Fix regressions inherited from Maruku trunk

The last commit was a real mess.

lib/maruku/attributes.rb

@@ -41,6 +41,8 @@ def read_attribute_list(src, con=nil, break_on_chars=nil)
         break if break_on_chars.include? src.cur_char
 
         case src.cur_char
+        when ':'
+          src.ignore_char
         when nil
           break      # we're done here.
         when '='     # error
@@ -69,7 +71,7 @@ def read_attribute_list(src, con=nil, break_on_chars=nil)
             next
           end
 
-          if src.cur_char != '='
+          if src.cur_char != '=' && key.length > 0
             al << [:ref, key]
             next
           end

lib/maruku/helpers.rb

@@ -70,15 +70,17 @@ def md_html(raw_html, al=nil)
       e = md_el(:raw_html, [], :raw_html => raw_html)
       e.extend HTMLElement
       begin
-        d = Nokogiri::HTML::Document.new
+        # Some might prefer an HTML::Document here. Alas, we need to preserve case
+        # on element names, so that's not suitable.
+        d = Nokogiri::XML::Document.new
 
         # Make sure the SVG namespace is known
         root = Nokogiri::XML::Element.new('html', d)
         root.add_namespace('svg', "http://www.w3.org/2000/svg" )
 
         # Set this as an attribute so it doesn't get included
         # in metadata comparisons
-        e.parsed_html = Nokogiri::HTML::DocumentFragment.new(d, raw_html, d)
+        e.parsed_html = Nokogiri::XML::DocumentFragment.new(d, raw_html, d)
       rescue => ex
         maruku_recover "Nokogiri cannot parse this block of HTML/XML:\n" +
           raw_html.gsub(/^/, '|').rstrip + "\n" + ex.inspect

lib/maruku/input/parse_block.rb

@@ -1,3 +1,5 @@
+require 'set'
+
 module MaRuKu; module In; module Markdown; module BlockLevelParser
 
   include Helpers
@@ -236,7 +238,8 @@ def read_xml_instruction(src, output)
     end
   end
 
-  HTML_INLINE_ELEMS = %w(a abbr acronym b big bdo br button cite code del dfn em i img input ins kbd label option q rb rbc rp rt rtc ruby samp select small span strong sub sup textarea tt var)
+  HTML_INLINE_ELEMS = Set.new %w[a abbr acronym b big bdo br button canvas cite code del dfn em i img input ins
+    kbd label option q rb rbc rp rt rtc ruby samp select small span strong sub sup textarea tt var] 
   def read_raw_html(src)
     extra_line = nil
     h = HTMLHelper.new
@@ -306,9 +309,9 @@ def read_list_item(src)
 
     indentation, ial = spaces_before_first_char(first)
     al = read_attribute_list(CharSource.new(ial, src)) if ial
-
+    ial_offset = ial ? ial.length + 3 : 0
     lines, want_my_paragraph =
-      read_indented_content(src, indentation, :ial, item_type)
+      read_indented_content(src, indentation, [], item_type, ial_offset)
 
     # add first line
     # Strip first '*', '-', '+' from first line
@@ -318,9 +321,7 @@ def read_list_item(src)
     src2 = LineSource.new(lines, src, parent_offset)
     children = parse_blocks(src2)
 
-    with_par = want_my_paragraph
-
-    md_li(children, with_par, al)
+    md_li(children, want_my_paragraph, al)
   end
 
   def read_abbreviation(src)
@@ -374,16 +375,17 @@ def read_footnote_text(src)
 
   # This is the only ugly function in the code base.
   # It is used to read list items, descriptions, footnote text
-  def read_indented_content(src, indentation, break_list, item_type)
+  def read_indented_content(src, indentation, break_list, item_type, ial_offset=0)
     lines = []
     # collect all indented lines
     saw_empty = false
     saw_anything_after = false
     break_list = Array(break_list)
+    len = indentation - ial_offset
 
     while src.cur_line
       num_leading_spaces = src.cur_line.number_of_leading_spaces
-      break if num_leading_spaces < indentation && ![:text, :empty].include?(src.cur_line.md_type)
+      break if num_leading_spaces < len && ![:text, :empty].include?(src.cur_line.md_type)
 
       line = strip_indent(src.cur_line, indentation)
       md_type = line.md_type
@@ -396,7 +398,7 @@ def read_indented_content(src, indentation, break_list, item_type)
       end
 
       # Unquestioningly grab anything that's deeper-indented
-      if md_type != :code && num_leading_spaces > indentation
+      if md_type != :code && num_leading_spaces > len
         lines << line
         src.shift_line
         next
@@ -405,7 +407,7 @@ def read_indented_content(src, indentation, break_list, item_type)
       # after a white line
       if saw_empty
         # we expect things to be properly aligned
-        break if num_leading_spaces < indentation
+        break if num_leading_spaces < len
         saw_anything_after = true
       else
         break if break_list.include?(md_type)

lib/maruku/input/parse_doc.rb

@@ -160,7 +160,7 @@ def substitute_markdown_inside_raw_html
       block_tags = ['div']
 
       # find span elements or elements with 'markdown' attribute
-      doc.css((["[markdown]"] + HTML_INLINE_ELEMS).join(",")).each do |e|
+      doc.css((["[markdown]"] + HTML_INLINE_ELEMS.to_a).join(",")).each do |e|
         # should we parse block-level or span-level?
 
         how = e['markdown']

lib/maruku/toc.rb

@@ -173,7 +173,10 @@ class MDElement
     # Generate an id for headers. Assumes @children is set.
     def generate_id
       raise "generate_id only makes sense for headers" unless node_type == :header
-      children_to_s.tr(' ', '_').downcase.gsub(/\W/, '').strip
+      #The generated id must be unique; not just because non-unique id's are invalid,
+      #but because the are functionally-broken (don't work as link targets).
+      @doc.id_counter += 1
+      children_to_s.tr(' ', '_').downcase.gsub(/\W/, '').strip + "_" + @doc.id_counter.to_s
     end
   end
 end

spec/block_docs/abbrev.md

@@ -413,7 +413,7 @@ md_el(:document,[
 		md_ref_def("jos", "jos", {:title=>nil})
 ],{},[])
 *** Output of to_html ***
-<h1 id="webkit_safari_31_and_the_css_fontface_declaration">WebKit (Safari 3.1) and the <abbr title="Cascading Style Sheets">CSS</abbr> @font-face declaration</h1>
+<h1 id="webkit_safari_31_and_the_css_fontface_declaration_1">WebKit (Safari 3.1) and the <abbr title="Cascading Style Sheets">CSS</abbr> @font-face declaration</h1>
 
 <p>I&#8217;m a big fan of typography in general. If you check out <a href="http://elliottcable.name">my homepage</a> or my <a href="http://elliottcable.name/contact.xhtml">contact elliottcable</a> page, and you&#8217;re using Safari/WebKit or Opera/Kestrel, you&#8217;ll notice the typefaces (fonts, as colloquialized) are <em>very</em> non-standard. (As of this writing, I&#8217;m using <a href="http://www.josbuivenga.demon.nl/museo.html" title="Jos Buivenga's Museo free typeface">Museo</a> and <a href="http://www.josbuivenga.demon.nl/diavlo.html" title="Jos Buivenga's free Diavlo typeface">Diavlo</a><sup id="fnref:1"><a href="#fn:1" rel="footnote">1</a></sup> heavily on both.)</p>
 
@@ -508,7 +508,7 @@ div#content {
 <p>To give Microsoft a little credit, something I rarely do&#8230; Yes, I&#8217;m aware Microsoft submitted EOT to the <abbr title="World Wide Web Consortium">W3C</abbr> as a proposal - the problem isn&#8217;t with their attempts to make it non-proprietary, but with the basic concept of making typefaces on the web DRMed. Look what such attempts have done to the music and video industry - simply decimated it. Do we really want to see the same thing happen to our beloved medium as typography moves into the 21st century? <a href="#fnref:2" rev="footnote">&#8617;</a></p>
 </li></ol></div>
 *** Output of to_latex ***
-\hypertarget{webkit_safari_31_and_the_css_fontface_declaration}{}\section*{{WebKit (Safari 3.1) and the CSS @font-face declaration}}\label{webkit_safari_31_and_the_css_fontface_declaration}
+\hypertarget{webkit_safari_31_and_the_css_fontface_declaration_1}{}\section*{{WebKit (Safari 3.1) and the CSS @font-face declaration}}\label{webkit_safari_31_and_the_css_fontface_declaration_1}
 
 I'm a big fan of typography in general. If you check out \href{http://elliottcable.name}{my homepage} or my \href{http://elliottcable.name/contact.xhtml}{contact elliottcable} page, and you're using Safari/WebKit or Opera/Kestrel, you'll notice the typefaces (fonts, as colloquialized) are \emph{very} non-standard. (As of this writing, I'm using \href{http://www.josbuivenga.demon.nl/museo.html}{Museo} and \href{http://www.josbuivenga.demon.nl/diavlo.html}{Diavlo}\footnote{These are fonts by \href{jos}{Jos Buivenga}, quite the amazing person. His (free) fonts are, uniquely, released for use on the web in {\colorbox[rgb]{1.00,0.93,1.00}{\tt \char64font\char45face}} declarations - unlike the vast majority of other (even free to download) typefaces, which have ridiculously restricting licenses and terms of use statements. Props, Jos - you're a pioneer, and deserve recognition as such.}  heavily on both.)
 

spec/block_docs/attributes/attributes.md

@@ -34,7 +34,7 @@ md_el(:document,[
 
 <h3 id="header2">Header with attributes</h3>
 
-<h3 id="header_no_attributes">Header no attributes</h3>
+<h3 id="header_no_attributes_3">Header no attributes</h3>
 
 <p id="par1">Paragraph with a.</p>
 
@@ -44,7 +44,7 @@ md_el(:document,[
 
 \hypertarget{header2}{}\subsubsection*{{Header with attributes}}\label{header2}
 
-\hypertarget{header_no_attributes}{}\subsubsection*{{Header no attributes}}\label{header_no_attributes}
+\hypertarget{header_no_attributes_3}{}\subsubsection*{{Header no attributes}}\label{header_no_attributes_3}
 
 Paragraph with a.
 

spec/block_docs/fenced_code_blocks.md

@@ -1,6 +1,6 @@
 Fenced code blocks
 *** Parameters: ***
-{ :fenced_code_blocks => true }
+{ :fenced_code_blocks => true, :html_use_syntax => false }
 *** Markdown input: ***
 ```ruby
 john = Twitter::Client.new(

spec/block_docs/footnotes2.md

@@ -0,0 +1,78 @@
+Write a comment abouth the test here.
+*** Parameters: ***
+require 'maruku/ext/math';{:html_math_engine => 'itex2mml'}
+*** Markdown input: ***
+Ruby on Rails is a web-framework[^framework]. It uses the MVC[^MVC] architecture pattern. It has its good points[^points].
+
+[^framework]: a reusable set of libraries
+[^MVC]: Model View Controller
+[^points]: Here are its good points
+     1. Ease of use
+     2. Rapid development
+
+That has nothing to do with putting equations in footnotes[^equations].
+
+[^equations]: Like this:
+$$
+  x = r\cos\theta
+$$
+*** Output of inspect ***
+md_el(:document,[
+	md_par([
+		"Ruby on Rails is a web-framework",
+		md_foot_ref("^framework"),
+		". It uses the MVC",
+		md_foot_ref("^MVC"),
+		" architecture pattern. It has its good points",
+		md_foot_ref("^points"),
+		"."
+	]),
+	md_el(:footnote, md_par("a reusable set of libraries"), {:footnote_id=>"^framework"}),
+	md_el(:footnote, md_par("Model View Controller"), {:footnote_id=>"^MVC"}),
+	md_el(:footnote, [
+		md_par("Here are its good points"),
+		md_el(:ol, [md_li("Ease of use", false), md_li("Rapid development", false)])
+		], {:footnote_id=>"^points"}),
+	md_par([
+		"That has nothing to do with putting equations in footnotes",
+		md_foot_ref("^equations"),
+		"."
+		]),
+	md_el(:footnote, [
+		md_par("Like this:"),
+		md_el(:equation, [], {:math=>"\nx = r\\cos\\theta\n\n", :label=>nil, :num=>nil})
+		], {:footnote_id=>"^equations"})
+],{},[])
+*** Output of to_html ***
+<p>Ruby on Rails is a web-framework<sup id="fnref:1"><a href="#fn:1" rel="footnote">1</a></sup>. It uses the MVC<sup id="fnref:2"><a href="#fn:2" rel="footnote">2</a></sup> architecture pattern. It has its good points<sup id="fnref:3"><a href="#fn:3" rel="footnote">3</a></sup>.</p>
+
+<p>That has nothing to do with putting equations in footnotes<sup id="fnref:4"><a href="#fn:4" rel="footnote">4</a></sup>.</p>
+<div class="footnotes"><hr /><ol><li id="fn:1">
+<p>a reusable set of libraries <a href="#fnref:1" rev="footnote">↩</a></p>
+</li><li id="fn:2">
+<p>Model View Controller <a href="#fnref:2" rev="footnote">↩</a></p>
+</li><li id="fn:3">
+<p>Here are its good points</p>
+
+<ol>
+<li>Ease of use</li>
+
+<li>Rapid development</li>
+</ol>
+<a href="#fnref:3" rev="footnote">↩</a></li><li id="fn:4">
+<p>Like this:</p>
+<div class="maruku-equation"><math xmlns="http://www.w3.org/1998/Math/MathML" display="block" class="maruku-mathml"><mi>x</mi><mo>=</mo><mi>r</mi><mi>cos</mi><mi>θ</mi></math><span class="maruku-eq-tex"><code style="display: none">x = r\cos\theta</code></span></div><a href="#fnref:4" rev="footnote">↩</a></li></ol></div>
+*** Output of to_latex ***
+Ruby on Rails is a web-framework\footnote{a reusable set of libraries} . It uses the MVC\footnote{Model View Controller}  architecture pattern. It has its good points\footnote{Here are its good points
+
+\begin{enumerate}%
+\item Ease of use
+\item Rapid development
+
+\end{enumerate}} .
+
+That has nothing to do with putting equations in footnotes\footnote{Like this:
+
+\begin{displaymath}
+x = r\cos\theta
+\end{displaymath}} .

spec/block_docs/header_after_par.md

@@ -25,27 +25,27 @@ md_el(:document,[
 *** Output of to_html ***
 <p>Paragraph</p>
 
-<h3 id="header_1">header 1</h3>
+<h3 id="header_1_1">header 1</h3>
 
 <p>Paragraph</p>
 
-<h2 id="header_2">header 2</h2>
+<h2 id="header_2_2">header 2</h2>
 
 <p>Paragraph</p>
 
-<h1 id="header_3">header 3</h1>
+<h1 id="header_3_3">header 3</h1>
 *** Output of to_latex ***
 Paragraph
 
-\hypertarget{header_1}{}\subsubsection*{{header 1}}\label{header_1}
+\hypertarget{header_1_1}{}\subsubsection*{{header 1}}\label{header_1_1}
 
 Paragraph
 
-\hypertarget{header_2}{}\subsection*{{header 2}}\label{header_2}
+\hypertarget{header_2_2}{}\subsection*{{header 2}}\label{header_2_2}
 
 Paragraph
 
-\hypertarget{header_3}{}\section*{{header 3}}\label{header_3}
+\hypertarget{header_3_3}{}\section*{{header 3}}\label{header_3_3}
 *** Output of to_md ***
 Paragraph
 

spec/block_docs/headers.md

@@ -20,17 +20,17 @@ md_el(:document,[
 	md_el(:header,["A title with ", md_em(["emphasis"])],{:level=>4},[])
 ],{},[])
 *** Output of to_html ***
-<h1 id="a_title_with_emphasis">A title with <em>emphasis</em></h1>
+<h1 id="a_title_with_emphasis_1">A title with <em>emphasis</em></h1>
 
-<h2 id="a_title_with_emphasis">A title with <em>emphasis</em></h2>
+<h2 id="a_title_with_emphasis_2">A title with <em>emphasis</em></h2>
 
-<h4 id="a_title_with_emphasis">A title with <em>emphasis</em></h4>
+<h4 id="a_title_with_emphasis_3">A title with <em>emphasis</em></h4>
 *** Output of to_latex ***
-\hypertarget{a_title_with_emphasis}{}\section*{{A title with \emph{emphasis}}}\label{a_title_with_emphasis}
+\hypertarget{a_title_with_emphasis_1}{}\section*{{A title with \emph{emphasis}}}\label{a_title_with_emphasis_1}
 
-\hypertarget{a_title_with_emphasis}{}\subsection*{{A title with \emph{emphasis}}}\label{a_title_with_emphasis}
+\hypertarget{a_title_with_emphasis_2}{}\subsection*{{A title with \emph{emphasis}}}\label{a_title_with_emphasis_2}
 
-\hypertarget{a_title_with_emphasis}{}\paragraph*{{A title with \emph{emphasis}}}\label{a_title_with_emphasis}
+\hypertarget{a_title_with_emphasis_3}{}\paragraph*{{A title with \emph{emphasis}}}\label{a_title_with_emphasis_3}
 *** Output of to_md ***
 # A title with *emphasis* #
 

spec/block_docs/html_inline.md

@@ -6,11 +6,17 @@ One
 <span></span>123
 
 <span></span>123
+
+<animateColor/>123
 *** Output of inspect ***
 md_el(:document,[
 	md_par(["One ", md_html("<span></span>"), "123"]),
-	md_par([md_html("<span></span>"), "123"])
+	md_par([md_html("<span></span>"), "123"]),
+	md_html("<animateColor/>"),
+	md_par("123")
 ],{},[])
 *** Output of to_html ***
 <p>One <span></span>123</p>
 <p><span></span>123</p>
+<animateColor></animateColor>
+<p>123</p>

spec/block_docs/issue20.md

@@ -6,4 +6,4 @@ Just an attribute list on its own in a header is probably really content. https:
 *** Output of inspect ***
 md_el(:document, md_el(:header, "{hi}", {:level=>1}))
 *** Output of to_html ***
-<h1 id="hi">{hi}</h1>
+<h1 id="hi_1">{hi}</h1>

spec/block_docs/issue29.md

@@ -6,4 +6,4 @@ Generating ids for unicode headers. Other markdown implementations drop the non-
 *** Output of inspect ***
 md_el(:document, md_el(:header, "Übersicht", {:level=>1}))
 *** Output of to_html ***
-<h1 id='bersicht'>Übersicht</h1>
+<h1 id='bersicht_1'>Übersicht</h1>

spec/block_docs/issue74.md

@@ -26,7 +26,7 @@ md_el(:document, [
      	md_el(:header, "Recents Post", {:level=>2})
      ])
 *** Output of to_html ***
-<h2 id="cool_title">Cool Title</h2>
+<h2 id="cool_title_1">Cool Title</h2>
 <ul>
     <li>Cool Text yada yada</li>
     <li>ICool Tex 2 yada yada</li>
@@ -35,4 +35,4 @@ md_el(:document, [
     <a href="" class="stumbleupon" target="_blank"></a>
     <a href="" class="stumbleupon" target="_blank"></a>
 </p>
-<h2 id="recents_post">Recents Post</h2>
+<h2 id="recents_post_2">Recents Post</h2>

spec/block_docs/issue79.md

@@ -1,13 +1,16 @@
-PENDING - Non-ASCII characters in a PRE tag. See https://github.com/bhollis/maruku/issues/79
+Encoding: utf-8
+Non-ASCII characters in a PRE tag. See https://github.com/bhollis/maruku/issues/79
 *** Parameters: ***
 {}
 *** Markdown input: ***
+<p>á é í ó ú</p>
+
 <pre>
 á é í ó ú
 </pre>
 *** Output of inspect ***
-md_el(:document, md_html("<pre>\ná é í ó ú\n</pre>"))
+md_el(:document, [md_html("<p>á é í ó ú</p>"), md_html("<pre>\ná é í ó ú\n</pre>")])
 *** Output of to_html ***
-<pre>
+<p>á é í ó ú</p><pre>
 á é í ó ú
 </pre>