require 'fileutils' require 'tmpdir' require 'test/unit' require 'rdoc/generator/texinfo' # From chapter 18 of the Pickaxe 3rd ed. and the TexInfo manual. class TestRDocInfoFormatting < Test::Unit::TestCase def setup @output_dir = File.join Dir.tmpdir, "test_rdoc_info_formatting_#{$$}" @output_file = File.join @output_dir, 'rdoc.texinfo' RDoc::RDoc.new.document(['--fmt=texinfo', '--quiet', File.expand_path(__FILE__), "--op=#{@output_dir}"]) @text = File.read @output_file # File.open('rdoc.texinfo', 'w') { |f| f.puts @text } end def teardown # FileUtils.rm_rf @output_dir end # Make sure tags like *this* do not make HTML def test_descriptions_are_not_html assert_no_match Regexp.new("\this\<\/b\>"), @text, "We had some HTML; icky!" end # Ensure we get a reasonable amount # # of space in between paragraphs. def test_paragraphs_are_spaced assert_match(/amount\n\n\nof space/, @text) end # @ and {} should be at-sign-prefixed def test_escaping assert_match(/@@ and @\{@\} should be at-sign-prefixed/) end # This tests that *bold* and bold me become @strong{bolded} def test_bold # Seems like a limitation of the Info format: @strong{bold} # becomes *bold* when read in Info or M-x info. highly lame! assert_match(/@strong\{bold\}/) assert_match(/@strong\{bold me\}/) end # Test that _italics_ and italicize me becomes @emph{italicized} def test_italics assert_match(/@emph\{italics\}/) assert_match(/@emph\{italicize me\}/) end # And that typewriter +text+ and typewriter me becomes @code{typewriter} def test_tt assert_match(/@code\{text\}/) assert_match(/@code\{typewriter me\}/) end # Check that # anything indented is # verbatim @verb{|foo bar baz|} def test_literal_code assert_match("@verb{| anything indented is verbatim @@verb@{|foo bar baz|@} |}") end # = Huge heading should be a @majorheading # == There is also @chapheading # === Everything deeper becomes a regular @heading # ====== Regardless of its nesting level def test_headings assert_match(/@majorheading Huge heading should be a @@majorheading/) assert_match(/@chapheading There is also @@chapheading/) assert_match(/@heading Everything deeper becomes a regular @@heading/) assert_match(/@heading Regardless of its nesting level/) end # * list item # * list item2 # # with a paragraph in between # # - hyphen lists # - are also allowed # and items may flow over lines def test_bullet_lists assert_match("@itemize @bullet @item list item @item list item2 @end itemize") assert_match("@itemize @bullet @item hyphen lists @item are also allowed and items may flow over lines @end itemize") end # 2. numbered lists # 8. are made by # 9. a digit followed by a period def test_numbered_lists end # a. alpha lists # b. should be parsed too def test_alpha_lists end # [cat] small domestic animal # [+cat+] command to copy standard input # to standard output def test_labelled_lists end # * First item. # * Inner item. # * Second inner item. # * Second outer item. def test_nested_lists assert_match("@itemize @bullet @item First item. @itemize @bullet @item Inner item. @item Second inner item. @end itemize @item Second outer item. @end itemize") end def test_internal_hyperlinks # be sure to test multi-word hyperlinks as well. end def test_hyperlink_targets end def test_web_links # An example of the two-argument form: The official # @uref{ftp://ftp.gnu.org/gnu, GNU ftp site} holds programs and texts. # produces: # The official GNU ftp site (ftp://ftp.gnu.org/gnu) # holds programs and texts. # and the HTML output is this: # The official GNU ftp site # holds programs and texts. end # three or more hyphens # ---- # should produce a horizontal rule def test_horizontal_rule # gah; not sure texinfo supports horizontal rules end private # We don't want the whole string inspected if we pass our own # message in. def assert_match(regex, string = @text, message = "Didn't find #{regex.inspect} in #{string}.") assert string[regex] #, message end end