From b5d2dcd8c7a51635d8bb11a69827c789d07e200e Mon Sep 17 00:00:00 2001
From: Kalil de Lima <kalil@rootstrap.com>
Date: Sat, 3 Oct 2020 22:43:51 -0300
Subject: [PATCH 1/2] add support for markdown text

---
 lib/reading_time.ex        | 44 ++++++++++++++++++++++++++++++++++++--
 test/reading_time_test.exs | 32 +++++++++++++++++++++++++++
 2 files changed, 74 insertions(+), 2 deletions(-)

diff --git a/lib/reading_time.ex b/lib/reading_time.ex
index a186113..07a8e35 100644
--- a/lib/reading_time.ex
+++ b/lib/reading_time.ex
@@ -5,11 +5,50 @@ defmodule ReadingTime do
 
   @split_pattern [" ", "\n", "\r", "\t"]
   @words_per_minute 200
+  @markdown_words [
+    # ignore headers
+    ~r/#+/,
+    ~r/(==)+/,
+    ~r/(--)+/,
+    # ingore boxed text
+    ~r/>+/,
+    # ignore unordered lists
+    ~r/(-)+/,
+    ~r/(\*)+/,
+    ~r/(\+)+/,
+    # ignore horizontal rules
+    ~r/----*/,
+    ~r/\*\*(\*)+/,
+    ~r/___+/,
+  ]
+
+  defp markdown_word?(word) do
+    Enum.any?(
+      @markdown_words,
+      fn md_word ->
+        Regex.match?(md_word, word)
+      end
+    )
+  end
+
+  defp ignore_format_words(words, :text) do
+    words
+  end
+
+  defp ignore_format_words(words, :markdown) do
+    words
+    |> Enum.filter(
+      fn word ->
+        !markdown_word?(word)
+      end
+    )
+  end
 
   @spec time(
           String.t(),
           words_per_minute: non_neg_integer(),
-          split_pattern: nonempty_list(String.t())
+          split_pattern: nonempty_list(String.t()),
+          format: atom()
         ) :: number
   @doc """
   Returns the time in minutes for a given string.
@@ -26,10 +65,11 @@ defmodule ReadingTime do
   def time(string, opts \\ []) do
     words_per_minute = Keyword.get(opts, :words_per_minute, @words_per_minute)
     split_pattern = Keyword.get(opts, :split_pattern, @split_pattern)
-
+    text_format = Keyword.get(opts, :text_format, :text)
     words =
       string
       |> String.split(split_pattern, trim: true)
+      |> ignore_format_words(text_format)
       |> length
 
     minutes =
diff --git a/test/reading_time_test.exs b/test/reading_time_test.exs
index d0327c8..bcad520 100644
--- a/test/reading_time_test.exs
+++ b/test/reading_time_test.exs
@@ -46,4 +46,36 @@ defmodule ReadingTimeTest do
                ) * long_text_multiplier
     end
   end
+
+  property "Markdown formatters get ignored" do
+    text = "
+    # This is a header1.
+
+    ## This is a header2.
+
+    This is also a header1.
+    ==
+
+    This is also a header2.
+    --
+
+    > This is a blockquote
+    >> This is a nested blockquote
+    >> # This is a header1 inside a blockquote
+
+    ---
+
+    - List1
+    + List2
+    * List3
+
+    ***
+
+    ___
+
+    "
+    actual_words = 37
+    assert ReadingTime.time(text, words_per_minute: 1, text_format: :markdown) == actual_words
+  end
+
 end

From f85ee77c84681bb34c2bb0b714657a85c6953ddd Mon Sep 17 00:00:00 2001
From: Kalil de Lima <kalil@rootstrap.com>
Date: Sun, 4 Oct 2020 17:37:40 -0300
Subject: [PATCH 2/2] use a stricter regex

---
 lib/reading_time.ex | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lib/reading_time.ex b/lib/reading_time.ex
index 07a8e35..35142ed 100644
--- a/lib/reading_time.ex
+++ b/lib/reading_time.ex
@@ -13,11 +13,11 @@ defmodule ReadingTime do
     # ingore boxed text
     ~r/>+/,
     # ignore unordered lists
-    ~r/(-)+/,
-    ~r/(\*)+/,
-    ~r/(\+)+/,
+    ~r/-/,
+    ~r/(\*)/,
+    ~r/(\+)/,
     # ignore horizontal rules
-    ~r/----*/,
+    ~r/---+/,
     ~r/\*\*(\*)+/,
     ~r/___+/,
   ]