* this library does not seem used so remove it and its dependency from the app
* use the flask version of sentry-sdk so it can install its dependencies
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* some better error logging when the background processor fails w/ burnettk
* handle process model not found issues better and do not raise in handle exception if it is missing w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* delete from process caller table explicitly before removing from reference cache to hopefully avoid integrity errors
* check if reference cache id list has items to avoid unnecessary db calls
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
Co-authored-by: Kevin Burnett <18027+burnettk@users.noreply.github.com>
* return a blank task list if the new instance has not started yet and does not have an associated bpmn process
* fixed minor typo
* use ruff instead of black for formatting w/ burnettk
* upgrade ruff and remove references to black w/ burnettk
* remove unnecessary packages for pre-commit w/ burnettk
* we do not use reorder lib w/ burnettk
* ruff is needed w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* added test to make sure process caller relationship table is properly updated and refactored clear reference cache to hopefully fix oddities when attempting to run it w/ burnettk
* fixed pyl w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* updated SpiffWorkflow and added test to make sure correct task gets the error in a multi-instance w/ burnettk
* use main of SpiffWorkflow w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* queue instructions after running a manual task as well
* avoid adding instructions for human tasks
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* call the proc index when we need it rather than passing it through w/ burnettk
* removed all uses of additional_processing_identifier w/ burnettk
* added comment about why we need current proc index and made coderabbit suggestion w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* Fix process data get subprocess (#1493)
* added test to make sure we can get the data object of a sub process
* avoid the processor altogether to get data objects but use the db directly
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* get the data object value from the bpmn process that defines it w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* added test to make sure we can get the data object of a sub process
* avoid the processor altogether to get data objects but use the db directly
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* Revert "Fix process data get subprocess (#1493)"
This reverts commit 84feef321d.
* added process model so we can test fixing loading task data with mi w/ burnettk
* added test and a potential fix w/ burnettk
* pyl is passing
* Update spiffworkflow-backend/tests/spiffworkflow_backend/unit/test_process_instance_processor.py
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
* put back api yml description
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
* added test to make sure we can get the data object of a sub process
* avoid the processor altogether to get data objects but use the db directly
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* added new table to handle called process relationships
* process caller api endpoint is working again w/ burnettk
* fixed more tests and mypy w/ burnettk
* unit tests are passing and commented out unrelated failing test w/ burnettk
* all tests are passing almost w/ burnettk
* uncommented flakey test w/ burnettk
* minor change while reviewing w/ burnettk
* some changes from coderabbit suggestions w/ burnettk
* commented out sampling change w/ burnettk
* name the foreign keys on the process caller table w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* wip related to giving access to groups when you have subgroups
* handle deny permissions and check normal permissions if pg parent method as well w/ burnettk
* reverted local dev perm file w/ burnettk
* a little test cleanup
---------
Co-authored-by: burnettk <burnettk@users.noreply.github.com>
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* added config to turn off task execution with threads w/ burnettk
* do not run tasks in parallel if they have gateway children w/ burnettk
* remove debugging code w/ burnettk
* attempting to fix flakey test w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* if you have access to things in a group, allow navigating through group
* lint
---------
Co-authored-by: burnettk <burnettk@users.noreply.github.com>
* added test to demonstrate data store issue when run in certain contexts w/ burnettk
* added method to add kkv data store getters back onto tasks if appropriate w/ burnettk
* removed debug code w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* load task data for the immediate completed parent and for cancelled tasks
* reverted query_tasks script
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
Co-authored-by: Kevin Burnett <18027+burnettk@users.noreply.github.com>
* updated import pi test to set task data w/ burnettk
* fixed persist test so it loads the processor with data
* load data and completed tasks when getting bpmn json for an instance
* save start and end times on tasks on import w/ burnettk
* verify we are saving start and end times for imported process instances w/ burnettk
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
* added basic support to store process instance file data on the file system w/ burnettk
* the files need to be in a hashed directory structure w/ burnettk
* fixed exception message w/ burnettk
* use the same entry method to the contents of data file w/ burnettk
* fixed hash method w/ burnettk
* make hash method easier to read w/ burnettk
* added a migrator and test for migration w/ burnettk
* fix a couple things while testing
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
Co-authored-by: burnettk <burnettk@users.noreply.github.com>
* check for mi tasks first before checking more newest in task show api w/ burnettk
* remove console
---------
Co-authored-by: jasquat <jasquat@users.noreply.github.com>
Co-authored-by: burnettk <burnettk@users.noreply.github.com>
jasquat
burnettk
Chris Daßler
Sky
jbirddog
dependabot[bot]
Kevin Burnett
jasquat