[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [cp-patches]: FYI: GapContent fix
From: |
Roman Kennke |
Subject: |
Re: [cp-patches]: FYI: GapContent fix |
Date: |
Wed, 07 Sep 2005 21:40:18 +0200 |
> > Lillian> - if (mark <= gapStart)
> > Lillian> + if (mark <= gapStart || mark <= gapEnd)
> >
> > Isn't gapEnd >= gapStart?
> > So the first test is redundant?
Sorry, I think the first version (mark <= gapStart) is correct as it
was. The mark should always be 0<=mark<gapStart || gapEnd <= mark <=
buffer.length. If it isn't then there is a problem at another place
(very likely in shiftGap or shiftEnd, where the marks are updated). This
is a typical case where I think an assert statement makes sense, that is
to check the precondition. Also this is a case where we seriously need a
Mauve test to check for regressions.
> > Also, in passing, I wonder whether GapContent should use an ArrayList
> > rather than a LinkedList to store the list of positions, for
> > efficiency.
Yeah, this is very well possible. I have not thought about this deep
enough when writing this code.
/Roman
signature.asc
Description: This is a digitally signed message part