discuss-gnustep
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: ANN: Price 0.4.0


From: Fred Kiefer
Subject: Re: ANN: Price 0.4.0
Date: Mon, 21 Jun 2004 13:25:10 +0200
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040114

Riccardo Mottola wrote:
It is a bug in GNUstep. Alexander Malmberg had a look at the time when I
discovered it and at least fixed the memory consumption... but no one really
fixed this.

The same code works as expected in MacOS-X.

The expected behavior is that typing in "100%" should yield you always the
original image size (sometimes it doesn't if the original image wasn't
72dpi, but it is at least consistent, typing 100% will bring the correct
image size back anytime after any zoomings, in and out) and from what I read
in the documentation it is the correct behavior and gnustep is worng, but
not everybody seems to agree (at least, when I posed the problem in the IRC
channel months ago there were disagreements) and the bug remains.


It really would help, if you could open a bug report for this on savannah. I, and perhaps other GNUstep developers, do not follow the discussions on the IRC, so would never get an update on this.

As far as I could analyse the problem comes from GNUstep changing the image size, when scaling the NSImageView (or rather the NSImageCell), whereas Cocoa seems to maintain the same image size. (Could you please verify this assumption on MacOSX?) The next time you compute the new view frame the already changed image size is used and so on.

As we currently dont support the method [NSImage -drawInRect:fromRect:operation:fraction:) in the backends the only workaround I see it so reset the image siez at the end of the (NSImageCell -drawInteriorWithFrame:inView:) method, which would be a horrible hack. Perhaps somebody steps up to implement the image drawing method to prevent this from happening?

Fred




reply via email to

[Prev in Thread] Current Thread [Next in Thread]