使用gsub和regex替换所有其他出现的字符

Question

I am designing a card game (spades). I want to store each hand as a hash.

This is a randomly generated string representing a single had with 13 cards where 10c is the key and 10 is the value

"10c 10 bj 18 2s 15 3s 3 8d 8 as 14 kh 13 8c 8 3c 3 6d 6 kc 13 5d 5 ad 14"

I want to replace the space ' ' between 10c and 10 with a => and the following space between 10 and bj with a comma... This shoulod look like 10c=>10,bj=>18 etc.

From there I will convert this to a hash by calling Hash#inspect; however, I haven't been able to find a way to replace the nth character of a string using gsub.

I've tried calling hand.gsub(' ', '=>') but this obviously replaces every instance of a space with =>

Answer 1

You may build the hash right away with

s = "10c 10 bj 18 2s 15 3s 3 8d 8 as 14 kh 13 8c 8 3c 3 6d 6 kc 13 5d 5 ad 14"
hash = Hash[s.scan(/(\w+) (\d+)/)].transform_values(&:to_i)

Or, for older Ruby versions:

hash = Hash[s.scan(/(\w+) (\d+)/)]
hash = Hash[ hash.map{ |a, b| [ a, b.to_i ] } ]
puts hash
# => {"10c"=>10, "bj"=>18, "2s"=>15, "3s"=>3, "8d"=>8, "as"=>14, "kh"=>13, "8c"=>8, "3c"=>3, "6d"=>6, "kc"=>13, "5d"=>5, "ad"=>14}

See the Ruby demo

The (\\w+) (\\d+) pattern matches and captures into Group 1 any one or more word chars (with (\\w+) , letters, digits or underscores, you may use [[:alnum:]]+ , or [^\\W_] , instead to match letters or digits), then matches a space, and then captures into Group 2 any one or more digits ( (\\d+) ). String#scan matches all non-overlapping occurrences, and only returns captured substrings that are leveraged into a hash that is built dynamically.

Note that when you run b.to_i , you need no additional checks since \\d+ will only match digit substrings (and they are of length 2 in the input).

Answer 2

I don't believe the use of a regular expression is the best way to create the hash. Instead, I suggest using the methods String#split , Hash::[] , Hash#transform_values and String#to_i . If s is the string given in the example,

Hash[*s.split].transform_values(&:to_i)
  #=> {"10c"=>10, "bj"=>18, "2s"=>15, "3s"=>3, "8d"=>8,
  #    "as" =>14, "kh"=>13, "8c"=> 8, "3c"=>3, "6d"=>6,
  #    "kc" =>13, "5d"=> 5, "ad"=>14} 

The steps are as follows.

a = s.split
  #=> ["10c", "10", "bj", ... , "14"] 
b = Hash[*a]
  #=> {"10c"=>"10", "bj"=>"18", "2s"=>"15", "3s"=>"3",
  #    "8d" => "8", "as"=>"14", "kh"=>"13", "8c"=>"8",
  #    "3c" => "3", "6d"=> "6", "kc"=>"13", "5d"=>"5",
  #    "ad" =>"14"} 
b.transform_values(&:to_i)
  #=> {"10c"=>10, "bj"=>18, "2s"=>15, "3s"=>3, "8d"=>8,
  #    "as" =>14, "kh"=>13, "8c"=> 8, "3c"=>3, "6d"=>6,
  #    "kc" =>13, "5d"=> 5, "ad"=>14} 

The splat operator ( * ), causes Hash[*a] to be evaluated as Hash["10c", "10", "bj", ... , "14"] .

b.transform_values(&:to_i) is here equivalent to b.transform_values { |v| v.to_i } b.transform_values { |v| v.to_i } .