Perfomance optimization - adding transients

[photogenic89]

Following up on the conversation in issue #74 , I added transients to your plugin. It is now not necessary anymore to loop through all cart sessions but only to consult the value of this transient.

Once a product is added to the cart, the transient is set. It will be updated on any quantity changes. The transient is removed after the cart item expires or when the order is generated from the cart items.
Expiration time is calculated in seconds based on what users set up as an expiration time. If Expire Items is unchecked, expiration time is at 0, thus without any.

I did some basic testing, nothing extensive.

Possible issues:

• You will find inside the method get_virtual_stock_available the WC method get_cart_item_quantities(). It can't be called in every scenario, I therefor added an !is_admin check. Not sure if this brings out faulty scenarios anywhere, it looked all fine on my end.
• If expiration time is set to 0, the database might be filled with a lot of transients. In my case it's not important to me but other users might need a new option to flush the transients? Maybe when carts are manually cleared in the WooCommerce options?

Looking forward to hear your thoughts on this.

[photogenic89]

Hej James,
have you had any time to look at this? It would be interesting for me to see what you think about the code.
Best
Jan

[jamesgol]

Sorry, other than a quick glance I have not had a chance yet. I'm hoping to have a chance sometime soon, but also build out some testing so the plugins functionality (and speed) can be compared.

[photogenic89]

Sounds great! I'm not really sure how to test code speed myself, so your research would help a lot.

[photogenic89]

Did you have a look at this pull request as well? If yes, was there any positive effect on speed? I can see you improved speed on your end with the newest version by unserializing session data only once.

[jamesgol]

Yes I did look over this and it's not the way I want to go with it, I only want to trust the WooCommerce count and don't want to be doing any additional math. I do have something like this mostly implemented but need to finish up a few things before I commit it. In my test with the added caching there was essentially no difference in speed between having 10 carts/sessions and having 15,000 (until it ran out of memory)