As part of the refactoring in #5119, some vim_strchr() were changed to
strchr(). However, vim_strchr() behaves differently than strchr() when
c is NUL, returning NULL instead of a pointer to the NUL.
Revert the strchr() calls where it isn't known whether c is NUL, since
this causes a semantic change the surrounding code doesn't expect. In
the case of #6650, this led to a heap overrun.
Closes#6650
Hope this will make people using feed_command less likely: this hides bugs.
Already found at least two:
1. msgpackparse() will show internal error: hash_add() in case of duplicate
keys, though it will still work correctly. Currently silenced.
2. ttimeoutlen was spelled incorrectly, resulting in option not being set when
expected. Test was still functioning somehow though. Currently fixed.
- Use this to properly test cursor shape events.
- tests: update screen_basic_spec to use `screen:expect` short form.
Clearer than using `screen:wait` directy.
- Vim "unix default" of 'noshowcmd' is serving few users. And it's
inconsistent.
- 'ruler' and 'belloff=all' improve the out-of-the-box experience.
- Continue to use 'noshowcmd' and 'noruler' by default in the functional
tests to keep them fast.
TODO: Add a "disable slow stuff" command or mapping to address the
use-case of a very slow terminal connection.
Let the terminal dictate the normal-mode cursor position. This will be
disorienting sometimes, but it is closer to what users expect vs always
going to the last line.
Previously alternate branches were not accounted for properly, with this
change g- after an undo to a branch point works.
The current sequence number b_u_seq_cur is used in undo_time(), in
u_doit() this was calculated by subtracting one from the curhead
sequence number.
The curhead header entry represents the change that was just undone, so
the sequence number we want is that of the change we have moved to. This
is the sequence number of the undo head that is the uh_next element of
this curhead. That sequence number is not always one less than the
curhead sequence number -- there may have been an alternate branch at
this point.
Instead of subtracting one, we now directly find the sequence number of
curhead->uh_next.
`find_command(s->ca.cmdchar) >= 0` was established near the start of
normal_execute(). And `unshift_special(&s->ca);` "should" not ever
result in s->ca.cmdchar containing a multibyte char.
So only an assert() is needed here.
Also give NVIM_TUI_ENABLE_CURSOR_SHAPE more granularity:
0 = do not change cursor shape
1 = non-blinking ("steady") cursor with mode-sensitive shape
2 = blinking cursor with mode-sensitive shape
Note: blink state is not changed for Konsole, instead user's terminal
preference makes the decision. (Can't do that for xterm-likes, DECSCUSR
forces us to choose blink-state.)
This is a temporary step until the TUI respects 'guicursor'
Ref: https://github.com/neovim/neovim/issues/2583#issuecomment-272988384
When there is a difference in expected vs. actual row count, the user
gets a confusing message about being unable to string concat a nil value
from screen:expect.
This assert makes it clear what the problem is rather than requiring
people to dig through the code of screen:expect to determine what
happened.
This ameliorates use-cases like:
:!cat foo.txt
:make
where the user is interested in the last few lines of output.
Try these shell-based ex-commands before/after this commit:
:grep -r '' *
:make
:!yes
:!grep -r '' *
:!git grep ''
:!cat foo
:!echo foo
:!while true; do date; done
:!for i in `seq 1 20000`; do echo XXXXXXXXXX $i; done
In all cases the last few lines of the command should always be shown,
regardless of where throttling was triggered.
Periodically skip :! spam. This is a "cheat" that works for all UIs and greatly
improves responsiveness when :! spams MB or GB of output:
:!yes
:!while true; do date; done
:!git grep ''
:grep -r '' *
After ~10KB of data is seen from a single :! invocation, output will be skipped
for ~1s and three dots "..." will pulse in the bottom-left. Thereafter the
behavior alternates at every:
* 10KB received
* ~1s throttled
This also avoids out-of-memory which could happen with large :! outputs.
Note: This commit does not change the behavior of execute(':!foo').
execute(':!foo') returns the string ':!foo^M', it captures *only* Vim
messages, *not* shell command output. Vim behaves the same way.
Use system('foo') for capturing shell command output.
Closes#1234
Helped-by: oni-link <knil.ino@gmail.com>
https://github.com/mpeterv/luacheck/pull/81#issuecomment-261099606
> If you really want to use bleeding-edge version you should get the
> rockspec from master branch, not a fixed commit ...
> The correct way to install from a specific commit is cloning that
> commit and running "luarocks make" from project directory. The reason
> is that running "install" or "build" on an scm rockspec fetches
> sources from master but uses build description from the rockspec
> itself, which may be outdated.
- Use a standard scratch buffer instead of a new 'buftype', functions
like curbufIsChanged() already have special handling for scratch bufs.
- Cleanup some stuff from the previous merge.
- Add support for :smagic, :snomagic. Closes#5578
Add tests for undotree().
Helped-by: Björn Linse <bjorn.linse@gmail.com>
When "curhead" points to a valid head, the value of "newhead" is
meaningless (and really should be set to null). The undo state for
a buffer is _logically_ the enum:
enum UndoState {
CurrentHead(head),
NewHead(head),
EmptyTree
}
nvim _represents_ this as: whenever `curbuf->b_u_curhead` is nonnull it
should be used as the current head, and `curbuf->b_u_newhead` is
ignored. If the there is a current head, then this will be redoed on the
next redo, and its parent will be undone on next undo. Only if
`b_u_curhead` is NULL, `b_u_newhead` will be used as the head to undo
(and it is not possible to redo). Also both can be NULL, to indicate an
empty undotree. (To be fair, this only strictly true when calling undo.c
from the outside, in some places _within_ a function in undo.c both
values might be meaningful)
Apparently `undotree()` breaks this non-abstraction, this _cosmetic_
issue can easily be fixed by `ex_substitute` also saving and restoring
`b_u_newhead`, but is doesn't reflect any error really how
`u_undo_and_forget` manipulates the _actual_ state of the undo tree.
'inccommand' allows us to expand the feature to other commands, such as:
:cdo
:cfdo
:global
Also rename "IncSubstitute" highlight group to "Substitute".
[inc_sub] is less obvious for users. Also, in the future we may want to
generalize the idea of a "preview buffer", or "incremental commands"
besides :substitute.
- Eliminate/isolate static/global variables
- Remove special-case parameter from buflist_new()
- Remove special-case ECMD_RESERVED_BUFNR
- To determine when u_undo_and_forget() should be done, check
b_changedtick instead of a heuristic.
- use mb_string2cells() instead of strlen() to measure the :sub patterns
- call ml_close() before buf_clear_file(). Avoids leaks caught by ASan.
Original patch by:
Robin Elrharbi-Fleury (Robinhola)
Audrey Rayé (Adrey06)
Philémon Hullot (DesbyP)
Aymeric Collange (aym7)
Clément Guyomard (Clement0)
When running the tests in a contained environment, the lingering
terminal processes prevent the unmounting of the filesystems in the
contained environment.
These tests fail on master, so it's not a regression. Changes in #4874
(parent commit) seem to work (and pass most CI), so skipping these tests
is better than blocking the changes.
- Links to Search by default
screen.c: Combine CursorLine with QuickFixLine
- HLF_QFL takes priority over HLF_CUL
docs: Updated to mention QuickFixLine
runtime: Added QuickFixLine to nvimHLGroup
tests: QuickFixLine highlight
- Use the default buffer text provided by before_each (avoids extra steps and
makes the tests more consistent with each other)
- Indent
- Adjust help doc
It is otherwise impossible to determine which test failed sanitizer/valgrind
check. test/functional/helpers.lua module return was changed so that tests which
do not provide after_each function to get new check will automatically fail.
Problem: When "syntax manual" was used switching between buffers removes
the highlighting.
Solution: Set the syntax option without changing the value. (Anton
Lindqvist)
885f24fbca
Patch applied cleanly to the nvim sources except for version.c.
- Code from Vim source.
- Removed the check for 'guioptions'
- mouse_spec.lua: test <ScrollWheelLeft> and <ScrollWheelRight>
- Move horizontal scroll logic to mouse.c
- Remove 'gui_' from the function names
- Renamed variables to be more specific (as opposed to generic p, w).
- Marked some functions as `static`
Adds support for:
- api:vim_input("<D-a>")
- ":nnoremap <C-D-S-...>" and permutations thereof
UIs must capture the modifier and send it as "<D-...>" to vim_input().
Note: Before this commit, any arbitrary ":nnoremap <{foo}-{bar}>"
mapping could already be invoked with feedkeys("\<{foo}-{bar}>"). This
commit supports "D-" as a modifier that can be combined with "C-", "A-",
"S-" in any order.
For non-GUI (terminal) support, user must:
:set <D-a>={CSI sequence}
then send the {CSI sequence} from their terminal. But this does not work
yet (regression #2204).
Closes#2190
Plugins (YCM, dispatch.vim) use gui_running to decide behavior; so do
some colorschemes. Up to now, nvim lied about gui_running for the
benefit of colorschemes and the detriment of all _other_ plugins that
check this condition. That's counterproductive: a user employs at most
_one_ colorscheme but may use many other plugins which expect the Vim
legacy semantics of gui_running.
Moreover, colorschemes usually don't _need_ to check gui_running: they
can instead set cterm{fg,bg} and gui{fg,bg} in the same :highlight call.
It is reasonable for users who want "true color" to modify their
colorscheme once (or request upstream to do so) in order to avoid
running into quirks in any other plugins.
Closes#2782
This is consistent with VIM behavior. When showing a visual selection,
VIM only extends it past the last character if eol is in listchars (even
if nolist is set).
Add a new special key that can be used by UIs to toggle the 'paste' option and
use it in the TUI instead of the user's 'pastetoggle' value.
Close#2843#2092
No one has taken a real interest in fixing this, so let's disable it for
now. The issue crops up most on the QB OS X node, but it has
periodically occurred under other nodes too.
This is a port of my original contribution to Vim, added in 7.4.687
(https://github.com/vim/vim/commit/v7-4-687). The TUI code has been
heavily refactored (see esp. 25ceadab37),
so this required some translation, but the logic is the same.
Currently, there are two functions in the UI API that are called when
the mode changes: insert_mode() and normal_mode(). These can be folded
into a single mode_change() entrypoint which can do whatever it wants
based on the mode it is passed, limited to INSERT and NORMAL for now.
Problem: When 'ruler' is set the preferred column is reset. (Issue 339)
Solution: Don't set curswant when redrawing the status lines.
https://code.google.com/p/vim/source/detail?r=v7-4-659
Helped-by: David Bürgin <676c7473@gmail.com>
Based on #2347 by @pvinis
Problem: When using 'incsearch' "2/pattern/e" highlights the first match.
Solution: Move the code to set extra_col inside the loop for count. (Ozaki
Kiichi)
https://github.com/vim/vim/releases/tag/v7-4-532
- Modify tty-test to allow easier control over the terminal
- Add a new directory with various terminal tests/specifications
- Remove a pending job/pty test.
- Flush stdout in Screen:snapshot_util() (avoid waiting for the test to finish)
- Replace libuv sigwinch watcher by a sigaction handler. libuv randomly fails to
deliver signals on OSX. Might be related to the problem fixed by
@bbcddc55ee1e5605657592644be0102ed3a5f104 (under the hoods, libuv uses a pipe
to deliver signals to the main thread, which might be blocking in some
situations)
This is can be used for spawning nvim outside a test context. Also refactor
screen.lua to use this function when loading the color map(It is better because
the GDB/VALGRIND environment variables are ignored)
Even though assuming nvim is busy most times is simpler, it has a problem: A lot
of unnecessary busy_start/busy_stop notifications are sent to the UI. That's
because in the majority of scenarios almost no time is spent between
`event_poll` calls.
This restores the normal behavior which is to call busy_start only when nvim is
going to perform some task that can take a significant amount of time. Also
improve the usage of buffering in the TUI when changing the cursor state.
Switching cursor off is only necessary in two occasions:
- When redrawing to avoid terminal flickering
- When the editor is busy
The first can now be handled by the TUI, so most calls to ui_cursor_off can be
removed from the core.
So, before this commit it was only necessary to switch the cursor off to notify
the user that nvim was running some long operation. Now the cursor_{on,off}
functions have been replaced by busy_{stop,start} which can be handled in a
UI-specific way(turning the cursor off or showing a busy indicator, for
example).
To make things even more simpler, nvim is always busy except when waiting for
user input or other asynchronous events: It automatically switches to a non-busy
state when the event loop is about to be entered for more than 100 milliseconds.
`ui_busy_start` can be called when its not desired to change the busy state in
the event loop (As its now done by functions that perform blocking shell
invocations).
