Module: Shydra

Defined in:
lib/shydra/hydra.rb,
lib/shydra.rb,
lib/shydra/batch.rb,
lib/shydra/request.rb,
lib/shydra/version.rb,
lib/shydra/resource.rb,
lib/shydra/response.rb,
lib/shydra/resources/product.rb,
lib/shydra/resources/variant.rb,
lib/shydra/resources/metafield.rb

Overview

Scenarios:

1. request that might return more than 250
  - limit: :none
  - adds callback that fetches another page if result size == SHOPIFY_API_MAX_LIMIT

2. request that will return several pages that needs to be fast
- use a hydra
- does count first
- queues request for each page, callback inserts into correct slot in result_array
- if too many pages for api limit, fetch as many pages as possible, then sleep for some amount of time
- time to sleep: requests-used-before-starting / API-LIMIT
- at the end flatten the result_array 

API LIMIT
- memcached
- cache {used: x, at: Time.now}
- before starting hydra:
  - count requests
  - check cache
  - if at < TUNEABLE_PARAM (like, 20 seconds) & enough requests (including TUNEABLE_BUFFER)
    - run everything
  - do_run: run 1 request. update cache
    - if enough requests, run everything
    - else bail or sleep(time allowed)
    - after sleeping time allowed, goto: do_run
  - because requests may have callbacks that queue other requests, the run_loop has to go through this alogrithm every time it pulls a new batch of requests off

Defined Under Namespace

Modules: Resources, Response Classes: Batch, Hydra, Request, Resource

Constant Summary collapse

VERSION = 
"0.4.2"

Class Attribute Summary collapse

Class Attribute Details

.max_concurrency ⇒ Object

Returns the value of attribute max_concurrency.



15
16
17
 
# File 'lib/shydra.rb', line 15

def max_concurrency
  @max_concurrency
end