocked IPs from $MANUAL_IPS_FILE..."
-  grep -E '^([0-9]{1,3}\.){3}[0-9]{1,3}(/[0-9]{1,2})?$' "$MANUAL_IPS_FILE" >> /tmp/cidr/all.txt
-fi
-
-# Format IPs as YAML list items
-sed 's/^/    - /' /tmp/cidr/all.txt > /tmp/cidr/ips_formatted.txt
-
-awk '
-BEGIN {
-  # Read formatted IPs into array
-  while ((getline line < "/tmp/cidr/ips_formatted.txt") > 0) {
-    ips[++count] = line
-  }
-  close("/tmp/cidr/ips_formatted.txt")
-  inside=0
-}
-
-/^  disallowed_clients:/ {
-  print
-  inside=1
-  next
-}
-
-/^  [^ ]/ && inside==1 {
-  # Insert all IPs here
-  for (i=1; i<=count; i++) print ips[i]
-  inside=0
-}
-
-{
-  if (!inside) print
-}
-
-END {
-  # If file ended while still inside disallowed_clients section
-  if (inside==1) {
-    for (i=1; i<=count; i++) print ips[i]
-  }
-}
-' "$ADGUARD_YAML" > "$TMP_YAML"
-
-mv "$TMP_YAML" "$ADGUARD_YAML"
-
-echo "Restarting $CONTAINER_NAME container..."
-curl -s -X POST "$DOCKER_API_URL/containers/$CONTAINER_NAME/restart" -o /dev/null
-
-echo "Done."