mirror of
https://github.com/StrawberryMaster/wayback-machine-downloader.git
synced 2025-12-17 09:46:05 +00:00
gzip support
This commit is contained in:
parent
576298dca8
commit
27dd619aa4
@ -9,6 +9,8 @@ require 'json'
|
||||
require 'time'
|
||||
require 'concurrent-ruby'
|
||||
require 'logger'
|
||||
require 'zlib'
|
||||
require 'stringio'
|
||||
require_relative 'wayback_machine_downloader/tidy_bytes'
|
||||
require_relative 'wayback_machine_downloader/to_regex'
|
||||
require_relative 'wayback_machine_downloader/archive_api'
|
||||
@ -485,16 +487,26 @@ class WaybackMachineDownloader
|
||||
request = Net::HTTP::Get.new(URI(wayback_url))
|
||||
request["Connection"] = "keep-alive"
|
||||
request["User-Agent"] = "WaybackMachineDownloader/#{VERSION}"
|
||||
request["Accept-Encoding"] = "gzip, deflate"
|
||||
|
||||
response = connection.request(request)
|
||||
|
||||
case response
|
||||
when Net::HTTPSuccess
|
||||
File.open(file_path, "wb") do |file|
|
||||
if block_given?
|
||||
yield(response, file)
|
||||
body = response.body
|
||||
if response['content-encoding'] == 'gzip' && body && !body.empty?
|
||||
begin
|
||||
gz = Zlib::GzipReader.new(StringIO.new(body))
|
||||
decompressed_body = gz.read
|
||||
gz.close
|
||||
file.write(decompressed_body)
|
||||
rescue Zlib::GzipFile::Error => e
|
||||
@logger.warn("Failure decompressing gzip file #{file_url}: #{e.message}")
|
||||
file.write(body)
|
||||
end
|
||||
else
|
||||
file.write(response.body)
|
||||
file.write(body) if body
|
||||
end
|
||||
end
|
||||
when Net::HTTPRedirection
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user