Open thomasmerz opened 5 months ago
Oh, wow, that's a really neat idea! The only downside I can think of, is that'd be to a detriment to the security aspect that Docker provides... e.g. if the $PWD happens to be ~, it'll just expose everything the user owns to whatever's in the container. Right now it has no ability to "infect" or "hijack" anything, but the -v
thing would eliminate that sense of security.
I could make it create a temporary directory, e.g. .docker-monolith-temp-39t84tXb
, put the -o
file there, move it out of there to one level above, and then clean up by removing .docker-monolith-temp-39t84tXb
after. That'd preserve the security, yet allow to use the -o
option.
I'm doing this (-v "$PWD:/mnt"
) already for running shellcheck
on my Macbook in a docker container - so it should be safe enough when someone normally wants the output to be written to a file relative to the current working directory 😄
docker run --rm -v "$PWD:/mnt" koalaman/shellcheck:latest -Calways --exclude=SC1090,SC1091 -S warning $*
Wow, that's some 1337 CLI stuff you have going in there. I'll try to do that thing with the temporary directory, will show what I have in a PR.
create a temporary directory
From a user's perspective: do whatever is neccessary to keep it as simple as possible when using it. So you would need to fix this line:
$DOCKER run --rm y2z/$PROG_NAME "$@"
to something like that?
$DOCKER run --rm -v "$PWD:/mnt" y2z/$PROG_NAME $url -o /mnt/$outputfile
I'm using
monolith
in docker container and followed your installation instructions. But when I run this:I'm getting no file written at all on my local disk because the docker container has no volume ($PWD) mounted 😞
This can be fixed by this: