Releases: nanos/FediFetcher
v4.1.2
v4.1.1
Change log
- Fetch remote replies to bookmarks (#19)
- File based locking to prevent overlapping executions of the script in cron mode (#18)
- Implement Support for Dead Man Switch monitoring (#20)
- Prevent schedule from being disabled after 60 days (#17)
Update instructions
If you are running from Git Hub Actions
- Update your fork
- If you wish to use dead man monitoring, provide the
ON_START
,ON_DONE
, and/orON_FAIL
environment variables as desired. - If you wish to use the option to fetch remote replies for bookmarks:
- Add the
read:bookmarks
scope to your Access Token - Re-generate your access token
- Supply the new
ACCESS_TOKEN
as an action secret - Supply
MAX_BOOKMARKS
as environment variable
- Add the
If you are running locally
- Fetch the latest version
- Provide the
--lock-hours
parameters as a value greater than0
- If you wish to use dead man monitoring, provide the
--on-start
,--on-done
, and/or--on-fail
argument, as desired. - If you wish to use the option to fetch remote replies for bookmarks:
- Add the
read:bookmarks
scope to your Access Token - Re-generate your access token
- Supply the new
ACCESS_TOKEN
as an action secret - Supply the
--max-bookmarks
argument to the script
- Add the
v4.1.0
Change log
New features
- This release adds the functionality to backfill profiles for your pending follow requests (#7)
Fixes
Update instructions
If you are running from Git Hub Actions
- Unless you wish to make use of the option to backfill profiles of pending follow requests, no action is required on your side, as the Action will always use the latest release.
- If you do wish to make use of the option to backfill profiles of pending follow requests:
- Update your fork of the repository.
- Add the
read:follows
scope to your access token, regenerate the token, and update theACCESS_TOKEN
Action Secret. - Provide the
MAX_FOLLOW_REQUESTS
environment variable.
If you are running locally
- Get the latest copy of
find_posts.py
- If you wish to make use of the option to backfill profiles of pending follow requests, you must add the
read:follows
scope to your access token, regenerate the token, and supply it, together with the new--max-follow-requests
flag
v4.0.4
v4.0.3
v4.0.2
Thanks to @nikdoof, this script is also available in a pre-packaged container, mastodon_get_replies. 🎉
- Pull the container from
ghcr.io
, using Docker or your container tool of choice:docker pull ghcr.io/nanos/mastodon_get_replies:latest
- Run the container, passing the command line arguments like running the script directly:
docker run -it ghcr.io/nanos/mastodon_get_replies:latest --access-token=<TOKEN> --server=<SERVER>
The same rules for running this as a cron job apply to running the container, don't overlap any executions.
An example Kubernetes CronJob for running the container is included in the examples
folder.
v4.0.1
v4.0.0
v3.0.2
Fixes two issues:
- Extremely large gaps during rate limit timeout / incorrect handling of time zones #8
- Workflow failing with “IndexError: list index out of range” #9
Users of GitHub actions don’t need to do anything, as changes are applied automatically.
For users who run the script manually, please fetch a new copy of get_context.py
v3.0.1
In this version you can also backfill your most recent followers' posts.
To enable this firstly update your fork to v3.0.1, then provide the following two variables at Settings > Environments > Mastodon:
MAX_FOLLOWERS
: An integer number representing how many of your last followers you want to backfill. (e.g.80
). Leave blank to disable this part of the script.USER
(if not already provided): The username of the user whose followings you want to pull in (e.g.Michael
for the user@[email protected]
). Leave blank to disable this part of the script.
For full details see the blog post: Pull missing posts from recently followed accounts into Mastodon