- Small bug fixes for the build and `deployment_utils`, see 88
- Releases "support" for Python 3.7
- Adds `EBDeployer` to `deployment_utils.py`, allowing us to run both normal deployments and indexer deployments from dcicutils
- Introduces `indexer` env compatibility + first version of VirtualApp + bug fixes
- Fixes problems with CGAP bucket resolution, merging 66 67 68
- Merges https://github.com/4dn-dcic/utils/pull/62
- NOTE: all previous versions are now incompatible with our production setup. You MUST use at least this version if you are interacting with production data.
* fixed a weird behavior newly introduced in s3_utils unzip_s3_to_s3
handling of zip files without a directory added to unzip_s3_to_s3 in s3utils.
- Merge 53 54 and 55
- Hotfix from 0.9.0 to account for bug in poetry
- Pin `colorama` version
- Hotfix: allow Limit in faceted_search
- See 51
- Add more user friendly search
- Add QC metric
Incorporates https://github.com/4dn-dcic/utils/pull/49 which adds code to work with namespacing changes on the portal.
Several minor changes.
- Use `beanstalk_utils.update_bs_config` to update existing EB environments in `beanstalk_utils.create_bs`
- Set up kwargs to make `ff_utils. generate_rand_accession` more flexible
- Remove `awcli` requirement
- Change `ff_utils.get_es_metadata` to use pre-existing auth when getting health page info
Change `s3_utils.s3Utils.get_access_keys` and related methods to use a new encryption key and new filenames. The idea is to use different sets of keys with different resources with s3Utils.
Also defined a list of installation dependencies in setup.py. Many of these are flexible range dependencies. Previously `install_requires` was set to the contents of requirements.txt.
One small tweak to `expand_es_metadata` to skip abstract schemas.