Add spec/generated_page_spec.rb, fix bugs

Also tweaked comments and reordered the arguments to GeneratedPage.new.
mike-burns · Jun 22, 2015 · 0ba4ea6 · 0ba4ea6
1 parent 6327669
commit 0ba4ea6
Show file tree

Hide file tree

Showing 4 changed files with 105 additions and 13 deletions.
diff --git a/bin/jekyll_pages_api b/bin/jekyll_pages_api
@@ -21,9 +21,10 @@ Arguments:
   title_prefix
     Prefix to strip from page titles
   body_element_tag
-    Tag (or tag prefix) identifying the main content element of each tag. Can
-    be a complete tag (ending in '>'), or the prefix of a longer tag. Used to
-    strip boilerplate out of the content exported via the API.
+    Tag (or tag prefix) identifying the main content element within the <body>
+    element of each document. Can be a complete tag (ending in '>'), or the
+    prefix of a longer tag. Used to strip boilerplate out of the content
+    exported via the API.
 END_USAGE
 
 if ARGV.length == 1 && ARGV[0] == '-h'

diff --git a/lib/jekyll_pages_api/generated_page.rb b/lib/jekyll_pages_api/generated_page.rb
@@ -8,20 +8,25 @@ module JekyllPagesApi
   class GeneratedPage
     attr_reader :path, :relative_path, :data, :content
 
+    # @param path [String] full path to the generated page's file
     # @param basedir see {GeneratedSite#initialize}
     # @param title_prefix see {GeneratedSite#initialize}
     # @param body_element_tag see {GeneratedSite#initialize}
-    # @param path [String] full path to the generated page's file
     # @param content [String] HTML content of the generated page's file
-    def initialize(basedir, title_prefix, body_element_tag, path, content)
+    # @raises [RuntimError] if path does not begin with basedir
+    def initialize(path, basedir, title_prefix, body_element_tag, content)
+      unless path.start_with? basedir
+        raise "#{path} does not start with #{basedir}"
+      end
+
       @path = path
       basedir_len = basedir.size
-      basedir_len += 1 if basedir.end_with? File::SEPARATOR
+      basedir_len -= File::SEPARATOR.size if basedir.end_with? File::SEPARATOR
 
       end_path = path.size
       index_suffix = File.join "", "index.html"
-      end_path -= (index_suffix.size + 1) if path.end_with? index_suffix
-      @relative_path = path[basedir_len..end_path]
+      end_path -= index_suffix.size if path.end_with? index_suffix
+      @relative_path = (path[basedir_len..end_path] || "")
       @data, @content = GeneratedPageParser.parse_generated_page(
         content, title_prefix, body_element_tag)
     end

diff --git a/lib/jekyll_pages_api/generated_site.rb b/lib/jekyll_pages_api/generated_site.rb
@@ -16,9 +16,9 @@ class GeneratedSite
     # @param basedir [String] Path to the generated site's root directory
     # @param title_prefix [String] Prefix to strip from page titles
     # @param body_element_tag [String] Tag (or tag prefix) identifying the
-    #   main content element of each tag. Can be a complete tag (ending in
-    #   '>'), or the prefix of a longer tag. Used to strip boilerplate out of
-    #   the content exported via the API.
+    #   main content element within the <body> element of each document. Can
+    #   be a complete tag (ending in '>'), or the prefix of a longer tag. Used
+    #   to strip boilerplate out of the content exported via the API.
     def initialize(baseurl, basedir, title_prefix, body_element_tag)
       @baseurl = baseurl
       @basedir = basedir
@@ -33,8 +33,8 @@ def each_site_file
       Dir.glob(File.join(self.basedir, '**', '*')) do |f|
         next unless f.end_with? '.html'
         begin
-          page = GeneratedPage.new(self.basedir, self.title_prefix,
-            self.body_element_tag, f, File.read(f))
+          page = GeneratedPage.new(f, self.basedir, self.title_prefix,
+            self.body_element_tag, File.read(f))
           yield page unless page.data['title'].nil?
         rescue
           $stderr.puts "Error while processing #{f}:"

diff --git a/spec/generated_page_spec.rb b/spec/generated_page_spec.rb
@@ -0,0 +1,86 @@
+module JekyllPagesApi
+  describe GeneratedPage do
+    describe '#initialize' do
+      it "handles all empty strings correctly" do
+        page = GeneratedPage.new "", "", "", "", ""
+        expect(page.path).to eq("")
+        expect(page.relative_path).to eq("")
+        expect(page.data).to eq({})
+        expect(page.content).to eq("")
+      end
+
+      it "parses the relative path when basedir ends in SEPARATOR" do
+        path = File.join "foo", "bar", "baz.html"
+        basedir = File.join "foo", ""
+        page = GeneratedPage.new path, basedir, "", "", ""
+        expect(page.path).to eq("foo/bar/baz.html")
+        expect(page.relative_path).to eq("/bar/baz.html")
+        expect(page.data).to eq({})
+        expect(page.content).to eq("")
+      end
+
+      it "parses the relative path when basedir doesn't end in SEPARATOR" do
+        path = File.join "foo", "bar", "baz.html"
+        basedir = "foo"
+        page = GeneratedPage.new path, basedir, "", "", ""
+        expect(page.path).to eq("foo/bar/baz.html")
+        expect(page.relative_path).to eq("/bar/baz.html")
+        expect(page.data).to eq({})
+        expect(page.content).to eq("")
+      end
+
+      it "raises RuntimeError when path does not begin with basedir" do
+        path = File.join "foo", "bar", "baz.html"
+        basedir = File.join "quux", ""
+        expect{GeneratedPage.new path, basedir, "", "", ""
+          }.to raise_error(
+            RuntimeError, "#{path} does not start with #{basedir}")
+      end
+
+      it "parses out index.html suffix and leaves trailing slash" do
+        path = File.join "foo", "bar", "index.html"
+        basedir = File.join "foo", ""
+        page = GeneratedPage.new path, basedir, "", "", ""
+        expect(page.path).to eq("foo/bar/index.html")
+        expect(page.relative_path).to eq("/bar/")
+        expect(page.data).to eq({})
+        expect(page.content).to eq("")
+      end
+
+      it "parses out index.html suffix and leaves trailing slash for root" do
+        path = File.join "foo", "index.html"
+        basedir = File.join "foo", ""
+        page = GeneratedPage.new path, basedir, "", "", ""
+        expect(page.path).to eq("foo/index.html")
+        expect(page.relative_path).to eq("/")
+        expect(page.data).to eq({})
+        expect(page.content).to eq("")
+      end
+
+      it "parses content correctly" do
+        path = File.join "foo", "bar", "index.html"
+        basedir = File.join "foo", ""
+        title_prefix = "18F &mdash; "
+        body_element_tag = "<div class='content'"
+        content = "<head><title>18F &mdash; Blah Blah Woof Woof</title>"+
+          "<meta name='skip-index' content='true'>"+
+          "<meta content='baz,quux,xyzzy,plugh' name=\"tags\" />"+
+          "</head>"+
+          "<body><div>header</div>"+
+          "<div class='content'>foobar</div>"+
+          "<div>footer</div></body>"
+
+        page = GeneratedPage.new(path, basedir, title_prefix,
+          body_element_tag, content)
+        expect(page.path).to eq("foo/bar/index.html")
+        expect(page.relative_path).to eq("/bar/")
+        expect(page.data).to eq(
+          "title" => "Blah Blah Woof Woof",
+          "skip-index" => "true",
+          "tags" => "baz,quux,xyzzy,plugh",
+        )
+        expect(page.content).to eq("foobar")
+      end
+    end
+  end
+end