From 44992458ac99e2183eed3e6d054afb54ed1c21f8 Mon Sep 17 00:00:00 2001
From: King_DuckZ <king_duckz@gmx.com>
Date: Sun, 1 Mar 2015 05:03:12 +0100
Subject: [PATCH] Quick dirty fix to avoid invalid characters in scripts.

Note that with this change scritps are stripped away, so
you won't find any <script></script> pair in the html.

Also print some more detailed info about errors.
---
 src/htmlretrieve.cpp | 30 ++++++++++++++++++++++++++----
 src/main.cpp         | 33 ++++++++++++++++++++++++++++++++-
 2 files changed, 58 insertions(+), 5 deletions(-)

diff --git a/src/htmlretrieve.cpp b/src/htmlretrieve.cpp
index d1206b8..c8982ae 100644
--- a/src/htmlretrieve.cpp
+++ b/src/htmlretrieve.cpp
@@ -8,14 +8,34 @@
 #include <cstring>
 #include <stack>
 #include <algorithm>
+#include <memory>
+#include <cassert>
 
 namespace duck {
 	namespace {
-		std::string cleanHTML(const std::string &html) {
+		void dropScriptTags (std::string& html) {
+			size_t open_index = 0;
+			const std::string open_tag("<script");
+			const std::string close_tag("</script>");
+
+			while (html.npos != (open_index = html.find(open_tag, open_index))) {
+				assert(open_index < html.size());
+				auto close_index = html.find(close_tag, open_index + open_tag.size());
+				if (close_index == html.npos)
+					close_index = html.size();
+				html.erase(open_index, std::min(html.size(), close_index + close_tag.size()) - open_index);
+			}
+		}
+
+		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;
-			std::memset(&tidyOutputBuffer, 0, sizeof(TidyBuffer));
+			tidyBufInit(&tidyOutputBuffer);
 
 			// Configure Tidy
 			// The flags tell Tidy to output XML and disable showing warnings
@@ -27,8 +47,10 @@ namespace duck {
 			int tidyResponseCode = -1;
 
 			// Parse input
-			if (configSuccess)
-				tidyResponseCode = tidyParseString(tidyDoc, html.c_str());
+			if (configSuccess) {
+				tidyBufAppend(&tidyOutputBuffer, html_copy.get(), html.size());
+				tidyResponseCode = tidyParseBuffer(tidyDoc, &tidyOutputBuffer);
+			}
 
 			// Process HTML
 			if (tidyResponseCode >= 0)
diff --git a/src/main.cpp b/src/main.cpp
index 58fde18..abb9160 100644
--- a/src/main.cpp
+++ b/src/main.cpp
@@ -4,6 +4,14 @@
 #include <string>
 #include <pugixml.hpp>
 #include <sstream>
+#include <utility>
+#include <ciso646>
+
+namespace {
+	typedef std::pair<int, int> LineColType;
+
+	LineColType line_col_from_offset ( ptrdiff_t parOffset, const std::string& parData );
+} //unnamed namespace
 
 int main (int argc, char* argv[]) {
 	if (argc != 3) {
@@ -30,7 +38,10 @@ int main (int argc, char* argv[]) {
 		std::istringstream iss(tidyHtml);
 		pugi::xml_parse_result result(doc.load(iss));
 		if (not result) {
-			std::cerr << "Error parsing the source XML";
+			auto line_col = line_col_from_offset(result.offset, tidyHtml);
+			std::cerr << "Error parsing the source XML at line " <<
+				line_col.first << " col " << line_col.second << ":\n" <<
+				result.description() << std::endl;
 			return 1;
 		}
 
@@ -47,3 +58,23 @@ int main (int argc, char* argv[]) {
 	}
 	return 0;
 }
+
+namespace {
+	LineColType line_col_from_offset (ptrdiff_t parOffset, const std::string& parData) {
+		size_t index = 0;
+		int line = 1;
+		int chara = 1;
+		while (parOffset and index < parData.size()) {
+			if (parData[index] == '\n') {
+				chara = 1;
+				++line;
+			}
+			else {
+				++chara;
+			}
+			++index;
+			--parOffset;
+		}
+		return std::make_pair(line, chara);
+	}
+} //unnamed namespace