nateklaiber

API

Here are the concerns:

• Routes: How do I get to the necessary resource?
• Configuration: What options do I want for my connection or requests
• Connection: How do I connect to the application? Typically this will be HTTP(S).
• Logging: We will have specific logging for the api, the requests, and applications.
• Errors: We will have specific exceptions we raise if any issues arise.
• Requests: These use the routes and connections to make a request to the service. The return from this is a response.
• Response: The response from a request is typically a schema that aligns with content negotiation (Content-Type or Accept)

nateklaiber

SDK Architecture

Rough Draft

This is still in rough draft form. I will continue to make edits as I receive them. I will collate the list of feedback here and make notes of the changelog as revisions are made.

What does it solve?

String Interpolation

No string interpolation for building your routes to your resources.

nateklaiber

Keybase proof

I hereby claim:

• I am nateklaiber on github.
• I am nateklaiber (https://keybase.io/nateklaiber) on keybase.
• I have a public key whose fingerprint is 6930 A573 358D 135A 2269 F9CC 7B5D D6D8 F811 B831

To claim this, I am signing this object:

nateklaiber

class FlatArray
  instance_methods.each do |m|
    undef_method(m) unless m =~ /(^__|^nil\?|^send$|^object_id$)/
  end

  def initialize(array)
    @target = array
  end

  def respond_to?(symbol, include_priv=false)

nateklaiber

API Client Design

When working with API clients there are several distinct areas of concern. Each of these happen in sequence and are used with the goal of receiving a Domain Model at the end.

• Routing
• Configuration
• Logging
• Connection Handling
• Error Handling
• Request Building

nateklaiber

t = 236 # seconds
Time.at(t).utc.strftime("%H:%M:%S")
=> "00:03:56"

# Reference
# http://stackoverflow.com/questions/3963930/ruby-rails-how-to-convert-seconds-to-time

nateklaiber

cat *.json |\
    jq 'map(select(.from.name=="Nate Klaiber"))' |\
    jq '.[].message' |\
    grep -v 'jq' |\
    sed -e 's/\([[:punct:]]\)//g' |\
    grep -v '^$' |\
    tr '\n' ' ' |\
    tr ' ' '\n' |\
    tr '[:upper:]' '[:lower:]' |\
    sort |\

nateklaiber

API_HOST='http://utilities.theklaibers.com'
CACHE_HOST='localhost:11211'

nateklaiber

I'll use Stripe as an example. The client is responsible for:

• Configuration
• Routing
• Connection Handling
• Request Handling
  • Raw requests with JSON
  • Object model requests that wrap the raw request (allow us to say Stripe::Request::Charge.get(id: 'sdfsdf'))
• Object Creation. This includes top level and associations.

nateklaiber

Given an object has an owner association (relation), when expanded by default it will be exposed as:

{
  "name": "service-production",
  "owner": {
    "id": "5d8201b0...",
    "name": "Alice",
    "email": "alice@heroku.com"
 },