From 6f568a03e15d4631bb23e49813256f59ec4bfb22 Mon Sep 17 00:00:00 2001
From: Shadowfacts <me@shadowfacts.net>
Date: Thu, 31 Oct 2019 16:52:01 -0400
Subject: [PATCH] Add whatever.scalzi.com extractor

---
 .../pipeline/extractor/daring_fireball.ex     |  4 +++
 .../pipeline/extractor/whatever_scalzi.ex     | 32 +++++++++++++++++++
 2 files changed, 36 insertions(+)
 create mode 100644 lib/frenzy/pipeline/extractor/whatever_scalzi.ex

diff --git a/lib/frenzy/pipeline/extractor/daring_fireball.ex b/lib/frenzy/pipeline/extractor/daring_fireball.ex
index d668f6d..5daf6f5 100644
--- a/lib/frenzy/pipeline/extractor/daring_fireball.ex
+++ b/lib/frenzy/pipeline/extractor/daring_fireball.ex
@@ -1,4 +1,8 @@
 defmodule Frenzy.Pipeline.Extractor.DaringFireball do
+  @moduledoc """
+  Extractor for https://daringfireball.net
+  """
+
   alias Frenzy.Pipeline.Extractor
   @behaviour Extractor
 
diff --git a/lib/frenzy/pipeline/extractor/whatever_scalzi.ex b/lib/frenzy/pipeline/extractor/whatever_scalzi.ex
new file mode 100644
index 0000000..279124a
--- /dev/null
+++ b/lib/frenzy/pipeline/extractor/whatever_scalzi.ex
@@ -0,0 +1,32 @@
+defmodule Frenzy.Pipeline.Extractor.WhateverScale do
+  @moduledoc """
+  Extractor for https://whatever.scalzi.com
+  """
+
+  alias Frenzy.Pipeline.Extractor
+  @behaviour Extractor
+
+  @impl Extractor
+  def extract(body) do
+    html_tree = Floki.parse(body)
+
+    case get_article_content(html_tree) do
+      nil ->
+        {:error, "no matching elements"}
+
+      elem ->
+        {:ok, Floki.raw_html(elem)}
+    end
+  end
+
+  defp get_article_content(html_tree) do
+    case Floki.find(html_tree, "article.post > div.entry-content") do
+      [content_elem | _] ->
+        # remove social media buttons that are included in the .entry-content element
+        Floki.filter_out(content_elem, "div#jp-post-flair")
+
+      _ ->
+        nil
+    end
+  end
+end