GaloreLib

[rose canopy]

Some bonus features (planned/implemented);

• Improved automatic property assigmend; localisation properties, data.RecipePrototype.allow_productivity, data.RecipePrototype.main_product, etc.
• A full procuderal icon library. (tiering, overlays, Galore-like icons)
• A enormous amount of high performance prototype manipulating functions. (Optimised for prototypes made through the vgal.data.extend function.)

And a example.

[wheat willow]

ok so prefixes right

it takes like 10 seconds to write a name function that prefixes things for you

and sure you have to go name("thing") in prototype defs but its easier than typing the prefix every time

[rose canopy]

But luckily you dont have to type the prefix every time, see linked example

[wheat willow]

how does it decide when to add the prefix?

[rose canopy]

And yea while a name function takes a few second to create, it did not take a few second for me to create the inherited prefix and domain system

[rose canopy]

The prefix on the end of the function call?

(from the example)

[wheat willow]

becuse surely youre not automatically prepending it to recipe ingredients etc

[rose canopy]

Nono

Only propotype names that get created in the function call

So the nested prototypes and ofc the main prototype

[normal raptor]

That sounds great. Might try it in a new project

[prisma tulip]

huh how does it do that? modifying the _G table to store a context variable?

Or does Lua have dynamic bindings that i don't know about

[rose canopy]

Just table manipulation. With "Only prototype names that get created in the function call" I meant only the names of prototypes created/to-be-extended in the function call, I should probably have worded that better

[still forge]

When does this preprocessing happen? At the end of your mod’s execution of each data stage? A combined preprocessing at some point in data-final-fixed?

E.g. if my mod using this lib adds a ItemPrototype with a recipe field, is the recipe added to data.raw alongside the item?

You also seem focused on performance a couple of times but I’ve literally never run into a case where performance in data stage is an issue. Even naively iterating through data.raw is completely fine in my experience.

[wheat willow]

during the galorelib extend call

[rose canopy]

When does this preprocessing happen?
In the function call. vgal.data.extend processes the input table and calls data:extend for the contained prototypes.

if my mod using this lib adds a ItemPrototype with a recipe field, is the recipe added to data.raw alongside the item?
Yes.

You also seem focused on performance a couple of times but I've literally never run into a case where performance in data stage is an issue. Even naively iterating through data.raw is completely fine in my experience.
Even though Lua's performance is great, I have seen some performance cost originating from looping through prototypes in combination with other actions. My best example is looping through prototypes and figuring out if they belong to a specific mod. Either through icon paths, trying to trim the prefix (if the mod has one) or other methods. (This is not needed for prototypes added by this library.)

But I must say that this library is more of a QOL lib than one focussed on increasing prototype loading performance, the performance improvements are just a bonus.