From 82dbddfd19e4a4d7e35910ad49e54cc83638459f Mon Sep 17 00:00:00 2001
From: Wouter Scherphof <wouter.scherphof@gmail.com>
Date: Thu, 21 Mar 2013 14:03:17 +0100
Subject: [PATCH] parse valueless attributes (e.g. itemscope)

And allow nested tags, which was a newly encountered bug
---
 htmlparser.lua | 25 ++++++++++++++-----------
 test.html      | 18 ++++++++++++++++++
 2 files changed, 32 insertions(+), 11 deletions(-)

diff --git a/htmlparser.lua b/htmlparser.lua
index be80013..bcc4ad0 100644
--- a/htmlparser.lua
+++ b/htmlparser.lua
@@ -16,14 +16,18 @@ local function parse(text)
 
     local tagst, apos = tag:gettext(), 1
     while true do
-      local start, k, quote, v
-      start, apos, k, quote = string.find(tagst, "%s+([^%s=]+)=(['\"]?)", apos)
+      local start, k, eq, quote, v
+      start, apos, k, eq, quote = string.find(tagst, "%s+([^%s=]+)(=?)(['\"]?)", apos)
       if not k then break end
-      local pattern = "=([^%s'\">]*)"
-      if quote ~= '' then
-        pattern = quote .. "([^" .. quote .. "]*)" .. quote
+      if eq == "" then
+        v = ""
+      else
+        local pattern = "=([^%s'\">]*)"
+        if quote ~= '' then
+          pattern = quote .. "([^" .. quote .. "]*)" .. quote
+        end
+        start, apos, v = string.find(tagst, pattern, apos)
       end
-      start, apos, v = string.find(tagst, pattern, apos)
       tag:addattribute(k, v)
     end
 
@@ -31,17 +35,16 @@ local function parse(text)
       descend = false
       tag:close()
     else
-      opentags[tag.name] = tag
+      opentags[tag.name] = opentags[tag.name] or {}
+      table.insert(opentags[tag.name], tag)
     end
 
     local closeend = tpos
     while true do
       local closestart, closing, closename
       closestart, closeend, closing, closename = string.find(root._text, "[^<]*<(/?)(%w+)", closeend)
-      closing = closing and closing ~= ''
-      if not closing then break end
-      tag = opentags[closename]
-      opentags[closename] = nil
+      if not closing or closing == "" then break end
+      tag = table.remove(opentags[closename])
       closestart = string.find(root._text, "<", closestart)
       tag:close(closestart, closeend + 1)
       node = tag.parent
diff --git a/test.html b/test.html
index 3e86fb5..c3ea16d 100644
--- a/test.html
+++ b/test.html
@@ -39,5 +39,23 @@
       </a>
     </li>
   </ul>
+<section itemscope itemtype="http://schema.org/Person"> 
+  Hello, my name is 
+  <span itemprop="name">John Doe</span>,
+  I am a 
+  <span itemprop="jobTitle">graduate research assistant</span> 
+  at the 
+  <span itemprop="affiliation">University of Dreams</span>. 
+  My friends call me 
+  <span itemprop="additionalName">Johnny</span>. 
+  You can visit my homepage at 
+  <a href="http://www.JohnnyD.com" itemprop="url">www.JohnnyD.com</a>. 
+  <section itemprop="address" itemscope itemtype="http://schema.org/PostalAddress">
+    I live at 
+    <span itemprop="streetAddress">1234 Peach Drive</span>,
+    <span itemprop="addressLocality">Warner Robins</span>,
+    <span itemprop="addressRegion">Georgia</span>.
+  </section>
+</section>
 </body>
 </html>
\ No newline at end of file