When it happens it wastes 2 seconds which is NOT included in the
normal busted timing info. It is hard to correct this, but we
can at least print a warning when this happens.
Problem:
Nvim fails to create tempfile "…/nvim6UJx04/7" when 'shell' is set to
pwsh (PowerShell Core). This breaks filtered shell commands ":{range}!".
With shell set to cmd, it works.
Solution:
PowerShell doesn't use "<" for stdin redirection. Instead, use
"-RedirectStandardInput".
Closes#15913
Problem:
- Unix sockets are created in random /tmp dirs.
- /tmp is messy, unclear when OSes actually clear it.
- The generated paths are very ugly. This adds friction to reasoning
about which paths belong to which Nvim instances.
- No way to provide a human-friendly way to identify Nvim instances in
logs or server addresses.
Solution:
- Store unix sockets in stdpath('state')
- Allow --listen "name" and serverstart("name") to given a name (which
is appended to a generated path).
TODO:
- is stdpath(state) the right place?
Problem:
1. Log messages (especially in CI) are hard to correlate with tests.
2. Since b353a5c05f#11886, dumplog() prints the logs next to test
failures. This is noisy and gets in the way of the test results.
Solution:
1. Associate an incrementing id with each test and include it in log
messages.
- FUTURE: add v:name so Nvim instances can be formally "named"?
2. Mention "child" in log messages if the current Nvim is a child (based
on the presence of $NVIM).
BEFORE:
DBG … 12345 UI: event
DBG … 12345 log_server_msg:722: RPC ->ch 1: …
DBG … 12345 UI: flush
DBG … 12345 inbuf_poll:444: blocking... events_enabled=1 events_pending=0
DBG … 23454 UI: stop
INF … 23454 os_exit:594: Nvim exit: 0
AFTER:
DBG … T57 UI: event
DBG … T57 log_server_msg:722: RPC ->ch 1: …
DBG … T57 UI: flush
DBG … T57 inbuf_poll:444: blocking... events_enabled=1 events_pending=0
DBG … T57/child UI: stop
INF … T57/child os_exit:594: Nvim exit: 0
PROBLEM
------------------------------------------------------------------------
$NVIM_LISTEN_ADDRESS has conflicting purposes as both a parameter ("the
current process should listen on this address") and a descriptor ("the
current process is a child of this address").
This contradiction means the presence of NVIM_LISTEN_ADDRESS is
ambiguous, so child Nvim always tries to listen on its _parent's_
socket. This is the cause of lots of "Failed to start server" spam in
our test/CI logs:
WARN 2022-04-30… server_start:154: Failed to start server: address already in use: \\.\pipe\nvim-4480-0
WARN 2022-04-30… server_start:154: Failed to start server: address already in use: \\.\pipe\nvim-2168-0
SOLUTION
------------------------------------------------------------------------
1. Set $NVIM to the parent v:servername, *only* in child processes.
- Now the correct way to detect a "parent" Nvim is to check for $NVIM.
2. Do NOT set $NVIM_LISTEN_ADDRESS in child processes.
3. On startup if $NVIM_LISTEN_ADDRESS exists, unset it immediately after
server init.
4. Open a channel to parent automatically, expose it as v:parent.
Fixes#3118Fixes#6764Fixes#9336
Ref https://github.com/neovim/neovim/pull/8247#issuecomment-380275696
Ref #8696
helpers.source() was a hack to work around the lack of anonymous
:source. Its "create tempfile" behavior is not a required part of most
tests that use it.
Some tests still need the old "create tempfile" behavior either because
they test SID behavior, or because of missing nvim_exec features: #16071
It was not possible to run the tests under the gdbserver because we were
not closing the old session before starting a new one. This caused the
server to not to be able to bind to the given address and crashing the
tests.
This commit closes the session before starting a new one.
Signed-off-by: Shreyansh Chouhan <chouhan.shreyansh2702@gmail.com>
Problem
- `redir_exec` is obsolete, but it keeps getting used in new tests
because people copy existing tests.
- Disadvantages of `redir_exec`:
- Captures extra junk before the actual error/message that we _want_ to test.
- Does not fail on error, unlike e.g. `command()`.
Solution
- Use new functions like `nvim_exec` and `pcall_err`.
The official developer documentation in in :h dev-lua-doc specifies to
use "--@" for special/magic tokens. However, this format is not
consistent with EmmyLua notation (used by some Lua language servers) nor
with the C version of the magic docstring tokens which use three comment
characters.
Further, the code base is currently split between usage of "--@",
"---@", and "--- @". In an effort to remain consistent, change all Lua
magic tokens to use "---@" and update the developer documentation
accordingly.
Ensure that
* Shell uses UTF8 input/output mode
* Stderr output is captured, in UTF8
* Program exit codes are correctly captured
Update functional test harness and add tests
for :make command.
Closes#13713
Opt in to this secret world using
set wildchar=0
" already the default, but remove if non-zero existing config:
set wildcharm=0
now you can map 'wildmode' just like any mode:
cnoremap <tab> <c-z>
function! Spacey()
return getcmdline()[-1:] == "/" ? "\<bs>" : ""
endfunc
cnoremap <expr> / wildmenumode() ? Spacey()."/<c-z>" : "/"
Possibly asked questions:
What about backwards compatibility?
====
Just do nothing and your existing 'wildchar' and 'wildcharm' will keep working.
Doesn't `<c-z>` mean suspend?
====
Not in cmdline mode. If it would then the recommended wildcharm would not
have been `<c-z>` to start with.
My config relies on `:<c-z>` being a synonym to `:<nop>`!
====
just no.
Problem: ruby#Detect() and node#Detect() don't return a [prog, err] pair
which means callers must special-case them.
Solution: align their return signatures with the perl/pythonx providers.
using just 'perl' isn't correct as it may not be the version requested.
ditto for 'cpanm', rather go through 'App::cpanminus' to find the latest
perl version
Having the cursor change column can be surprising.
Force startofline in functional and old tests.
Remove the functional breakindent test, as it's a subset of the oldtest one.
Having the cursor change column can be surprising.
Force startofline in functional and old tests.
Remove the functional breakindent test, as it's a subset of the oldtest one.
- We already find ourselves renaming nvim_execute_lua in tests and
scripts, which suggests "exec" is the verb we actually want.
- Add "exec" verb to `:help dev-api`.
- Running out of retries, or unexpected screen state should make the
test FAIL, not ERROR.
- Uses levels to report the location of the caller.
- Improve message with retry-failure (formatting).
Before:
[ RUN ] test: 103.53 ms ERR
test/functional/helpers.lua:388:
retry() attempts: 1
test/functional/ui/screen.lua:587: Row 1 did not match.
Expected:
|*X^ |
|{0:~ }|
|{0:~ }|
| |
Actual:
|*^ |
|{0:~ }|
|{0:~ }|
| |
To print the expect() call that would assert the current screen state, use
screen:snapshot_util(). In case of non-deterministic failures, use
screen:redraw_debug() to show all intermediate screen states.
stack traceback:
test/functional/helpers.lua:388: in function 'retry'
test/functional/test_spec.lua:24: in function <test/functional/test_spec.lua:23>
After:
[ RUN ] test: 105.22 ms FAIL
test/functional/test_spec.lua:24: stopping after 1 retry() attempts.
test/functional/test_spec.lua:25: Row 1 did not match.
Expected:
|*X^ |
|{0:~ }|
|{0:~ }|
| |
Actual:
|*^ |
|{0:~ }|
|{0:~ }|
| |
To print the expect() call that would assert the current screen state, use
screen:snapshot_util(). In case of non-deterministic failures, use
screen:redraw_debug() to show all intermediate screen states.
stack traceback:
test/functional/helpers.lua:389: in function 'retry'
test/functional/test_spec.lua:24: in function <test/functional/test_spec.lua:23>
