I'm attempting to use a grammar originally generated to Java, which contains a number of lexer actions that behave differently depending on the current mode / contents of the mode stack. This patch simply exposes the mode_stack field, just as it would be exposed to subclasses in a generated Java lexer.
I'm attempting to use a grammar originally generated to Java, which contains a number of lexer actions that behave differently depending on the current mode / contents of the mode stack. This patch simply exposes the
mode_stackfield, just as it would be exposed to subclasses in a generated Java lexer.