The scrollbar in my application is just taking 2 positions and nothing in between, that make it look silly to be honest... is there a way to raise its precision?
First position
Second position
Have a look at https://doc.qt.io/archives/qt-4.8/qabstractitemview.html#horizontalScrollMode-prop There you can set QAbstractItemView::ScrollPerPixel
The answer of V.K. is correct, I just want to give you the link to the current release document for Qt 5 to setHorizontalScrollMode and use as parameter QAbstractItemView::ScrollPerPixel
Related
I found this interesting animation and wonder if it's possible to repeat it in the PaperJS? It seems like it could be made of a group of circles, rotated around another circle.
But the problem is - when these objects are in a group I can't use sendToBack() per each, it's not going to work. Here is the Sketch example, that doesn't work.
P.S. Here is a simplified version of how it could be achieved, but it works only for 1 element, and only for animation along one axis.
If you still want to have your object grouped, a basic solution could be to have 2 groups: one below and one above the circle and toggle items visibility.
Just to demonstrate the idea, this sketch extends your code in this way.
I think that the way you currently decide if the item should be below or above can be improved though for a better result (you could try to detect intersection between the item and the circle rather than only checking the position).
I don't know if there is some calculation or something like that, but the first / last line (depending on the orientation setting) when groups are used becomes higher when an entry is added.
First group (line) is higher
When an entry is in other line, the height is as it should be
is this known ? why it happens ?
thanks a lot for any help
regards
Marek
thanks to Jos de Jong personally, I got a solution. It's answered here:
https://github.com/almende/vis/issues/1388
Jos, thanks! ;)
I'm trying to make a subclass of QSlider that allows the user to insert "bookmarks" so they can remember significant locations on the slider. I've run into a problem painting the tabs on the slider - using QStyle.sliderPositionFromValue, I get a value but it is slightly inaccurate. If I'm on the left side of the slider, the tab is painted too far left, and on the right side it is painted too far right. I believe this is because QSlider.width() returns the width of the whole object, including the small offsets at the left and right. So width() might return 630 pixels, when the length of the slider itself is really 615.
This is the code I'm using to get the pixel position and draw a line across the slider.
pos = QStyle.sliderPositionFromValue(self.minimum(),self.maximum(),sliderIndex,self.width())
painter.drawLine(pos,0,pos,self.height())
I've been looking at the QT Source here starting on line 2699 and it seems like I need to be using the PixelMetric class from QStyle. I've tried using:
self.style().pixelMetric(QStyle.PM_SliderSpaceAvailable)
But that returns 0, which is clearly not the value I need.
I'd appreciate any advice.
Edit: As suggested in the comments, I changed the call to:
self.style().pixelMetric(QStyle.PM_SliderSpaceAvailable, QStyleOption(1,QStyleOption.SO_Slider),self)
This however, returns -14, which also doesn't match for the value of the offsets (I tried using self.width()-14 but the offset remains.
I want to draw a tick mark for a qslider in paintEvent, so I should get the exact position of each tick mark. As you know, the handle of slider takes some space, so the first tick mark does not sit at the left/top position of the slider, there is offset of several pixels. Same thing happens to the last tick mark which indicates the maximum value. I want to know how many pixels of the space? (On win and mac, the handle are not in the same width)
If you have not yet done so, download the Qt source code and copy how they do it. You will want to look in the various Q...Style classes, i.e QMacStyle, QWindowsXPStyle, etc. Some of the key calculations come from:
QStyle::sliderPositionFromValue() method
QStyle::pixelMetric(QStyle::PM_SliderTickmarkOffset, ...)
Look in the various drawComplexControl methods for case CC_Slider:, where slider controls are drawn. In the Qt 4.7 code, this starts on line 2699 in qwindowsxpstyle.cpp, for example.
I want to generate a cube where each face is divided into bits, like the following image:
http://img59.imageshack.us/img59/2504/gridcube165c3.jpg
Now, I can do this pretty simply if I'm just rendering quads, by just spacing vertices along each face plane at regular intervals, but my problem comes in when I want to turn the whole thing into a triangle strip. I've just got no idea how to unwrap it programmatically- is there some pattern to unwrapping that I'd follow?
I'm thinking of starting with the vertex at the top left corner as Row 0 Column 0 (R0C0), I'd want (first triangle) R0C0, ROC1, R1C1, (second triangle) R0C0, R1C0, R1C1 and so forth, and then when I reach the end of a row I guess I'd use a degenerate triangle to move to the next row, and then when I reach the end of the face I'd do the same to start a new face.
My main problem is that I can't visualize the program loop that would do this. I can reason out which vertex comes next visually, which is how I worked out the order above, but when I try to think programmatically I just stare blankly.
Even worse, with the end product I want the generated cube to be UV-mapped with a simple cube-map unwrap (the kind that looks like a T or t).
I guess, really, the best solution would be to find a library that already does this for me.
You could take a look at Ignacio Castaño's 'Optimal Grid Rendering' even though it's not triangle strips, it may inspire you.
Otherwise, you could use NVTriStrip library and be done with it.