OutputManager configuration file keys

runmodes: comma separated list of runmodes (single, master ensemble)

Runner configuration file keys

opts*: The command line options used for running the script, multiple keys matching opts* can be given
single*, master*, ensemble*: Additional options for the specific runmodes. Multiple keys matching <runmode>* can be given.

Example usage:

# The options used for running the scripts, multiple keys can be given if they match opts*
opts=--etat 8 --sdf 3
opts1=--dc 0 --Dt 0.1 --NDt 10

# runmode specific options
single=...
single1=...
ensemble=...
master=...

Verifier configuration file keys

The Verifier compares some test 'this' to another test 'other'.

this: Test name of 'this', by default the current test if missing
other: Testname of 'other', by default the results from the directory of expected results (OutputManager::expecteddir)
verify: Verify that both trajectories are exactly equal (default if this key is missing or verify=full), or verify that the last outcome of the simulation is equal, e.g. timesteps may differ (verify=outcome)

If this=some_test is specified, it is probably also a good idea to import=some_test to keep the runmodes in sync. Currently the directory of expected results is Testing/expected, it is kept under version control so that changes in the output of the scripts are noticed.

GenericContinuer configuration file keys

firstrun: script options for the first run
secondrun: script options for the second run

Continuer configuration file keys

See GenericContinuer keys.

PythonContinuer configuration file keys

See GenericContinuer keys.

CompileTarget configuration file keys

error: Turn on "failure mode". The error message which is expected in the output.
dependencies: Space separated list of dependencies to compile first. These are always required to succeed, independent of the presence of error.

StateComparer configuration file keys

trajectories: List of comma-separated trajectories which should be tested.
function: A meta-function which should return the actual test function. The actual test function should accept the state array and return some epsilon value (the measure of the test).
parameters: Tuple of function parameters passed to the meta function.

The following configuration keys are read from the 'target'-sections.

runmodes_<test>: For the compare test <test>, only use these runmodes.
epsilon_<runmode>_<test>: Acceptable deviation for the given runmode and comparison test.

TrajectoryComparer configuration file keys

pdf: Save plots to this pdf file.
reference: Section of reference trajectory
trajectories: List of comma-separated trajectories which should be compared to the reference.

The following configuration keys are read from the 'target'-sections.

runmodes_<test>: For the compare test <test>, only use these runmodes.
columns_<test>: Use these columns of the output files for the comparison.
epsilon_<runmode>_<test>: List of acceptable deviations for the given runmode and comparison test.
postprocess_local: Name of a global function which expects the data array as input and postprocesses the data.
format_local: specifies which columns are floats (f) and which are complex numbers (c). Example: "f+f+c+c" will result in 6 columns, the two complex number columns are split into real and imaginary parts.
start_<test>: The first row of the data lines to consider for the comparison test <test>.
length_<test>: How many lines of data to consider for the comparison test <test>.

FunctionComparer configuration file keys

reference_function: Name of a global function, which should return a tuple of a unary function and a label used in plots.

The following configuration keys are read from the 'target'-sections.

paramters_<test>: List of tuples or single tuple which are passed to the reference function. Example: [(1,5,3),(2,2,1)] or (1,5,3). If this is a list, each entry corresponds to a column of the data file, otherwise the same parameters are used for all columns.