You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
node-s3 is a simple web application for creating and manipulating buckets and files on Amazon Web Services S3. With slight modifications, you can also use it to proxy file uploads directly to AWS.
Just edit `config.jsq and place your own configuration values into each of the possible environment variables.
var dev = {
ports:[8000],
s3:{
key: "somekeyfordevelopment",
secret: "somesecretfordevelopment",
reviewer: "bastard@samurai.com",
upload_directory:'./tmp/' // relative to the app root
}
}
var prod = {
ports:[8000],
s3:{
key: "somekeyforproduction",
secret: "somesecretforproduction",
reviewer: "bastard@samurai.com",
upload_directory:'./tmp/'
}
}
If you need to create additional environment variable options, then remember to modify the config#init function accordingly. The port value is an array so you can define multiple ports to listen on. When the application starts, it will create a new app instance per port.
Fire It Up, Man
To fire up the application on each of the designated port options:
node app.js --env=[env option (defaults to dev)]
Upload Options
/upload (parses and loads the uploaded file in memory before putting to s3)
/stream_upload (parses and streams to s3)
/save_then_stream_upload (stores in memory, writes to disk, and then streams the file to s3.)
/stream_save_stream_upload (streams the file to disk and then streas the file to s3)
Notes
Chunked encoding isn't supported by s3.
I experience the following error when trying to stream large files (>10MB) to s3. Not exactly sure what the problem is.
TypeError: Cannot call method 'flush' of undefined
at Client. (http:514:20)
at node.js:810:9