Problem: When 'hlsearch' is set and matching with the last search pattern
is very slow, Vim becomes unusable. Cannot quit search by
pressing CTRL-C.
Solution: When the search times out set a flag and don't try again. Check
for timeout and CTRL-C in NFA loop that adds states.
fbd0b0af68
wp->w_height_inner now contains the "inner" size, regardless if the
window has been drawn yet or not. It should be used instead of
wp->w_grid.Rows, for stuff that is not directly related to accessing
the allocated grid memory, such like cursor movement and terminal size
There is various places where 'conceallevel' and 'concealcursor'
necessitates additional redraws. This tries to separate the different
cases and handle each accordingly:
- Share code with 'cursorline' for the common case: vertical move of
cursor within the same window (concealcursor not active)
- Improve the logic for managing 'concealcursor' and switching modes:
test for the case where the new mode behaves differently from the
last one.
- Clarify the special case for horizontal movement within a line when
'concealcursor' is active, now there is an if-statement only for this
and not hidden in larger check mostly for the first point.
- Keep the special case for moving between windows as is.
update_single_line() was only used for 'concealcursor'. But 'cursorline'
has very similiar characteristics (redraw both lines on move cursor
between lines) and works without its own special entry point to the
redraw subsystem.
Later on 'concealcursor' and 'cursorline' could share more logic, but for
now make the former use standard redrawWinline(). Make sure it is called
before update_screen(), so that it is immediately visible.
Get rid of update_prepare() and update_finish(), and all issues from
them and their callsites not being in sync with changes to
update_screen()
Problem: When test_edit fails 'insertmode' may not be reset and the next
test may get stuck. (James McCoy)
Solution: Always reset 'insertmode' after executing a test. Avoid that an
InsertCharPre autocommand or a 'complete' function can change the
state. (closesvim/vim#3768)
8ad16da729
Problem: Characters deleted on completion. (Adrià Farrés)
Solution: Also check the last item for the ORIGINAL_TEXT flag. (Christian
Brabandt, closesvim/vim#1645)
e87edf3b85
Problem: Bogus characters appear when indenting kicks in while doing a
visual-block append.
Solution: Recompute when indenting is done. (Christian Brabandt)
e2e69e4813
Problem: Using setline() in TextChangedI splits undo.
Solution: Use another solution for undo not working properly.
9fa9506853
This deviates from Vim in the handling of the CursorHoldI event. In
Vim, any buffer changes are merged into the insert. In Neovim,
CursorHoldI is handled via the multiqueue, and the point at which
the cursor hold is implemented (in input.c) doesn't know enough about it.
Making all queued events merge into the insert seems more wronger since
changes by other asynchronous events really should be separately
undoable.
