Bring word-count up to date with canonical data (#659)

Co-authored-by: Angelika Tyborska <[email protected]>
Co-authored-by: Tim Austin <[email protected]>

Author: angelikatyborska

config.json

@@ -539,7 +539,7 @@
         "name": "Word Count",
         "uuid": "4a24ba2f-ae92-4095-be53-64bc881422ea",
         "prerequisites": [],
-        "difficulty": 2,
+        "difficulty": 3,
         "topics": [
           "lists",
           "maps",

exercises/practice/word-count/.meta/example.ex

@@ -1,23 +1,34 @@
 defmodule WordCount do
-  @ascii_punctuation ~r/!|"|\#|\$|%|&|'|\(|\)|\*|\+|,|\.|\/|:|;|<|=|>|\?|@|\[|\\|]|\^|_|`|\{|\||}|~/
+  @doc """
+  Count the number of words in the sentence.
 
+  Words are compared case-insensitively.
+  """
+  @spec count(String.t()) :: map
   def count(sentence) do
     sentence
-    |> String.downcase()
-    |> remove_punctuation
-    |> to_words
-    |> summarize
+    |> normalize()
+    |> split
+    |> frequencies()
   end
 
-  defp remove_punctuation(string), do: String.replace(string, @ascii_punctuation, " ")
-
-  defp to_words(sentence), do: List.flatten(String.split(sentence))
+  defp normalize(string) do
+    string
+    |> String.downcase()
+    |> String.replace("_", " ", global: true)
+  end
 
-  defp summarize(words) do
-    Enum.reduce(words, %{}, &add_count/2)
+  defp split(string) do
+    ~r/\b[\w'-]+\b/u
+    |> Regex.scan(string)
+    |> List.flatten()
   end
 
-  defp add_count(word, counts) do
-    Map.update(counts, word, 1, &(&1 + 1))
+  defp frequencies(strings) do
+    # Enum.frequencies is only available from Elixir 1.10
+    strings
+    |> Enum.reduce(%{}, fn word, count ->
+      Map.update(count, word, 1, &(&1 + 1))
+    end)
   end
 end

exercises/practice/word-count/.meta/version

@@ -17,6 +17,18 @@ defmodule WordCountTest do
     assert WordCount.count("one fish two fish red fish blue fish") == expected
   end
 
+  @tag :pending
+  test "handles cramped list" do
+    expected = %{"one" => 1, "two" => 1, "three" => 1}
+    assert WordCount.count("one,two,three") == expected
+  end
+
+  @tag :pending
+  test "handles expanded list" do
+    expected = %{"one" => 1, "two" => 1, "three" => 1}
+    assert WordCount.count("one,\ntwo,\nthree") == expected
+  end
+
   @tag :pending
   test "ignore punctuation" do
     expected = %{"car" => 1, "carpet" => 1, "as" => 1, "java" => 1, "javascript" => 1}
@@ -43,13 +55,65 @@ defmodule WordCountTest do
 
   @tag :pending
   test "normalize case" do
-    expected = %{"go" => 3}
-    assert WordCount.count("go Go GO") == expected
+    expected = %{"go" => 3, "stop" => 2}
+    assert WordCount.count("go Go GO Stop stop") == expected
+  end
+
+  @tag :pending
+  test "with apostrophes" do
+    expected = %{"first" => 1, "don't" => 2, "laugh" => 1, "then" => 1, "cry" => 1}
+    assert WordCount.count("First: don't laugh. Then: don't cry.") == expected
+  end
+
+  @tag :pending
+  test "with quotations" do
+    expected = %{"joe" => 1, "can't" => 1, "tell" => 1, "between" => 1, "large" => 2, "and" => 1}
+    assert WordCount.count("Joe can't tell between 'large' and large.") == expected
+  end
+
+  @tag :pending
+  test "with quotations at the end at the beginning of the sentence" do
+    expected = %{"venti" => 1, "in" => 1, "italian" => 1, "means" => 1, "twenty" => 1}
+    assert WordCount.count("'venti' in Italian means 'twenty'") == expected
+  end
+
+  @tag :pending
+  test "substrings from the beginning" do
+    expected = %{
+      "joe" => 1,
+      "can't" => 1,
+      "tell" => 1,
+      "between" => 1,
+      "app" => 1,
+      "apple" => 1,
+      "and" => 1,
+      "a" => 1
+    }
+
+    assert WordCount.count("Joe can't tell between app, apple and a.") == expected
+  end
+
+  @tag :pending
+  test "multiple spaces not detected as a word" do
+    expected = %{"multiple" => 1, "whitespaces" => 1}
+    assert WordCount.count(" multiple   whitespaces") == expected
+  end
+
+  @tag :pending
+  test "alternating word separators not detected as a word" do
+    expected = %{"one" => 1, "two" => 1, "three" => 1}
+    assert WordCount.count(",\n,one,\n ,two \n 'three'") == expected
   end
 
   @tag :pending
   test "German" do
     expected = %{"götterfunken" => 1, "schöner" => 1, "freude" => 1}
     assert WordCount.count("Freude schöner Götterfunken") == expected
   end
+
+  @tag :pending
+  test "Polish" do
+    expected = %{"a" => 1, "w" => 1, "piątym" => 1, "stoi" => 1, "sześć" => 1, "fortepianów" => 1}
+    assert WordCount.count("A w piątym stoi sześć fortepianów") == expected
+  end
 end