def lambda_handler(event, context):
print('Event Received: {}'.format(json.dumps(event)))
extra_params = {
'beaconId': BEACON_ID,
}
if event['httpMethod'] == 'POST':
event_body = event.get('body')
if not event_body:
return bad_request('No body sent with request.', extra_params)
try:
parameters = json.loads(event_body)
except ValueError:
return bad_request('Error parsing request body, Expected JSON.',
extra_params)
else: # method == 'GET'
parameters = event['queryStringParameters']
if not parameters:
return bad_request('No query parameters sent with request.',
extra_params)
multi_values = event['multiValueQueryStringParameters']
parameters['datasetIds'] = multi_values.get('datasetIds')
for int_field in ('start', 'end', 'startMin', 'startMax', 'endMin',
'endMax'):
if int_field in parameters:
try:
parameters[int_field] = int(parameters[int_field])
except ValueError:
# Cannot be formatted as an integer, handle in validation
pass
return query_datasets(parameters)
def lambda_handler(event, context):
print('Event Received: {}'.format(json.dumps(event)))
extra_params = {
'beaconId': BEACON_ID,
}
if event['httpMethod'] == 'POST':
event_body = event.get('body')
if not event_body:
return bad_request('No body sent with request.', extra_params)
try:
parameters = json.loads(event_body)
except ValueError:
return bad_request('Error parsing request body, Expected JSON.',
extra_params)
parameters['queries'] = get_queries(parameters)
else: # method == 'GET'
parameters = event['queryStringParameters']
if not parameters:
return bad_request('No query parameters sent with request.',
extra_params)
multi_values = event['multiValueQueryStringParameters']
parameters['queries'] = get_queries(multi_values)
parameters['datasetIds'] = multi_values.get('datasetIds')
parameters['sampleFields'] = multi_values.get('sampleFields')
for int_field in ('page', 'pageSize', 'variantsDescending', 'similar'):
if int_field in parameters:
parameters[int_field] = int_or_self(parameters[int_field])
return query_datasets(parameters, context)
def lambda_handler(event, context):
print('Event Received: {}'.format(json.dumps(event)))
event_body = event.get('body')
if not event_body:
return bad_request('No body sent with request.')
try:
body_dict = json.loads(event_body)
except ValueError:
return bad_request('Error parsing request body, Expected JSON.')
method = event['httpMethod']
return submit_dataset(body_dict, method)
def lambda_handler(event, context):
print('Event Received: {}'.format(json.dumps(event)))
time_assigned = (context.get_remaining_time_in_millis()-MILLISECONDS_BEFORE_SPLIT)
print("time assigned",time_assigned)
timeStart = time.time()
event_body = event.get('body')
if not event_body:
return bad_request('No body sent with request.')
try:
body_dict = json.loads(event_body)
requestID = event['requestContext']['requestId']
location = body_dict['location']
vcf_regions = get_translated_regions(location)
except ValueError:
return bad_request('Error parsing request body, Expected JSON.')
batchID = ''
print(vcf_regions)
for index,region in enumerate(vcf_regions):
if( (time.time() - timeStart)*1000 > time_assigned):
newRegions = vcf_regions[index:]
batchID = ''
print("New Regions ",newRegions)
#publish Sns for itself!
kwargs = {
'TopicArn': QUERY_VCF_EXTENDED_SNS_TOPIC_ARN,
}
kwargs['Message'] = json.dumps({'regions' : newRegions,'requestID' : requestID, 'location': location})
print('Publishing to SNS: {}'.format(json.dumps(kwargs)))
response = sns.publish(**kwargs)
print('Received Response: {}'.format(json.dumps(response)))
break
else:
chrom, start_str = region.split(':')
regionID = chrom+"_"+start_str
start = round(1000000 * float(start_str) + 1)
end = start + round(1000000 * SLICE_SIZE_MBP - 1)
all_coords, all_changes = get_regions_and_variants(location, chrom, start, end, time_assigned)
total_coords = [all_coords[x:x+RECORDS_PER_SAMPLE] for x in range(0, len(all_coords), RECORDS_PER_SAMPLE)]
total_changes = [all_changes[x:x+RECORDS_PER_SAMPLE] for x in range(0, len(all_changes), RECORDS_PER_SAMPLE)]
batchID = submitQueryGTF(total_coords,total_changes,requestID,regionID)
if(batchID != ''):
print("sending for concat")
kwargs = {
'TopicArn': CONCAT_SNS_TOPIC_ARN,
}
kwargs['Message'] = json.dumps({'APIid' : requestID,'lastBatchID' : batchID})
print('Publishing to SNS: {}'.format(json.dumps(kwargs)))
response = sns.publish(**kwargs)
print('Received Response: {}'.format(json.dumps(response)))
return bundle_response(200, "Process started")
def submit_dataset(body_dict, method):
new = method == 'POST'
validation_error = validate_request(body_dict, new)
if validation_error:
return bad_request(validation_error)
if 'vcfLocations' in body_dict:
errors = check_vcf_locations(body_dict['vcfLocations'])
if errors:
return bad_request(errors)
summarise = True
else:
summarise = False
if new:
create_dataset(body_dict)
else:
update_dataset(body_dict)
if summarise:
summarise_dataset(body_dict['id'])
return bundle_response(200, {})
def query_datasets(parameters, context):
response_dict = {
'beaconId': BEACON_ID,
'apiVersion': None,
'alleleRequest': parameters,
}
validation_error = validate_request(parameters)
if validation_error:
return bad_request(validation_error, response_dict)
responses = Caches(
dynamodb_client=dynamodb,
lambda_client=aws_lambda,
s3_client=s3,
)
datasets = get_datasets(
parameters['assemblyId'],
[query['referenceName'] for query in parameters['queries']],
parameters.get('datasetIds'),
)
query_details_list = get_query_details_list(parameters['queries'])
page_details = {
'page': parameters.get('page', 1),
'page_size': parameters.get('pageSize'),
'sortby': parameters.get('variantsSortby', 'pos'),
'desc': bool(parameters.get('variantsDescending')),
}
include_datasets = parameters.get('includeDatasetResponses', 'NONE')
iupac = parameters.get('iupac', 'True')
similar = bool(parameters.get('similar'))
call_collate_queries(datasets, query_details_list,
parameters.get('queryCombination'), page_details,
parameters.get('sampleFields'), include_datasets,
iupac, similar, responses)
dataset_responses = []
exists = False
for response in responses.collect_responses():
result = response.result
if not result or 'exists' not in result:
# function errored out, ignore
continue
if not exists and result['exists']:
exists = True
if include_datasets == 'NONE':
break
if result.pop('include'):
dataset_responses.append(result)
dataset_responses.sort(key=lambda r: r['datasetId'])
response_dict.update({
'exists': exists,
'datasetAlleleResponses': dataset_responses or None,
})
response_dict = check_size(response_dict, context)
return bundle_response(200, response_dict)
def query_datasets(parameters):
response_dict = {
'beaconId': BEACON_ID,
'apiVersion': None,
'alleleRequest': parameters,
}
validation_error = validate_request(parameters)
if validation_error:
return bad_request(validation_error, response_dict)
datasets = get_datasets(parameters['assemblyId'],
parameters.get('datasetIds'))
reference_name = parameters['referenceName']
vcf_chromosomes = get_vcf_chromosome_map(datasets, reference_name)
start = parameters.get('start')
if start is None:
region_start = parameters['startMin']
region_end = parameters['startMax']
end_min = parameters['endMin']
end_max = parameters['endMax']
else:
region_start = region_end = start
end = parameters.get('end')
if end is None:
end = start
end_min = end_max = end
reference_bases = parameters['referenceBases']
# account for the 1-based indexing of vcf files
region_start += 1
region_end += 1
end_min += 1
end_max += 1
if reference_bases != 'N':
# For specific reference bases region may be smaller
max_offset = len(reference_bases) - 1
end_max = min(region_end + max_offset, end_max)
region_start = max(end_min - max_offset, region_start)
if end_min > end_max or region_start > region_end:
# Region search will find nothing, search a dummy region
region_start = 2000000000
region_end = region_start
end_min = region_start + max_offset
end_max = end_min
alternate_bases = parameters.get('alternateBases')
variant_type = parameters.get('variantType')
include_datasets = parameters.get('includeDatasetResponses', 'NONE')
responses = queue.Queue()
threads = []
for dataset in datasets:
dataset_id = dataset['id']['S']
vcf_locations = {vcf: vcf_chromosomes[vcf]
for vcf in dataset['vcfLocations']['SS']
if vcf_chromosomes[vcf]}
t = threading.Thread(target=perform_query,
kwargs={
'dataset_id': dataset_id,
'vcf_locations': vcf_locations,
'reference_bases': reference_bases,
'region_start': region_start,
'region_end': region_end,
'end_min': end_min,
'end_max': end_max,
'alternate_bases': alternate_bases,
'variant_type': variant_type,
'include_datasets': include_datasets,
'responses': responses,
})
t.start()
threads.append(t)
num_threads = len(threads)
processed = 0
dataset_responses = []
exists = False
while processed < num_threads and (include_datasets != 'NONE'
or not exists):
response = responses.get()
processed += 1
if 'exists' not in response:
# function errored out, ignore
continue
exists = exists or response['exists']
if response.pop('include'):
dataset_responses.append(response)
response_dict.update({
'exists': exists,
'datasetAlleleResponses': dataset_responses or None,
})
return bundle_response(200, response_dict)
