def _iter_collect_organic_insights(
cls, data_iter: Generator[Dict[str, Any], None, None],
job_scope: JobScope) -> Generator[Dict[str, Any], None, None]:
raw_store = batch_store.NormalStore(
job_scope,
bucket_type=ColdStoreBucketType.RAW_BUCKET,
custom_namespace=NAMESPACE_RAW)
orig_store = batch_store.NormalStore(
job_scope, bucket_type=ColdStoreBucketType.ORIGINAL_BUCKET)
common_vendor_data = {
'ad_account_id':
job_scope.ad_account_id,
'entity_type':
job_scope.report_variant,
'report_type':
job_scope.report_type,
ORGANIC_DATA_ENTITY_ID_MAP[job_scope.report_variant]:
job_scope.entity_id,
}
data = list(data_iter)
raw_record = {
'payload':
data,
'page_id':
job_scope.ad_account_id,
ORGANIC_DATA_ENTITY_ID_MAP[job_scope.report_variant]:
job_scope.entity_id,
}
vendor_data_raw = report_type_vendor_data_raw_extractor_map[
job_scope.report_type](raw_record, **common_vendor_data)
raw_record = add_vendor_data(raw_record, **vendor_data_raw)
raw_store.store(raw_record)
if len(data):
# then, transpose it to correct form
final_record = {
'page_id':
job_scope.ad_account_id,
ORGANIC_DATA_ENTITY_ID_MAP[job_scope.report_variant]:
job_scope.entity_id,
}
for param_datum in data:
final_record[
param_datum['name']] = param_datum['values'][0]['value']
vendor_data = report_type_vendor_data_extractor_map[
job_scope.report_type](raw_record, **common_vendor_data)
final_record = add_vendor_data(final_record, **vendor_data)
orig_store.store(final_record)
yield final_record
def iter_collect_entities_per_page_post(
job_scope: JobScope) -> Generator[Dict[str, Any], None, None]:
"""
Collects an arbitrary entity for a page post
"""
entity_type = job_scope.report_variant
page_token_manager = PageTokenManager.from_job_scope(job_scope)
with PlatformApiContext(
page_token_manager.get_best_token(
job_scope.ad_account_id)) as fb_ctx:
root_fb_entity = fb_ctx.to_fb_model(job_scope.entity_id,
Entity.PagePost)
entities = iter_native_entities_per_page_post(root_fb_entity, entity_type)
record_id_base_data = job_scope.to_dict()
record_id_base_data.update(entity_type=entity_type, report_variant=None)
del record_id_base_data['entity_id']
with ChunkDumpStore(job_scope, chunk_size=DEFAULT_CHUNK_SIZE) as store:
for entity in entities:
entity_data = entity.export_all_data()
entity_data = add_vendor_data(entity_data,
id=generate_universal_id(
entity_id=entity_data.get('id'),
**record_id_base_data))
entity_data['page_id'] = job_scope.ad_account_id
entity_data['page_post_id'] = job_scope.entity_id
# Store the individual datum, use job context for the cold
# storage thing to divine whatever it needs from the job context
store(entity_data)
yield entity_data
def collect_page(job_scope: JobScope, _job_context: JobContext):
"""
Collect a single facebook page
"""
if job_scope.report_variant != Entity.Page:
raise ValueError(
f"Report level {job_scope.report_variant} specified is not: {Entity.Page}"
)
token = job_scope.token
if not token:
raise ValueError(
f"Job {job_scope.job_id} cannot proceed. No platform tokens provided."
)
# We don't use it for getting a token. Something else that calls us does.
# However, we use it to report usages of the token we got.
token_manager = PlatformTokenManager.from_job_scope(job_scope)
with PlatformApiContext(token) as fb_ctx:
page_inst = page.Page(fbid=job_scope.entity_id, api=fb_ctx.api)
page_fetched = page_inst.api_get(fields=get_default_fields(Page))
report_job_status_task.delay(ExternalPlatformJobStatus.DataFetched,
job_scope)
token_manager.report_usage(token, 2)
record_id_data = job_scope.to_dict()
record_id_data.update(entity_type=Entity.Page,
entity_id=job_scope.entity_id,
report_variant=None)
entity_data = page_fetched.export_all_data()
entity_data = add_vendor_data(
entity_data, id=generate_universal_id(**record_id_data))
store = NormalStore(job_scope)
store.store(entity_data)
def collect_pages_from_business(job_scope: JobScope,
_job_context: JobContext) -> int:
"""
Collect all facebook pages that are active
"""
if job_scope.report_variant != Entity.Page:
raise ValueError(
f"Report level {job_scope.report_variant} specified is not: {Entity.Page}"
)
token = job_scope.token
if not token:
raise ValueError(
f"Job {job_scope.job_id} cannot proceed. No platform tokens provided."
)
# We don't use it for getting a token. Something else that calls us does.
# However, we use it to report usages of the token we got.
token_manager = PlatformTokenManager.from_job_scope(job_scope)
with PlatformApiContext(token) as fb_ctx:
fb_req = FacebookRequest(node_id="me",
method="GET",
endpoint="/businesses",
api=fb_ctx.api,
api_type='EDGE',
target_class=Business)
businesses = fb_req.execute()
report_job_status_task.delay(ExternalPlatformJobStatus.DataFetched,
job_scope)
token_manager.report_usage(token)
entity_type = Entity.Page
record_id_base_data = job_scope.to_dict()
record_id_base_data.update(entity_type=entity_type, report_variant=None)
cnt = 0
for biz in businesses:
client_pages = list(
biz.get_client_pages(fields=get_default_fields(Page)))
owned_pages = list(
biz.get_owned_pages(fields=get_default_fields(Page)))
pages_list = client_pages + owned_pages
for page_inst in pages_list:
entity_data = page_inst.export_all_data()
record_id_base_data.update(entity_id=entity_data.get('id'))
entity_data = add_vendor_data(
entity_data, id=generate_universal_id(**record_id_base_data))
store = NormalStore(job_scope)
store.store(entity_data)
cnt += 1
report_job_status_task.delay(ExternalPlatformJobStatus.Done, job_scope)
return cnt
def collect_adaccount(job_scope: JobScope) -> Dict[str, Any]:
"""
Collects ad account data for a AA specific JobScope definition.
:param JobScope job_scope: The JobScope as we get it from the task itself
"""
if job_scope.report_variant != Entity.AdAccount:
raise ValueError(
f"Report level {job_scope.report_variant} specified is not: {Entity.AdAccount}"
)
token = job_scope.token
if not token:
raise ValueError(
f"Job {job_scope.job_id} cannot proceed. No platform tokens provided."
)
assert (
job_scope.ad_account_id == job_scope.entity_id
), f'This is an ad account entity job, account_id should be equal to entity_id'
# Used to report token usage by this job
token_manager = PlatformTokenManager.from_job_scope(job_scope)
with PlatformApiContext(token) as fb_ctx:
ad_account = fb_ctx.to_fb_model(job_scope.ad_account_id,
Entity.AdAccount)
fields = get_default_fields(ad_account.__class__)
ad_account_with_selected_fields = ad_account.api_get(
fields=fields) # Read just the fields we need
ad_account_data_dict = ad_account_with_selected_fields.export_all_data(
) # Export the object to a dict
token_manager.report_usage(token)
job_scope_base = {
# Duplicate the job_scope data to avoid mutating it
**job_scope.to_dict(),
'entity_type': Entity.AdAccount,
'report_variant': None,
}
augmented_ad_account_data = add_vendor_data(
# Augment the data returned from the remote API with our vendor data
ad_account_data_dict,
id=generate_universal_id(**job_scope_base),
)
feedback_entity_task.delay(ad_account_data_dict,
job_scope.report_variant)
store = NormalStore(job_scope)
store.store(augmented_ad_account_data)
# TODO: feedback account? this probably wouldn't make sense at the moment
# because ad accounts are discovered from console and their lifecycle is controlled from there.
return ad_account_data_dict
def test_add_new_vendor_block(self):
data = {'a': 1}
data_should_be = {'a': 1, '__oprm': {'id': 5}}
data_actual = add_vendor_data(data, id=5)
assert data_actual is data, 'we did not repackage the instance. Same instance'
assert data_actual == data_should_be
def test_update_existing_vendor_block(self):
data = {'a': 1, '__oprm': {'id': 5}}
data_should_be = {'a': 1, '__oprm': {'id': 5, 'extra_attr': 7}}
data_actual = add_vendor_data(data, extra_attr=7)
assert data_actual is data, 'we did not repackage the instance. Same instance'
assert data_actual == data_should_be
def iter_collect_entities_per_page(
job_scope: JobScope) -> Generator[Dict[str, Any], None, None]:
"""
Collects an arbitrary entity for a page
"""
token, entity_type, root_fb_entity = _extract_token_entity_type_parent_entity(
job_scope, [Entity.PagePost, Entity.PageVideo], Entity.Page,
'ad_account_id')
entities = iter_native_entities_per_page(root_fb_entity, entity_type)
record_id_base_data = job_scope.to_dict()
record_id_base_data.update(entity_type=entity_type, report_variant=None)
token_manager = PlatformTokenManager.from_job_scope(job_scope)
with ChunkDumpStore(
job_scope, chunk_size=DEFAULT_CHUNK_SIZE) as store, ChunkDumpStore(
job_scope,
chunk_size=DEFAULT_CHUNK_SIZE,
bucket_type=ColdStoreBucketType.RAW_BUCKET,
custom_namespace=NAMESPACE_RAW,
) as raw_store:
cnt = 0
for entity in entities:
entity_data = entity.export_all_data()
entity_data = add_vendor_data(entity_data,
id=generate_universal_id(
entity_id=entity_data.get('id'),
**record_id_base_data))
entity_data['page_id'] = job_scope.ad_account_id
if entity_type == Entity.PagePost:
# store raw version of response (just to remain consistent)
raw_store(entity_data)
entity_data = _augment_page_post(entity_data)
# Store the individual datum, use job context for the cold
# storage thing to divine whatever it needs from the job context
store(entity_data)
# Signal to the system the new entity
feedback_entity_task.delay(entity_data, entity_type)
yield entity_data
cnt += 1
if cnt % 1000 == 0:
# default paging size for entities per parent
# is typically around 200. So, each 200 results
# means about 5 hits to FB
token_manager.report_usage(token, 5)
token_manager.report_usage(token)
def iter_collect_entities_per_page_graph(
job_scope: JobScope) -> Generator[Dict[str, Any], None, None]:
"""
Collects an arbitrary entity for a page using graph API
"""
page_token_manager = PageTokenManager.from_job_scope(job_scope)
with PlatformApiContext(
page_token_manager.get_best_token(
job_scope.ad_account_id)) as fb_ctx:
page_root_fb_entity = fb_ctx.to_fb_model(job_scope.ad_account_id,
Entity.Page)
entity_type = job_scope.report_variant
# page size reduced to avoid error:
# "Please reduce the amount of data you're asking for, then retry your request"
entities = iter_native_entities_per_page_graph(page_root_fb_entity,
entity_type,
page_size=30)
record_id_base_data = job_scope.to_dict()
record_id_base_data.update(entity_type=entity_type, report_variant=None)
with ChunkDumpStore(
job_scope, chunk_size=DEFAULT_CHUNK_SIZE) as store, ChunkDumpStore(
job_scope,
chunk_size=DEFAULT_CHUNK_SIZE,
bucket_type=ColdStoreBucketType.RAW_BUCKET,
custom_namespace=NAMESPACE_RAW,
) as raw_store:
for entity in entities:
entity_data = entity.export_all_data()
entity_data = add_vendor_data(entity_data,
id=generate_universal_id(
entity_id=entity_data.get('id'),
**record_id_base_data))
entity_data['page_id'] = job_scope.ad_account_id
if entity_type == Entity.PagePostPromotable:
# store raw version of response (just to remain consistent)
raw_store(entity_data)
entity_data = _augment_page_post(entity_data)
# Store the individual datum, use job context for the cold
# storage thing to divine whatever it needs from the job context
store(entity_data)
# Signal to the system the new entity
feedback_entity_task.delay(entity_data, entity_type)
yield entity_data
def iter_collect_entities_per_adaccount(
job_scope: JobScope) -> Generator[Dict[str, Any], None, None]:
"""
Collects an arbitrary entity for an ad account
"""
token, entity_type, root_fb_entity = _extract_token_entity_type_parent_entity(
job_scope, Entity.AA_SCOPED, Entity.AdAccount, 'ad_account_id')
entities = iter_native_entities_per_adaccount(root_fb_entity, entity_type)
record_id_base_data = job_scope.to_dict()
record_id_base_data.update(entity_type=entity_type, report_variant=None)
token_manager = PlatformTokenManager.from_job_scope(job_scope)
with ChunkDumpStore(job_scope, chunk_size=DEFAULT_CHUNK_SIZE) as store:
for cnt, entity in enumerate(entities):
entity_data = entity.export_all_data()
entity_data = add_vendor_data(
entity_data,
id=generate_universal_id(
# FIXME: add a bug to facebook ads (get_ad_videos doesnt return ad videos but AbstractCrudObject)
# FIXME so it is unable to access entity.Field.id then (only a problem for ad videos)
entity_id=entity_data.get('id'),
**record_id_base_data,
),
)
# Store the individual datum, use job context for the cold
# storage thing to divine whatever it needs from the job context
store(entity_data)
# Signal to the system the new entity
feedback_entity_task.delay(entity_data, entity_type)
yield entity_data
if cnt % 1000 == 0:
# default paging size for entities per parent
# is typically around 200. So, each 200 results
# means about 5 hits to FB
token_manager.report_usage(token, 5)
# Report on the effective task status
token_manager.report_usage(token)
def __init__(self, job_scope: JobScope, report_entity_api_kind: str):
if job_scope.report_type not in ReportType.ALL_METRICS:
raise ValueError(
f"Report type {job_scope.report_type} specified is not one of supported values: "
+ ReportType.ALL_METRICS)
# cool. we are in the right place...
self.report_params = {
'fields':
DEFAULT_REPORT_FIELDS,
'action_attribution_windows': [
# https://developers.facebook.com/docs/marketing-api/reference/adgroup/insights/
# https://developers.facebook.com/docs/marketing-api/insights#sample
# 'actions' and 'action_values' can contain values per different measurement window
# In case of 'actions', default 'value' is always 1d_view PLUS 28d_click and cannot be removed.
# In case of 'action_values', default 'value' is some weighted sum of
# 1d_view AND 28d_click $ values, that may be smaller than raw 1d_view PLUS 28d_click $ values.
# Many customers interpret their conversions / actions in different attribution windows.
# The more windows we ask the data for, the less reliably it returns reports.
# Be super conservative about asking for more / all.
AdsInsights.ActionAttributionWindows.value_1d_view,
AdsInsights.ActionAttributionWindows.value_7d_view,
AdsInsights.ActionAttributionWindows.value_28d_view,
AdsInsights.ActionAttributionWindows.value_1d_click,
AdsInsights.ActionAttributionWindows.value_7d_click,
AdsInsights.ActionAttributionWindows.value_28d_click,
],
}
# Next is (a) vs (b) - abstraction level determination
is_per_parent_report = not job_scope.entity_id and job_scope.report_variant in Entity.ALL
if is_per_parent_report:
entity_id = job_scope.ad_account_id
entity_type = Entity.AdAccount
entity_type_reporting = job_scope.report_variant
if report_entity_api_kind == ReportEntityApiKind.Ad:
self.report_params.update(
level=ENUM_LEVEL_MAP[job_scope.report_variant])
else:
# direct, per-entity report
entity_id = job_scope.entity_id
entity_type = job_scope.entity_type
entity_type_reporting = job_scope.report_variant
if report_entity_api_kind == ReportEntityApiKind.Ad:
self.report_params.update(
level=ENUM_LEVEL_MAP[entity_type_reporting])
# Now, (c), (d), (e), (f), (g) choices
# we already checked above that this is one of metrics report types
# So we know it will be either lifetime or day-with-breakdown type
# TODO: add fields listings appropriate for each type
if job_scope.report_type == ReportType.lifetime:
self.report_params.update(
date_preset=AdsInsights.DatePreset.lifetime)
elif job_scope.report_type in REPORT_TYPE_FB_BREAKDOWN_ENUM: # some day-with-breakdown type
self.report_params.update(
time_increment=1, # group by calendar day (in AA tz)
time_range={
'since':
_convert_and_validate_date_format(job_scope.range_start),
# No value for job_scope.range_end means 1-day report for range_start day
'until':
_convert_and_validate_date_format(
job_scope.range_end or job_scope.range_start),
},
breakdowns=REPORT_TYPE_FB_BREAKDOWN_ENUM[
job_scope.report_type],
)
else:
raise ValueError(
f"Report type {job_scope.report_type} does not have a mapped Platform-side breakdown value."
)
# Indicates that datum returned in a per-parent report is by itself
# naturally mapped to some single normative job ,
# meaning each element can be stored separately
# but only under normative ID computed on the fly
# from the datum.
# This must be accompanied by a transform fn that
# derives a normative ID from data.
# special case.
# when report type is per-specific-single-entity-ID
# AND one of per-day-with-breakdown
# per-Entity-ID-per-day bundle with 24 records before saving it.
# This results in a single write to the cold store under
# single normative ID.
is_whole_report_bundle_write = (
# must be one of those per-day reports
job_scope.report_type in ReportType.ALL_DAY_BREAKDOWNS and
# except for DMA-based data, as these can be very long,
# - 10s of thousands of records per day
job_scope.report_type not in [
ReportType.day_dma, ReportType.day_region,
ReportType.day_country
] and
# and the report is per single entity ID
job_scope.entity_id and not job_scope.report_variant and
# and report is for a single calendar day
# ReportType.ALL_DAY_BREAKDOWNS means there must be a non-Null
# value in time_range, but we check anyway
self.report_params['time_range']['since']
and self.report_params['time_range']['since']
== self.report_params['time_range']['until'])
# a more complex variant of whole_report_bundle_write
# where, while we canNOT spool entire report into memory to
# write it as one bundle, we cannot really write each
# individual result out either, as there will be a shit-load of them
# and we have to write is some sort of batching mode, but
# cannot cleanly group the bundles into per-normative-ID bundles,
# and instead will write under effective ID, but with a suffix
# indicating the monotonically-increasing chunk number.
# Disabled but kept for reference to compare to shorter version immediately below
# These represent good range of choices for cold store handlers.
# When / if there is value to it, steal from this commented out code.
# if is_naturally_normative_child:
# self.datum_handler = batch_store.NaturallyNormativeChildStore(job_scope)
# elif is_whole_report_bundle_write:
# self.datum_handler = batch_store.MemorySpoolStore(job_scope)
# elif is_chunk_write:
# self.datum_handler = batch_store.ChunkDumpStore(job_scope)
# else:
# self.datum_handler = batch_store.NormalStore(job_scope)
# let's be more aggressive about doing bundled writes to cold store
# and (temporarily) get away from "normative" and single-datum writes
# There are two ways we can get closer to bundled writes:
# - spool entire report in memory and flush out at the end, when we know we can tolerate that
# - spool large chunks of report in memory and flush them periodically if we fear large sizes in report.
if is_whole_report_bundle_write:
self.datum_handler = batch_store.MemorySpoolStore(job_scope)
else:
self.datum_handler = batch_store.ChunkDumpStore(job_scope,
chunk_size=200)
with PlatformApiContext(job_scope.token) as fb_ctx:
self.report_root_fb_entity = fb_ctx.to_fb_model(
entity_id, entity_type)
# here we configure code that will augment each datum with record ID
vendor_data_extractor = report_type_vendor_data_extractor_map[
job_scope.report_type]
if job_scope.report_type == ReportType.day_hour:
# hour report type's ID extractor function needs extra leading arg - timezone
vendor_data_extractor = functools.partial(
vendor_data_extractor, job_scope.ad_account_timezone_name)
aux_data = {
'ad_account_id': job_scope.ad_account_id,
'entity_type': entity_type_reporting,
'report_type': job_scope.report_type,
}
self.augment_with_vendor_data = lambda data: add_vendor_data(
data, **vendor_data_extractor(data, **aux_data))
