Hi, I've realized that the scrolling behavior in View() is unlike most other applications. It seems that the minimum scroll size is very large for large objects( i.e., a vector of 1M integers), and it might be that this scroll is proportional to the size of the object. Try this: a = 1:1000000 View(a) Click once on the scroll arrow down. It will move to item 1001. It could be that the scroll size is 1/1000 object length. If this is so, it's a bit of a problem, since there's no way to View say cell 500... unless one does a View(head(a, 500)), which kind of defeats the point of view. Is this behavior by design? If so, wouldn't it be better to make it optional (as it breaks GUI conventions of mostly any OS for scrolling behavior)? An argument for view to turn this off would be great. Thanks! Best, -Jose -- Jose Quesada, PhD. Max Planck Institute, Center for Adaptive Behavior and Cognition -ABC-, Lentzeallee 94, office 224, 14195 Berlin http://www.josequesada.name/
On 07/06/2009 9:50 AM, Jose Quesada wrote:> Hi, > > I've realized that the scrolling behavior in View() is unlike most other > applications. It seems that the minimum scroll size is very large for > large objects( i.e., a vector of 1M integers), and it might be that this > scroll is proportional to the size of the object. > Try this: > a = 1:1000000 > View(a) > Click once on the scroll arrow down. It will move to item 1001. > > It could be that the scroll size is 1/1000 object length. > > If this is so, it's a bit of a problem, since there's no way to View say > cell 500... unless one does a View(head(a, 500)), which kind of defeats > the point of view. > > Is this behavior by design? If so, wouldn't it be better to make it > optional (as it breaks GUI conventions of mostly any OS for scrolling > behavior)? An argument for view to turn this off would be great.You didn't say which version or platform you were using. I see it on Windows 2.9.0 and R-devel. The behaviour you describe (scrolling 1/1000 of the object) happens when the object has more than 10000 items in it. If I temporarily disable it, things are worse: then there is no easy way I can find to get to the bottom of the object. There appear to be some 16 bit overflows happening if you try to drag the scrollbar slider down too far. The reason for this is that Windows only provides 16 bits of information in the scrolling message; however, there are workarounds to get 32 bit positions. I'll see if I can add them to our code. Duncan Murdoch
On 07/06/2009 9:50 AM, Jose Quesada wrote:> Hi, > > I've realized that the scrolling behavior in View() is unlike most other > applications. It seems that the minimum scroll size is very large for > large objects( i.e., a vector of 1M integers), and it might be that this > scroll is proportional to the size of the object. > Try this: > a = 1:1000000 > View(a) > Click once on the scroll arrow down. It will move to item 1001. > > It could be that the scroll size is 1/1000 object length. > > If this is so, it's a bit of a problem, since there's no way to View say > cell 500... unless one does a View(head(a, 500)), which kind of defeats > the point of view. > > Is this behavior by design? If so, wouldn't it be better to make it > optional (as it breaks GUI conventions of mostly any OS for scrolling > behavior)? An argument for view to turn this off would be great.This should be fixed now. Please try R-patched dated June 8 or later, and let me know if there are any more problems. Duncan Murdoch
Reasonably Related Threads
- alternative to built-in data editor
- a simple trick to get autoclose parenthesis on windows
- dgTMatrix --- [, , drop=F] strange behavior, Matrix 0.999375-20
- How do I set the working directory when using ESS and R under windows?
- Rmysql linking to an old-ish mysql build