Jobs

class saagieapi.jobs.Jobs(saagie_api)

Bases: object

count_deletable_instances_by_date(job_id: str, date_before: str)

Count deletable job instances by date

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• date_before (str) – Instances before this date will be counted. The date must be in this format : ‘%Y-%m-%dT%H:%M:%S%z’

Returns:

Dict of number of job instances before the given date

Return type:

dict

Examples

>>> saagie_api.jobs.count_deletable_instances_by_date(job_id=job_id, date_before="2023-06-01T00:00:00+01:00")
{
    'countJobInstancesByDate': 3
}

count_deletable_instances_by_status(job_id)

Count deletable job instances by status

Parameters:

job_id (str) – UUID of your job (see README on how to find it)

Returns:

Dict of number of job instances by status

Return type:

dict

Examples

>>> saagie_api.jobs.count_deletable_instances_by_status(job_id=job_id)
{
    'countJobInstancesBySelector': [
        {'selector': 'ALL', 'count': 0},
        {'selector': 'SUCCEEDED', 'count': 0},
        {'selector': 'FAILED', 'count': 0},
        {'selector': 'STOPPED', 'count': 0},
        {'selector': 'UNKNOWN', 'count': 0}
    ]
}

count_instances_by_status(job_id)

Count deletable job instances by status

Parameters:

job_id (str) – UUID of your job (see README on how to find it)

Returns:

Dict of number of job instances by status

Return type:

dict

Examples

>>> saagie_api.jobs.count_instances_by_status(job_id=job_id)
{
    'countJobInstancesBySelector': [
        {'selector': 'ALL', 'count': 0},
        {'selector': 'SUCCEEDED', 'count': 0},
        {'selector': 'FAILED', 'count': 0},
        {'selector': 'STOPPED', 'count': 0},
        {'selector': 'UNKNOWN', 'count': 0}
    ]
}

Deprecated since version 2.9.0: This function is deprecated and will be removed in a future version. Please use count_deletable_instances_by_status() instead.

create(job_name: str, project_id: str, file: str = None, description: str = '', category: str = 'Processing', technology: str = 'python', technology_catalog: str = 'Saagie', runtime_version: str = '3.10', command_line: str = 'python {file} arg1 arg2', release_note: str = '', extra_technology: str = None, extra_technology_version: str = None, cron_scheduling: str = None, schedule_timezone: str = 'UTC', resources: Dict = None, emails: List = None, status_list: List = None, source_url: str = '') → Dict

Create job in given project

Note

• Only work for gql>=3.0.0

Parameters:

• job_name (str) – Name of the job. Must not already exist in the project

• project_id (str) – UUID of your project (see README on how to find it)

• file (str, optional) – Local path of the file to upload

• description (str, optional) – Description of the job

• category (str, optional) – Category to create the job into. Must be ‘Extraction’, ‘Processing’ or ‘Smart App’

• technology (str, optional) – Technology label of the job to create.

• technology_catalog (str, optional) – Technology catalog containing the technology to use for this job

• runtime_version (str, optional) – Technology version of the job, the ID of the context

• command_line (str, optional) – Command line of the job

• release_note (str, optional) – Release note of the job

• extra_technology (str, optional) – Extra technology when needed (spark jobs). If not needed, leave to empty string or the request will not work

• extra_technology_version (str, optional) – Version of the extra technology. Leave to empty string when not needed

• cron_scheduling (str, optional) – Scheduling CRON format Example: “0 0 * * *” (for every day At 00:00)

• schedule_timezone (str, optional) – Timezone of the scheduling Example: “UTC”, “Pacific/Pago_Pago”

• resources (dict, optional) – CPU, memory limit and requests Example: {“cpu”:{“request”:0.5, “limit”:2.6},”memory”:{“request”:1.0}}

• emails (List[String], optional) – Emails to receive alerts for the job, each item should be a valid email

• status_list (List[String], optional) – Receive an email when the job status change to a specific status Each item of the list should be one of these following values: “REQUESTED”, “QUEUED”, “RUNNING”, “FAILED”, “KILLED”, “KILLING”, “SUCCEEDED”, “UNKNOWN”, “AWAITING”, “SKIPPED”

• source_url (str, optional) – URL of the source code used for the job (link to the commit for example)

Returns:

Dict of job information

Return type:

dict

Examples

>>> saagieapi.jobs.create(
...     job_name="my job",
...     project_id="860b8dc8-e634-4c98-b2e7-f9ec32ab4771",
...     file="/tmp/test.py",
...     description='My description',
...     category='Extraction',
...     technology='python',# technology id corresponding to your context.id in your technology catalog definition
...     technology_catalog='Saagie',
...     runtime_version='3.9',
...     command_line='python {file}',
...     release_note='First release',
...     extra_technology='',
...     extra_technology_version='',
...     cron_scheduling='0 0 * * *',
...     schedule_timezone='Europe/Paris',
...     resources={"cpu": {"request": 0.5, "limit": 2.6}, "memory": {"request": 1.0}},
...     emails=['email1@saagie.io', 'email2@saagie.io'],
...     status_list=["FAILED", "KILLED"],
...     source_url="",
... )
{
    "data":{
        "createJob":{
            "id":"60f46dce-c869-40c3-a2e5-1d7765a806db",
            "versions":[
                {
                    "number":1,
                    "__typename":"JobVersion"
                }
            ],
            "__typename":"Job"
        }
    }
}

create_or_upgrade(job_name: str, project_id: str, file: str = None, use_previous_artifact: bool = True, description: str = None, category: str = None, technology: str = None, technology_catalog: str = None, runtime_version: str = None, command_line: str = None, release_note: str = None, extra_technology: str = None, extra_technology_version: str = None, is_scheduled: bool = None, cron_scheduling: str = None, schedule_timezone: str = 'UTC', resources: Dict = None, emails: List = None, status_list: List = None, source_url: str = '') → Dict

Create or upgrade a job

Parameters:

• job_name (str) – Name of your job

• project_id (str) – UUID of your project

• file (str (optional)) – Path to your file

• use_previous_artifact (bool (optional)) – Use previous artifact

• description (str (optional)) – Description of your job

• category (str (optional)) – Category of your job

• technology (str (optional)) – Technology of your job

• technology_catalog (str (optional)) – Technology catalog of your job

• runtime_version (str (optional)) – Runtime version

• command_line (str (optional)) – Command line

• release_note (str (optional)) – Release note

• extra_technology (str (optional)) – Extra technology when needed (spark jobs). If not needed, leave to None or the request will not work

• extra_technology_version (str (optional)) – Version of the extra technology. Leave to None when not needed

• is_scheduled (bool (optional)) – True if the job is scheduled, False to deactivate scheduling

• cron_scheduling (str (optional)) – Cron scheduling

• schedule_timezone (str (optional)) – Schedule timezone

• resources (dict (optional)) – Resources

• emails (list (optional)) – Emails

• status_list (list (optional)) – Status list

• source_url (str (optional)) – URL of the source code used for the job (link to the commit for example)

Returns:

Either the same dict as create_job, or the one returned by concatenation of upgrade_job and edit_job

Return type:

dict

Examples

>>> saagieapi.jobs.create_or_upgrade(
...     job_name="my job",
...     project_id="860b8dc8-e634-4c98-b2e7-f9ec32ab4771",
...     file="/tmp/test.py",
...     use_previous_artifact=False,
...     description='My description',
...     category='Extraction',
...     technology='python',# technology id corresponding to your context.id in your technology catalog definition
...     technology_catalog='Saagie',
...     runtime_version='3.9',
...     command_line='python {file}',
...     release_note='First release',
...     extra_technology='',
...     extra_technology_version='',
...     cron_scheduling='0 0 * * *',
...     schedule_timezone='Europe/Paris',
...     resources={"cpu": {"request": 0.5, "limit": 2.6}, "memory": {"request": 1.0}},
...     emails=['email1@saagie.io', 'email2@saagie.io'],
...     status_list=["FAILED", "KILLED"],
...     source_url="",
... )
{
    "data":{
        "createJob":{
            "id":"60f46dce-c869-40c3-a2e5-1d7765a806db",
            "versions":[
                {
                    "number":1,
                    "__typename":"JobVersion"
                }
            ],
            "__typename":"Job"
        }
    }
}

delete(job_id: str) → Dict

Delete a given job

Parameters:

job_id (str) – UUID of your job (see README on how to find it)

Returns:

Dict of deleted job

Return type:

dict

Examples

>>> saagieapi.jobs.delete(job_id="f5fce22d-2152-4a01-8c6a-4c2eb4808b6d")
{
    "deleteJob": True
}

delete_instances(job_id, job_instances_id)

Delete given job’s instances NB: You can only delete an instance not associated to a pipeline instance Also you can only delete instances if they aren’t processing by the orchestrator

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• job_instances_id ([str]) – List of UUID of instances to delete (see README on how to find it)

Returns:

Dict of deleted instances

Return type:

dict

Examples

>>> saagie_api.jobs.delete_instances(
...     job_id=job_id,
...     job_instances_id=["c8f156bc-78ab-4dda-acff-bbe828237fd9", "7e5549cd-32aa-42c4-88b5-ddf5f3087502"]
... )
{
    'deleteJobInstances': [
        {'id': '7e5549cd-32aa-42c4-88b5-ddf5f3087502', 'success': True},
        {'id': 'c8f156bc-78ab-4dda-acff-bbe828237fd9', 'success': True}
    ]
}

delete_instances_by_date(job_id: str, date_before: str, exclude_instances_id: List = None, include_instances_id: List = None)

Delete given job’s instances by date NB: You can only delete an instance not associated to a pipeline instance. Also you can only delete instances if they aren’t processing by the orchestrator

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• date_before (str) – Instances before this date will be deleted. The date must be in this format : ‘%Y-%m-%dT%H:%M:%S%z’

• exclude_instances_id ([str]) – List of UUID of instances of your job to exclude from the deletion

• include_instances_id ([str]) – List of UUID of instances of your job to include from the deletion

Returns:

Return the number of instances deleted

Return type:

Dict

Examples

>>> saagie_api.jobs.delete_instances_by_date(
...     job_id=job_id,
...     date_before="2023-06-01T00:00:00+01:00",
...     exclude_instances_id=["478d48d4-1609-4bf0-883d-097d43709aa8"],
...     include_instances_id=["47d3df2c-5a38-4a5e-a49e-5405ad8f1699"]
... )
{
    'deleteJobInstancesByDate': 1
}

delete_instances_by_selector(job_id, selector, exclude_instances_id: List = None, include_instances_id: List = None)

Delete given job’s instances by selector NB: You can only delete an instance not associated to a pipeline instance. Also you can only delete instances if they aren’t processing by the orchestrator

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• selector (str) – Name of status to select in this list : ALL, SUCCEEDED, FAILED, STOPPED, UNKNOWN

• exclude_instances_id ([str]) – List of UUID of instances of your job to exclude from the deletion

• include_instances_id ([str]) – List of UUID of instances of your job to include from the deletion

Returns:

Return the number of instances deleted

Return type:

Dict

Examples

>>> saagie_api.jobs.delete_instances_by_selector(
...     job_id=job_id,
...     selector="FAILED",
...     exclude_instances_id=["478d48d4-1609-4bf0-883d-097d43709aa8"],
...     include_instances_id=["47d3df2c-5a38-4a5e-a49e-5405ad8f1699"]
... )
{
    'deleteJobInstancesBySelector': 1
}

delete_versions(job_id, versions)

Delete given job’s versions NB: You can only delete a version not associated to a pipeline instance

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• versions ([str]) – List of version numbers to delete

Returns:

Dict of deleted versions with their number and success status

Return type:

dict

Examples

>>> saagie_api.jobs.delete_versions(
...     job_id=job_id,
...     versions=["1"]
... )
{
    'deleteJobVersions': [
        {'number': 1, 'success': True}
    ]
}

duplicate(job_id)

Duplicate a given job

Parameters:

job_id (str) – UUID of your job (see README on how to find it)

Returns:

Dict of duplicate job with its id and name

Return type:

dict

Examples

>>> saagie_api.jobs.duplicate(job_id=job_id)
{
    'duplicateJob': {
        'id': '29cf1b80-6b9c-47bc-a06c-c20897257097',
        'name': 'Copy of my_job 2'
    }
}

edit(job_id: str, job_name: str = None, description: str = None, is_scheduled: bool = None, cron_scheduling: str = None, schedule_timezone: str = 'UTC', resources: Dict = None, emails: List = None, status_list: List = None) → Dict

Edit a job

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• job_name (str, optional) – Job name If not filled, defaults to current value, else it will change the job’s name

• description (str, optional) – Description of job if not filled, defaults to current value, else it will change the description of the pipeline

• is_scheduled (bool, optional) – True if the job is scheduled, else False if not filled, defaults to current value

• cron_scheduling (str, optional) – Scheduling CRON format When is_scheduled is set to True, it will be mandatory to fill this value if not filled, defaults to current value Example: “0 0 * * *” (for every day At 00:00)

• schedule_timezone (str, optional) – Timezone of the scheduling Example: “UTC”, “Pacific/Pago_Pago”

• resources (dict, optional) – CPU, memory limit and requests if not filled, defaults to current value Example: {“cpu”:{“request”:0.5, “limit”:2.6},”memory”:{“request”:1.0}}

• emails (List[String], optional) – Emails to receive alerts for the job, each item should be a valid email, If you want to remove alerting, please set emails to [] or list() if not filled, defaults to current value

• status_list (List[String], optional) – Receive an email when the job status change to a specific status Each item of the list should be one of these following values: “REQUESTED”, “QUEUED”, “RUNNING”, “FAILED”, “KILLED”, “KILLING”, “SUCCEEDED”, “UNKNOWN”, “AWAITING”, “SKIPPED”

Returns:

Dict of job information

Return type:

dict

Examples

>>> saagieapi.jobs.edit(
...     job_id="60f46dce-c869-40c3-a2e5-1d7765a806db",
...     job_name="newname",
...     description="new desc",
...     is_scheduled=True,
...     cron_scheduling='0 * * * *',
...     schedule_timezone='Europe/Paris',
...     resources={"cpu": {"request": 1.5, "limit": 2.2}, "memory": {"request": 2.0}},
...     emails=['email1@saagie.io'],
...     status_list=["FAILED", "QUEUED"]
... )
{
    "editJob": {
        "id": "60f46dce-c869-40c3-a2e5-1d7765a806db",
        "name": "newname",
        "alias": "newname",
        "description": "new desc",
        "isScheduled": True,
        "cronScheduling": "0 * * * *",
        "scheduleTimezone": "Europe/Paris",
        "resources": {
            "cpu": {
                "request": 1.5,
                "limit": 2.2
            },
            "memory": {
                "request": 2.0,
                "limit": None
            }
        },
        "alerting": {
            "emails": [
                "email1@saagie.io"
            ],
            "statusList": [
                "FAILED",
                "QUEUED"
            ]
        }
    }
}

export(job_id: str, output_folder: str, error_folder: str | None = '', versions_limit: int | None = None, versions_only_current: bool = False) → bool

Export the job in a folder

Parameters:

• job_id (str) – Job ID

• output_folder (str) – Path to store the exported job

• error_folder (str, optional) – Path to store the job ID in case of error. If not set, job ID is not write

• versions_limit (int, optional) – Maximum limit of versions to fetch per job. Fetch from most recent to the oldest

• versions_only_current (bool, optional) – Whether to only fetch the current version of each job

Returns:

True if job is exported False otherwise

Return type:

bool

Examples

>>> saagieapi.jobs.export(
...    job_id="f5fce22d-2152-4a01-8c6a-4c2eb4808b6d",
...    output_folder="./output/job/",
...    error_folder="./output/error/",
...    versions_only_current=True
... )
True

generate_description_by_ai(job_id: str)

Generate a description for a job using AI. Be careful, by calling this function the code contained in the job package will be sent to OpenAI and thus will not be secured anymore by Saagie DataOps Platform. Otherwise, the function returns an error if the description is already the one generated by AI.

Parameters:

job_id (str) – UUID of your job (see README on how to find it)

Returns:

Dict of the generated description

Return type:

dict

Examples

>>> saagie_api.jobs.generate_description_by_ai(job_id=job_id)
{
    'editJobWithAiGeneratedDescription': {
        'id': 'bfa25e4a-1796-4ebb-8c3d-138f74146973',
        'description': 'The purpose of this code is to display the message "Hello World" on the screen.',
        'aiDescriptionVersionNumber': 1
    }
}

get_id(job_name: str, project_name: str) → str

Get the job id with the job name and project name

Parameters:

• job_name (str) – Name of your job

• project_name (str) – Name of your project

Returns:

Job UUID

Return type:

str

Examples

>>> saagieapi.jobs.get_id(
...     project_name="Test project",
...     job_name="Python test job"
... )
"f5fce22d-2152-4a01-8c6a-4c2eb4808b6d"

get_info(job_id: str, instances_limit: int | None = None, versions_limit: int | None = None, versions_only_current: bool = False, pprint_result: bool | None = None) → Dict

Get job’s info

Parameters:

• job_id (str) – UUID of your job

• instances_limit (int, optional) – Maximum limit of instances to fetch per job. Fetch from most recent to oldest

• versions_limit (int, optional) – Maximum limit of versions to fetch per job. Fetch from most recent to oldest

• versions_only_current (bool, optional) – Whether to only fetch the current version of each job

• pprint_result (bool, optional) – Whether to pretty print the result of the query, default to saagie_api.pprint_global

Returns:

Dict of job’s info

Return type:

dict

Examples

>>> saagieapi.jobs.get_info(
...     job_id="f5fce22d-2152-4a01-8c6a-4c2eb4808b6d",
...     instances_limit=2
... )
{
    "job": {
        "id": "f5fce22d-2152-4a01-8c6a-4c2eb4808b6d",
        "name": "Python test job",
        "description": "Amazing python job",
        "alerting": None,
        "countJobInstance": 5,
        "instances": [
            {
                "id": "61f6175a-fd38-4fac-9fa9-a7b63554f14e",
                "status": "SUCCEEDED",
                "history": {
                    "currentStatus": {
                        "status": "SUCCEEDED",
                        "details": None,
                        "reason": None
                    }
                },
                "startTime": "2022-04-19T13:46:40.045Z",
                "endTime": "2022-04-19T13:46:47.708Z",
                "version": {
                    "number": 1,
                    "releaseNote": "",
                    "runtimeVersion": "3.7",
                    "commandLine": "python {file} arg1 arg2",
                    "isMajor": False,
                    "doesUseGPU": False
                }
            },
            {
                "id": "befe73b2-81ab-418f-bc2f-9d012102a895",
                "status": "SUCCEEDED",
                "history": {
                    "currentStatus": {
                        "status": "SUCCEEDED",
                        "details": None,
                        "reason": None
                    }
                },
                "startTime": "2022-04-19T13:45:49.783Z",
                "endTime": "2022-04-19T13:45:57.388Z",
                "version":{
                    "number": 1,
                    "releaseNote": "",
                    "runtimeVersion": "3.7",
                    "commandLine": "python {file} arg1 arg2",
                    "isMajor": False,
                    "doesUseGPU": False
                }
            }
        ],
        "versions": [
            {
                "number": 1,
                "creationDate": "2022-04-26T08:16:20.681Z",
                "releaseNote": "",
                "runtimeVersion": "3.7",
                "commandLine": "python {file} arg1 arg2",
                "packageInfo": {
                    "name": "test.py",
                    "downloadUrl": "/projects/api/platform/6/project/860b8dc8-e634-4c98-b2e7-f9ec32ab4771/job/f5fce22d-2152-4a01-8c6a-4c2eb4808b6d/version/1/artifact/test.py"
                },
                "dockerInfo": None,
                "extraTechnology": None,
                "isCurrent": True,
                "isMajor": False
            }
        ],
        "category": "Extraction",
        "technology": {
            "id": "0db6d0a7-ad4b-45cd-8082-913a192daa25"
        },
        "isScheduled": False,
        "cronScheduling": None,
        "scheduleStatus": None,
        "scheduleTimezone": "UTC",
        "isStreaming": False,
        "creationDate": "2022-04-26T08:16:20.681Z",
        "migrationStatus": None,
        "migrationProjectId": None,
        "isDeletable": True,
        "graphPipelines": [],
        "doesUseGPU": False,
        "resources": None
    }
}

get_info_by_alias(project_id: str, job_alias: str, instances_limit: int | None = None, versions_limit: int | None = None, versions_only_current: bool = False, pprint_result: bool | None = None) → Dict

Get job’s info

Parameters:

• project_id (str) – UUID of the project of your job

• job_alias (str) – Alias of your job

• instances_limit (int, optional) – Maximum limit of instances to fetch per job. Fetch from most recent to oldest

• versions_limit (int, optional) – Maximum limit of versions to fetch per job. Fetch from most recent to oldest

• versions_only_current (bool, optional) – Whether to only fetch the current version of each job

• pprint_result (bool, optional) – Whether to pretty print the result of the query, default to saagie_api.pprint_global

Returns:

Dict of job’s info

Return type:

dict

Examples

>>> saagieapi.jobs.get_info_by_alias(
...     project_id="7c199d29-676a-483f-b28b-112ec71fcf81",
...     job_alias="Python_test_job"
...     instances_limit=2
... )
{
    "jobByAlias": {
        "id": "f5fce22d-2152-4a01-8c6a-4c2eb4808b6d",
        "name": "Python test job",
        "description": "Amazing python job",
        "alerting": None,
        "countJobInstance": 5,
        "instances": [
            {
                "id": "61f6175a-fd38-4fac-9fa9-a7b63554f14e",
                "status": "SUCCEEDED",
                "history": {
                    "currentStatus": {
                        "status": "SUCCEEDED",
                        "details": None,
                        "reason": None
                    }
                },
                "startTime": "2022-04-19T13:46:40.045Z",
                "endTime": "2022-04-19T13:46:47.708Z",
                "version": {
                    "number": 1,
                    "releaseNote": "",
                    "runtimeVersion": "3.7",
                    "commandLine": "python {file} arg1 arg2",
                    "isMajor": False,
                    "doesUseGPU": False
                }
            },
            {
                "id": "befe73b2-81ab-418f-bc2f-9d012102a895",
                "status": "SUCCEEDED",
                "history": {
                    "currentStatus": {
                        "status": "SUCCEEDED",
                        "details": None,
                        "reason": None
                    }
                },
                "startTime": "2022-04-19T13:45:49.783Z",
                "endTime": "2022-04-19T13:45:57.388Z",
                "version":{
                    "number": 1,
                    "releaseNote": "",
                    "runtimeVersion": "3.7",
                    "commandLine": "python {file} arg1 arg2",
                    "isMajor": False,
                    "doesUseGPU": False
                }
            }
        ],
        "versions": [
            {
                "number": 1,
                "creationDate": "2022-04-26T08:16:20.681Z",
                "releaseNote": "",
                "runtimeVersion": "3.7",
                "commandLine": "python {file} arg1 arg2",
                "packageInfo": {
                    "name": "test.py",
                    "downloadUrl": "/projects/api/platform/6/project/860b8dc8-e634-4c98-b2e7-f9ec32ab4771/job/f5fce22d-2152-4a01-8c6a-4c2eb4808b6d/version/1/artifact/test.py"
                },
                "dockerInfo": None,
                "extraTechnology": None,
                "isCurrent": True,
                "isMajor": False
            }
        ],
        "category": "Extraction",
        "technology": {
            "id": "0db6d0a7-ad4b-45cd-8082-913a192daa25"
        },
        "isScheduled": False,
        "cronScheduling": None,
        "scheduleStatus": None,
        "scheduleTimezone": "UTC",
        "isStreaming": False,
        "creationDate": "2022-04-26T08:16:20.681Z",
        "migrationStatus": None,
        "migrationProjectId": None,
        "isDeletable": True,
        "graphPipelines": [],
        "doesUseGPU": False,
        "resources": None
    }
}

get_instance(job_instance_id: str, pprint_result: bool | None = None) → Dict

Get the given job instance

Parameters:

• job_instance_id (str) – UUID of your job instance (see README on how to find it)

• pprint_result (bool, optional) – Whether to pretty print the result of the query, default to saagie_api.pprint_global

Returns:

Dict of instance information

Return type:

dict

Examples

>>> saagieapi.jobs.get_instance(job_instance_id="befe73b2-81ab-418f-bc2f-9d012102a895")
{
    "jobInstance": {
        "id": "befe73b2-81ab-418f-bc2f-9d012102a895",
        "number": 1,
        "status": "SUCCEEDED",
        "history": {
            "currentStatus": {
                "status": "SUCCEEDED",
                "details": None,
                "reason": None
            }
        },
        "startTime": "2022-04-19T13:45:49.783Z",
        "endTime": "2022-04-19T13:45:57.388Z",
        "jobId": "f5fce22d-2152-4a01-8c6a-4c2eb4808b6d",
        "version":{
            "number": 1,
            "releaseNote": "",
            "runtimeVersion": "3.7",
            "commandLine": "python {file} arg1 arg2",
            "isMajor": False,
            "isCurrent": True
        }
        "executionGlobalVariablesInput": [
            {
                "key": "TEST_PASSWORD",
                "value": None,
                "isPassword": True
            },
            {
                "key": "TEST_PROJECT",
                "value": "TEST_PROJECT",
                "isPassword": False
            }
        ],
        "executionVariablesInput": [
            {
                "parentJobInstanceId": None,
                "parentJobId": None,
                "parentJobAlias": None,
                "isDirectParent": None,
                "executionVariables": [
                    {
                        "key": "TEST_PASSWORD",
                        "value": None,
                        "isPassword": True
                    },
                    {
                        "key": "TEST_PROJECT",
                        "value": "TEST_PROJECT",
                        "isPassword": False
                    }
                ],
                "isGlobalVariables": True
            }
        ],
        "executionVariablesOutput": None,
        "executionVariablesByKey": []
    }
}

import_from_json(project_id: str, path_to_folder: str) → bool

Import a job from JSON format

Parameters:

• project_id (str) – Project ID to import the job

• path_to_folder (str) – Path to the folder of the job to import

Returns:

True if job is imported False otherwise

Return type:

bool

Examples

>>> saagieapi.jobs.import_from_json(
...     project_id="860b8dc8-e634-4c98-b2e7-f9ec32ab4771",
...     path_to_package="/path/to/the/package/of/the/job"
... )
True

list_for_project(project_id: str, instances_limit: int | None = None, versions_limit: int | None = None, versions_only_current: bool = False, pprint_result: bool | None = None) → Dict

List jobs in the given project with their instances. NB: You can only list jobs if you have at least the viewer role on the project

Parameters:

• project_id (str) – UUID of your project (see README on how to find it)

• instances_limit (int, optional) – Maximum limit of instances to fetch per job. Fetch from most recent to oldest

• versions_limit (int, optional) – Maximum limit of versions to fetch per job. Fetch from most recent to oldest

• versions_only_current (bool, optional) – Whether to only fetch the current version of each job

• pprint_result (bool, optional) – Whether to pretty print the result of the query, default to saagie_api.pprint_global

Returns:

Dict of jobs information

Return type:

dict

Examples

>>> saagieapi.jobs.list_for_project(
...     project_id="860b8dc8-e634-4c98-b2e7-f9ec32ab4771",
...     instances_limit=2
... )
{
    "jobs": [
        {
            "id": "fc7b6f52-5c3e-45bb-9a5f-a34bcea0fc10",
            "name": "Python test job 1",
            "description": "Amazing python job",
            "alerting": None,
            "countJobInstance": 0,
            "instances": [],
            "versions": [
                {
                    "number": 1,
                    "creationDate": "2022-04-26T12:08:15.286Z",
                    "releaseNote": "",
                    "runtimeVersion": "3.7",
                    "commandLine": "python {file} arg1 arg2",
                    "packageInfo": {
                        "name": "_tmp_test.py",
                        "downloadUrl": "/projects/api/platform/6/project/860b8dc8-e634-4c98-b2e7-f9ec32ab4771/job/fc7b6f52-5c3e-45bb-9a5f-a34bcea0fc10/version/1/artifact/_tmp_test.py"
                    },
                    "dockerInfo": None,
                    "extraTechnology": None,
                    "isCurrent": True,
                    "isMajor": False
                }
            ],
            "category": "Extraction",
            "technology": {
                "id": "0db6d0a7-ad4b-45cd-8082-913a192daa25"
            },
            "isScheduled": False,
            "cronScheduling": None,
            "scheduleTimezone": "UTC",
            "scheduleStatus": None,
            "isStreaming": False,
            "creationDate": "2022-04-26T12:08:15.286Z",
            "migrationStatus": None,
            "migrationProjectId": None,
            "isDeletable": True,
            "pipelines": [],
            "graphPipelines": [],
            "doesUseGPU": False,
            "resources": None
        }
    ]
}

list_for_project_minimal(project_id: str) → Dict

List only job names and ids in the given project . NB: You can only list jobs if you have at least the viewer role on the project

Parameters:

project_id (str) – UUID of your project (see README on how to find it)

Returns:

Dict of jobs ids and names

Return type:

dict

Examples

>>> saagieapi.jobs.list_for_project_minimal(project_id="860b8dc8-e634-4c98-b2e7-f9ec32ab4771")
{
    "jobs": [
        {
            "id": "fc7b6f52-5c3e-45bb-9a5f-a34bcea0fc10",
            "name": "Python test job 1",
            "alias": "Python_test_job_1"
        },
        {
            "id": "e92ed170-50d6-4041-bba9-098a8e16f444",
            "name": "Python test job 2",
            "alias": "Python_test_job_2"
        }
    ]
}

move_job(job_id: str, target_platform_id: int, target_project_id: str)

Move a job to another project in the same platform or another one

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• target_platform_id (int) – Id of the platform to move the job to

• target_project_id (str) – UUID of the project to move the job to

Returns:

Dict of the moved job with its new id

Return type:

dict

Examples

>>> saagie_api.jobs.move_job(job_id=job_id, target_platform_id=1, target_project_id=project_id)
{
    'moveJob': '29cf1b80-6b9c-47bc-a06c-c20897257097',
}

rollback(job_id: str, version_number: str)

Rollback a given job to the given version

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• version_number (str) – Number of the version to rollback

Returns:

Dict of rollback job

Return type:

dict

Examples

>>> saagie_api.jobs.rollback(
...     job_id="58870149-5f1c-45e9-93dc-04b2b30a732c",
...     version_number=3
... )
{
    "rollbackJobVersion": {
        "id": "58870149-5f1c-45e9-93dc-04b2b30a732c",
        "versions": [
            {
                "number": 4,
                "isCurrent": False
            },
            {
                "number": 3,
                "isCurrent": True
            },
            {
                "number": 2,
                "isCurrent": False
            },
            {
                "number": 1,
                "isCurrent": False
            }
        ]
    }
}

run(job_id: str) → Dict

Run a given job

Parameters:

job_id (str) – UUID of your job (see README on how to find it)

Returns:

Dict of the given job information

Return type:

dict

Examples

>>> saagieapi.jobs.run(job_id="f5fce22d-2152-4a01-8c6a-4c2eb4808b6d")
{
    "runJob":{
        "id":"5b9fc971-1c4e-4e45-a978-5851caef0162",
        "status":"REQUESTED"
    }
}

run_with_callback(job_id: str, freq: int = 10, timeout: int = -1) → Dict

Run a job and wait for the final status (KILLED, FAILED, UNKNOWN or SUCCESS). Regularly check (default to 10s) the job’s status.

Parameters:

• job_id (str) – UUID of your job (see README on how to find it)

• freq (int, optional) – Seconds to wait between two state checks

• timeout (int, optional) – Seconds before timeout for a status check call

Returns:

(Final state of the job, job instance id)

Return type:

(str, str)

Raises:

TimeoutError – When the status check is not responding

Examples

>>> saagieapi.jobs.run_with_callback(
...        job_id="f5fce22d-2152-4a01-8c6a-4c2eb4808b6d",
...        freq=5,
...        timeout=60
... )
("SUCCEEDED", "5b9fc971-1c4e-4e45-a978-5851caef0162")

stop(job_instance_id: str) → Dict

Stop a given job instance

Parameters:

job_instance_id (str) – UUID of your job instance (see README on how to find it)

Returns:

Job instance information

Return type:

dict

Examples

>>> saagieapi.jobs.stop(job_instance_id="8e9b9f16-4a5d-4188-a967-1a96b88e4358")
{
    "stopJobInstance":{
        "id":"8e9b9f16-4a5d-4188-a967-1a96b88e4358",
        "number":17,
        "status":"KILLING",
        "history": {
            "currentStatus": {
                "status": "SUCCEEDED",
                "details": None,
                "reason": None
            }
        },
        "startTime":"2022-04-29T08:38:49.344Z",
        "endTime":None,
        "jobId":"e92ed472-50d6-4041-bba9-098a8e16f444"
    }
}

upgrade(job_id: str, file: str = None, use_previous_artifact: bool = True, runtime_version: str = None, command_line: str = None, release_note: str = '', extra_technology: str = None, extra_technology_version: str = None, source_url: str = '') → Dict

Upgrade a job

Parameters:

• job_id (str) – UUID of your job

• file (str (optional)) – Path to your file

• use_previous_artifact (bool (optional)) – Use previous artifact

• runtime_version (str (optional)) – Runtime version, the ID of the context Example: “3.10”

• command_line (str (optional)) – Command line used to run the job Example: “python3 {file} arg1 arg2”

• release_note (str (optional)) – Release note

• extra_technology (str (optional)) – Extra technology when needed (spark jobs). If not needed, leave to None or the request will not work

• extra_technology_version (str (optional)) – Version of the extra technology. Leave to None when not needed

• source_url (str (optional)) – URL of the source code used for the job (link to the commit for example)

Returns:

Dict with version number

Return type:

dict

Examples

>>> saagieapi.jobs.upgrade(
...     job_id="60f46dce-c869-40c3-a2e5-1d7765a806db",
...     use_previous_artifact=True,
...     runtime_version='3.8',
...     command_line='python {file} new_arg',
...     release_note="Second version"
... )
{
    "data":{
        "addJobVersion":{
            "number":2,
            "__typename":"JobVersion"
        }
    }
}

upgrade_by_name(job_name: str, project_name: str, file=None, use_previous_artifact: bool = True, runtime_version: str = None, command_line: str = None, release_note: str = None, extra_technology: str = None, extra_technology_version: str = None, source_url: str = '') → Dict

Upgrade a job

Parameters:

• job_name (str) – Name of your job

• project_name (str) – Name of your project

• file (str (optional)) – Path to your file

• use_previous_artifact (bool (optional)) – Use previous artifact

• runtime_version (str (optional)) – Runtime version

• command_line (str (optional)) – Command line

• release_note (str (optional)) – Release note

• extra_technology (str (optional)) – Extra technology when needed (spark jobs). If not needed, leave to None or the request will not work

• extra_technology_version (str (optional)) – Version of the extra technology. Leave to None when not needed

• source_url (str (optional)) – URL of the source code used for the job (link to the commit for example)

Returns:

Dict with version number

Return type:

dict

Examples

>>> saagieapi.jobs.upgrade_by_name(
...     job_name="my job",
...     project_name="My project",
...     use_previous_artifact=True,
...     runtime_version='3.8',
...     command_line='python {file} new_arg',
...     release_note="Second version"
... )
{
    "data":{
        "addJobVersion":{
            "number":3,
            "__typename":"JobVersion"
        }
    }
}