Bug Tranform in a Macro

[erdi12]

The Transform function uses the center point as the coordinates. 
If it is recorded as a macro, the upper left corner is used as the reference point. 
This means that when the macros are played with another file and a differently dimensioned object, 
a completely different transformation takes place. 
The reference point in the macro recording must also be the center point in order to correct the error.

[walt.farrell]

What kind of transformation are you performing?

I ask because I think you can only transform to a fixed size, no matter what dimension the objects have. Macros in Photo generally record the results of any calculation, not the calculation itself. So, for example, if you have an object that is 500px wide, and you record the macro using w/2, it will record 250px. That makes macros not very useful for resizing files or objects unless they use Filters > Distort > Equations instead of the Transform panel.

I can't comment about the transformation origin, but that problem may not be meaningful to you at this point if you can't use a macro at all for what you want to do.

[erdi12]

The transformation ist only a rotation. Rotation for example 15deg. No additanal calculation is used.

And the Filter Distort with an small objekt on a layer ist not usefull.

[walt.farrell]

Thanks for that added data. Yes, with a rotation it should be OK, if that other problem didn't interfere. Perhaps someone will have an idea about that, and the added detail will provide more information.

[erdi12]

I added a screenshot. The problem is that the reference point of the rotation relative to the object is not recorded and taken into account. A pure rotation in the panel results in a rotation, a distortion and a displacement in the macro. (See screenshot). If this is applied to another object or document, the result is no longer a pure rotation. The relative pivot point selected in the Transformation panel must be recorded and taken into account when playing the macro. The same error also leads to the same behavior in the affine transformation filter if the layer was transformed before the filter was applied. The wrong coordinate system is then used.

[Gabe]

Thanks. Issue logged.