GitLab

restored the semaphored version to its original simplicity (the -wait_for rule was accidentally checked in earlier)

control_rules are now centrally stored in controlled Analysis object; -wait_for rule added for testing

pipeline_name is now automatically computed from ClassName; simplified workshop's example files and slides

added support for -pipeline_url as an input parameter (no need to supply hive_driver or password in this case)

bugfix: substituting the accu signature on demand from the very depths of emitting job's param_stack

Dataflowing minimal information out of Runnables, relying on templates in PipeConfig file to extend it if needed

Start using the new db_conn.pl script instead of building driver-specific commands and running them.

Slow the example down a bit and allow 2 Workers. In "-can_specialize 1" mode the two Workers will complete the whole pipeline.

free 'Start' from dealing with 'a_multiplier' by using an input_id_template in PipeConfig instead; renamed 'Start' to 'DigitFactory' to reflect that

change of default behaviour: hive_capacity is now off by default (=NULL); setting hive_capacity=0 or analysis_capacity=0 stops scheduling AND specialization to a particular analysis

added a new per-analysis "analysis_capacity" limiter for cases where users want to limit analyses independently

Added API and schema support for analysis_base.meadow_type / Analysis->meadow_type(), which will be NULL/undef by default

Dropped 'BLOCKED' job status and introduced 'SEMAPHORED' status that is maintained in sync with semaphore_counts; less confusing and more efficient (with new 3-part index)

Removed maximise_concurrency and added analysis_stats.priority to guide the scheduler; improved scheduler and LSF meadow

Incorporate ENV hash into the tree of possible options in order to be able to "require" a value. And a bit of config inheritance cleanup.

Branch names proved to be too confusing. The functionality stays, but is removed from the example to follow.