AGS 4: Script API: add "split options" for String.Split method

[ivan-mogilko]

I propose to add optional "split options" as a set of flags for the String.Split methods, equivalent or similar to the C# library function: https://learn.microsoft.com/en-us/dotnet/api/system.stringsplitoptions?view=net-8.0

Something like

enum StringSplitOptions {
    eStrSplit_None = 0,
    eStrSplit_RemoveEmpty = 0x0001,
    eStrSplit_Trim = 0x0002
};

import String[] Split(const string separator, StringSplitOptions options = eStrSplit_None);

[AlanDrake]

I agree, it's a good convenience for several cases.

[ericoporto]

It misses the actual flags, but I believe the gist of the changes would be something like this

https://github.com/ericoporto/ags/commit/de5d29cd1d2f339dcaa8ca75336fe98b933a2e12

Not sure if there is a pretty way to get the trim in a single function.

[ericoporto]

Uhm, I wonder if a string like " , , , ," is passed with both flags, what the result should be, would it be null or does an empty array exists in AGS script?

[ivan-mogilko]

The trim implementation from above commit looks fine, similar to what I've imagined. (I'd only use camelcase function names similar to all names around, according to our code style)

Uhm, I wonder if a string like " , , , ," is passed with both flags, what the result should be, would it be null or does an empty array exists in AGS script?

AGS does not currently allow creating a zero-length dynamic array in script (with the new operator). But if this is a matter of convenience, we may investigate if it's possible to support such array, and return one. In fact, maybe it's already technically possible.

[ericoporto]

I put a PR in #2461, please test it!

[ivan-mogilko]

Resolved by #2461