Command line interface

When installed, Smttask adds the following commands to the shell.

• smttask project init

  Run the initialization wizard for SumatraTask.

• smttask run

  Run a task description file. This is the most common way of running a task, especially for batched runs.

  The command supports basic multiprocessing and queueing of multiple tasks. Also supports using a separate temporary record store; this is useful when running many parallel jobs on a compute cluster where the only available DB backend is SQLite, to avoid file lock conflicts.

• smttask store find_output

  Find output files for a previously run task, given its task description file.

• smttask store rebuild

  Rebuild the input datastore.

  This is useful if e.g. an update has caused all of the Task digests to change, in order for previous computations to be found by future Tasks.

• smttask store create_surrogates

  Create surrogate records for outputs without records

  This allows routines which query the record store for outputs to work as expected, but of course statistics like run time for surrogate records are undefined.

• smttask store merge

  Merge entries from the SOURCES record store(s) into the target record store.

  Intended usage is for combining run data that was recorded in separate record stores with the –record-store option of smttask run.