ecsjobs.jobs.base module

class ecsjobs.jobs.base.Job(name, schedule, summary_regex=None, cron_expression=None)

Bases: object

Base class for all Job types/classes.

Parameters:

• name (str) – unique name for this job
• schedule (str) – the name of the schedule this job runs on
• summary_regex (string or None) – A regular expression to use for extracting a string from the job output for use in the summary table. If there is more than one match, the last one will be used.
• cron_expression (str) – A cron-like expression parsable by cronex specifying when the job should run. This has the effect of causing runs to skip this job unless the expression matches. It’s recommended not to use any minute specifiers and not to use any hour specifiers if the total runtime of all jobs is more than an hour.

_schema_dict = {'properties': {'class_name': {'type': 'string'}, 'cron_expression': {'type': 'string'}, 'name': {'type': 'string'}, 'schedule': {'type': 'string'}, 'summary_regex': {'type': 'string'}}, 'required': ['name', 'schedule', 'class_name'], 'title': 'Configuration for base Job class', 'type': 'object'}

Dictionary describing the configuration file schema, to be validated with jsonschema.

duration

Return the duration/runtime of the job, or None if the job did not run.

Returns:job duration Return type:datetime.timedelta or None

error_repr

Return a detailed representation of the job state for use in error reporting.

Returns:detailed representation of job in case of error Return type:str

exitcode

For Job subclasses that result in a command exit code, return the integer exitcode. For Job subclasses that result in a boolean (success / failure) status, return 0 on success or 1 on failure. Returns -1 if the Job has not completed.

Returns:Job exit code or (0 / 1) status Return type:int

is_finished

Return whether or not the Job is finished.

Returns:whether or not the Job is finished Return type:bool

is_started

Return whether or not the Job has been started.

Returns:whether or not the Job has been started Return type:bool

name

Return the Job Name.

Returns:Job name Return type:str

output

Return the output of the Job as a string, or None if the job has not completed.

Returns:Job output Return type:str

poll()

For asynchronous jobs (is_started is True but is_finished is False), check if the job has finished yet. If not, return False. If the job has finished, update self._finish_time, self._exit_code, self._output and self._finished and then return True.

This method should never raise exceptions; recoverable exceptions should be handled via internal retry logic on subsequent poll attempts. Retries should be done on the next call of this method; we never want to sleep during this method. Unrecoverable exceptions should set self._exit_code, self._output and self._finished.

Returns:is_finished Return type:bool

report_description()

Return a one-line description of the Job for use in reports.

Return type:str

run()

Run the job.

This method sets self._started and self._start_time. If the Job runs synchronously, this method also sets self._finished, self._exit_code, self._finish_time and self._output.

In the case of an exception, this method must still set those attributes as appropriate and then raise the exception.

Returns:True if job finished successfully, False if job finished but failed, or None if the job is still running in the background.

schedule_name

Return the configured schedule name for this job.

Returns:schedule name Return type:str

skip

Either None if the job should not be skipped, or a string reason describing why the Job should be skipped.

Return type:None or str

summary()

Retrieve a simple one-line summary of the Job output/status.

Returns:Job one-line summary. Return type:str