The "length-of-one" trick, where the last element of a struct is an
array of size 1, but extra size is allocated when calling malloc where
it uses more than 1 element in the array, cause problems with some
compilers. Some compilers set _FORTIFY_SOURCE=2 by default which
incorrectly considers it as an overflow. More information:
https://github.com/neovim/neovim/issues/223#issuecomment-1413828554
Using flexible array members allows us to to properly convey to the
compiler that its size may be larger than 1. This also enables us to
remove lengthy workarounds that are unreliable, as they depend on
CMAKE_BUILD_TYPE which isn't defined for multi-config generators.
Closes: https://github.com/neovim/neovim/issues/223
Problem:
When a TUI client is suspended it still receives UI events from the
server, and has to process these accumulated events when it is resumed.
With mulitple TUI clients this is a bigger problem, considering the
following steps:
1. A TUI client is attached.
2. CTRL-Z is pressed and the first client is suspended.
3. Another TUI client is attached.
4. CTRL-Z is pressed and a "suspend" event is sent to both clients. The
second client is suspended, while the first client isn't able to
process the event because it has already been suspended.
5. The first client is resumed. It processes the accumulated "suspend"
event and suspends immediately.
Solution:
Make a TUI client detach on suspend and re-attach on resume.
Problem: The 'statuscolumn' width is being estimated without the
proper context. In particular, this resulted in the fact
that a custom fold column could be included in the estimated
`number_width()`, and doubly added when actually drawing the
statuscolumn due to `win_col_off()` also adding the
`'foldcolumn'` width. Resulting in a status column that is
`'foldcolumn'` cells wider than necessary.
Solution: Estimate 'statuscolumn' width in `get_statuscol_str()` when
a buffer's line count has changed.
Make it more clear that on unix the "pipe" mode of sockconnect uses unix
local domain sockets, not named pipes (FIFOs) which are not currently
supported.
See discussion in #22080.
Signed-off-by: Thayne McCombs
Problem: "1v" may select block with wrong size. (Evgeni Chasnovski)
Solution: Compute "curswant" in the right line. (closesvim/vim#11925)
8f531662e2
Co-authored-by: Bram Moolenaar <Bram@vim.org>
Problem: Allocating more memory than needed for extended structs.
Solution: Use offsetof() instead of sizeof(). (Dominique Pelle,
closesvim/vim#4786)
47ed553fd5
Problem: .clangd and .stylelintrc files don't get a filetype.
Solution: Use yaml for .clangd and json for .stylelintrc files. (Mark
Skelton, closesvim/vim#11916)
9c51798a1f
Co-authored-by: Mark Skelton <mdskelton99@gmail.com>
This value can not be relied on as it doesn't work for
multi-configuration generators. I don't think this undocumented option
is used much, if at all, so I think we should remove it.
When I run ./scripts/bump_deps.lua I get an error:
Vim:E475: Invalid value for argument cmd: 'command' is not executable
Running command -v in shell fixes this.
Currently, only the nvim target uses ccache but not libnvim or
unittests. It is generally a good idea to operate on targets rather than
globally, but this is an exception as there isn't a target where we
don't want to use ccache on.
Problem: NetworkManager connection files are not recognized.
Solution: Add a pattern for NetworkManager connection files. (closesvim/vim#11893)
04e4f1d985
Co-authored-by: ObserverOfTime <chronobserver@disroot.org>
Problem: Changing 'virtualedit' does not have immediate effect.
Solution: Correct how is checked for a changed value. (closesvim/vim#11878)
8fe5b9c8c1
Co-authored-by: Bram Moolenaar <Bram@vim.org>
Problem: Some injections (like markdown) allow specifying arbitrary
language names for code blocks, which may be lead to errors when
looking for a corresponding parser in runtime path.
Solution: Validate that the language name only contains alphanumeric
characters and `_` (e.g., for `c_sharp`) and error otherwise.
Initial benchmarks show that this gives a nice 50% compile time
reduction for neovim. This shouldn't affect users or CI, but it's a nice
speedup for local development. The C_COMPILER_LAUNCHER target property
is only supported by Makefiles and Ninja generators only, so this won't
give a speedup when using the Xcode and Visual Studio generators even if
ccache is available.
Find modules should only link to libraries defined in the find module,
and not the main project. This helps the find modules be more self-contained and easier to understand.
