TODO

• Finish scraping code for TODO to be added to this list.

• Add documentation for more classes.

• Change away from using import ispyb.model.__future__.

• Reconsider package/module placement for remex keywords.

• Use asyncio wait or sentinel for all started processes to be dead.

• Centralize definition of configuration_keyword symbol.

• Use ImportlibMetadataVersion to get package versions.

• Sanitize input to sql query when looking for data label in ispyb.

• Put lock around computing datacollection_number when creating new catalog workflow run.

• When updating ispyb autoproc progarm status, determine if job is blocked due to error.

• Add ispby file attachment type and rank.

• Figure out some good things to do in the gda parser base class.

• Fix hierarchy crossover where dataface imports job constants.

• Use filestore object to discover all available files in task runtime directory.

• Compose news details to open under click or hover.

• Parallelize context process alive/dead checking.

• Fix early-send problem when setting up news server/client context.

• Set permission on sqlite file from configuration.

• Properly escape literal label field in sql query.

• Improved storage efficiency by separate lookup for task error lines in get_bx_jobs_bx_tasks_bx_gates.

• Die if inserting a new job when its directory already exists.

• Gui should get system health in parallel network requests.

• Iron out where data_label comes from when submitting a new workflow from the gui.

• Move sizewatch inside particular ux that needs it.

• Figure out a way to put the package’s version into the version.js file.

• In aiohttp BxLauncher, give special http status to indicate task count max exceeded.

• Make the task specification contain pythonpath, module, conda or venv.

• Use the utilpack library for building the main.sh bash script.