[antlr-interest] Segfault in C target on EOF error reporting

[Marco Trudel]

On 03.02.2011 17:01, Jim Idle wrote:
> I have already fixed this I think.

No. libantlr3c-3.3-SNAPSHOT downloaded 30min ago still crashes.

> It is because the EOF token is trying
> to be duplicated or otherwise modified.

It's because recognizer->state->tokenNames is accessed with the EOF 
token type (-1). The garbage returned is then handled as a string and 
strlen segfaults. My patch just specially handles the EOF type. No idea 
if that's the right fix.

> However, the runtime error message
> routine is just an example - you are expected to implement your own that
> does something sensible ;-)

I have seen you telling that to people over and over again. This 
surprises me a little. I would expect that a majority of people just 
want to have default error message and have no time/interest in diving 
into the antlr internals to write it themselves. If someone still has to 
do that, he can invest the time and energy to achieve it.
Actually in 2007 you seemed to have the same idea: "... I am going to 
produce an error message handler that will do for most things ..." [1].

Anyway, actually do have my own "displayRecognitionError" routines. But 
it's just a copy-paste from yours with the difference that I put 
everything into a pANTLR3_STRING so that I can show it to the user in 
the GUI.


Thanks
Marco

[1] http://markmail.org/message/27ezjtsvfiglnuyk


> Jim
>
>> -----Original Message-----
>> From: antlr-interest-bounces at antlr.org [mailto:antlr-interest-
>> bounces at antlr.org] On Behalf Of Marco Trudel
>> Sent: Thursday, February 03, 2011 5:58 AM
>> To: antlr-interest at antlr.org
>> Subject: [antlr-interest] Segfault in C target on EOF error reporting
>>
>> Dear Jim
>>
>> If there's an EOF in the grammar, the C target crashes on input with
>> certain syntax errors. I wrote this quick fix. I hope it at least helps
>> you pinpointing the problem.
>>
>> Any thoughts/comments?
>>
>>
>> Thanks
>> Marco
>
> List: http://www.antlr.org/mailman/listinfo/antlr-interest
> Unsubscribe: http://www.antlr.org/mailman/options/antlr-interest/your-email-address