Check asset encoding for valid UTF-8

assets.rake

namespace :assets do

  desc "Check that all assets have valid encoding"
  task  :check => :environment do

    paths = ["app/assets", "lib/assets", "vendor/assets"]
    extensions = ["js", "coffee", "css", "scss"]

    paths.each do |path|
      dir_path = Rails.root + path

      if File.exists?(dir_path)
        dir_files = File.join(dir_path, "**")

        Dir.glob(dir_files + "/**.{#{extensions.join(',')}}").each do |file|

            # make sure we're not trying to process a directory
            unless File.directory?(file)
              # read the file and check its encoding
              data = File.read(file)
              unless data.valid_encoding?
                puts "Invalid encoding: #{ file }"
              end
            end

        end # end Dir.glob

      end #end File.exists
    end # end paths.each

  end

end

Thanks for the excellent script! Here is a slightly modified version that checks coffee and scss too.

# check that all assets have the correct encoding
# https://gist.github.com/1301199
namespace :assets do

  desc "Check that all assets have valid encoding"
  task  :check => :environment do

    paths = ["app/assets", "lib/assets", "vendor/assets"]
    extensions = ["js", "coffee", "css", "scss"]

    paths.each do |path|
      dir_path = Rails.root + path

      if File.exists?(dir_path)
        dir_files = File.join(dir_path, "**")

        Dir.glob(dir_files + "/**.{#{extensions.join(',')}}").each do |file|

            # make sure we're not trying to process a directory
            unless File.directory?(file)
              # read the file and check its encoding
              data = File.read(file)
              unless data.valid_encoding?
                puts "Invalid encoding: #{ file }"
              else
                puts "Valid: #{ file }"
              end
            end

        end # end Dir.glob

      end #end File.exists
    end # end paths.each

  end

end

Thank @shedd, I'll incorporate that in my version though! Cheers.

thanks a lot

Thanks for sharing!

Does anyone know how I could update this task so that it also checks included gems assets?

Thanks!

For the record, after using this rake task to track the files that were not UTF-8, I was able to convert them using recode, as recommended here http://stackoverflow.com/questions/64860/best-way-to-convert-text-files-between-character-sets