research/chex/dump-strings

#!/usr/bin/env ruby

require "json"

def read_block(file)
        result = ""

        4.times do
                break if file.eof?
                c = file.readbyte
                result += sprintf("%c", c)
        end

        end_index = result.index("\0")
        if end_index
                result = result[0, end_index]
        end

        result
end

def valid_block?(block)
        for i in 0...block.length
                c = block[i, 1]

                if c !~ /[a-zA-Z0-9_= :\\\'\"\-\,\.\(\)\[\]\%\!\n\?\<\>\%\#\@\/\&]/
                        return false
                end
        end

        true
end

File.open(ARGV[0]) do |file|
        current_str = ""

        until file.eof?
                block = read_block(file)

                if valid_block?(block)
                        current_str += block

                        # End of string?
               
                        if block.length < 4
                                if current_str.length >= 4
                                        puts current_str.to_json
                                end
                                current_str = ""
                        end
                else
                        # Invalid string; start again
  
#                        if current_str.length >= 4
#                                puts "INVALID: #{current_str + block}"
#                        end

                        current_str = ""
                end
        end
end