def harvest_job_create(context, data_dict):
log.info('Harvest job create: %r', data_dict)
check_access('harvest_job_create', context, data_dict)
source_id = data_dict['source_id']
# Check if source exists
source = HarvestSource.get(source_id)
if not source:
log.warn('Harvest source %s does not exist', source_id)
raise NotFound('Harvest source %s does not exist' % source_id)
# Check if the source is active
if not source.active:
log.warn(
'Harvest job cannot be created for inactive source %s',
source_id)
raise Exception('Can not create jobs on inactive sources')
# Check if there already is an unrun or currently running job for this
# source
exists = _check_for_existing_jobs(context, source_id)
if exists:
log.warn(
'There is already an unrun job %r for this source %s',
exists,
source_id)
raise HarvestJobExists('There already is an unrun job for this source')
job = HarvestJob()
job.source = source
job.save()
log.info('Harvest job saved %s', job.id)
return harvest_job_dictize(job, context)
def harvest_source_show_status(context, data_dict):
'''
Returns a status report for a harvest source
Given a particular source, returns a dictionary containing information
about the source jobs, datasets created, errors, etc.
Note that this information is already included on the output of
harvest_source_show, under the 'status' field.
:param id: the id or name of the harvest source
:type id: string
:rtype: dictionary
'''
p.toolkit.check_access('harvest_source_show_status', context, data_dict)
model = context.get('model')
source = harvest_model.HarvestSource.get(data_dict['id'])
if not source:
raise p.toolkit.ObjectNotFound(
'Harvest source {0} does not exist'.format(
data_dict['id']))
out = {
'job_count': 0,
'last_job': None,
'total_datasets': 0,
}
jobs = harvest_model.HarvestJob.filter(source=source).all()
job_count = len(jobs)
if job_count == 0:
return out
out['job_count'] = job_count
# Get the most recent job
last_job = harvest_model.HarvestJob.filter(source=source) \
.order_by(harvest_model.HarvestJob.created.desc()).first()
if not last_job:
return out
out['last_job'] = harvest_job_dictize(last_job, context)
# Overall statistics
packages = model.Session.query(model.Package) \
.join(harvest_model.HarvestObject) \
.filter(harvest_model.HarvestObject.harvest_source_id == source.id) \
.filter(harvest_model.HarvestObject.current) \
.filter(model.Package.state == u'active') \
.filter(model.Package.private == False)
out['total_datasets'] = packages.count()
return out
def harvest_job_show(context, data_dict):
check_access('harvest_job_show', context, data_dict)
id = data_dict.get('id')
attr = data_dict.get('attr', None)
job = HarvestJob.get(id, attr=attr)
if not job:
raise NotFound
return harvest_job_dictize(job, context)
def harvest_job_list(context, data_dict):
check_access('harvest_job_list', context, data_dict)
model = context['model']
session = context['session']
source_id = data_dict.get('source_id', False)
status = data_dict.get('status', False)
query = session.query(HarvestJob)
if source_id:
query = query.filter(HarvestJob.source_id == source_id)
if status:
query = query.filter(HarvestJob.status == status)
query = query.order_by(HarvestJob.created.desc())
jobs = query.all()
context['return_error_summary'] = False
return [harvest_job_dictize(job, context) for job in jobs]
