getparty

HTTP Download Manager with multi-parts

Installation

Homebrew

$ brew install getparty

AUR

$ paru -S getparty

From source

$ git clone --depth 1 https://github.com/vbauerster/getparty.git
$ cd getparty/cmd/getparty && go build

Usage

Usage:
  getparty [OPTIONS] [<url>]

Application Options:
  -p, --parts=n                                    number of parts (default: 1)
  -r, --max-retry=n                                max retries per each part, 0 for infinite (default: 10)
      --max-follow=n                               max redirections to follow, 0 for infinite (default: 10)
  -t, --timeout=sec                                timeout to read 64KiB; on exceed chunk size is reduced (default: 5)
  -l, --speed-limit=[1|2|3|4|5]                    speed limit (default: 0 = no limit; 5 = max limit)
  -s, --session=FILE                               json file of previous incomplete download session
  -U, --user-agent=[chrome|firefox|safari|edge]    User-Agent header (default: getparty/ver)
      --username=                                  basic http auth username
      --password=                                  basic http auth password
  -H, --header=key:value                           http header, can be specified more than once
  -x, --proxy=<[scheme://]host[:port]>             proxy to use, if scheme is empty http is assumed
  -q, --quiet                                      quiet mode, progress bars are disabled
  -d, --debug                                      debug mode, output debug info to stderr
  -v, --version                                    show version

Https Options:
  -c, --certs-file=certs.crt                       root certificates to use when verifying server certificates
      --no-check-cert                              don't verify the server's certificate chain and host name

Output Options:
  -o, --output.name=FILE                           output file name
  -f, --output.overwrite                           overwrite existing file silently
      --output.use-path                            resolve name from url path first (default: Content-Disposition header)

Best-mirror Options:
  -m, --mirror.list=FILE|-                         mirror list input
      --mirror.method=[HEAD|GET]                   http method to use (default: HEAD)
      --mirror.doget=KiB                           get first n KiB, implies method=GET
  -w, --mirror.workers=n                           max concurrent workers (default: number of logical CPUs)
      --mirror.top=n                               list top n mirrors without downloading unless n=1 (default: 1)
      --mirror.pass=n                              query each mirror n times to get average result (default: 1)
      --mirror.timeout=sec                         query timeout (default: 5)

Expose progress (host:port/debug/vars):
      --expose.host=host                           host (optional)
      --expose.port=port                           port (required)

Help Options:
  -h, --help                                       Show this help message

Arguments:
  <url>:                                           http location

Best mirror example:

read mirrors.txt from a file and proceed to download from the best mirror:

getparty -m mirrors.txt

read mirrors.txt from a stdin and proceed to download from the best mirror:

cat mirrors.txt | getparty -m -

just list top 3 mirrors without any download:

getparty -m mirrors.txt --mirror.top 3

License

BSD 3-Clause