| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
|
|
| |
Avoid consecutive RPC requests involving :startinsert or :stopinsert,
because consecutive RPC requests may be processed together, before the
:startinsert or :stopinsert takes effect.
Also change some feed_command() to command() to make tests faster.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Special keys are negative integers, so sending them to terminal leads to
strange behavior.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: nvim_buf_set_text(), nvim_open_term() and termopen() all change buffer
text, which is forbidden during textlock. Additionally, nvim_open_term() and
termopen() may be used to convert the cmdwin buffer into a terminal buffer,
which is weird.
Solution: Allow nvim_buf_set_text() and nvim_open_term() in the cmdwin, but
disallow nvim_open_term() from converting the cmdwin buffer into a terminal
buffer. termopen() is not allowed in the cmdwin (as it always operates on
curbuf), so just check text_locked().
Also happens to improve the error in #21055: nvim_buf_set_text() was callable
during textlock, but happened to check textlock indirectly via u_save();
however, this caused the error to be overwritten by an unhelpful "Failed to
save undo information" message when msg_list == NULL (e.g: an `<expr>` mapping
invoked outside of do_cmdline()).
|
| |
|
|
|
|
|
|
|
| |
Problem: Message is cleared when removing mode message (Gary Johnson).
Solution: Do not clear the command line after displaying a message.
https://github.com/vim/vim/commit/800cdbb7caeb5dd4379c6cb071bb12391f20bcf3
Co-authored-by: Bram Moolenaar <Bram@vim.org>
|
| | |
|
| |
|
|
|
|
|
|
|
| |
libvterm v0.3 supports reflow of terminal buffer when Nvim is resized
Since v0.3 is now a required dependency, enable it by default to find
(and fix) possible issues.
Note: Neovim's scrollback buffer does not support reflow (yet), so lines
vanishing into the buffer due to a too small window will be restored
without reflow.
|
| |
|
|
| |
`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.
|
| |
|
|
|
| |
Co-authored-by: zeertzjq <zeertzjq@outlook.com>
Co-authored-by: famiu <famiuhaque@protonmail.com>
|
| | |
|
| | |
|
| |\
| |
| | |
fix(tui): redraw on SIGWINCH even if size didn't change
|
| | | |
|
| |/
|
|
|
|
|
| |
Problem: When setting a shell size smaller than the containing
terminal window through `:winsize` or `:set lines/columns`
the screen is not properly cleared.
Solution: Clear the tui dimensions rather than the grid dimensions.
|
| |
|
| |
Check for the [No Name] after wiping the buffer.
|
| |
|
|
|
|
|
|
| |
Problem:
Codebase inconsistently binds vim.api onto a or api.
Solution:
Use api everywhere. a as an identifier is too short to have at the
module level.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
"tmux 3.2a" (output from "tmux -V") is not parsed easily.
Solution:
With `strict=false`, discard everything before the first digit.
- rename Semver => Version
- rename vim.version.version() => vim.version._version()
- rename matches() => has()
- remove `opts` from cmp()
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
On Windows, pasting multiple lines on a terminal buffer cause all the
lines to appear on the same line, i.e., the line breaks are lost.
Cause:
Windows shells expect "\r\n" as line break but "terminal_paste" function
uses "\n".
Solution:
Use "\r\n" as line break for pasting in terminal buffer on Windows.
Note:
Although this issue was reported with powershell set as 'shell', it
occurs in cmd too.
Fixes #14621
|
| |
|
|
| |
Some tests that were previously not working have started to work again
for unspecified reasons, so let's enable them.
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Problem:
has('gui_running') is still common in the wild and our answer has
changed over time, causing frustration.
https://github.com/vimpostor/vim-tpipeline/commit/95a6ccbe9f33bc42dd4cee45731d8bc3fbcd92d1
Solution:
Use stdin_tty/stdout_tty to decide if a UI is (not) a GUI.
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Problem:
nvim_list_uis does not report all ":help ui-option" fields.
Solution:
Store ":help ui-option" fields on the `UI` object and update ui_array.
|
| |/ |
|
| |
|
|
| |
This avoids changing cmdline and fixes a warning.
|
| |
|
|
| |
Include the rest of the line and allow multiple {MATCH:} patterns.
|
| | |
|
| | |
|
| |
|
|
| |
When a TUI client has already stopped, handling UI events will cause a
heap-use-after-free, so ignore them.
|
| |
|
|
| |
This prevents the TUI from doing unexpected things when receiving a
deadly signal or running out of memory.
|
| |
|
| |
Also fix the race between output and resize in the test for this.
|
| |
|
|
| |
There may be a race between resizes, so don't expect exact screen.
Expect window width instead.
|
| |
|
|
|
| |
Actually execute the command.
Don't use assert_alive() as it checks the host session instead of the
child session. Use screen:expect() instead.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Included patches:
821. By Paul "LeoNerd" Evans on 2022-12-29
Don't bother to emit the unrecognised sequence in DECRQSS response as it provides an echo roundtrip possibility
820. By Paul "LeoNerd" Evans on 2022-11-26
erase_internal() should only set fg/bg colour, resetting other attributes (especially RV)
819. By Paul "LeoNerd" Evans on 2022-11-09
Added vterm_screen_set_default_colors(), which repaints the cells in the buffer(s)
818. By Paul "LeoNerd" Evans on 2022-11-09
Permit either colour argument to be NULL to vterm_state_set_default_colors()
817. By Paul "LeoNerd" Evans on 2022-10-01
Delete the mk_wcswidth functions as they're not used; guard the CJK-wide one with an ifdef as by default we don't use it
816. By Paul "LeoNerd" Evans on 2022-10-01
Make sure to supply empty (void) prototype to functions that take no arguments in bin/vterm-ctrl.c
|
| |
|
| |
Using Unicode makes the test flaky because of #16245.
|
| | |
|
| |
|
|
| |
fixes #21610
|
| |
|
|
|
|
|
|
| |
Problem: When unibi_format() modifies params and data->buf overflows,
unibi_format() is called again, causing the params to be
modified twice. This can happen for escapes sequences that
use the %i terminfo format specifier (e.g. cursor_address),
which makes unibi_format() increase the param by 1.
Solution: Make a copy of data->params before calling unibi_format().
|
| |
|
|
|
| |
Now that the TUI calls nvim_paste and nvim_input through RPC, the data
race is much more likely, as nvim_paste is deferred while nvim_input is
not. Add an expect_child_buf_lines() call to avoid the race.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
| |
Problem: `chansend()` on Windows sends lines in reverse order.
Cause: Using \n instead of \r\n for newlines on Windows.
Solution: on Windows, use CRLF newline characters.
Fixes #18501
|
| |
|
|
|
| |
Extend the capabilities of is_os to detect more platforms such as
freebsd and openbsd. Also remove `iswin()` helper function as it can be
replaced by `is_os("win")`.
|
| | |
|
| |
|
|
|
|
|
| |
* refactor(test): use exec_lua
* fix(test): fix failing tui_spec tests
test is failing when path of tty-test does not fit cmdline
|
