Ergonomic, hermetic, user-friendly software build system for Python, Go, Java, Kotlin, Protobuf, Scala, Shell. Pants lets you fearlessly scale up your codebase!

Pants

```-to_sort=$(mktemp -t changed.isort.XXXXX)
-trap "rm -f ${to_sort}" EXIT
-if [[ "$(./pants changed | tee ${to_sort})" != "" ]]
-then
-  ./pants -q --target-spec-file=${to_sort} fmt.isort -- ${isort_args[@]}
-fi
+./pants -q --changed-parent=HEAD fmt.isort -- ${isort_args[@]}```

<@U0AR45336>: could probably use that in ci... `./pants --changed-parent=master filter-minimize`

anyone have any clues as to where the stdout string `100% .................................................. 18153 KB 0.592s` might be coming from?

seems to be getting intermittently outputted to stdout in an integration test that attempts to parse the pants output as json since the v2-engine land

```                     failed to parse stdout:
                     100% .................................................. 18153 KB 0.592s
                     {
                         "preferred_jvm_distributions": {
                             "java7": {
                                 "strict": "/usr/lib/jvm/java-7-openjdk-amd64",
```

yeah possible. the `preferred_jvm_distributions` part should be from `./pants export`

yeah it’s happening in `tests/python/pants_test/core_tasks/test_deferred_sources_mapper_integration.py::DeferredSourcesMapperIntegration::test_deferred_sources_export_successfully`

best practice would be `./pants export —output-file=..`

should be able to pipe things around freely tho

oh you mean that `.................................................. 18153 KB 0.592s` should be in stderr?

right that’ll be ideal. if you need to something quick, `—output-file` should do

likely came from <https://github.com/wisechengyi/pants/blob/96e0187c96cc799e2135597859585a85d0a33758/src/python/pants/net/http/fetcher.py#L182-L182>

filed <https://github.com/pantsbuild/pants/issues/4370>

it appears that you can’t Ctl+C the engine even without the daemon

Depending on when in doing work things are it either segfaults or hangs

```
$ time   ./pants list ::
^CFrom cffi callback &lt;function extern_val_for at 0x101e46758&gt;:
[2]+  Done                    say $cmd_name $status
Segmentation fault: 11
```

late, you get into the pre-existing Ctl+C failure mode of killing multiprocess worker threads