schrodinger.test.scival module

Tools used by SciVal test writers. Of most common use will be’ JobControlJob and Reporter. Every group of SciVal tests should probably have a Reporter.

class schrodinger.test.scival.JobControlJob(command: List[str], suite: str, name: str, command_dir: Optional[str] = None, max_retries: Optional[int] = None, launch_env_variables: Optional[Dict[str, str]] = None, **kwargs)

Bases: schrodinger.job.queue.JobControlJob

__init__(command: List[str], suite: str, name: str, command_dir: Optional[str] = None, max_retries: Optional[int] = None, launch_env_variables: Optional[Dict[str, str]] = None, **kwargs)

Initialize a job object for SciVal jobs, which also includes suite and name attributes.

Parameters

• command – commandline for launching job

• suite – name of scival suite

• name – job name, for annotation for JobDJ before job is running

• command_dir – directory to launch job from, default to CWD

• max_retries – Number of times to retry failed jobs, default is globally controlled by JobDJ.

• launch_env_variabes – Extra environment variables to add to existing environment for job launching.

kill(*args, **kwargs)

Send kill request to jobcontrol managed job

addFinalizer(function: Callable[[BaseJob], None], run_dir: str = None)

Add a function to be invoked when the job completes successfully.

See also the add_multi_job_finalizer function.

addGroupPrereq(job: schrodinger.job.queue.BaseJob)

Make all jobs connected to job prerequisites of all jobs connected to this Job.

addPrereq(job: schrodinger.job.queue.BaseJob)

Add a job that is an immediate prerequisite for this one.

cancel()

Send kill request to jobcontrol managed job. This method will eventually deprecate JobControlJob.kill

cancelSubmitted() → bool

If the job is still in the ‘submitted’ state, cancel it, purge the jobrecord and set the job handle to None.

Return True if this was successful, False otherwise.

doCommand(host: str, local: bool = False)

Launch job on specified host using jobcontrol.launch_job().

Parameters

• host – Host on which the job will be executed.

• local – Removed in JOB_SERVER.

finalize()

Clean up after a job successfully runs.

genAllJobs(seen: Set[BaseJob] = None) → Generator[schrodinger.job.queue.BaseJob, None, None]

A generator that yields all jobs connected to this one.

genAllPrereqs(seen=None) → Generator[schrodinger.job.queue.BaseJob, None, None]

A generator that yields all jobs that are prerequisites on this one.

getCommand() → List[str]

Return the command used to run this job.

getCommandDir() → str

Return the launch/command directory name. If None is returned, the job will be launched in the current directory.

getDuration() → Optional[int]

Return the duration of the Job as recorded by job server. The duration does not include queue wait time.

If the job is running or has not launched, returns None.

Note that this method makes a blocking call to the job server.

getJob() → Optional[schrodinger.job.jobcontrol.Job]

Return the job record as a schrodinger.job.jobcontrol.Job instance.

Returns None if the job hasn’t been launched.

getJobDJ() → schrodinger.job.queue.JobDJ

Return the JobDJ instance that this job has been added to.

getPrereqs()

Return a set of all immediate prerequisites for this job.

getStatusStrings() → Tuple[str, str, str]

Return a tuple of status strings for printing by JobDJ.

The strings returned are (status, jobid, host).

hasExited() → bool

Returns True if this job finished, successfully or not.

hasStarted() → bool

Returns True if this job has started (not waiting)

init_count = 0

isComplete() → bool

Returns True if this job finished successfully

maxFailuresReached(msg: str)

Print an error summary, including the last 20 lines from each log file in the LogFiles list of the job record.

postCommand()

A method to restore things to the pre-command state.

preCommand()

A method to make pre-command changes, like cd’ing to the correct directory to run the command in.

retryFailure(max_retries: int = 0) → bool

This method will be called when the job has failed, and JobDJ needs to know whether the job should be retried or not.

JobDJ’s value for the max_retries parameter is passed in, to be used when the job doesn’t have its own max_retries value.

Return True if this job should be retried, otherwise False.

run(*args, **kwargs)

Run the job.

The steps taken are as follows:

1. Execute the preCommand method for things like changing the working directory.

2. Call the doCommand to do the actual work of computation or job launching.

3. Call the postCommand method to undo the changes from the preCommand that need to be undone.

runsLocally() → bool

Return True if the job runs on the JobDJ control host, False if not. Jobs that run locally don’t need hosts.

There is no limit on the number of locally run jobs.

setup()

A method to do initial setup; executed after preCommand, just before doCommand.

property state

Return the current state of the job.

Note that this method can be overridden by subclasses that wish to provide for restartability at a higher level than unpickling BaseJob instances. For example, by examining some external condition (e.g. presence of output files) the state JobState.DONE could be returned immediately and the job would not run.

update()

Checks for changes in job status, and updates the object appropriately (marks for restart, etc).

Raises

RuntimeError – if an unknown Job Status or ExitStatus is encountered.

usesJobServer() → bool

Detect, by looking at the jobId, whether this job uses a job server.

class schrodinger.test.scival.Reporter(buildtype: Optional[str] = None, build_id: Optional[str] = None)

Bases: abc.ABC

Report results to the performance database when all jobs are complete. Based on queue._MultiJobFinalizer.

NAME = None

PRODUCT = None

DESCRIPTION = None

__init__(buildtype: Optional[str] = None, build_id: Optional[str] = None)

Parameters

• buildtype – NB or OB, or None if reporting not enabled

• build_id – build-XXX or None if reporting not enabled

addJob(job: schrodinger.test.scival.JobControlJob)

Add a job to the reporter.

addAsFinalizer(job: schrodinger.job.queue.BaseJob)

addResult(name: str, value: Union[int, float], units: Optional[str] = None)

Add a result to be reported. When not reporting results, print them.

Parameters

• name – name of metric to report

• value – value of metric

• units – Optional units for value

analyze(jobs: List[schrodinger.test.scival.JobControlJob])

If all jobs completed, loop over self.tests and use self.addResult() to generate data to be reported.

class schrodinger.test.scival.JobDJ(*args, **kwargs)

Bases: schrodinger.job.queue.JobDJ

__init__(*args, **kwargs)

Constructor.

Parameters

• hosts – A list of (<hostname>, <maximum_concurrent_subjobs>) tuples, where <hostname> is a string and <maximum_concurrent_subjobs> is an integer. The default value of None is determined automatically and is usually desired.

• local – No longer functional in JOB_SERVER.

• max_retries – Number of allowed retries per subjob. If this is set, it is never overridden by the SCHRODINGER_MAX_RETRIES environment variable. If it is not set, the value in default_max_retries is used, and SCHRODINGER_MAX_RETRIES is allowed to override. If you wish to disable restarting altogether, set this value to zero.

• default_max_retries – Number of allowed retries per subjob. This value can always be overridden by the SCHRODINGER_MAX_RETRIES environment variable. Default is zero.

• max_failures – Total number of allowed subjob failures before JobDJ exits. If it is not defined, a default of zero will be used (exit on any failure after attempting to restart), but this can be overridden with the SCHRODINGER_MAX_FAILURES environment variable. To allow an unlimited number of subjob failures, set max_failures to the module level NOLIMIT constant.

• verbosity – There are three allowed verbosity levels: “quiet” - only warnings and errors are printed; “normal” - JobDJ progress is printed; and “verbose” - additional debugging info is printed. Default is “quiet”.

• job_class – The class to use as the default job constructor when the addJob argument is not a BaseJob instance.

• update_delay – The number of seconds to wait between job control database reads for JobControlJob jobs. (This delay is for an individual job, not for any job database read.) Default is None, which causes the module level constant UPDATE_DELAY to be used.

run()

Call this method to run all jobs that have been added. The method will return control when all jobs have completed.

Parameters

• status_change_callback – A function to call every time a job status changes. For example, JobState.RUNNING->JobState.DONE. This function takes a single argument of a schrodinger.job.queue.BaseJob object.

• periodic_callback – A command to call periodically, regardless of whether job status has changed or not. The function will be called without any arguments.

• callback_interval – The interval at which the periodic interval will be called. This time is only approximately enforced and will depend on the timing delay settings (e.g. MONITOR_DELAY).

• restart_failed – True (default) if previously failed jobs should be restarted, False if not.

printStatus(job: Optional[schrodinger.job.queue.BaseJob] = None, action: Optional[str] = None)

Override base printStatus. Change formatting, replace printing the jobname with printing the test name (usually the same, but…), add printing of suite.

Parameters

• job – job object to print the status of, otherwise prints general status

• action – custom string to print as status of a job, otherwise use standard status

addJob(job: schrodinger.test.scival.JobControlJob, add_connected: bool = False, **kwargs)

Add a job to run. If job is not a BaseJob instance, a BaseJob instance is constructed with job as the first argument. The default BaseJob class for the JobDJ instance can be specified in the constructor for JobDJ.

Additional keyword arguments are passed on to the job constructor.

All job prerequisites and dependencies need to be specified before adding a job to JobDJ.

Parameters

add_connected – If True, for jobs with dependencies only one job per connected group should be added and all connected jobs will be discovered and added automatically. If False, it is the user’s responsibility to make sure that any prerequisites of a job are also added.

addTests(item: _pytest.nodes.Item, build_id: Optional[str] = None, buildtype: Optional[str] = None)

Call the add_tests function in the Pytest Item.

Parameters

• item – Current node that represents an add_tests function

• build_id – build-XXX matching current build, can be None if not reporting

• buildtype – NB or OB, can be None if not reporting

setItem(pytest_item: _pytest.nodes.Item)

Set the current Pytest Item.

Jobs added to the jobdj while the pytest_item is set to pytest_item will be associated with it for error reporting &c.

clearItem()

Clear the current Pytest Item.

getCommandDir() → Optional[str]

The default job launching directory

(matches BaseJob interface)

Returns

directory of job launching, or None if CWD

killAllJobsForSuite(suite: str)

Kill all running jobs using the named suite, remove all waiting jobs for with the named suite.

property active_jobs

property all_jobs

disableSmartDistribution()

Disable smart distribution of jobs.

Smart distribution allows subjobs to run on the machine that JobDJ is running on when JobDJ itself is running under a queuing system. This is usually desirable since the JobDJ process doesn’t generally consume significant computational resources and you don’t want to leave a queue slot mostly idle.

property done_jobs

Successfully completed jobs, sorted into the order they were marked as completed by JobDJ.

dump(filename: pathlib.Path)

Pickle the JobDJ instance to the specified file name.

property failed_jobs

getActiveProcCounts() → Dict[str, int]

Return a dictionary containing the number of active jobs on each host.

hasStarted() → bool

Returns True if JobDJ has started already

isComplete() → bool

Returns True if JobDJ has completed, False otherwise.

killJobs()

Kill all active jobs

markForRestart(job: schrodinger.job.queue.BaseJob, action: str)

Mark a job as dead, but make sure that it gets restarted.

Parameters

action – Describes the reason the job is being restarted.

setHostList(host_list: List[Tuple[str, int]])

Define compute hosts to run subjobs on.

Active jobs are not affected by a change in the host list.

Parameters

host_list – A list of (<host_entry_name>, <maximum_concurrent_subjobs>) tuples, where <host_entry_name> is a string and <maximum_concurrent_subjobs> is an integer.

property total_active

The number of jobs currently running.

property total_added

The number of individual jobs that have been added to the JobDJ instance.

property total_failed

The number of jobs that have failed.

property total_finished

The number of jobs that have finished successfully.

property waiting_jobs

Jobs waiting to be started.

schrodinger.test.scival.call_and_report(item: _pytest.nodes.Item, when: str, fxn: Callable, *args)

Execute a function and report its result to pytest.

A slightly modified version from _pytest.runner to deal with functions that don’t return the word “passed”.