We already have an extensive suite of static analysis tools we use,
which causes a fair bit of redundancy as we get duplicate warnings. PVS
is also prone to give false warnings which creates a lot of work to
identify and disable.
REP_NODELAY was added because REP delayed too much. This changes REP to
only add a delay on every 100th line instead.
This helps to cover the additional pulse steps with
out_data_decide_throttle, which would have required to change
REP_NODELAY anyway.
Fix flaky "shell command :! throttles shell-command output greater than ~10KB:":
[ RUN ] shell command :! throttles shell-command output greater than ~10KB:
warning: Screen changes were received after the expected state. This indicates
indeterminism in the test. Try adding screen:expect(...) (or wait()) between
asynchronous (feed(), nvim_input()) and synchronous API calls.
- Use screen:redraw_debug() to investigate; it may find relevant intermediate
states that should be added to the test to make it more robust.
- If the purpose of the test is to assert state after some user input sent
with feed(), adding screen:expect() before the feed() will help to ensure
the input is sent when Nvim is in a predictable state. This is preferable
to wait(), for being closer to real user interaction.
- wait() can trigger redraws and consequently generate more indeterminism.
Try removing wait().
ERR
test/functional/ui/screen.lua:579: Failed to match any screen lines.
Expected (anywhere): "
%."
Actual:
|XXXXXXXXXX 591 |
|XXXXXXXXXX 592 |
|XXXXXXXXXX 593 |
|XXXXXXXXXX 594 |
| |
| |
|{3:-- TERMINAL --} |
stack traceback:
test/functional/ui/screen.lua:579: in function '_wait'
test/functional/ui/screen.lua:367: in function 'expect'
test/functional/ui/output_spec.lua:63: in function <test/functional/ui/output_spec.lua:53>
Log: https://travis-ci.org/neovim/neovim/jobs/569082705#L5355
(gcc-functionaltest-lua)
Typically most shell output is the result of non-trivial work, so it
would not blast stdout instantaneously. To more closely simulate that
typical scenario, change `shell-test REP` to wait 1 millisecond between
iterations.
Also fix V576: use width specification
> Incorrect format. Consider checking the third actual argument of the
> 'sscanf' function. It's dangerous to use string specifier without width
> specification. Buffer overflow is possible.
Tokenize p_sh if used as default in ex_terminal(). Previously p_sh was
used as the first arg in a list when calling termopen(), this would try
to call an untokenized version of shell, meaning if you had an argument
in 'shell':
set shell=/bin/bash\ --login
the command would fail.
Helped-by: oni-link <knil.ino@gmail.com>
Closes#3999
Old behaviour: termopen('cmd') would run `&shell &shcf "cmd"`, which
caused the functional tests to fail on some systems due to the process
not "owning" the terminal. Also, it is inconsistent with jobstart().
Modify termopen() so that &shell is not invoked, but maintain the old
behaviour with :terminal. Factor the common code for building the
argument vector from jobstart() and modify the functional tests to call
termopen() instead of :terminal (fixes#2354).
Also:
* Add a 'name' option for termopen() so that `:terminal {cmd}` produces
a buffer named "term//{cwd}/{cmd}" and termopen() users can customize
the name.
* Update the documentation.
* Add functional tests for `:terminal` sinse its behaviour now differs
from termopen(). Add "test/functional/fixtures/shell-test.c" and move
"test/functional/job/tty-test.c" there, too.
Helped-by: Justin M. Keyes <@justinmk>
