def lambda_handler(event, context):
print("event: {}".format(event))
result = {}
documentBucket = os.environ['CONTENT_BUCKET']
sampleBucket = os.environ['SAMPLE_BUCKET']
if ('resource' in event):
request = {}
request["elasticsearchDomain"] = os.environ['ES_DOMAIN']
request["outputTable"] = os.environ['OUTPUT_TABLE']
request["documentsTable"] = os.environ['DOCUMENTS_TABLE']
if (event['resource'] == '/search'):
if ('queryStringParameters' in event
and 'k' in event['queryStringParameters']):
request["keyword"] = event['queryStringParameters']['k']
if ('documentId' in event['queryStringParameters']):
request["documentId"] = event['queryStringParameters'][
'documentId']
result = search(request)
elif (event['resource'] == '/documents'):
if ('queryStringParameters' in event
and event['queryStringParameters']
and 'nexttoken' in event['queryStringParameters']):
request["nextToken"] = event['queryStringParameters'][
'nexttoken']
if ('queryStringParameters' in event
and event['queryStringParameters']
and 'type' in event['queryStringParameters']):
request["type"] = event['queryStringParameters']['type']
result = getDocuments(request)
elif (event['resource'] == '/document'):
if (event['httpMethod'] == 'GET'):
if ('queryStringParameters' in event
and event['queryStringParameters']):
if ('documentid' in event['queryStringParameters']):
request["documentId"] = event['queryStringParameters'][
'documentid']
result = getDocument(request)
elif ('bucketname' in event['queryStringParameters']
and 'objectname' in event['queryStringParameters']):
request["bucketName"] = event['queryStringParameters'][
'bucketname']
request["objectName"] = event['queryStringParameters'][
'objectname']
result = createDocument(request)
elif (event['httpMethod'] == 'POST'):
body = json.loads(event['body'])
if ('objects' in body):
results = []
for obj in body['objects']:
request["bucketName"] = sampleBucket if obj[
'sample'] else documentBucket
request["objectName"] = obj['key']
results.append(createDocument(request))
result = results
else:
request["bucketName"] = sampleBucket if body[
'sample'] else documentBucket
request["objectName"] = body['key']
result = createDocument(request)
elif (event['httpMethod'] == 'DELETE'):
if ('documentid' in event['queryStringParameters']):
request["documentId"] = event['queryStringParameters'][
'documentid']
result = deleteDocument(request)
deleteESItem(request["elasticsearchDomain"],
request["documentId"])
elif (event['resource'] == '/redact'):
params = event['queryStringParameters'] if 'queryStringParameters' in event else {}
request["params"] = params
result = redact(request)
return {
"isBase64Encoded": False,
"statusCode": 200,
'body': json.dumps(result),
"headers": {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Headers': '*',
'Access-Control-Allow-Methods': 'GET,POST,OPTIONS'
}
}
def lambda_handler(event, context):
eventCopy = event
print("Redacted Event: {}".format(redactHeadersFromLambdaEvent(event)))
result = {}
status_code = 200
documentBucket = os.environ['CONTENT_BUCKET']
sampleBucket = os.environ['SAMPLE_BUCKET']
if ('resource' in event):
request = {}
request["elasticsearchDomain"] = os.environ['ES_DOMAIN']
request["outputTable"] = os.environ['OUTPUT_TABLE']
request["documentsTable"] = os.environ['DOCUMENTS_TABLE']
# search Elasticsearch handler
if (event['resource'] == '/search'):
if ('queryStringParameters' in event
and 'k' in event['queryStringParameters']):
request["keyword"] = event['queryStringParameters']['k']
if ('documentId' in event['queryStringParameters']):
request["documentId"] = event['queryStringParameters'][
'documentId']
result = search(request)
# search Kendra if available
elif (event['resource'] == '/searchkendra'
and event['httpMethod'] == 'POST'):
if 'KENDRA_INDEX_ID' in os.environ:
kendraClient = KendraHelper()
result = kendraClient.search(os.environ['KENDRA_INDEX_ID'],
event['body'])
# Kendra search result feedback for relevance boosting
elif (event['resource'] == '/feedbackkendra'
and event['httpMethod'] == 'POST'):
if 'KENDRA_INDEX_ID' in os.environ:
kendraClient = KendraHelper()
result = kendraClient.submitFeedback(
os.environ['KENDRA_INDEX_ID'], event['body'])
elif (event['resource'] == '/documents'):
if ('queryStringParameters' in event
and event['queryStringParameters']
and 'nexttoken' in event['queryStringParameters']):
request["nextToken"] = event['queryStringParameters'][
'nexttoken']
if validate_get_documents_request(request):
result = getDocuments(request)
else:
status_code = 400
result.append("Bad request, nexttoken is not valid")
elif (event['resource'] == '/document'):
if (event['httpMethod'] == 'GET'):
if ('queryStringParameters' in event
and event['queryStringParameters']):
if ('documentid' in event['queryStringParameters']):
request["documentId"] = event['queryStringParameters'][
'documentid']
if validate_get_document_request(request):
result = getDocument(request)
else:
status_code = 400
result.append(
"Bad request, documentId is not valid")
elif ('bucketname' in event['queryStringParameters']
and 'objectname' in event['queryStringParameters']):
request["bucketName"] = event['queryStringParameters'][
'bucketname']
request["objectName"] = event['queryStringParameters'][
'objectname']
if validate_create_document_request(request):
result = createDocument(request)
else:
status_code = 400
result.append(
"Bad request, object key does not exist in bucket"
)
elif (event['httpMethod'] == 'POST'):
body = json.loads(event['body'])
if ('objects' in body):
results = []
for obj in body['objects']:
request["bucketName"] = sampleBucket if obj[
'sample'] else documentBucket
request["objectName"] = obj['key']
successes = failures = 0
if validate_create_document_request(request):
results.append(createDocument(request))
successes += 1
else:
result.append(
f"Object key {request['objectName']} does not exist in bucket"
)
failures += 1
status_code = 207 if successes > 0 and failures > 0 else status_code
status_code = 400 if successes == 0 and failures > 0 else status_code
result = results
else:
request["bucketName"] = sampleBucket if body[
'sample'] else documentBucket
request["objectName"] = body['key']
if validate_create_document_request(request):
result = createDocument(request)
else:
status_code = 400
result.append(
"Bad request, object key does not exist in bucket")
elif (event['httpMethod'] == 'DELETE'):
if ('documentid' in event['queryStringParameters']):
request["documentId"] = event['queryStringParameters'][
'documentid']
result = deleteDocument(request)
deleteESItem(request["elasticsearchDomain"],
request["documentId"])
# remove it from Kendra's index too if present
if 'KENDRA_INDEX_ID' in os.environ:
kendraClient = KendraHelper()
kendraClient.deindexDocument(
os.environ['KENDRA_INDEX_ID'],
request["documentId"])
elif (event['resource'] == '/redact'):
params = event['queryStringParameters'] if 'queryStringParameters' in event else {
}
request["params"] = params
result = redact(request)
return {
"isBase64Encoded": False,
"statusCode": 200,
'body': json.dumps(result),
"headers": {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Headers': '*',
'Access-Control-Allow-Methods': 'GET,POST,OPTIONS'
}
}