Ruby non consistent results with scanned string's length - ruby

Question

I may not be having the whole picture here but I am getting inconsistent results with a calculation: I am trying to solve the run length encoding problem so that if you get an input string like "AAABBAAACCCAA" the encoding will be: "3A2B3A3C2A" so the functions is:
def encode(input)
res = ""
input.scan(/(.)\1*/i) do |match|
res << input[/(?<bes>#{match}+)/, "bes"].length.to_s << match[0].to_s
end
res
end
The results I am getting are:
irb(main):049:0> input = "AAABBBCCCDDD"
=> "AAABBBCCCDDD"
irb(main):050:0> encode(input)
(a) => "3A3B3C3D"
irb(main):051:0> input = "AAABBBCCCAAA"
=> "AAABBBCCCAAA"
irb(main):052:0> encode(input)
(b) => "3A3B3C3A"
irb(main):053:0> input = "AAABBBCCAAA"
=> "AAABBBCCAAA"
irb(main):054:0> encode(input)
(c) => "3A3B2C3A"
irb(main):055:0> input = "AAABBBCCAAAA"
=> "AAABBBCCAAAA"
irb(main):056:0> encode(input)
(d) => "3A3B2C3A"
irb(main):057:0> input = 'WWWWWWWWWWWWBWWWWWWWWWWWWBBBWWWWWWWWWWWWWWWWWWWWWWWWB'
=> "WWWWWWWWWWWWBWWWWWWWWWWWWBBBWWWWWWWWWWWWWWWWWWWWWWWWB"
irb(main):058:0> encode(input)
(e) => "12W1B12W1B12W1B"
As you can see, results (a) through (c) are correct, but results (d) and (e) are missing some repetitions and the resulting code is several letters short, can you give a hint as to where to check, please? (I am learning to use 'pry' right now)

Answer 1

Regular expressions are great, but they're not the golden hammer for every problem.
str = "AAABBAAACCCAA"
str.chars.chunk_while { |i, j| i == j }.map { |a| "#{a.size}#{a.first}" }.join
Breaking down what it does:
str = "AAABBAAACCCAA"
str.chars # => ["A", "A", "A", "B", "B", "A", "A", "A", "C", "C", "C", "A", "A"]
.chunk_while { |i, j| i == j } # => #<Enumerator: #<Enumerator::Generator:0x007fc1998ac020>:each>
.to_a # => [["A", "A", "A"], ["B", "B"], ["A", "A", "A"], ["C", "C", "C"], ["A", "A"]]
.map { |a| "#{a.size}#{a.first}" } # => ["3A", "2B", "3A", "3C", "2A"]
.join # => "3A2B3A3C2A"
to_a is there for illustration, but isn't necessary:
str = "AAABBAAACCCAA"
str.chars
.chunk_while { |i, j| i == j }
.map { |a| "#{a.size}#{a.first}" }
.join # => "3A2B3A3C2A"
how do you get to know such methods as Array#chunk_while? I am using Ruby 2.3.1 but cannot find it in the API docs, I mean, where is the compendium list of all the methods available? certainly not here ruby-doc.org/core-2.3.1/Array.html
Well, this is off-topic to the question but it's useful information to know:
Remember that Array includes the Enumerable module, which contains chunk_while. Use the search functionality of http://ruby-doc.org to find where things live. Also, get familiar with using ri at the command line, and try running gem server at the command-line to get the help for all the gems you've installed.
If you look at the Array documentation page, on the left you can see that Array has a parent class of Object, so it'll have the methods from Object, and that it also inherits from Enumerable, so it'll also pull in whatever is implemented in Enumerable.

Answer 2

You only get the count of the matched symbol repetitions that occur first. You need to perform a replacement within a gsub and pass the match object to a block where you can perform the necessary manipulations:
def encode(input)
input.gsub(/(.)\1*/) { |m| m.length.to_s << m[0] }
end
See the online Ruby test.
Results:
"AAABBBCCCDDD" => 3A3B3C3D
"AAABBBCCCAAA" => 3A3B3C3A
"AAABBBCCAAA" => 3A3B2C3A
"AAABBBCCAAAA" => 3A3B2C4A
"WWWWWWWWWWWWBWWWWWWWWWWWWBBBWWWWWWWWWWWWWWWWWWWWWWWWB" => 12W1B12W3B24W1B