def main():
if len(sys.argv) == 1:
print(f"Usage: {sys.argv[0]} YYYY-mm-dd")
sys.exit(1)
date = sys.argv[1]
job_bucket = os.environ["PQ_FLATTENER_GLUE_JOB_BUCKET"]
iam_role = os.environ["PQ_FLATTENER_JOB_IAM_ROLE"]
source_path = os.environ["PQ_FLATTENER_SOURCE_PATH"]
dest_path = os.environ["PQ_FLATTENER_DEST_PATH"]
job = GlueJob(
"v1/glue_jobs/pq_flattener",
bucket=job_bucket,
job_role=iam_role,
job_arguments={
"--date": date,
"--s3_source": source_path,
"--s3_dest": dest_path,
},
)
job.job_name = f"pq_flattener"
# Run job on AWS Glue
print(f'Starting job "{job.job_name}"...')
try:
job.run_job()
job.wait_for_completion()
finally:
job.cleanup()
def main():
iam_role = os.environ["IAM_ROLE"]
github_tag = os.environ["GITHUB_TAG"]
snapshot_date = os.environ["SNAPSHOT_DATE"]
# Get job parameters for specific glue job
job_args = {"--github_tag": github_tag, "--snapshot_date": snapshot_date}
job = GlueJob(f"glue_jobs/example_job/",
bucket=job_bucket,
job_role=iam_role,
job_arguments=job_args)
print(f'Starting job "{job.job_name}"...')
job.run_job()
job.wait_for_completion(verbose=True)
if job.job_run_state == 'SUCCEEDED':
print('Job successful - cleaning up')
job.cleanup()
def main(job_role):
package_name = "gluejobutils"
to_path = f"test/glue_test/glue_py_resources/{package_name}.zip"
zf = zipfile.ZipFile(to_path, "w")
zf.write(os.path.join(package_name, "__init__.py"))
zf.write(os.path.join(package_name, "datatypes.py"))
zf.write(os.path.join(package_name, "s3.py"))
zf.write(os.path.join(package_name, "utils.py"))
zf.write(os.path.join(package_name, "record_datetimes.py"))
zf.write(os.path.join(package_name, "df_transforms.py"))
zf.write(os.path.join(package_name, "data/data_type_conversion.json"))
zf.close()
g = GlueJob("test/glue_test/",
bucket="alpha-gluejobutils",
job_role=job_role)
g.job_name = "gluejobutils_unit_test"
g.run_job()
g.wait_for_completion(True)
if g.job_run_state == "SUCCEEDED":
print("cleaning up job...")
g.cleanup()
def main(job_role):
package_name = 'gluejobutils'
to_path = f'test/glue_test/glue_py_resources/{package_name}.zip'
zf = zipfile.ZipFile(to_path, "w")
zf.write(os.path.join(package_name, '__init__.py'))
zf.write(os.path.join(package_name, 'datatypes.py'))
zf.write(os.path.join(package_name, 's3.py'))
zf.write(os.path.join(package_name, 'utils.py'))
zf.write(os.path.join(package_name, 'record_datetimes.py'))
zf.write(os.path.join(package_name, 'df_transforms.py'))
zf.write(os.path.join(package_name, 'data/data_type_conversion.json'))
zf.close()
g = GlueJob('test/glue_test/',
bucket='alpha-gluejobutils',
job_role=job_role)
g.job_name = 'gluejobutils_unit_test'
g.run_job()
g.wait_for_completion(True)
if g.job_run_state == 'SUCCEEDED':
print("cleaning up job...")
g.cleanup()
except:
raise Exception("You must provide a role name")
bucket = 'alpha-data-linking'
job = GlueJob(
'match/',
bucket=bucket,
job_role=ROLE,
job_arguments={
"--test_arg": 'some_string',
"--conf":
'spark.jars.packages=graphframes:graphframes:0.6.0-spark2.3-s_2.11',
'--enable-spark-ui': 'true',
'--spark-event-logs-path':
's3://alpha-data-linking/glue_test_delete/logsdelete',
'--enable-continuous-cloudwatch-log': 'true'
})
job.job_name = '1m_p_50_e_6'
print(job._job_definition())
job.allocated_capacity = 2
try:
job.run_job()
job.wait_for_completion()
finally:
pass
# job.cleanup()