def process_non_facility_warehouse_data(location,
start_date,
end_date,
runner=None,
strict=True):
facs = get_non_archived_facilities_below(location, end_date)
start_date = datetime(start_date.year, start_date.month, 1)
end_date = datetime(end_date.year, end_date.month, 1)
if runner:
runner.location = location
runner.save()
fac_ids = [f.location_id for f in facs]
logging.info("processing non-facility %s (%s), %s children" %
(location.name, str(location.location_id), len(facs)))
prods = SQLProduct.objects.filter(domain=location.domain,
is_archived=False)
sub_summaries = OrganizationSummary.objects.filter(
location_id__in=fac_ids,
date__range=(start_date, end_date),
average_lead_time_in_days__gt=0).values('date').annotate(
average_time=Avg('average_lead_time_in_days'))
sub_summaries = {(subsummary['date'].year, subsummary['date'].month):
subsummary
for subsummary in sub_summaries}
sub_prods = ProductAvailabilityData.objects.filter(
location_id__in=fac_ids,
date__range=(start_date, end_date)).values('product', 'date').annotate(
total_sum=Sum('total'),
with_stock_sum=Sum('with_stock'),
without_stock_sum=Sum('without_stock'),
)
sub_prods = {((sub_prod['date'].year, sub_prod['date'].month),
sub_prod['product']): sub_prod
for sub_prod in sub_prods}
sub_group_summaries = GroupSummary.objects.filter(
org_summary__location_id__in=fac_ids,
org_summary__date__range=(start_date, end_date)).values(
'title',
'org_summary__date').annotate(total_sum=Sum('total'),
responded_sum=Sum('responded'),
on_time_sum=Sum('on_time'),
complete_sum=Sum('complete'))
sub_group_summaries = {((sub_group_summary['org_summary__date'].year,
sub_group_summary['org_summary__date'].month),
sub_group_summary['title']): sub_group_summary
for sub_group_summary in sub_group_summaries}
total_orgs = len(facs)
for year, month in months_between(start_date, end_date):
window_date = datetime(year, month, 1)
org_summary = OrganizationSummary.objects.get_or_create(
location_id=location.location_id, date=window_date)[0]
org_summary.total_orgs = total_orgs
# lead times
if (year, month) in sub_summaries:
sub_summary = sub_summaries[year, month]
org_summary.average_lead_time_in_days = sub_summary['average_time']
else:
org_summary.average_lead_time_in_days = 0
org_summary.save()
# product availability
for p in prods:
product_data = ProductAvailabilityData.objects.get_or_create(
product=p.product_id,
location_id=location.location_id,
date=window_date)[0]
sub_prod = sub_prods.get(((year, month), p.product_id), {})
product_data.total = sub_prod.get('total_sum', 0)
if strict:
assert product_data.total == total_orgs, \
"total should match number of sub facilities %s-%s" % (product_data.total, total_orgs)
product_data.with_stock = sub_prod.get('with_stock_sum', 0)
product_data.without_stock = sub_prod.get('without_stock_sum', 0)
product_data.without_data = product_data.total - product_data.with_stock - product_data.without_stock
product_data.save()
dg = DeliveryGroups(month=month, facs=facs)
for status_type in const.NEEDED_STATUS_TYPES:
gsum = GroupSummary.objects.get_or_create(org_summary=org_summary,
title=status_type)[0]
sub_sum = sub_group_summaries.get(((year, month), status_type), {})
gsum.total = sub_sum.get('total_sum', 0)
gsum.responded = sub_sum.get('responded_sum', 0)
gsum.on_time = sub_sum.get('on_time_sum', 0)
gsum.complete = sub_sum.get('complete_sum', 0)
gsum.save()
if status_type == SupplyPointStatusTypes.DELIVERY_FACILITY:
expected = len(dg.delivering())
elif status_type == SupplyPointStatusTypes.R_AND_R_FACILITY:
expected = len(dg.submitting())
elif status_type == SupplyPointStatusTypes.SOH_FACILITY \
or status_type == SupplyPointStatusTypes.SUPERVISION_FACILITY:
expected = len(facs)
if gsum.total != expected:
logging.info("expected %s but was %s for %s" %
(expected, gsum.total, gsum))
for alert_type in [
const.RR_NOT_SUBMITTED, const.DELIVERY_NOT_RECEIVED,
const.SOH_NOT_RESPONDING, const.RR_NOT_RESPONDED,
const.DELIVERY_NOT_RESPONDING
]:
sub_alerts = Alert.objects.filter(location_id__in=fac_ids,
date=window_date,
type=alert_type)
aggregate_response_alerts(location.location_id, window_date,
sub_alerts, alert_type)
update_historical_data_for_location(location)
def process_non_facility_warehouse_data(org, start_date, end_date, strict=True):
facs = get_nested_children(org)
fac_ids = [f._id for f in facs]
logging.info("processing non-facility %s (%s), %s children" % (org.name, str(org._id), len(facs)))
for year, month in months_between(start_date, end_date):
window_date = datetime(year, month, 1)
org_summary = OrganizationSummary.objects.get_or_create(supply_point=org._id, date=window_date)[0]
org_summary.total_orgs = len(facs)
sub_summaries = OrganizationSummary.objects.filter(date=window_date, supply_point__in=fac_ids)
subs_with_lead_time = [s for s in sub_summaries if s.average_lead_time_in_days]
# lead times
if subs_with_lead_time:
days_sum = sum([s.average_lead_time_in_days for s in subs_with_lead_time])
org_summary.average_lead_time_in_days = days_sum / len(subs_with_lead_time)
else:
org_summary.average_lead_time_in_days = 0
org_summary.save()
# product availability
prods = Product.ids_by_domain(org.domain)
for p in prods:
product_data = ProductAvailabilityData.objects.get_or_create(product=p,
supply_point=org._id,
date=window_date)[0]
sub_prods = ProductAvailabilityData.objects.filter(product=p,
supply_point__in=fac_ids,
date=window_date)
product_data.total = sum([p.total for p in sub_prods])
if strict:
assert product_data.total == len(facs), \
"total should match number of sub facilities"
product_data.with_stock = sum([p.with_stock for p in sub_prods])
product_data.without_stock = sum([p.without_stock for p in sub_prods])
product_data.without_data = product_data.total - product_data.with_stock - product_data.without_stock
product_data.save()
dg = DeliveryGroups(month=month, facs=facs)
for type in NEEDED_STATUS_TYPES:
gsum = GroupSummary.objects.get_or_create(org_summary=org_summary, title=type)[0]
sub_sums = GroupSummary.objects.filter(title=type, org_summary__in=sub_summaries).all()
# TODO: see if moving the aggregation to the db makes it
# faster, if this is slow
gsum.total = sum([s.total for s in sub_sums])
gsum.responded = sum([s.responded for s in sub_sums])
gsum.on_time = sum([s.on_time for s in sub_sums])
gsum.complete = sum([s.complete for s in sub_sums])
# gsum.missed_response = sum([s.missed_response for s in sub_sums])
gsum.save()
if type == SupplyPointStatusTypes.DELIVERY_FACILITY:
expected = len(dg.delivering())
elif type == SupplyPointStatusTypes.R_AND_R_FACILITY:
expected = len(dg.submitting())
elif type == SupplyPointStatusTypes.SOH_FACILITY \
or type == SupplyPointStatusTypes.SUPERVISION_FACILITY:
expected = len(facs)
if gsum.total != expected:
logging.info("expected %s but was %s for %s" % (expected, gsum.total, gsum))
for alert_type in ['rr_not_submitted', 'delivery_not_received',
'soh_not_responding', 'rr_not_responded', 'delivery_not_responding']:
sub_alerts = Alert.objects.filter(supply_point__in=fac_ids, date=window_date, type=alert_type)
aggregate_response_alerts(org, window_date, sub_alerts, alert_type)
def process_non_facility_warehouse_data(location, start_date, end_date, runner=None, strict=True):
if runner:
runner.location = location.sql_location
runner.save()
facs = get_non_archived_facilities_below(location)
fac_ids = [f._id for f in facs]
logging.info("processing non-facility %s (%s), %s children"
% (location.name, str(location.location_id), len(facs)))
for year, month in months_between(start_date, end_date):
window_date = datetime(year, month, 1)
org_summary = OrganizationSummary.objects.get_or_create(
location_id=location.location_id, date=window_date
)[0]
org_summary.total_orgs = len(facs)
sub_summaries = OrganizationSummary.objects.filter(date=window_date, location_id__in=fac_ids)
subs_with_lead_time = [s for s in sub_summaries if s.average_lead_time_in_days]
# lead times
if subs_with_lead_time:
days_sum = sum([s.average_lead_time_in_days for s in subs_with_lead_time])
org_summary.average_lead_time_in_days = days_sum / len(subs_with_lead_time)
else:
org_summary.average_lead_time_in_days = 0
org_summary.save()
# product availability
prods = SQLProduct.objects.filter(domain=location.domain, is_archived=False)
for p in prods:
product_data = ProductAvailabilityData.objects.get_or_create(product=p.product_id,
location_id=location.location_id,
date=window_date)[0]
sub_prods = ProductAvailabilityData.objects.filter(product=p.product_id,
location_id__in=fac_ids,
date=window_date)
product_data.total = sum([p.total for p in sub_prods])
if strict:
assert product_data.total == len(facs), \
"total should match number of sub facilities"
product_data.with_stock = sum([p.with_stock for p in sub_prods])
product_data.without_stock = sum([p.without_stock for p in sub_prods])
product_data.without_data = product_data.total - product_data.with_stock - product_data.without_stock
product_data.save()
dg = DeliveryGroups(month=month, facs=facs)
for status_type in const.NEEDED_STATUS_TYPES:
gsum = GroupSummary.objects.get_or_create(org_summary=org_summary, title=status_type)[0]
sub_sums = GroupSummary.objects.filter(title=status_type, org_summary__in=sub_summaries).all()
# TODO: see if moving the aggregation to the db makes it
# faster, if this is slow
gsum.total = sum([s.total for s in sub_sums])
gsum.responded = sum([s.responded for s in sub_sums])
gsum.on_time = sum([s.on_time for s in sub_sums])
gsum.complete = sum([s.complete for s in sub_sums])
# gsum.missed_response = sum([s.missed_response for s in sub_sums])
gsum.save()
if status_type == SupplyPointStatusTypes.DELIVERY_FACILITY:
expected = len(dg.delivering())
elif status_type == SupplyPointStatusTypes.R_AND_R_FACILITY:
expected = len(dg.submitting())
elif status_type == SupplyPointStatusTypes.SOH_FACILITY \
or status_type == SupplyPointStatusTypes.SUPERVISION_FACILITY:
expected = len(facs)
if gsum.total != expected:
logging.info("expected %s but was %s for %s" % (expected, gsum.total, gsum))
for alert_type in [const.RR_NOT_SUBMITTED, const.DELIVERY_NOT_RECEIVED,
const.SOH_NOT_RESPONDING, const.RR_NOT_RESPONDED, const.DELIVERY_NOT_RESPONDING]:
sub_alerts = Alert.objects.filter(location_id__in=fac_ids, date=window_date, type=alert_type)
aggregate_response_alerts(location.location_id, window_date, sub_alerts, alert_type)
def process_non_facility_warehouse_data(location, start_date, end_date, runner=None, strict=True):
start_date = datetime(start_date.year, start_date.month, 1)
end_date = datetime(end_date.year, end_date.month, 1)
if runner:
runner.location = location.sql_location
runner.save()
facs = get_non_archived_facilities_below(location)
fac_ids = [f._id for f in facs]
logging.info("processing non-facility %s (%s), %s children"
% (location.name, str(location.location_id), len(facs)))
prods = SQLProduct.objects.filter(domain=location.domain, is_archived=False)
sub_summaries = OrganizationSummary.objects.filter(
location_id__in=fac_ids, date__range=(start_date, end_date), average_lead_time_in_days__gt=0
).values('date').annotate(average_time=Avg('average_lead_time_in_days'))
sub_summaries = {
(subsummary['date'].year, subsummary['date'].month): subsummary
for subsummary in sub_summaries
}
sub_prods = ProductAvailabilityData.objects.filter(
location_id__in=fac_ids, date__range=(start_date, end_date)
).values('product', 'date').annotate(
total_sum=Sum('total'),
with_stock_sum=Sum('with_stock'),
without_stock_sum=Sum('without_stock'),
)
sub_prods = {
((sub_prod['date'].year, sub_prod['date'].month), sub_prod['product']): sub_prod for sub_prod in sub_prods
}
sub_group_summaries = GroupSummary.objects.filter(
org_summary__location_id__in=fac_ids,
org_summary__date__range=(start_date, end_date)
).values('title', 'org_summary__date').annotate(
total_sum=Sum('total'),
responded_sum=Sum('responded'),
on_time_sum=Sum('on_time'),
complete_sum=Sum('complete')
)
sub_group_summaries = {
((sub_group_summary['org_summary__date'].year, sub_group_summary['org_summary__date'].month), sub_group_summary['title']): sub_group_summary
for sub_group_summary in sub_group_summaries
}
total_orgs = len(facs)
for year, month in months_between(start_date, end_date):
window_date = datetime(year, month, 1)
org_summary = OrganizationSummary.objects.get_or_create(
location_id=location.location_id, date=window_date
)[0]
org_summary.total_orgs = total_orgs
# lead times
if (year, month) in sub_summaries:
sub_summary = sub_summaries[year, month]
org_summary.average_lead_time_in_days = sub_summary['average_time']
else:
org_summary.average_lead_time_in_days = 0
org_summary.save()
# product availability
for p in prods:
product_data = ProductAvailabilityData.objects.get_or_create(product=p.product_id,
location_id=location.location_id,
date=window_date)[0]
sub_prod = sub_prods.get(((year, month), p.product_id), {})
product_data.total = sub_prod.get('total_sum', 0)
if strict:
assert product_data.total == total_orgs, \
"total should match number of sub facilities %s-%s" % (product_data.total, total_orgs)
product_data.with_stock = sub_prod.get('with_stock_sum', 0)
product_data.without_stock = sub_prod.get('without_stock_sum', 0)
product_data.without_data = product_data.total - product_data.with_stock - product_data.without_stock
product_data.save()
dg = DeliveryGroups(month=month, facs=facs)
for status_type in const.NEEDED_STATUS_TYPES:
gsum = GroupSummary.objects.get_or_create(org_summary=org_summary, title=status_type)[0]
sub_sum = sub_group_summaries.get(((year, month), status_type), {})
gsum.total = sub_sum.get('total_sum', 0)
gsum.responded = sub_sum.get('responded_sum', 0)
gsum.on_time = sub_sum.get('on_time_sum', 0)
gsum.complete = sub_sum.get('complete_sum', 0)
gsum.save()
if status_type == SupplyPointStatusTypes.DELIVERY_FACILITY:
expected = len(dg.delivering())
elif status_type == SupplyPointStatusTypes.R_AND_R_FACILITY:
expected = len(dg.submitting())
elif status_type == SupplyPointStatusTypes.SOH_FACILITY \
or status_type == SupplyPointStatusTypes.SUPERVISION_FACILITY:
expected = len(facs)
if gsum.total != expected:
logging.info("expected %s but was %s for %s" % (expected, gsum.total, gsum))
for alert_type in [const.RR_NOT_SUBMITTED, const.DELIVERY_NOT_RECEIVED,
const.SOH_NOT_RESPONDING, const.RR_NOT_RESPONDED, const.DELIVERY_NOT_RESPONDING]:
sub_alerts = Alert.objects.filter(location_id__in=fac_ids, date=window_date, type=alert_type)
aggregate_response_alerts(location.location_id, window_date, sub_alerts, alert_type)
update_historical_data_for_location(location)
def process_non_facility_warehouse_data(location,
start_date,
end_date,
runner,
strict=True):
runner.location = location.sql_location
runner.save()
facs = get_non_archived_facilities_below(location)
fac_ids = [f._id for f in facs]
logging.info("processing non-facility %s (%s), %s children" %
(location.name, str(location._id), len(facs)))
for year, month in months_between(start_date, end_date):
window_date = datetime(year, month, 1)
org_summary = OrganizationSummary.objects.get_or_create(
location_id=location._id, date=window_date)[0]
org_summary.total_orgs = len(facs)
sub_summaries = OrganizationSummary.objects.filter(
date=window_date, location_id__in=fac_ids)
subs_with_lead_time = [
s for s in sub_summaries if s.average_lead_time_in_days
]
# lead times
if subs_with_lead_time:
days_sum = sum(
[s.average_lead_time_in_days for s in subs_with_lead_time])
org_summary.average_lead_time_in_days = days_sum / len(
subs_with_lead_time)
else:
org_summary.average_lead_time_in_days = 0
org_summary.save()
# product availability
prods = SQLProduct.objects.filter(domain=location.domain,
is_archived=False)
for p in prods:
product_data = ProductAvailabilityData.objects.get_or_create(
product=p.product_id,
location_id=location._id,
date=window_date)[0]
sub_prods = ProductAvailabilityData.objects.filter(
product=p.product_id,
location_id__in=fac_ids,
date=window_date)
product_data.total = sum([p.total for p in sub_prods])
if strict:
assert product_data.total == len(facs), \
"total should match number of sub facilities"
product_data.with_stock = sum([p.with_stock for p in sub_prods])
product_data.without_stock = sum(
[p.without_stock for p in sub_prods])
product_data.without_data = product_data.total - product_data.with_stock - product_data.without_stock
product_data.save()
dg = DeliveryGroups(month=month, facs=facs)
for status_type in const.NEEDED_STATUS_TYPES:
gsum = GroupSummary.objects.get_or_create(org_summary=org_summary,
title=status_type)[0]
sub_sums = GroupSummary.objects.filter(
title=status_type, org_summary__in=sub_summaries).all()
# TODO: see if moving the aggregation to the db makes it
# faster, if this is slow
gsum.total = sum([s.total for s in sub_sums])
gsum.responded = sum([s.responded for s in sub_sums])
gsum.on_time = sum([s.on_time for s in sub_sums])
gsum.complete = sum([s.complete for s in sub_sums])
# gsum.missed_response = sum([s.missed_response for s in sub_sums])
gsum.save()
if status_type == SupplyPointStatusTypes.DELIVERY_FACILITY:
expected = len(dg.delivering())
elif status_type == SupplyPointStatusTypes.R_AND_R_FACILITY:
expected = len(dg.submitting())
elif status_type == SupplyPointStatusTypes.SOH_FACILITY \
or status_type == SupplyPointStatusTypes.SUPERVISION_FACILITY:
expected = len(facs)
if gsum.total != expected:
logging.info("expected %s but was %s for %s" %
(expected, gsum.total, gsum))
for alert_type in [
const.RR_NOT_SUBMITTED, const.DELIVERY_NOT_RECEIVED,
const.SOH_NOT_RESPONDING, const.RR_NOT_RESPONDED,
const.DELIVERY_NOT_RESPONDING
]:
sub_alerts = Alert.objects.filter(location_id__in=fac_ids,
date=window_date,
type=alert_type)
aggregate_response_alerts(location._id, window_date, sub_alerts,
alert_type)
