def store(event, context):
"""store file"""
# check the store on load configuration
if runtime_context.STORE:
return {
'statusCode':
400,
'body':
json.dumps({
'message':
'Configured to store files at the time of upload.'
})
}
# get the list of files from the request
file_ids = json.loads(event.get('body'))
file_ids = file_ids[:runtime_context.
QUERY_LIMIT] # limit the number of files to store
stored_file_ids = []
dt = datetime.utcnow()
for file_id in file_ids:
FileModel.update({
'id': file_id,
'stored_at': dt,
})
LOGGER.debug(
'Files item updated (stored). service=ddb method=update_item id={}'
.format(file_id))
stored_file_ids.append(file_id)
return {"statusCode": 200, "body": json.dumps(stored_file_ids)}
def is_long(self, event):
"""determine if the document is too long and needs summarizing"""
# NOTE: this is a mock implementation
import string
val = (string.ascii_lowercase + string.digits).find(self.id[0]) < 18
LOGGER.debug('DocumentReviewMachine.is_long = {}'.format(val))
return val
def info(event, context):
"""get the info for a list of files"""
query_string_parameters = event.get('queryStringParameters') or {}
include_deleted = query_string_parameters.get('deleted') == 'yes'
include_nonstored = query_string_parameters.get('nonstored') == 'yes'
LOGGER.debug('File info includes deleted={} includes non-stored={}'.format(
include_deleted, include_nonstored))
file_ids = json.loads(event.get('body'))
file_ids = file_ids[:runtime_context.
QUERY_LIMIT] # limit the number of files queried
files = FileModel.get_by_ids(file_ids)
for file in files:
if not file.get('deleted_at') and file.get(
'stored_at'): # only return not deleted and stored
file['url'] = get_presigned_url_for_download(file)
else:
if (file.get('deleted_at')
and include_deleted) or (not file.get('stored_at')
and include_nonstored):
pass
else: # remove deleted or non-stored
files.remove(file)
return {
"statusCode": 200,
"body": json.dumps(files, cls=utils.DateTimeEncoder)
}
def enter(self, event_data):
"""Extends `transitions.core.State.enter`
see examples: https://github.com/pytransitions/transitions/blob/master/transitions/extensions/states.py
"""
LOGGER.debug('LambdaCall.enter - call: {}'.format(self.lambda_call))
if self.lambda_call: # process calls if there is any defined
self._process_lambda_call(event_data)
super(LambdaCall, self).enter(event_data)
def delete(event, context):
"""delete files"""
file_ids = json.loads(event.get('body'))
deleted_file_ids = []
for file_id in file_ids:
# NOTE: there is no check if file has already been deleted
FileModel.update({'id': file_id, 'deleted_at': datetime.utcnow()})
LOGGER.debug(
'Files item updated (deleted). service=ddb method=update_item id={}'
.format(file_id))
S3_CLIENT.delete_object(Bucket=runtime_context.BUCKET_NAME,
Key=file_id)
LOGGER.debug(
'S3 object deleted. service=s3 method=delete_object id={}'.format(
file_id))
deleted_file_ids.append(file_id)
return {"statusCode": 200, "body": json.dumps(deleted_file_ids)}
def expire(event, context):
"""remove files that are uploaded, not stored, and older than the expiration time
scheduled event
"""
# scan the database for expired files
expiry_at = datetime.utcnow() - runtime_context.NONSTORED_TIMEOUT
files = FileModel.list_expired(expiry_at)
# remove all files and all items one-by-one
for file in files:
file_id = file['id']['S']
FileModel.update({'id': file_id, 'deleted_at': datetime.utcnow()})
LOGGER.debug(
'Files item updated (expired). service=ddb method=update_item id={}'
.format(file_id))
S3_CLIENT.delete_object(Bucket=runtime_context.BUCKET_NAME,
Key=file_id)
LOGGER.debug(
'S3 object deleted. service=s3 method=delete_object id={}'.format(
file_id))
def get_presigned_url_for_download(file):
"""get presigned url for download"""
url = S3_CLIENT.generate_presigned_url(
ClientMethod='get_object',
Params={
'Bucket':
runtime_context.BUCKET_NAME,
'Key':
file['id'],
'ResponseContentDisposition':
'attachment; filename="{}"'.format(file['name']),
'ResponseContentType':
file['type']
},
ExpiresIn=runtime_context.EXPIRATION)
LOGGER.debug(
'Presigned URL generated. service=s3 method=get_object id={}'.format(
file['id']))
return url
def uploaded(event, context):
"""S3 event triggers when file is uploaded
event: https://docs.aws.amazon.com/AmazonS3/latest/dev/notification-content-structure.html
"""
dt = datetime.utcnow()
# NOTE: the event might include multiple records
for r in event['Records']:
file_id = r['s3']['object']['key']
file = {
'id': file_id,
'size': r['s3']['object']['size'],
'type': get_s3_file_type(file_id),
'uploaded_at': dt,
}
if runtime_context.STORE:
file['stored_at'] = dt
FileModel.update(file)
LOGGER.debug(
'Files item updated (uploaded). service=ddb method=update_item id={}'
.format(file_id))
return {"statusCode": 200}
def get_top_story(event, context):
"""retrieve the top story from HackerNews
https://github.com/HackerNews/API
"""
LOGGER.debug('get_top_story - ENTRY')
top_stories = requests.get(
'https://hacker-news.firebaseio.com/v0/topstories.json')
assert top_stories.status_code == 200, 'Failed to get top stories [{}]'.format(
top_stories.status_code)
top_story_id = top_stories.json()[0]
LOGGER.info('get_top_story - top_story_id={}'.format(top_story_id))
top_story = requests.get(
'https://hacker-news.firebaseio.com/v0/item/{}.json'.format(
top_story_id))
assert top_story.status_code == 200, 'Failed to get the top story [{}]'.format(
top_stories.status_code)
story = top_story.json()
LOGGER.debug('get_top_story - EXIT')
return {
'statusCode':
200,
'body':
'"{}" by {} on {}'.format(story.get('title'), story.get('by'),
story.get('url')),
'headers': {
'content-type': 'text/plain'
}
}
def preprocess(event, context):
"""Sample pure Lambda function
event: https://docs.aws.amazon.com/apigateway/latest/developerguide/set-up-lambda-proxy-integrations.html#api-gateway-simple-proxy-for-lambda-input-format
context: https://docs.aws.amazon.com/lambda/latest/dg/python-context-object.html
Returns: https://docs.aws.amazon.com/apigateway/latest/developerguide/set-up-lambda-proxy-integrations.html
"""
# create file in DDB
file_id = utils.generate_id()
file_request = json.loads(event.get('body'))
FileModel.create({'id': file_id, 'name': file_request.get('name')})
LOGGER.debug(
'Files item created. service=ddb method=put_item id={}'.format(
file_id))
# generate signed URL for posting file
url = S3_CLIENT.generate_presigned_url(
ClientMethod='put_object',
Params={
'Bucket': runtime_context.BUCKET_NAME,
'Key': file_id
},
ExpiresIn=runtime_context.EXPIRATION)
LOGGER.debug(
'Presigned URL generated. service=s3 method=put_object id={}'.format(
file_id))
# send back the signed url
return {
"statusCode": 200,
"body": json.dumps({
'id': file_id,
'url': url
}),
# CORS header
"headers": {
"Access-Control-Allow-Origin": "*"
}
}
def notify_reviewer(event, context):
"""notify reviewer function"""
LOGGER.info('Notify reviewer')
LOGGER.info(event.get('body'))
return {'statusCode': 200}
def archive_document(event, context):
"""archive original document function"""
LOGGER.info('Archive original document')
LOGGER.info(event.get('body'))
return {'statusCode': 200}
def delete_document(event, context):
"""delete document function"""
LOGGER.info('Delete document')
LOGGER.info(event.get('body'))
return {'statusCode': 200}
def summarize_document(event, context):
"""summarize document function"""
LOGGER.info('Summarize document')
LOGGER.info(event.get('body'))
return {'statusCode': 200}