Download resume with NET::HTTP

Ruby
1

I’m wondering is there any way to do interrupt file download and then
resume it from the same spot?
The way I download file:

    Net::HTTP.start("server.com") { |http|
       response = http.request_get('/some/path') {|response|
           file_length =  response['Content-Length']
           File.open('video.avi', 'w') {|f|
               response.read_body do |str|   # read body now
                  puts  str.length
                  f.write str
               end
           }
        }
    }

I’m not using open-url as I found that open() method first downloads
whole file and then starts to write it. Not for me…

Cheers Alex

2

On Fri, May 9, 2008 at 8:18 AM, Alex Dj [email protected] wrote:

    Net::HTTP.start("server.com") { |http|
       response = http.request_get('/some/path') {|response|

create your request by hand:

req = Net::HTTP::Get.new(path)
req.range = …

In HTTP 1.1 range has the form “[lower]-[upper]”, so “[lower]-” is
valid. I’m not sure what range= takes, you get to try it :slight_smile:

Marcelo

3

Thanks Marcelo,

In HTTP 1.1 range has the form “[lower]-[upper]”, so “[lower]-” is
valid. I’m not sure what range= takes, you get to try it :slight_smile:

I found that its “Range: bytes=10000-”, something like this. I wonder if
I can use this header to download the same file from different mirrors
i.e. have several
req = Net::HTTP::Get.new(path)
but with different ranges and write them to the same file with random
access. Would it be a right technique for this idea?

Thanks Alex