Hi Guys,
The !makingRoomForNew clause needs to be there. For this
case, since the buffer is being re-initialized, InitBuffer
has to be called (which happens on line 569) instead of
ActivateBuffer.
Actually I don't think OnSwitchFile should be called in this
case, as it is not supposed to be called for new buffers -
only for switching to an existing buffer. So I think the
correct fix is probably on line 645:
if (extender && !closingLast && !makingRoomForNew) {
extender->OnSwitchFile(filePath.AsFileSystem());
}
Regards,
Bruce
"Neil Hodgson" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> [Redirecting to mailing list as it looks like a Bruce
> Dodson issue]
>
> Kein-Hong Man:
>
>> Got a reproducible test case:
>> (1) Open the maximum number of buffers
>> (2) Create a new buffer
>>
>> Bug is due to the following:
>> (1) SciTEBuffers.cxx[549], SciTEBase::New:
>> Close() called with makingRoomForNew=true
>> (2) SciTEBuffers.cxx[595-599], SciTEBase::Close:
>> In 596, extender->RemoveBuffer is called:
>> In LuaExtension.cxx, curBufferIndex set to -1 in
>> anticipation of ActivateBuffer call.
>> In 599, extender->ActivateBuffer is called, but due
>> to the condition (extender && !makingRoomForNew),
>> this
>> never happens and curBufferIndex is still -1.
>
> It looked to me like removing the !makingRoomForNew
> clause would
> fix this but its an explicit piece of code (that is, it
> has no point
> unless it avoids the call) and I'm not sure about the
> reasoning behind
> it so it would be best for Bruce to OK removing it.
>
>> (3) So upon buffer initialization, the error message
>> from line 311 in LuaExtension.cxx appears.
>
> Neil
_______________________________________________
Scite-interest mailing list
[email protected]
http://mailman.lyra.org/mailman/listinfo/scite-interest
