Merge remote-tracking branch 'origin/topic/bernhard/input-fixes'

* origin/topic/bernhard/input-fixes:
  fix problem with possible access to unititialized memory (thanks robin :) )
  and just to be a little bit careful - add check if the field description is long enough. Otherwise there might possibly be an access of uninitialized memory, when someone reads a file that contains just #fields without any following field descriptions.
  and like nearly always - forgot the baseline.
  Input framework now accepts escaped ascii values as input.
  make reading ascii logfiles work when the input separator is different from \t.

src/input/readers/Ascii.cc

@@ -144,7 +144,7 @@ bool Ascii::ReadHeader(bool useCached)
 		pos++;
 		}
 
-	//printf("Updating fields from description %s\n", line.c_str());
+	// printf("Updating fields from description %s\n", line.c_str());
 	columnMap.clear();
 
 	for ( int i = 0; i < NumFields(); i++ )
@@ -199,7 +199,7 @@ bool Ascii::GetLine(string& str)
 		if ( str[0] != '#' )
 			return true;
 
-		if ( str.compare(0,8, "#fields\t") == 0 )
+		if ( ( str.length() > 8 ) && ( str.compare(0,7, "#fields") == 0 ) && ( str[7] == separator[0] ) )
 			{
 			str = str.substr(8);
 			return true;
@@ -438,6 +438,8 @@ bool Ascii::DoUpdate()
 			if ( ! getline(splitstream, s, separator[0]) )
 				break;
 
+			s = get_unescaped_string(s);
+
 			stringfields[pos] = s;
 			pos++;
 			}

src/util.cc

@@ -42,6 +42,44 @@
 #include "Net.h"
 #include "Reporter.h"
 
+/**
+ * Takes a string, unescapes all characters that are escaped as hex codes
+ * (\x##) and turns them into the equivalent ascii-codes. Returns a string
+ * containing no escaped values
+ *
+ * @param str string to unescape
+ * @return A str::string without escaped characters.
+ */
+std::string get_unescaped_string(const std::string& arg_str)
+	{
+	const char* str = arg_str.c_str();
+	char* buf = new char [arg_str.length() + 1]; // it will at most have the same length as str.
+	char* bufpos = buf;
+	size_t pos = 0;
+
+	while ( pos < arg_str.length() )
+		{
+		if ( str[pos] == '\\' && str[pos+1] == 'x' &&
+		     isxdigit(str[pos+2]) && isxdigit(str[pos+3]) )
+			{
+				*bufpos = (decode_hex(str[pos+2]) << 4) +
+					decode_hex(str[pos+3]);
+
+				pos += 4;
+				bufpos++;
+			}
+		else
+			*bufpos++ = str[pos++];
+		}
+
+	*bufpos = 0;
+	string outstring(buf, bufpos - buf);
+
+	delete [] buf;
+
+	return outstring;
+	}
+
 /**
  * Takes a string, escapes characters into equivalent hex codes (\x##), and
  * returns a string containing all escaped values.
@@ -53,7 +91,7 @@
  * @return A std::string containing a list of escaped hex values of the form
  * \x## */
 std::string get_escaped_string(const std::string& str, bool escape_all)
-{
+	{
 	char tbuf[16];
 	string esc = "";
 
@@ -71,7 +109,7 @@ std::string get_escaped_string(const std::string& str, bool escape_all)
 		}
 
 	return esc;
-}
+	}
 
 char* copy_string(const char* s)
 	{

src/util.h

@@ -90,6 +90,7 @@ void delete_each(T* t)
 		delete *it;
 	}
 
+std::string get_unescaped_string(const std::string& str);
 std::string get_escaped_string(const std::string& str, bool escape_all);
 
 extern char* copy_string(const char* s);

testing/btest/Baseline/scripts.base.frameworks.input.binary/out

@@ -0,0 +1,6 @@
+abc^J\xffdef
+DATA2
+abc|\xffdef
+DATA2
+abc\xff|def
+DATA2