Added String.count method

[Chaosus]

This method returns the amount of identical substrings in string, Its particularly useful for text parsing purposes + there is simular function in Python API and in the Array class of Godot

example

	print(str("".count("Test"))) # 0
	print(str("Test".count(""))) # 0
	print(str("Test".count("test"))) # 0
	print(str("Test".count("TEST"))) # 0
	print(str("TEST".count("TEST"))) # 1
	print(str("Test".count("Test"))) # 1
	print(str("aTest".count("Test"))) # 1
	print(str("Testa".count("Test"))) # 1
	print(str("TestTestTest".count("Test"))) # 3
	print(str("TestTestTest".count("TestTest"))) # 1
	print(str("TestGodotTestGodotTestGodot".count("Test"))) # 3
	
	print(str("TestTestTestTest".count("Test", 4, 8))) # 1
	print(str("TestTestTestTest".count("Test", 4, 12))) # 2
	print(str("TestTestTestTest".count("Test", 4, 16))) # 3
	print(str("TestTestTestTest".count("Test", 4))) # 3

Fix #27749

[escargot-sans-gluten]

Shouldn't "TestTestTest".count("TestTest") return 2? ("TestTest Test" and "Test TestTest") Is this behavior unwanted?

[Chaosus]

Shouldn't "TestTestTest".count("TestTest") return 2?

I think you should use a regular expression for that kind of precision :) However you right - TestTest exist only one time if you looks on it from the one side, but 2 times from another

[Chaosus]

Also, In Python it will be 1, you can test it on https://repl.it/languages/python3

s = "TestTestTest"
print(s.count("TestTest",0,12))

[hpvb]

I agree that returning overlapping results would not be what would be expected in most cases.

[aaronfranke]

This should have a corresponding case-insensitive countn similar to matchn, replacen, and rfindn

[Chaosus]

Added countn

[bojidar-bg]

I think the function should support negative indices if possible. Also, default value for p_from should be 0, not -1.

[Chaosus]

I think the function should support negative indices if possible.

I dont think it really need for count + it will complicate the code. rcount and rcountn could be implemented for that in the future, but for now I want to implement it in simple form.

Also, default value for p_from should be 0, not -1.

Changed

[Chaosus]

I guess I need to write a C# version for this.

[Chaosus]

Added C# version, cc @neikeq

[aaronfranke]

Avoid double negatives. Instead of "ignoreCase", it should be "caseSensitive". Do a Ctrl+F in this file for bool caseSensitive = true to see examples of this.

[akien-mga]

It's pretty common for case-sensitive-by-default APIs to have a flag for ignore case, not for enable case sensitivity (which is default). See e.g. any kind of grepping tool with -i flag.

ignoreCase = false is not a double negative, dontIgnoreCase = true would be.

That being said, if other extensions in this case favor caseSensitive = true, I'm not against that.

[Chaosus]

Well, I changed to caseSensitive because the other functions in that file is using it

[aaronfranke]

IMO "ignore" is itself a negative, because it means "don't care about it", so ignoreCase = false would be "don't not care about case".

In C#, we specify in the method declaration which one is the default with true or false, so we don't need to depend on a flag existing or not - whichever makes the most readable code is best.

However, if this was changed to ignoreCase, it should be done for all of the other methods too.

[akien-mga]

Code seems good to me so it could be merged, but I think it should likely be added to GDNative too? If so, I believe it would have to be added to the 1.2 extension (CC @karroffel to confirm).

[karroffel]

This could be added to GDNative, yes. If there is an issue about things that should be added to the 1.2 core extension then it could be mentioned there and doesn't need to block this PR.

[Chaosus]

@karroffel last time when I tried to push to GDNative - #27452 you said it will break the binary compatibility.. what exactly I must do ?

[karroffel]

@karroffel last time when I tried to push to GDNative - #27452 you said it will break the binary compatibility.. what exactly I must do ?

The problem was that new code was added to a stabilized extension. Right now the GDNative core extension 1.2 (I think?) is not stabilized yet, so the code could be added there.

So for example the 1.0 GDNative core must not be touched, but 1.2 can be touched as nobody is expected to already depend on it.

[Chaosus]

Ok, so I've added a GDNative code similar to what I did in #27452.. Please check @karroffel

[karroffel]

Not quite :) The new things should be added here: https://github.com/godotengine/godot/blob/89dfcbc6f7f57db4cd8ce3daa10043b40c0b5231/modules/gdnative/gdnative_api.json#L47

[Chaosus]

And now ?

[karroffel]

GDNative seems okay!

[Chaosus]

I've fixed documentation a bit - I guess mention "default" for parameters is not neccessary now (these values is clearly visible in editor docs).

[akien-mga]

Thanks!