Class: AnagramSolver::Permutator

Inherits:
Object
  • Object
show all
Defined in:
lib/anagram_solver/permutator.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(word_list) ⇒ Permutator

Initializes word_list instance variable, precomputed_list assigning it a hash whose default values is an empty array.

Precomputes word_list in underneath the hood. (i.e A in-process thread that AsyncConsumer offers. )



46
47
48
49
50
51
 
# File 'lib/anagram_solver/permutator.rb', line 46

def initialize(word_list)
  @word_list        = word_list
  @precomputed_list = Hash.new([])
  @bg_process       = AsyncConsumer.bg_process(self) { |s| s.precompute }
  @precomputed_list_old = Hash.new([])
end

Instance Attribute Details

#bg_process ⇒ Object (readonly)

Returns the value of attribute bg_process.



37
38
39
 
# File 'lib/anagram_solver/permutator.rb', line 37

def bg_process
  @bg_process
end

#precomputed_list ⇒ Object (readonly)

Returns the value of attribute precomputed_list.



36
37
38
 
# File 'lib/anagram_solver/permutator.rb', line 36

def precomputed_list
  @precomputed_list
end

#precomputed_list_old ⇒ Object (readonly)

Returns the value of attribute precomputed_list_old.



37
38
39
 
# File 'lib/anagram_solver/permutator.rb', line 37

def precomputed_list_old
  @precomputed_list_old
end

#word_list ⇒ Object (readonly)

Returns the value of attribute word_list.



36
37
38
 
# File 'lib/anagram_solver/permutator.rb', line 36

def word_list
  @word_list
end

Instance Method Details

#precompute ⇒ Object

Generates a hash whose keys are sorted alphabetically and values (i.e rhd ) are words in its normal order.

Given a list pots, stop, opts When I precompute it Then I should have a hash like: { “opst” => [ “pots”, “stop”, “opts” ] }

NOTE: precompute only gets the first word it ‘slices’ off a line. Therefore if a line was: “pots, opts” It would get the first word, which is pots.

However if it was: “, Deviation’s, pots”; It would slice Deviation’s

It is also slices words with accents, such as: émigré‘s Ångström

If you’re on UNIX-like there might exist a dict word in see /usr/share/dict/words

Want to learn more about Regexp? Fear not: open-std.org/jtc1/sc22/wg21/docs/papers/2003/n1500.html



95
96
97
98
99
100
101
 
# File 'lib/anagram_solver/permutator.rb', line 95

def precompute
  word_list.each do |line|
    # word = line.chomp.slice(/\b['\wÀ-ÖÙ-Üà-öù-ü]+/i)
    word = line.chomp
    precomputed_list[sort!(word)] += [word]
  end
end

#precompute_old ⇒ Object

Used for benchmarking



56
57
58
59
60
61
62
 
# File 'lib/anagram_solver/permutator.rb', line 56

def precompute_old
  warn "This method should not be used, please use precompute instead."
  word_list.rewind.each do |line|
    word = line.chomp
    precomputed_list_old[word.split('').sort.join.freeze] += [word]
  end
end

#sort!(word) ⇒ Object

Orders any given word in alphabetical order by unpacking them (i.e spliting ) sort and packing them again ( i.e joining ). This facilitates when trying to find combinations for words, as we only have to sort once and return results by calling a hash with a sorted key.

ruby-doc.org/core-2.0.0/Array.html#method-i-pack See: en.wikipedia.org/wiki/UTF-8 Unpacks a string using: Integer UTF-8 character, as it can represent every character in the Unicode char set. Such as émigré‘s

  • means all remaining arrays elements will be

converted.

Freezes string, otherwise Ruby would generate another String object instead of keeping the one already supplied. So that we would have two instances of the same thing in memory.



125
126
127
 
# File 'lib/anagram_solver/permutator.rb', line 125

def sort!(word)
  word.unpack("U*").sort.pack("U*").freeze
end