Re: What should be check in Lexical Analyzer along with generating tokens?

"Joachim Durchholz" <joachim_d@gmx.de> writes:


> Clint Olsen wrote:
> >
> > /* parse !$#%$# this! :) */
> >
> > I don't see how by your example the above snippet would work.
>
> The expletive would probably go into the <error> token. Which is there
> exactly for the purpose of parsing stuff that the scanner doesn't like.


Here's a harder example:




/* parse "this */ 10 /* and this" */




It will scan as four tokens:


      /*
      parse
      "this */ 10 /* and this"
      */




The parser will be in trouble now if it wants to pull out the "10".
Thus, you seem to need a special scanner state to handle the meat of a
comment.


-Lex

Return to the comp.compilers page.
Search the comp.compilers archives again.