I guess it's in there somewhere but I have to admit my knowledge of regular expressions and how to define this type of syntax in the language definition is somewhat limited. Can you give an example of how this would be accomplished?
I see this sort of syntax in the C# defintion but not sure how I would adapt this to allow the \' to be considered part of the string and not a string delimiter itself:
<State Key="StringState" Token="StringDefaultToken" Style="StringDefaultStyle">
<Scopes>
<Scope>
<ExplicitPatternGroup Type="StartScope" Token="StringStartToken" Style="StringDelimiterStyle" PatternValue=""" />
<RegexPatternGroup Type="EndScope" Token="StringEndToken" Style="StringDelimiterStyle" PatternValue="[\"\n]" />
</Scope>
</Scopes>
<PatternGroups>
<RegexPatternGroup Token="StringEscapedCharacterToken" PatternValue="\\x {HexDigitMacro}{1,4}" />
<RegexPatternGroup Token="StringEscapedCharacterToken" PatternValue="\\u {HexDigitMacro}{4,4}" />
<RegexPatternGroup Token="StringEscapedCharacterToken" PatternValue="\\U {HexDigitMacro}{8,8}" />
<RegexPatternGroup Token="StringEscapedCharacterToken" PatternValue="\\." />
<RegexPatternGroup Token="StringWhitespaceToken" PatternValue="{WhitespaceMacro}+" IsWhitespace="True" />
<RegexPatternGroup Token="StringWordToken" PatternValue="\w+" />
<RegexPatternGroup Token="StringDefaultToken" PatternValue="[^\"\n]" />
</PatternGroups>
</State>