Automation on Canvas Grid
Hi Web Application we are trying to Automate has a grid which as per Spy is Canvas Object. Object Spy can only see the whole Canvas as one object and cant see anything at row level or Cell level. Doe...
Forum Discussion
Hi mspatel
In continuation to my previous reply add your java fx class name under
Tools->CurrentProjectProperties->OpenApplications->JavaFx . Click on Add button and add your canvas class(ex: javafx.scene.canvas.Canvas) .Click on the checkboxes under active column for the canvas class and click on derived classes checkbox. Save the project refresh object browser. Hope this might be of some help.
Refer to the screenshot attached
I'm not sure about the JavaFX version, but Canvas objects in Delphi (where I have to work with them) are essentially just an image file. It has no child nodes or objects. All the "controls" (which are just part of the one large image) within it are all virtualized and use a reference file. I can't extract anything much useful from the Canvas itself.
I ended up having to use OCR to do some basic checks against it. But, thankfully, I don't have to interact with it too much.
Maybe it's different for JavaFX ones. But they do still appear to be drawn objects.
The support article for Java FX doesn't mention Canvas as being a class supported by TestComplete, so if it is an image, you may not get much more out of it.
Yea...
TestComplete's documentation mentions only image-based comparisons for canvas objects (for HTML5). And a quick googling returned a set of (useful) articles about canvas. But all these articles outlines that canvas is closed approach that heavily relies on scripting and exposes nothing. (On the contrary to, say, SVG that has a means to access and reuse the data.)
But there is still a chance:) that Helen (HKosova) will provide some inspiring idea.
"TestComplete's documentation mentions only image-based comparisons for canvas objects (for HTML5)."
AlexKaras .... I would also include OCR, regardless what the documentation says.
For my Delphi canvas objects, they are actually usable (with limitations) using OCR. I do also use a couple of tiny image compares within them, but I get far more useful things done with them using OCR.
