This seems like a pixel rounding issue in the swf9 kernel when applying
resources. In the future, feel free to file bugs in JIRA - it's a great
place to attach files, etc. That way we can get it prioritized and in
the queue. I filed a bug here:
http://jira.openlaszlo.org/jira/browse/LPP-8764
Regards,
Max Carlson
OpenLaszlo.org
On 9/11/10 11:58 AM, jamesr wrote:
Stableborderlayout also exhibits the same issues in swf9 and DHTML while
operating correctly in swf8. How do people make three piece buttons in
laszlo if not this way? How to make collections of tabs - for instance -
lay side by side without gaps? (that issue - using buttons to make tabs,
has been provided as a screen shot attached, and are all side effects of
the same issue) Anyone want to share an alternate composition technique?
note that in this screen shot, as the size of the canvas changes the
vertical defects you see in the tabs and in the button go away and come
back on canvas size, even when the size of the child views is constant,
so it is not possible to have dynamically sized buttons or components in
swf9 or dhtml and a variable canvas width at the same time. Would some
one care to refute that? Show a method to have have views laid out live
side by side and stay that way under a variable canvas size? It would be
welcome.
-jamesr
P T Withington wrote:
I suspect an issue with stretches vs. resources (there have been a number of bugs in
that area -- not my expertise, so perhaps someone else can comment). But I also
wonder why you are not using<stableborderlayout> (or your own variation on
it), since that will be much more efficient than building your buttons with
constraints?
On 2010-09-10, at 22:20, jamesr wrote:
Having issues with pixellock? I might be, and whether or not it's so, i need to
know what i can do about it. An in-depth example attached -- if gmail lets the
zip attachment through, you should be able to run the example easily by
dropping the folder anywhere in your laszlo my-apps folder or equal.
My apologies if the email does not get to the list intact. I'm trying :)
The problem is summed up with a screen shot (also attached) where what you see
looks good -- so long as you don't resize the window. Then you variably get
this depending on whether the screen is odd/even in the x/y axis (where below,
both are off)
As you drag the window, the bug will become very apparent. I hope that the code
is clear, and hope that someone catches something that i didn't with what
should be (i think!) a common layout algorithm. Is it? is it not? Why do i
get... The Weird Lines? The desired outcome is easy to guess - when the window
is resized, the button should not jigsaw around, esp. since it's width and
height are constant. Perhaps one of the dons of ol could have a look and see if
this is a bug, a feature, or a work-around. Any further clarification will be
answered asap.
Thanks for taking time to look at this, as it seems to be big building block to
building skinnable (not in the css sense) laszlo components. It's been driving
me crazy for some time now.
best,
jamesr
p.s. it's great to finally have a question and a good test-case for the list!
overall the documentation is very good for laszlo in my experience, so my
questions are usually fringe..
<pixellock.zip><scrshot.jpg>
