lsst-rsp is versioned with semver.
Find changes for the upcoming release in the project's changelog.d directory.
- added
get_query_history()
function
- Pass in session as a keyword argument to initialization of TAP classes
- If the auto notebook pulls from GitHub time out, log an error and keep going rather than crashing.
- Set the default
DAF_BUTLER_CACHE_DIRECTORY
path to a sibling directory of theTMPDIR
directory if moved into the scratch directory instead of using a subdirectory.
- When using a scratch directory as the root of directory of temporary files, set the mode on that directory to 0700 if it has to be created.
- Conditionally set environment variable
DAF_BUTLER_CACHE_DIRECTORY
- Create user-specific TMPDIR under /scratch if applicable.
- Allow for the relative location of the logging source used in the two-Python model.
- Handle
RESET_USER_ENV
insidelsst.rsp.startup
. - Add utility functions to get values that were constants in the single-Python model but now are not.
- Stop forcing the value of
FIREFLY_HTML
, since newer versions of the Firefly plugin do not want it to be set. It must be set in the Nublado configuration if needed.
- Fix creation of the parent directory for the logging configuration.
- Add
RSPClient
, a configured HTTP client for services in the same RSP instance.
- Create the path to the logging profile directory if it doesn't exist.
- If the user credentials directory
$HOME/.lsst
does not exist, create it.
- Fix working directory for
git config --get
when checking whether a local Git repository needs to be updated.
- Add an
lsst.rsp.startup
module andlaunch-rubin-jupyterlab
entry point to replace most of therunlab.sh
script in sciplat-lab. This handles startup and configuration of the Rubin-customized JupyterLab.
- Raise
ValueError
for invalid arguments to utility functions rather than bareException
.