So I have major reservations with API V2 in regards to getting all episode data for a series.
Speed of getting complete episode information for all episodes associated with a series is horrendous, and amounts to serious performance issues versus current (but to be depreciated) api V1.
Here's my example.
User wishes to populate their database with all the episodes of say CSI: Crime Scene Investigation
There are 337 episodes.
In API V1, we would download the series XML file (for this volume of episode) in under 10 seconds. One API request.
This has complete Series information, as well as complete episode information.
We then pass this file locally, populating the episode information. And make One download request per episode matched to download the episode thumbnail.
Total 1 api requests, and 337 image calls.
In API V2,
We query the series: 1 api call.
Populate the episodes. Now here the episodes are given in pages of 100 episodes, that's four api calls. Total time approx 3 seconds.
Then we need to populate all episode details. 337 api calls. (each api call is taking just under half a second) Total time approx 98 seconds.
And download the episode thumbnails. 337 image calls
Total 342 api calls, and 337 image calls.
My questions is: If we populate the series with episodes:
can we not specify to load all episode details at that time, instead of having to call
In regards to performance, my application can probably be improved to cut down the time.
But, 1 api call, over 342 can not be better on your server.
The above example for CSI was actually performed by myself, and while populating a whole series is not a regular occurrence, it does happen. And whole seasons would be quite often. Followed by single episode very often.