From 00150938dd9a22db637dd1c4e1742b1183a03e8b Mon Sep 17 00:00:00 2001
From: King_DuckZ <king_duckz@gmx.com>
Date: Mon, 28 Sep 2015 22:59:09 +0200
Subject: [PATCH] Fix the html cleaning code that was not really cleaning.

---
 src/htmlretrieve.cpp | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/src/htmlretrieve.cpp b/src/htmlretrieve.cpp
index ac15768..b560d4f 100644
--- a/src/htmlretrieve.cpp
+++ b/src/htmlretrieve.cpp
@@ -29,13 +29,10 @@ namespace duck {
 
 		std::string cleanHTML (std::string&& html) {
 			dropScriptTags(html);
-			std::unique_ptr<char[]> html_copy(new char[html.size()]);
-			std::copy(html.begin(), html.end(), html_copy.get());
 
 			// Initialize a Tidy document
 			TidyDoc tidyDoc = tidyCreate();
-			TidyBuffer tidyOutputBuffer;
-			tidyBufInit(&tidyOutputBuffer);
+			TidyBuffer tidyOutputBuffer = {nullptr, nullptr, 0, 0, 0};
 
 			// Configure Tidy
 			// The flags tell Tidy to output XML and disable showing warnings
@@ -48,14 +45,18 @@ namespace duck {
 
 			// Parse input
 			if (configSuccess) {
-				tidyBufAppend(&tidyOutputBuffer, html_copy.get(), html.size());
-				tidyResponseCode = tidyParseBuffer(tidyDoc, &tidyOutputBuffer);
+				tidyResponseCode = tidyParseString(tidyDoc, html.c_str());
 			}
 
 			// Process HTML
 			if (tidyResponseCode >= 0)
 				tidyResponseCode = tidyCleanAndRepair(tidyDoc);
 
+			if (tidyResponseCode >= 0)
+				tidyResponseCode = tidyRunDiagnostics(tidyDoc);
+			if (tidyResponseCode > 1)
+				tidyResponseCode = (tidyOptSetBool(tidyDoc, TidyForceOutput, yes) ? tidyResponseCode : -1);
+
 			// Output the HTML to our buffer
 			if (tidyResponseCode >= 0)
 				tidyResponseCode = tidySaveBuffer(tidyDoc, &tidyOutputBuffer);