#!/usr/bin/env python3

# https://eax.me/ full offline copy
# Aleksander Alekseev, 2024-2025
# For more details see:
# - https://eax.me/blog-offline-copy/

from datetime import datetime as dt
import subprocess
import glob
import sys
import os
import re

dirname = dt.today().strftime('%Y-%m-%d')
os.makedirs(dirname+"/files/github", exist_ok=True)

code = subprocess.call("wget -r -l 0 -k -nH https://eax.me/ -P " +
  dirname +" 2>&1 | tee wget.log", shell=True)
if code != 0:
  print("Wget returned {}, check wget.log for 404s," +
        " 301s, etc".format(code))
  sys.exit(1)
os.remove("wget.log")

for page_path in glob.glob(dirname + '/**/index.html', recursive=True):
  page_data = ""
  with open(page_path) as page_f:
    page_data = page_f.read()

    re_str = """<a[ ]+href=['"]https?://github.com/afiskon/([^'"/]+)"""
    for m in re.finditer(re_str, page_data):
      repo_name = m.group(1)
      print("Found repo: {} in {}".format(repo_name, page_path))
      repo_backup_dir = dirname+"/files/github/"+repo_name
      if not os.path.isdir(repo_backup_dir):
        print("  New one, cloning...")
        code = subprocess.call(
          "git clone --depth 1 https://github.com/afiskon/" +
          repo_name + ".git " + repo_backup_dir, shell=True)
        if code != 0:
          print("Failed to clone {}, git returned {}".format(
                 repo_name, code))
          sys.exit(1)

    # Some articles may reference files in code fragments, e.g.:
    # - https://eax.me/fbg6-klipper-screen/
    # - https://eax.me/flying-bear-s1-camera/
    # Make sure such files are copied too.
    re_str = """https?://eax.me(/files/[^'"<>]+)"""
    for m in re.finditer(re_str, page_data):
      file_name = m.group(1)
      file_backup_name = dirname + file_name
      if not os.path.exists(file_backup_name):
        print("  Downloading {}".format(file_name))
        code = subprocess.call(
          "wget 'https://eax.me" + file_name + "' -O " + file_backup_name,
          shell=True)
        if code != 0:
          print("Failed to download {}, wget returned {}".format(
                 file_name, code))
          sys.exit(1)

  page_data = re.sub(
    """(<a[ ]+href=["'])https?://github.com/afiskon/([^'"/]+)""" +
      """(/tree/master|/tree/main|/blob/master|/blob/main)?""" +
      """([^'"]*["'])""", 
    r"\1../files/github/\2\4",
    page_data)

  # remove LiveInternet JavaScript code
  page_data = re.sub(
    """(?s)<script type="text/javascript">.*?</script>""",
    "", page_data)

  print("Updating {}".format(page_path))
  with open(page_path, 'w') as page_f:
    page_f.write(page_data)

subprocess.check_call(
  "rm -rf "+dirname+"/files/github/*/.git",
  shell=True)

subprocess.check_call(
  "tar -cvzf "+dirname+".tgz "+dirname,
  shell=True)

subprocess.check_call(
  "rm -rf "+dirname,
  shell=True)

print("DONE!")