So you connect with SSH and run Node scripts since you can use HTTP ?
Regards,
Richard Schoen
Director of Document Management
e. richard.schoen@xxxxxxxxxxxxxxx
p. 952.486.6802
w. helpsystems.com
----------------------------------------------------------------------
message: 1
date: Tue, 20 Mar 2018 09:55:12 -0500
from: Bradley Stone <bvstone@xxxxxxxxx>
subject: Re: [WEB400] [EXTERNAL] Re: Rise of Node
On Tue, Mar 20, 2018 at 9:23 AM, Nathan Andelin <nandelin@xxxxxxxxx> wrote:
In the case of reading files from the local file system, Node.js
async-await adds 10-25 times to complete the read cycle in comparison
to synchronous processes written in Python. Imagine how much longer
that would take if a batch process were working against a remote
database which entails not only reading from a file system, but add to
that the requirements of inter-system communications.
To me, that would be a strong case for using RPG to implement batch
processes against IBM i databases as opposed to force-fitting that
into a Node.js service.
I agree, and in my case I normally would use RPG if it's an option. But, sometimes it's not.
Example, as Aaron knows I am working on a project for a customer that uses Braintree (which I will hold my tongue about their sandbox.. contact me
offline for a good laugh). He is too! A different customer though.
Anyhow, Braintree doesn't allow traditionally HTTP requests. They only allow use of their SDKs, which are available in a host of languages, node being one of them, and the one I chose to use (I love JS...)
Sure, I may be able to spend days parsing their requests, finding the endpoints used in the SDKs, etc, and use RPG, but I decided to use this as a real POC project for node. All my others weren't real production, just samples.
http://nodejs.bvstools.com:5250/
My customer really wanted me to use RPG (GETURI, YAJL, etc). Well, I couldn't in this case.
So while there are instances where I wish things were sync, yes, I had to "force" them. I was planning on putting node 8 on for the new promise features but decided for now, I'll leave it as is.
I got the project done (still in testing) and learned a lot in the process, some good, some bad, about the evolution of node and how we can use it on the IBM i in real world cases.
One day when I can collect my thoughts and see how the process actually performs ITRW I shall pen an article about my journey at fieldexit.com.
midrange.com
