[antlr-interest] Lexer bug?

Loring Craymer lgcraymer at yahoo.com
Mon Oct 22 12:08:43 PDT 2007


Jim--

I think that you represent the pragmatic "make it work now" view fairly well.  I also think that the exchange

> Jim Idle wrote:


> > This isn't a bug.


> 


> Nonsense. Any lexer that consumes characters that aren't a legal token,


> and announces a legal token without flagging an error, has a bug.


represents a real frustration with the defect recognition process.  It took Ter well over a year to fix the premature optimization of synpreds during which time the mantra was "This isn't a bug"; meanwhile, there were grammars which could not be implemented in ANTLR 3.  I suspect that Austin's, Gavin's and my reaction would be very different if the response had been "Hmm.  I do not see how to fix this right now and am not sure I want to, but I will add it to the bug list.  In the meantime ...".  The point of making bug reports is not "Help!  I need a workaround!" as a rule, although that may be part of it, but rather "If this were fixed, it would improve the utility and user-friendliness of ANTLR".  The LL* algorithm is very nice, but there are still some rough edges with the implementation that need to be recognized and fixed; these rough edges do not have to be smoothed right away (there are workarounds), but they do need to be recognized so that they
 can eventually be fixed.


--Loring



__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.antlr.org/pipermail/antlr-interest/attachments/20071022/a9e75959/attachment.html 


More information about the antlr-interest mailing list