< Previous by Date Date Index Next by Date >
< Previous in Thread Thread Index Next in Thread >

Re: [reSIProcate] ParseBuffer::assertNotEof() ?




Jason Fischl wrote:

On 10/25/06, Daniel Pocock <daniel@xxxxxxxxxxxxxxxxxxxxx> wrote:



Byron Campen wrote:

>     That is not valid syntax, no. However, it would make sense to
> interpret this as the case where there is no tag parameter, and if
> this is something that a TU cannot live with, then it can be rejected
> there. Any objections?
>
I'm still seeing this behaviour in the latest code from SVN.

Does anyone object if I patch this to allow the tag= nothing syntax, or
is there another preferred solution?


How/where are you going to address the issue? I don't think that the
parser should be changed to allow tag= syntax but I agree that dum
should reject the request.


Ok, I'm quite happy to have it reject it - and hopefully log something too. Should assertNotEof() be modified to throw an exception instead of just calling assert(0)?

Which exception should be thrown, or should I just create one? Is there already a try{} catch{} block to handle this, or does this need to be done too?

Unfortunately I've already deleted my core dumps from today - next time this happens I'll post a backtrace.