VRChronicle 1.2.3

Small tag boxes stay the size you drew them

Reported as "with more than four tags on a photo the fifth is always a fixed size — at least sometimes". It was neither intermittent nor about the number of tags.

The test for was that a click or a drag? was written as a fraction of the image rather than as a gesture in screen pixels. On a 3840×2160 screenshot displayed in a roughly 1550 px viewer, that meant any box under about 31 px wide or 17 px tall was treated as a click and replaced with a head-sized default box. The more people are in the shot, the smaller each head is — which is exactly why it showed up once there were several tags to place.

Zooming in to draw a small box made it worse rather than better: the box grows on screen, but its fraction of the image stays the same, so it still snapped.

Click versus drag is a question about the gesture, so it is now measured where the gesture happens — 5 screen pixels. Anything above that keeps exactly the box you drew, however small it is on the photo, and a one-axis drag is padded so it cannot be stored as a sliver. A genuine click still gets the head-sized default.

Tests 92 → 95. On the old rule a 12 px drag came back as a 110 px box.