#!/usr/bin/env python3 """calculate_mirror_size.py Usage: calculate_mirror_size.py -H -R calculate_mirror_size.py -h Examples: calculate_mirror_size.py -H "https://mirrors.servercentral.com/voidlinux/" -R "rsync://repo-sync.voidlinux.org/voidlinux/" Options: -H HTTP URL of the mirror -R rsync URL of the mirror -h, --help show this help message and exit """ import subprocess from tempfile import TemporaryDirectory import requests from bs4 import BeautifulSoup from docopt import docopt from rich.console import Console from rich.text import Text def human_bytes(bites: int) -> str: B = float(bites) KiB = float(1024) MiB = float(KiB**2) GiB = float(KiB**3) TiB = float(KiB**4) match B: case B if B < KiB: return "{0} {1}".format(B, "bytes" if 0 == B > 1 else "byte") case B if KiB <= B < MiB: return "{0:.2f} KiB".format(B / KiB) case B if MiB <= B < GiB: return "{0:.2f} MiB".format(B / MiB) case B if GiB <= B < TiB: return "{0:.2f} GiB".format(B / GiB) case B if TiB <= B: return "{0:.2f} TiB".format(B / TiB) case _: return "" if __name__ == "__main__": args = docopt(__doc__) # type: ignore repo_shorthand = Text(args["-R"].split("//")[1]) print() console = Console() with console.status( "[bold magenta]Calculating mirror size...", spinner="aesthetic" ) as status: response = requests.get(args["-H"], timeout=60) soup = BeautifulSoup(response.text, "html.parser") mirror_dirs = [] for node in soup.find_all("a"): if not node.get("href").startswith(".") and node.get("href").endswith("/"): mirror_dirs.append(node.get("href")) console.log( f"Summing up the sizes of each directory in [bold blue]{repo_shorthand}[/bold blue]." ) with TemporaryDirectory() as tmp_dir: dir_sizes = [] max_dir_len = len(max(mirror_dirs, key=len)) for dir in mirror_dirs: rsync_cmd = f"rsync -a -n --stats {args['-R']}/{dir}/ | grep '^Total file size' | tr -d ','" output = subprocess.run(rsync_cmd, shell=True, capture_output=True) logstr = ( dir.rjust(max_dir_len) + " " + human_bytes(int(output.stdout.split()[3])) ) console.log(logstr) dir_sizes.append(int(output.stdout.split()[3])) console.print( f"\n[bold blue]{repo_shorthand}[/bold blue]: " + human_bytes(sum(dir_sizes)) )