aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/undo.h
diff options
context:
space:
mode:
authorRobin Haberkorn <robin.haberkorn@googlemail.com>2024-09-16 13:05:20 +0200
committerRobin Haberkorn <robin.haberkorn@googlemail.com>2024-09-16 13:05:20 +0200
commitbc859a06a13b24daea12d35df2a1ec5114b42180 (patch)
treeaf5ca4cb84e0b6e1c47a5837ab9d0724181c68e1 /src/undo.h
parent966d3efddcb1da6496dd99ee06bce7a985a5419f (diff)
downloadsciteco-bc859a06a13b24daea12d35df2a1ec5114b42180.tar.gz
fixed rubout of empty forward kill (FK)
Test case: IF$ J IX$ FKF$ ^W The range to delete is empty, Scintilla would not generate an undo action, but SCI_UNDO would still be exected on rubout which removes the "X" too early. * We should really get rid of Scintilla undo actions as they are a source of trouble and complexity. There could be a custom undo token to undo SCI_DELETERANGE that automatically fetches the text that's going to be deleted and stores it in the token's data. This could replace most uses of SCI_UNDO. The rest is to undo insertions, which can easily be replaced with undo__teco_interface_ssm(SCI_DELETERANGE...). * We should really allow rubout tests in the test suite...
Diffstat (limited to 'src/undo.h')
0 files changed, 0 insertions, 0 deletions