Martin Braun
Martin Braun
Right now we emulate migrations via the compose_cli. These are not transactional, though and if something goes wrong we leave the dataseries broken and might have to do data recovery...
- [ ] Prototype UX for a compose_client command and library that polls a dataseries using ?changes_since (without deletion events) - [ ] library functionality - [ ] Design Interface...
Currently the compose client cli is intended to be used as a component of bash scripts to handle the logic around diffing then pushing changes of dataseries for migrations. While...
In 2.1.0 we left in some code that should allow for read only access to removed backend data. This code was left untested as all test code related to that...
We currently don't have an opensourced example project that shows how to use NF Compose and especially with the compose_client.
Currently we only support the rather wasteful pattern of generating canonical ids for DataPoints using this logic: https://github.com/neuroforgede/nfcompose/blob/main/skipper/skipper/dataseries/storage/uuid.py ```python def _gen_uuid(data_series_id: Union[uuid.UUID, str], external_id: str) -> str: computed_id = hashlib.sha256(external_id.encode('UTF-8')).hexdigest()...
see https://packaging.python.org/en/latest/tutorials/packaging-projects/
To unpack we can use this: see https://crazymax.dev/undock/usage/examples/ see https://github.com/crazy-max/undock/ To pack, we can simply use scratch docker images. nothelm.py should do this automatically though so that users do not...
it could be a good idea to include jinja plugins similar to how ansible does this. This ticket is here to track ideas regarding this.
This project currently somewhat has a hard requirement for Ansible 2.8.5 because we have not tested it with anything else. We should add support for newer Ansible versions. This should...