awslabs/chalice

Hello Everyone,

I am looking to extend a current application that is build on Chalice to serve its Web Assets through the AWS API Gateway.

Currently the Web Assets are in an S3 bucket.
I have been able to get a function going that will serve the index.html page through the API Gateway which is a great start.

But what I am running into is trouble figuring out how to match all of the possible files in the S3 bucket.

I.E. pseudo code

WORKS

www.example.com/ui -> API Gateway ( /ui )

@app.route('/ui')
def ui_index():
file_body = get_file_from_s3('index.html')
return file_body

WORKS

www.example.com/ui/script.js -> API Gateway ( /ui/{path} )

@app.route('/ui/{path}')
def ui_path(path):
file_body = get_file_from_s3(path)
return file_body

FAILS

www.example.com/ui/images/logo.png -> API Gateway ( /ui/{path} )

@app.route('/ui/{path}')
def ui_path(path):
file_body = get_file_from_s3(path)
return file_body

FAILS

www.example.com/ui/really/long/path/into/random/library/app.js -> API Gateway ( /ui/{path} )

@app.route('/ui/{path}')
def ui_path(path):
file_body = get_file_from_s3(path)
return file_body

Looking at the AWS API Gateway docs there is a proxy feature that allows for a 'greedy' api resource {proxy+}
https://docs.aws.amazon.com/apigateway/latest/developerguide/api-gateway-set-up-simple-proxy.html
See: HTTP proxy integration with a proxy resource

The above seems to half work when I configure the following in the Chalice App

www.example.com/ui/really/long/path/into/random/library/app.js -> API Gateway ( /ui/{path+} )

@app.route('/ui/{path+}')
def ui_path(path):
file_body = get_file_from_s3(path)
return file_body

But obviously path+ is not a valid python variable name so I can't have that as a parameter for the function.

Does anyone know if we have access to the whole requested URL inside of the function?
I was thinking that could be a good workaround to this problem