Allow specification of starting index in str::find

[xales]

The find family of string functions currently do not have a way of specifying the starting index. This makes it difficult to find the nth occurrence of something.

cc @cmr

[huonw]

The API is designed to be composable: string.slice_from(n).find(...).

Is that what you are describing?

(I think the .match_offsets (can't remember the exact name) iterator is probably a better way of iterating through the matches of a string, too.)

[xales]

Yes. In particular

let y = x.find('"').unwrap()+1; x.slice(y, y + x.slice_from(y+1).find('"').unwrap()+1)

(intended to find string between first two double-quote characters)
could be simplified as

let y = x.find('"').unwrap()+1; x.slice(y, x.find('"', y).unwrap())

While it's possible to do using composition, it feels somewhat clumsy, and using an iterator probably overkill for small matches like this.

[huonw]

I'm personally nervous about adding more arguments for things like this; e.g. why don't we have an argument counting how many matches should be ignored; and the ending index, and then the same for the reverse versions.

This specific case may be fixed by something like adding a string.offsets(char) which returns Iterator<uint> where the uints are the positions of the found codepoints:

let mut it = x.offsets('"');
x.slice(it.next().unwrap(), it.next().unwrap() + 1)

That said, this sort of stuff is not great Unicode-practice.

[emberian]

I was just thinking a find_from. But, it is really awful Unicode hygiene...
I hate strings.

On Sun, Feb 2, 2014 at 7:29 AM, Huon Wilson notifications@github.202132.xyzwrote:

I'm personally nervous about adding more arguments for things like this;
e.g. why don't we have an argument counting how many matches should be
ignored; and the ending index, and then the same for the reverse versions.

This specific case may be fixed by something like adding a
string.offsets(char) which returns Iterator where the uints are the
positions of the found codepoints:

let mut it = x.offsets('"');x.slice(it.next().unwrap(), it.next().unwrap() + 1)

That said, this sort of stuff is not great Unicode-practice.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/11986#issuecomment-33899182
.

[treeman]

Triage bump. Needs an RFC as find is reachable by the prelude?

[steveklabnik]

This is bad unicode practice, and changes to the libraries will need an RFC, so yes, closing.

[stej]

I tried to use the proposed solution, but it looks that the API changed significantly. What is the current way how to find a string starting from given index?