trim prefix

[tty2]

trim prefix method:

• cuts n cells from the beginning of the string
• adds prefix to string if set

[mattn]

I don't understand why it's xxx... instead of ...xxx

[tty2]

Cause it's a tail.
To be honest I'm not really sure if tail is necessary in this function. I just wanted to keep similarity with Truncate. But in my case it's not necessary.
Do you mean that it's more obvious to set tail as a prefix? It's easy to implement as well.

[tty2]

Yes. I took a look at it more careful. You are right. It's stupid of me. Let me refactor this.

[tty2]

Done!
Check it again please.

[mattn]

Thanks. I'll look this in later.

[mattn]

Remove _

[mattn]

Thanks your contribution. I prefer the tests should be table-driven tests. See other tests.

[tty2]

Sure. I should have written them similar.
Changed.

[mattn]

Hmm, I'm still wondering what is this use-case. Could you please explain how this can be useful?

[tty2]

:)
Check this PR.
It's one of the way how to implement horizontal scroll.
And the final goal is my personal tui app :) the problem is mentioned in this issue

[tty2]

So instead of using getStringWithIndent function that is written by me in the PR I mentioned above, I'm gonna use this TrimPrefix function that will take care of cell not runes.

[mattn]

So TrimPrefix("あいうえお", 5, "") must be " えお"

あいうえお
Aあいうえお
あいうえお
Aあいうえお

Result of trimming 5 cells should be

 えお
うえお
 えお
うえお

[tty2]

So TrimPrefix("あいうえお", 5, "") must be " えお"

あいうえお
Aあいうえお
あいうえお
Aあいうえお

Result of trimming 5 cells should be

 えお
うえお
 えお
うえお

Nope. TrimPrefix("あいうえお", 5, "") returns "うえお"

And the result of TrimPrefix for this:

あいうえお
Aあいうえお
あいうえお
Aあいうえお

will be the same. I mean "うえお". Because part that we cut is not bigger than 5.

あい:	6 bytes, 2 runes, 4 cells, 2 grapheme clusters
Aあい:	7 bytes, 3 runes, 5 cells, 3 grapheme clusters
あい:	6 bytes, 2 runes, 4 cells, 2 grapheme clusters
Aあい:	7 bytes, 3 runes, 5 cells, 3 grapheme clusters

[tty2]

Or...
You probably wanted to say that there is a way how to improve the logic? Aren't you?
Add one empty cell if there is a difference between w and pos == 1.

If so, on line 235 inside if condition we can change logic to:

...
	if width+chWidth > w {
		if width < w {
			_, pos = g.Positions()
			prefix += strings.Repeat(" ", w-width)
		} else {
			pos, _ = g.Positions()
		}

		break
	}
...

[tty2]

I've pushed the implementation with logic from above comment.
And I decided to change the function name to TruncateLeft. It's more obvious, especially because there is a Truncate function.
Another reason is that there is a function in standard library strings.TrimPrefix that cut string relying on prefix (if it exists). For users this name similarity can play a bad game.
WDYT?

[mattn]

Look good to me.

[mattn]

Thank you

[tty2]

Thank you too