-
Notifications
You must be signed in to change notification settings - Fork 12.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
AsRef and related conversions for CString #30616
Conversation
r? @brson (rust_highfive has picked a reviewer for you, use r? to override) |
cc @rust-lang/libs |
Yes. |
@@ -522,6 +522,37 @@ impl ToOwned for CStr { | |||
} | |||
} | |||
|
|||
#[unstable(feature = "cstring_asref", reason = "recently added", issue = "0")] | |||
impl ops::Index<ops::RangeFull> for CString { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Probably should implement indexing and slicing with other kind of ranges as well then. Seems strange to only support slicing by RangeFull. Also, these should probably be implemented on CStr and relied on Deref to be found?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OsString does the same thing, though that's intentionally an opaque type so the situation is slightly different in that slicing it is actually "impossible"...
I'm iffy on whether we should allow slicing into a CStr, just because it doesn't seem like a very common operation. It's also not necessarily clear whether you actually want to slice as_bytes()
or as_bytes_with_nul()
, and &cstr[..1]
is impossible to implement without panicking if it doesn't cover the entire string. It also hides the cost of the strlen
which is supposed to be necessary in the future... I'm not entirely opposed to it though!
This impl was meant as a conversion trait (the good old &deref/&index[..]/AsRef/.as_slice()
bunch), the main advantage being that it isn't type-ambiguous. The discussion at #27729 might be a bit relevant there?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Index<RangeTo>
and Index<Range>
can't really work, you would have to insert a 0 byte in the middle of the string. And implementing only Index<RangeFrom>
also seems strange (though potentially useful).
LGTM. Regarding the |
Pushed up the non-generic |
Sorry, one other thing -- can you remove the |
@aturon yeah wasn't sure about those... All the other trait impls have |
@arcnmx Oh hm, it's possible that the stability system requires you to have some marker, even though it's effectively ignored for impls. So yes, mark as stable, version 1.7.0. Thanks! |
@aturon how does that look? |
Looks great, thanks! @bors: r+ rollup |
📌 Commit 53878e7 has been approved by |
Are trait impls still insta-stable? Considering that this design has been around for a long time on `String` and `OsString` it probably doesn't matter much... The `From` impl is a bit strange to me. It's stolen from `OsString` but I'm not really sure about it... `String` just impls `From<&str>` instead, would that make more sense?
Are trait impls still insta-stable? Considering that this design has been around for a long time on
String
andOsString
it probably doesn't matter much...The
From
impl is a bit strange to me. It's stolen fromOsString
but I'm not really sure about it...String
just implsFrom<&str>
instead, would that make more sense?