#So how do .env files work?
1 messages · Page 1 of 1 (latest)
Just found this https://docs.dagger.io/features/local-defaults/ following that now.
this, exactly. Ping if you have any questions 🙏
I've tried it out and I'm not able to see files changing in my running services. I'm using this pattern:
FilteredDefaultDirectory = Annotated[
dagger.Directory,
dagger.DefaultPath("."),
dagger.Ignore(["!frontend", "!backend"]),
]
@object_type
class AwesomeThing:
@function
def something(self, source: FilteredDefaultDirectory):
pass
Do DefaultPath and Ignore play nice together?
Just tried it without the Ignore and still got no love
the logs do say noCache=true tho...
I just did this minimal sample ```FilteredDefaultDirectory = Annotated[
dagger.Directory,
dagger.DefaultPath(
".",
),
dagger.Ignore(["!frontend", "!backend"]),
]
@object_type
class ReloadDefaultPath:
@function
def try_reload(self, source: FilteredDefaultDirectory) -> dagger.Service:
# Use GitLab mirror in CI, public registry locally
base_image = "public.ecr.aws/docker/library/python:3.12-slim"
container = dag.container().from_(base_image).with_directory("/app", source)
return container.as_service(args=["sh", "-c", "while true; do sleep 2; done"])
and ran it with `dagger call try-reload up --ports 1234:1234` and used the CLI to navigate to the service and get a terminal with `t`. I `tail`d a file, changed it in on the host, then `tail`d it again:
dagger / $ tail app/backend/src/app.py
exit(1)
uvicorn.run(
"app:app",
host="0.0.0.0",
port=8000,
reload=True,
)
hello
dagger / $ tail app/backend/src/app.py
exit(1)
uvicorn.run(
"app:app",
host="0.0.0.0",
port=8000,
reload=True,
)
hello
dagger / $
No change.
Oh man I just totally lost the plot on this thread.
my b
@cloud linden sorry for the late reply. Dagger doesn't have a concept of bind mounts yet. This means that files are transferred to your pipeline only once and they get snapshotted for reproducibility purposes. I understand your use case and we so plan to support it soon but there's currently no way to have live-syncing while your pipeline is running
Is there a convenient way to do Dev work using dagger, or are we stuck with docker compose?