[Pcgen-github] [PCGen/pcgen] 14473b: fix: expand JTreeTable nodes on a single disclosur...

[Vest <no...@gi...>]

  Branch: refs/heads/master
  Home:   https://github.com/PCGen/pcgen
  Commit: 14473b5799bfc73a2d3ff51370a27dadcb2633c3
      https://github.com/PCGen/pcgen/commit/14473b5799bfc73a2d3ff51370a27dadcb2633c3
  Author: Vest <Ve...@us...>
  Date:   2026-09-24 (Thu, 24 Sep 2026)

  Changed paths:
    M code/src/java/pcgen/gui2/util/JTreeTable.java
    A code/src/test/pcgen/gui2/util/JTreeTableExpandControlHitTest.java

  Log Message:
  -----------
  fix: expand JTreeTable nodes on a single disclosure-triangle click (#7744)

The tree column is edited by forwarding the mouse event to the embedded
JTree. Some platform L&Fs (notably macOS Aqua) do not treat that synthetic
forwarded click as a hit on the expand/collapse control, so a single click
on a node's disclosure triangle did nothing and only a double-click (or a
click on the row text) would toggle it. This showed up in the Spells tab,
where level nodes such as "0" would not expand.

Add JTreeTable.expandControlPathForClick(tree, x, y), which returns the
branch path whose disclosure control is under a forwarded (tree-space)
point, and have the editor toggle that path directly. The node's expanded
state is captured before dispatch and the direct toggle is applied only if
the dispatch itself did not change it, so L&Fs that already handle the
click are not double-toggled.

The end-to-end behaviour is L&F-specific and not reproducible in the
headless (Metal) test harness, so JTreeTableExpandControlHitTest covers the
pure hit-testing decision (control strip vs. content vs. off-row vs. leaf)
plus the editor toggle round-trip.



To unsubscribe from these emails, change your notification settings at https://github.com/PCGen/pcgen/settings/notifications