Analysys of uncompess on input of length N words (where each line
is bounded by 40 words max -> max 80 chars per line, which would
be worst case would be alternating alpha's separted by spaces:

  Q  |  #Q  |  time Q  |  #A  |  time A  |  total
-------------------------------------------------
look-up:
  1  |  1   |  O(1)    |  1   |  O(1)    |  O(1) 
  2  |  1   |  O(1)    |  1   |  O(1)    |  O(1)        
-------------------------------------------------
     |      |          |      |          |  O(1)

Add-to-Book:
  1  |  1   |  O(1)    |  1   |  O(1)    |  O(1) 
  2  |  1   |  O(1)    |  1   |  O(80)   |  O(1)        
  3  |  1   |  O(1)    |  1   |  O(1)    |  O(1)
-------------------------------------------------
     |      |          |      |          |  O(1)

compess:
[same as compress-first]
(per one line of characters):
  Q  |  #Q  |  time Q  |  #A  |  time A  |  total
-------------------------------------------------
  1  |  80  |  O(1)    |  1   |  O(1)    |  O(1)
  2  |  80  |  O(1)    |  80  |  O(1)    |  O(1)
  3  |  80  |  O(1)    |  80  |  O(1)    |  O(1)
  4  |  80  |  O(1)    |  80  |  0(1)    |  O(1)     
  5  |  80  |  O(1)    |  1   |  O(1)    |  O(1)      
  6  |  1   |  O(1)    |  1   |  O(1)    |  O(1) 
-------------------------------------------------
Multiplied by maximum number of lines->  |  O(1)*O(N)
     |      |          |      |          |  =O(N)