Open GoogleCodeExporter opened 9 years ago
It's very easy to fix this (I spotted the 2 lines in AGDefTable responsible for
this behavior). However, the question
is: do we want to fix it? The initialisation expression for table definitions
was designed to be a side-effect-free
expression. On the other hand, it does not really hurt to make a separate
call-frame per evaluation of this
expression. It causes less errors (like the one you mention above), but at the
expense of performance (evaluation
now requires only 1 call frame, as opposed to table.length call frames)
Original comment by tvcut...@gmail.com
on 16 Feb 2009 at 9:19
This feature is not very often... Do we still spend time on it? General
consensus is to kick it out.
Original comment by Botje.linux@gmail.com
on 24 May 2013 at 1:19
This hasn't bitten me over the past few years. I think if we change it, it's
harmless, so if someone wants to take on this bug go ahead. Otherwise let's
just close it and move on.
Original comment by tvcut...@gmail.com
on 24 May 2013 at 2:11
Original issue reported on code.google.com by
smost...@gmail.com
on 14 Feb 2009 at 6:26