closes: https://github.com/coder/internal/issues/1331
Fixes up an issue in the test where we end up calling `FailNow` outside
the main test goroutine. Also adds the ability to name a `ptytest.PTY`
for cases like this one where we start multiple commands. This will help
debugging if we see the issue again.
This doesn't address the root cause of the failure, but I think we
should close the flake issue. I think we'd need like a stacktrace of all
goroutines at the point of failing the test, but that's way too much
effort unless we see this again.
On Windows, `pty.New()` was creating a `ConPTY` (`PseudoConsole`) even
when no process would be attached. `ConPTY` requires a real process to
function correctly - without one, the pipe handles become invalid
intermittently, causing flaky test failures like `read |0: The handle is
invalid.`
This affected tests using the `ptytest.New()` + `Attach()` pattern for
in-process CLI testing.
The fix splits Windows PTY creation into two paths:
- `newPty()` now returns a simple pipe-based PTY for the `Attach()` use
case
- `newConPty()` creates a real `ConPTY`, called by `Start()` when a
process will be attached
AFAICT this will result in no change in behaviour outside of tests.
Fixescoder/internal#1277
_Disclaimer: investigated and implemented by Claude Opus 4.5, reviewed
by me._
---------
Signed-off-by: Danny Kopping <danny@coder.com>
Fixes all our Go file imports to match the preferred spec that we've _mostly_ been using. For example:
```
import (
"context"
"time"
"github.com/prometheus/client_golang/prometheus"
"golang.org/x/xerrors"
"gopkg.in/natefinch/lumberjack.v2"
"cdr.dev/slog/v3"
"github.com/coder/coder/v2/codersdk/agentsdk"
"github.com/coder/serpent"
)
```
3 groups: standard library, 3rd partly libs, Coder libs.
This PR makes the change across the codebase. The PR in the stack above modifies our formatting to maintain this state of affairs, and is a separate PR so it's possible to review that one in detail.
Closes https://github.com/coder/internal/issues/1173,
https://github.com/coder/internal/issues/1174
Currently these two tests are flaky because the contexts were created
before a potentially long-running process. By the time the context was
actually used, it may have timed out - leading to confusion.
Additionally, the `ExpectMatch` calls were not using the test context -
but rather a background context. I've marked that func as deprecated
because we should always tie these to the test context.
Special thanks to @mafredri for the brain probe 🧠
---------
Signed-off-by: Danny Kopping <danny@coder.com>
Resolved an invalid parameter error (-2147024809) during PTY creation on
Windows 11 22H2 (but not only) when connecting via JetBrains Toolbox
which spawns the native SSH client with `-tt` forcing PTY allocation
even though there is no "terminal" on the client side to query its size.
CreatePseudoConsole doesn't accept a 0x0 (zero width and zero height)
console size and unfortunately, there is NO explicit documentation in
the official Microsoft documentation that states the minimum valid
values or explicitly prohibits 0x0.
Looking at real-world implementations in the search results, all
examples use reasonable non-zero values.
I tested this with a local Windows VM registered to dev.coder.com i.e.
externally managed workspace.
Fixes: #20468
- Update go.mod to use Go 1.24.1
- Update GitHub Actions setup-go action to use Go 1.24.1
- Fix linting issues with golangci-lint by:
- Updating to golangci-lint v1.57.1 (more compatible with Go 1.24.1)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <claude@anthropic.com>
The experimental functions in `golang.org/x/exp/slices` are now
available in the standard library since Go 1.21.
Reference: https://go.dev/doc/go1.21#slices
Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>
- Adds `testutil.GoleakOptions` and consolidates existing options to
this location
- Pre-emptively adds required ignore for this Dependabot PR to pass CI
https://github.com/coder/coder/pull/16066
Fixes a race in TestUpdateValidateRichParameters where the parameter is sent prior to the prompt.
Causes errors like: https://github.com/coder/coder/actions/runs/11681622439/job/32527173007
```
ptytest.go:132: 2024-11-05 10:02:18.819: cmd: "bool_parameter"
ptytest.go:167: 2024-11-05 10:02:18.819: cmd: matched "bool_parameter" = "bool_parameter"
update_test.go:440: 2024-11-05 10:02:18.819: cmd: stdin: "cat\r\n"
ptytest.go:132: 2024-11-05 10:02:18.819: cmd: "> Enter a value (default: \"\"): can't validate build parameter \"bool_parameter\": boolean value can be either \"true\" or \"false\""
ptytest.go:167: 2024-11-05 10:02:18.819: cmd: matched "boolean value can be either" = "\n> Enter a value (default: \"\"): can't validate build parameter \"bool_parameter\": boolean value can be either"
update_test.go:440: 2024-11-05 10:02:18.819: cmd: stdin: "\r\n"
ptytest.go:167: 2024-11-05 10:02:18.819: cmd: matched "Enter a value" = " \"true\" or \"false\"\n> Enter a value"
update_test.go:440: 2024-11-05 10:02:18.819: cmd: stdin: "false\r\n"
ptytest.go:132: 2024-11-05 10:02:18.821: cmd: "> Enter a value (default: \"\"): can't validate build parameter \"bool_parameter\": boolean value can be either \"true\" or \"false\""
```
* chore: add /v2 to import module path
go mod requires semantic versioning with versions greater than 1.x
This was a mechanical update by running:
```
go install github.com/marwan-at-work/mod/cmd/mod@latest
mod upgrade
```
Migrate generated files to import /v2
* Fix gen
It looks like it is possible for screen to use control sequences instead
of literal newlines which fails the tests.
This reuses the existing readUntil function used in other pty tests.
Writing to stdin for `coder ssh` too early could result in the input
being discarded. To work around this we add a new `ptytest` method
called `ReadRune` that lets us read one character of output. This will
indicate the command is ready to accept input.
It could be one character of the prompt, or of the loading message
waiting for connection to be established.
To ensure ptytest closure always happens the same way, we now define a
new `Close` function on `PTY` and always call the close function instead
of manually closing read/writers.
A few additional log messages have been added as well, to better
understand the shutdown process in case of errors.
There are sporadic flakes in some tests on Windows related to the use of
`ptytest`. Since it's not clear why they fail, this commit adds some
more logging and timeouts to the cleanup methods.
Since we were not failing tests with `require` the error output was
somewhat hidden in the stream of log messages. This change standardizes
`ptytest` logging and failing to improve visibility.
