Closed Klubas closed 1 year ago
I am a bot, here are the test results for this PR: https://ci-tests.linuxserver.io/lspipepr/transmission/3.00-r5-pkg-e8beb90f-pr-202/index.html https://ci-tests.linuxserver.io/lspipepr/transmission/3.00-r5-pkg-e8beb90f-pr-202/shellcheck-result.xml
I am a bot, here are the test results for this PR: https://ci-tests.linuxserver.io/lspipepr/transmission/3.00-r5-pkg-e8beb90f-pr-202/index.html https://ci-tests.linuxserver.io/lspipepr/transmission/3.00-r5-pkg-e8beb90f-pr-202/shellcheck-result.xml
This option was existing in the past and was removed.
The bigger problem here is that each time the container is started, the default complete and incomplete dirs are created. Therefore, no matter what you put in these variables, the default directories will be created:
# make folders
mkdir -p \
/downloads/{complete,incomplete} /watch
I opened this pull request here which is meant to solve your problem as well.
IMHO, you should configure your directories in the settings.json file and deploy the file at the same time as your container. I'm personally really skeptical about adding more and more environment variables for elements that are already present in the settings.json configuration file.
I didn't though that creating the default dirs despite the env vars being set was a big problem, that's why I left it there, but I agree it's not the ideal scenario. I could change the script to only create the folders that will be really used, but your solution is definetely more elegant.
This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
Is there still someone maintaining this repository?
This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
It makes the most sense to me to make an ENV var that dictates whether or not the user is specifying their own settings.json file, that way the rest of the configuration needed can be done by the user without opening more PRs for every setting that needs changed.
I see no benefit to this PR. downloads directory can already be set within the transmission webui and any transmission remote client. Editing settings.json can also accomplish this. Adding more vars would just confuse the majority of our users.
in reply to https://github.com/linuxserver/docker-transmission/pull/202#issuecomment-1259876081 a user can already stop the container and drop in a settings.json to the config path they've selected, there is no need for a variable to accomplish this.
I do see value in this. The benefit for setting downloads-dir and incomplete-dir through environment variables would be the ease of configuration. I could setup 1 docker-compose file (or unraid docker template) without having to do extra steps separately in the settings.json file or any remote client. It would start and run as needed.
See the feature added here as an example: https://haugene.github.io/docker-transmission-openvpn/config-options/
"You may still override Transmission options by setting environment variables if that's your thing. The variables are named after the transmission config they target but are prefixed with TRANSMISSION_
, capitalized, and - is converted to _."
Description:
Added support for configuring the default download directory and the directory used to store incomplete downloads via the DOWNLOAD_DIR and INCOMPLETE_DIR environment variables.
Benefits of this PR and context:
With this change its possible to set de default download dir just setting the env var on docker-compose.yml, without needing any additional files and a custom Dockerfile.
How Has This Been Tested?
Run the container locally using a docker-compose.yml file to see if the changes where applied correctly to the settings.json file.
Source / References: