environment/lib/archive.rb
2025-08-05 10:48:11 +02:00

103 lines
2.8 KiB
Ruby

require 'net/http'
require 'uri'
require 'open3'
require 'fileutils'
require 'open-uri'
require 'shellwords'
module Archive
MAGIC_NUMBERS = {
"\x1F\x8B" => 'gzip',
"\x50\x4B\x03\x04" => 'zip',
"\x28\xB5\x2F\xFD" => 'zstd',
"\xFD\x37\x7A\x58" => 'xz'
}
def self.fetch_and_extract(url, cache_path, destination_dir)
uri = URI.parse(url)
file_name = File.basename(uri.path)
archive_path = File.join(destination_dir, file_name)
format = detect_format_from_url(url)
puts "format #{format}"
download_file(uri, archive_path)
unless format
puts "case 1"
format = detect_format_from_headers(uri)
end
unless format
puts "case 2"
format = detect_format_from_magic(archive_path)
end
raise "Could not determine archive format" unless format
extract_archive(archive_path, format)
end
private
def self.detect_format_from_url(url)
case File.extname(url)
when '.gz', '.tgz' then 'gzip'
when '.zip' then 'zip'
when '.zst' then 'zstd'
when '.xz' then 'xz'
else nil
end
end
def self.detect_format_from_headers(uri)
response = Net::HTTP.get_response(uri)
content_type = response['content-type']
case content_type
when /gzip/ then 'gzip'
when /zip/ then 'zip'
when /zstd/ then 'zstd'
when /xz/ then 'xz'
else nil
end
end
def self.detect_format_from_magic(file_path)
File.open(file_path, 'rb') do |f|
bytes = f.read(4)
MAGIC_NUMBERS.each do |magic, format|
return format if bytes.start_with?(magic)
end
end
nil
end
def self.download_file(uri, output_path)
URI.open(uri) do |input|
File.open(output_path, 'wb') do |output|
IO.copy_stream(input, output)
end
end
end
def self.extract_archive(file_path, format)
case format
when 'gzip'
system("tar -xzf #{Shellwords.escape(file_path)}")
when 'zip'
system("unzip #{Shellwords.escape(file_path)}")
when 'zstd'
system("unzstd #{Shellwords.escape(file_path)}")
when 'xz'
# if file_path.end_with?('.tar.xz')
if `file #{Shellwords.escape(file_path)}`.include?('tar archive')
system("tar -xJf #{Shellwords.escape(file_path)}")
else
system("xz -dk #{Shellwords.escape(file_path)}")
end
else
raise "Unsupported archive format: #{format}"
end
end
end
# Example usage:
# fetch_and_extract("https://codeberg.org/forgejo/forgejo/releases/download/v12.0.1/forgejo-12.0.1-linux-amd64.xz")