Problem:
With vim.treesitter.foldexpr, `o`-ing two lines above a folded region
opens the fold. This does not happen with legacy foldexprs. For example,
make a markdown file with the following text (without indentation),
enable treesitter fold, and follow the instruction in the text.
put cursor on this line and type zoo<Esc>
initially folded, revealed by zo
# then this fold will be opened
initially folded, revealed by o<Esc>
Analysis:
* `o` updates folds first (done in `changed_lines`), evaluating
foldexpr, and then invokes `on_bytes` (done in `extmark_splice`).
* Treesitter fold allocates the foldinfo for added lines (`add_range`)
on `on_bytes`.
* Therefore, when treesitter foldexpr is invoked while running `o`, it
sees outdated foldinfo.
Solution:
`extmark_splice`, and then `changed_lines`. This seems to be the
standard order in other places, e.g., `nvim_buf_set_lines`.
When tabstop and shiftwidth are not equal, tabs are inserted as individual
spaces and then rewritten as tab characters in a second pass. That second pass
did not call changed_bytes which resulted in events being omitted.
Fixes#25092
`nvim_(get|set)_option_value` pick the current buffer / window by default for buffer-local/window-local (but not global-local) options. So specifying `buf = 0` or `win = 0` in opts is unnecessary for those options. This PR removes those to reduce code clutter.
When a buffer update callback is called, textlock is active so buffer
text cannot be changed, but cursor can still be moved. This can cause
problems when the buffer update is in the middle of an operator, like
the one mentioned in #16729. The solution is to save cursor position and
restore it afterwards, like how cursor is saved and restored when
evaluating an <expr> mapping.
When :undo! was introduced to Nvim the implementation of 'inccommand'
preview callback hasn't been fully decided yet, so not notifying buffer
update callbacks made sense for 'inccommand' preview callback in case it
needs to undo the changes itself.
Now it turns out that the undo-and-forget is done automatically for
'inccommand', so it doesn't make sense for :undo! to avoid notifying
buffer update callbacks anymore.
There can be other places that access window buffer info (e.g.
`tabpagebuflist()`), so checking `w_closing` in `win_findbuf()` doesn't
solve the crash in all cases, and may also cause Nvim's behavior to
diverge from Vim.
Fix#14998
New versions of neovim will not read undofiles written by nvim before this merge (there will be an error message about incopmatible version). Nvim 0.4 (or an master up to bda12927be ) can be used to recover older undofiles, so if you worried about unsaved changes lurking around in undofiles it would make sense to keep such version around somewhere to recover them.
This is a necessary change to keep tree-sitter and plugins dependent on byte-level buffer change events fully working with undo states from a undofile. If there is a clear demand we might implement reading of the old format. Such recovered buffers will not be be fully functional with plugins relying on buffer updates or tree-sitter, however.
