Merge pull request #1656 from kmuto/review-header-listener

Extract EPUBMaker#parse_headlines to add a test
kmuto · Jan 27, 2021 · e248f10 · e248f10
2 parents 725d0d4 + 164c3e5
commit e248f10
Show file tree

Hide file tree

Showing 4 changed files with 105 additions and 5 deletions.
diff --git a/lib/review/epubmaker.rb b/lib/review/epubmaker.rb
@@ -422,12 +422,19 @@ def detect_properties(path)
       properties
     end
 
-    def write_info_body(basetmpdir, _id, filename, ispart = nil, chaptype = nil)
+    def parse_headlines(path)
       headlines = []
+
+      File.open(path) do |htmlio|
+        REXML::Document.parse_stream(htmlio, ReVIEWHeaderListener.new(headlines))
+      end
+
+      headlines
+    end
+
+    def write_info_body(basetmpdir, _id, filename, ispart = nil, chaptype = nil)
       path = File.join(basetmpdir, filename)
-      htmlio = File.new(path)
-      REXML::Document.parse_stream(htmlio, ReVIEWHeaderListener.new(headlines))
-      htmlio.close
+      headlines = parse_headlines(path)
 
       if headlines.empty?
         warn "#{filename} is discarded because there is no heading. Use `=[notoc]' or `=[nodisp]' to exclude headlines from the table of contents."

diff --git a/lib/review/epubmaker/reviewheaderlistener.rb b/lib/review/epubmaker/reviewheaderlistener.rb
@@ -7,6 +7,15 @@
 #
 module ReVIEW
   class EPUBMaker
+    # Listener class to scan HTML and get heading information
+    #
+    # The heading information this listener will retrieve is as follows:
+    #
+    # * level: Heading level (1..6)
+    # * id: HTMl ID attribute. Basically the `id` attribute of the h(1-6) element, but if there is an `a` element within the h(1-6) element, it will be its `id` attribute.
+    # * title: The title string of the headline. Usually, it is the text within the h(1-6) element, but if there is an `img` element, it will be the text with its `alt` attribute.
+    # * notoc: The `notoc` attribute of the headline element.
+    #
     class ReVIEWHeaderListener
       include REXML::StreamListener
       def initialize(headlines)
@@ -22,7 +31,7 @@ def tag_start(name, attrs)
           @level = $1.to_i
           @id = attrs['id'] if attrs['id'].present?
           @notoc = attrs['notoc'] if attrs['notoc'].present?
-        elsif !@level.nil?
+        elsif @level.present? # if in <hN> tag
           if name == 'img' && attrs['alt'].present?
             @content << attrs['alt']
           elsif name == 'a' && attrs['id'].present?

diff --git a/test/assets/header_listener.html b/test/assets/header_listener.html
@@ -0,0 +1,35 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xmlns:ops="http://www.idpf.org/2007/ops" xml:lang="ja">
+<head>
+  <meta charset="UTF-8" />
+  <link rel="stylesheet" type="text/css" href="style.css" />
+  <meta name="generator" content="Re:VIEW" />
+  <title>first chapter</title>
+</head>
+<body>
+<h1><a id="h1"></a><span class="secno">第1章　</span>first chapter</h1>
+
+<h2><a id="h1-1"></a><span class="secno">1.1　</span>first section</h2>
+
+<h3><a id="h1-1-1"></a>first <img src="images/icon1.jpg" alt="subsection" /></h3>
+
+<h2><a id="h1-2"></a><span class="secno">1.2　</span>second section</h2>
+
+<h3 id="dummy1"><a id="h1-2-1"></a>dummy subsection</h3>
+
+<h2><a id="h1-3"></a><span class="secno">1.3　</span>third section</h2>
+
+<h2 id="ch01_nonum1" notoc="true">notoc section</h2>
+
+<h2 id="dummy2" notoc="true">notoc section</h2>
+
+<a id="ch01_nonum3" /><h2 id="ch01_nonum3" hidden="true">nodisp section</h2>
+
+<a id="dummy3" /><h2 id="dummy3" hidden="true">nodisp section</h2>
+
+<h2 id="ch01_nonum5">nonum section</h2>
+
+<h2 id="dummy4">nonum section</h2>
+</body>
+</html>
diff --git a/test/test_reviewheaderlistener.rb b/test/test_reviewheaderlistener.rb
@@ -0,0 +1,49 @@
+require 'test_helper'
+require 'rexml/document'
+require 'rexml/streamlistener'
+require 'review/epubmaker'
+
+class ReVIEWHeaderListenerTest < Test::Unit::TestCase
+  def setup
+    @epubmaker = ReVIEW::EPUBMaker.new
+  end
+
+  def teardown
+  end
+
+  def test_epubmaker_parse_headlines
+    # original Re:VIEW source:
+    #
+    # = first chapter
+    # == first section
+    # === first @<embed>{<img src="images/icon1.jpg" alt="subsection" />}
+    # == second section
+    # ==={dummy1} dummy subsection
+    # == third section
+    # ==[notoc] notoc section
+    # ==[notoc]{dummy2} notoc section
+    # ==[nodisp] nodisp section
+    # ==[nodisp]{dummy3} nodisp section
+    # ==[nonum] nonum section
+    # ==[nonum]{dummy4} nonum section
+    Dir.mktmpdir do |_dir|
+      path = File.join(assets_dir, 'header_listener.html')
+      headlines = @epubmaker.parse_headlines(path)
+
+      expected = [{ 'id' => 'h1', 'level' => 1, 'notoc' => nil, 'title' => '第1章　first chapter' },
+                  { 'id' => 'h1-1', 'level' => 2, 'notoc' => nil, 'title' => '1.1　first section' },
+                  { 'id' => 'h1-1-1', 'level' => 3, 'notoc' => nil, 'title' => 'first subsection' },
+                  { 'id' => 'h1-2', 'level' => 2, 'notoc' => nil, 'title' => '1.2　second section' },
+                  { 'id' => 'h1-2-1', 'level' => 3, 'notoc' => nil, 'title' => 'dummy subsection' },
+                  { 'id' => 'h1-3', 'level' => 2, 'notoc' => nil, 'title' => '1.3　third section' },
+                  { 'id' => 'ch01_nonum1', 'level' => 2, 'notoc' => 'true', 'title' => 'notoc section' },
+                  { 'id' => 'dummy2', 'level' => 2, 'notoc' => 'true', 'title' => 'notoc section' },
+                  { 'id' => 'ch01_nonum3', 'level' => 2, 'notoc' => nil, 'title' => 'nodisp section' },
+                  { 'id' => 'dummy3', 'level' => 2, 'notoc' => nil, 'title' => 'nodisp section' },
+                  { 'id' => 'ch01_nonum5', 'level' => 2, 'notoc' => nil, 'title' => 'nonum section' },
+                  { 'id' => 'dummy4', 'level' => 2, 'notoc' => nil, 'title' => 'nonum section' }]
+
+      assert_equal expected, headlines
+    end
+  end
+end