mirror of
https://github.com/StrawberryMaster/wayback-machine-downloader.git
synced 2025-12-17 17:56:44 +00:00
92 lines
3.2 KiB
Ruby
Executable File
92 lines
3.2 KiB
Ruby
Executable File
#!/usr/bin/env ruby
|
|
|
|
require_relative '../lib/wayback_machine_downloader'
|
|
require 'optparse'
|
|
require 'pp'
|
|
|
|
options = {}
|
|
option_parser = OptionParser.new do |opts|
|
|
opts.banner = "Usage: wayback_machine_downloader http://example.com"
|
|
|
|
opts.separator ""
|
|
opts.separator "Download an entire website from the Wayback Machine."
|
|
|
|
opts.separator ""
|
|
opts.separator "Optional options:"
|
|
|
|
opts.on("-d", "--directory PATH", String, "Directory to save the downloaded files into", "Default is ./websites/ plus the domain name") do |t|
|
|
options[:directory] = t
|
|
end
|
|
|
|
opts.on("-s", "--all-timestamps", "Download all snapshots/timestamps for a given website") do |t|
|
|
options[:all_timestamps] = true
|
|
end
|
|
|
|
opts.on("-f", "--from TIMESTAMP", Integer, "Only files on or after timestamp supplied (ie. 20060716231334)") do |t|
|
|
options[:from_timestamp] = t
|
|
end
|
|
|
|
opts.on("-t", "--to TIMESTAMP", Integer, "Only files on or before timestamp supplied (ie. 20100916231334)") do |t|
|
|
options[:to_timestamp] = t
|
|
end
|
|
|
|
opts.on("-e", "--exact-url", "Download only the url provied and not the full site") do |t|
|
|
options[:exact_url] = t
|
|
end
|
|
|
|
opts.on("-o", "--only ONLY_FILTER", String, "Restrict downloading to urls that match this filter", "(use // notation for the filter to be treated as a regex)") do |t|
|
|
options[:only_filter] = t
|
|
end
|
|
|
|
opts.on("-x", "--exclude EXCLUDE_FILTER", String, "Skip downloading of urls that match this filter", "(use // notation for the filter to be treated as a regex)") do |t|
|
|
options[:exclude_filter] = t
|
|
end
|
|
|
|
opts.on("-a", "--all", "Expand downloading to error files (40x and 50x) and redirections (30x)") do |t|
|
|
options[:all] = true
|
|
end
|
|
|
|
opts.on("-c", "--concurrency NUMBER", Integer, "Number of multiple files to download at a time", "Default is one file at a time (ie. 20)") do |t|
|
|
options[:threads_count] = t
|
|
end
|
|
|
|
opts.on("-p", "--maximum-snapshot NUMBER", Integer, "Maximum snapshot pages to consider (Default is 100)", "Count an average of 150,000 snapshots per page") do |t|
|
|
options[:maximum_pages] = t
|
|
end
|
|
|
|
opts.on("-l", "--list", "Only list file urls in a JSON format with the archived timestamps, won't download anything") do |t|
|
|
options[:list] = true
|
|
end
|
|
|
|
opts.on("-r", "--rewritten", "Downloads the rewritten Wayback Machine files instead of the original files") do |t|
|
|
options[:rewritten] = true
|
|
end
|
|
|
|
opts.on("--reset", "Delete state files (.cdx.json, .downloaded.txt) and restart the download from scratch") do |t|
|
|
options[:reset] = true
|
|
end
|
|
|
|
opts.on("--keep", "Keep state files (.cdx.json, .downloaded.txt) after a successful download") do |t|
|
|
options[:keep] = true
|
|
end
|
|
|
|
opts.on("-v", "--version", "Display version") do |t|
|
|
options[:version] = t
|
|
end
|
|
end.parse!
|
|
|
|
if (base_url = ARGV[-1])
|
|
options[:base_url] = base_url
|
|
wayback_machine_downloader = WaybackMachineDownloader.new options
|
|
if options[:list]
|
|
wayback_machine_downloader.list_files
|
|
else
|
|
wayback_machine_downloader.download_files
|
|
end
|
|
elsif options[:version]
|
|
puts WaybackMachineDownloader::VERSION
|
|
else
|
|
puts "You need to specify a website to backup. (e.g., http://example.com)"
|
|
puts "Run `wayback_machine_downloader --help` for more help."
|
|
end
|