summaryrefslogtreecommitdiffstats
path: root/tests/test_config.cpp
diff options
context:
space:
mode:
authorDanilo M. <danix@danix.xyz>2026-08-03 10:09:26 +0200
committerDanilo M. <danix@danix.xyz>2026-08-04 12:52:59 +0200
commit229eecbf1893ad002d700f83ee7d8a2babbd595e (patch)
tree183d6a734c0265d9c10af4f9b42d69dd9a1d2015 /tests/test_config.cpp
parent6caea6aabe74014f9bdd2b8743925362bad19fe9 (diff)
downloadqtmaildir-229eecbf1893ad002d700f83ee7d8a2babbd595e.tar.gz
qtmaildir-229eecbf1893ad002d700f83ee7d8a2babbd595e.zip
fix: let key bindings win over the thread list's type-to-search
With the thread list focused, 'h' jumped to the next thread whose subject began with "h" instead of toggling HTML, and j, k, a, d, N, F, u and G were swallowed the same way. The keymap only worked when focus happened to be somewhere else. installEventFilter(this) was on the MainWindow, and a window-level filter only sees key presses the focused child did not consume. QAbstractItemView consumes plain letters for its type-to-search feature, so it took them first. The filter is now installed on the thread view as well, which puts the keymap ahead of that search. The existing query-bar guard in eventFilter() still keeps ordinary typing working there. Found by the maintainer while walking task 13 item 14, and confirmed fixed on screen. No regression test: a QTest::keyClick attempt passed both with and without the fix, because synthetic key posting does not reproduce the focus and consumption path that causes the bug. A test that cannot fail is worse than none, so it was dropped rather than kept for appearances. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Diffstat (limited to 'tests/test_config.cpp')
0 files changed, 0 insertions, 0 deletions