Search code examples
rubycurb

Get filename on server while using Ruby gem Curb

Is there a way to get the filename of the file being downloaded (without having to parse the url provided)? I am hoping to find something like: 

c = Curl::Easy.new("http://google.com/robots.txt")
c.perform
File.open( c.file_name, "w") { |file| file.write c.body_str }
Solution

  • Unfortunately, there's nothing in the Curb documentation regarding polling the filename. I don't know whether you have a particular aversion to parsing, but it's a simple process if using the URI module:

    require 'uri'
url = 'http://google.com/robots.txt'
uri = URI.parse(url)
puts File.basename(uri.path)
#=> "robots.txt"

    UPDATE:

    In the comments to this question, the OP suggests using split() to split the URL by slashes (/). While this may work in the majority of situations, it isn't a catch-all solution. For instance, versioned files won't be parsed correctly:

    url = 'http://google.com/robots.txt?1234567890'
puts url.split('/').last
#=> "robots.txt?1234567890"

    In comparison, using URI.parse() guarantees the filename – and only the filename – is returned:

    require 'uri'
url = 'http://google.com/robots.txt?1234567890'
uri = URI.parse(url)
puts File.basename(uri.path)
#=> "robots.txt"

    In sum, for optimal coherence and integrity, it's wise to use the URI library to parse universal resources – it's what it was created for, after all.