module Hippo

    module Strings

        # list of characters that should not be used
        BAD_RAND_CHARS  = %w{ B 8 o O 0 i l I 1 }

        # characters that should be randomly chosen from
        RAND_CHARS = ( ('A'..'Z').to_a + ('a'..'z').to_a + ('0'..'9').to_a ) - BAD_RAND_CHARS

        # Generates a random string without using commonly confused numbers and letters.
        # It's intended for use in situations where the code will have to be
        # read and re-entered later.
        #
        # The characters in BAD_RAND_CHARS are avoided since they can be confused with one another:
        # @return [String] a random string <length> characters long.
        def self.random( length = 12 )
            1.upto(length).each_with_object(''){|i,s| s << RAND_CHARS[ rand(RAND_CHARS.length) ] }
        end

        # Attempts to shorten a string while keeping it somewhat
        # like it's original self.
        # Useful for suggesting a code for a given entity name.
        # @param string [String] the string to shorten.
        # @param length [Integer] how long the returned code should be
        # @param padding [String] What to pad the code with if it is shorter than length.  False indicates no padding should be performed
        # @return [String] a shortened version of the name with:
        #  * Upper cased
        #  * Spaces stripped
        #  * Made up of a proportional number of chars from each word
        # #### Examples:
        #     Hippo::Strings.to_code_identifier('General Acme Corp')       #=> 'GENACMCORP'
        #     Hippo::Strings.code_identifier('General I B', padding:false) #=> 'GENERALIB' *(no trailing X)*
        #     Hippo::Strings.to_code_identifier('Top Cat', maxlen: 5 )     #=> 'TOCAT'
        def self.code_identifier( string, length:10, padding: 'X' )
            stripped = string.gsub(/\W/, '').upcase
            if stripped.length < length
                return padding.present? ? stripped.ljust( length, padding ) : stripped
            end
            result = ''
            words = string.gsub(/[^\w\s-]/,'').split(/\s+|-|_/)
            0.upto(words.length + 1).each_with_index do | attempt_num |
                remainder = length
                result = ''
                string = words.each_with_index do | word, index |
                    chars = ( remainder / (words.length-index) + attempt_num )
                    result << word[0...chars]
                    remainder = length - result.length
                end
                break if result.length >= length
            end
            result[0...length].upcase
        end

    end
end