visgl / deck.gl / 29114138740

83%

push

github

feat(core): upgrade mjolnir.js for DoubleClickDrag recognizer (#10416)

* upgrade mjolnir.js

* fix(core): register dblclickdrag recognizer to recognize with click/dblclick (#10418)

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

* fix(core): eliminate 300ms click delay and fix runtime controller prop changes (#10422)

* fix(core): disable doubleClickZoom by default to eliminate 300ms click delay

The click recognizer's requireFailure dependency on dblclick causes
mjolnir.js to defer click events by 300ms. Disabling doubleClickZoom
by default removes this delay while keeping doubleClickDragZoom
available for zoom-by-drag gestures.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix(core): also disable doubleClickDragZoom by default

Same principle — once #10416 lands, dblclickdrag should be in click's
requireFailure list to prevent spurious clicks before a drag-zoom.
That would reintroduce the 300ms delay, so disable by default too.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix(core): add dblclickdrag to click's requireFailure list

Prevents spurious click events from firing before a double-click-drag
zoom gesture is recognized. Without this, a click would emit on the
first tap before the recognizer determines the user intends to drag.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* test(core): add click delay unit tests for controller gesture config

Tests verify through the full EventManager → recognizer pipeline that:
- Click fires immediately when doubleClickZoom and doubleClickDragZoom
  are both disabled (the new default)
- Click is delayed ~300ms when doubleClickZoom is enabled
- Click is delayed ~300ms when doubleClickDragZoom is enabled

Also reorders RECOGNIZERS so dblclickdrag is registered before click,
which is required for click's requireFailure reference to resolve.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix(core): ensure dblclick recognizer is disabled when double... (continued)

8038 of 10148 branches covered (79.21%)

Branch coverage included in aggregate %.

28 of 28 new or added lines in 2 files covered. (100.0%)

14350 of 16772 relevant lines covered (85.56%)

19147.66 hits per line