10 releases (4 breaking)
0.5.1 | Feb 24, 2024 |
---|---|
0.5.0 | Feb 24, 2024 |
0.4.1 | Dec 27, 2023 |
0.3.0 | Oct 2, 2023 |
0.1.1 | Aug 19, 2023 |
#153 in Algorithms
29,334 downloads per month
Used in 2 crates
77KB
1.5K
SLoC
byteyarn
byteyarn
- Space-efficient byte strings π§Άπββ¬
A Yarn
is a highly optimized string type that provides a number of
useful properties over String
:
- Always two pointers wide, so it is always passed into and out of functions in registers.
- Small string optimization (SSO) up to 15 bytes on 64-bit architectures.
- Can be either an owned buffer or a borrowed buffer (like
Cow<str>
). - Can be upcast to
'static
lifetime if it was constructed from a known-static string. Option<Yarn>
has the same size and ABI asYarn
.
The main caveat is that Yarn
s cannot be easily appended to, since they
do not track an internal capacity, and the slice returned by
Yarn::as_slice()
does not have the same pointer stability properties as
String
(these are rarely needed, though).
Yarns are useful for situations in which a copy-on-write string is necessary
and most of the strings are relatively small. Although Yarn
itself is
not Copy
, there is a separate YarnRef
type that is. These types
have equivalent representations, and can be cheaply cast between each other.
The easiest way to create a yarn is with the yarn!()
macro, which is similar to format!()
.
// Create a new yarn via `fmt`ing.
let yarn = yarn!("Answer: {}", 42);
// Convert that yarn into a reference.
let ry: YarnRef<str> = yarn.as_ref();
// Try up-casting the yarn into an "immortal yarn" without copying.
let copy: YarnRef<'static, str> = ry.immortalize().unwrap();
assert_eq!(yarn, copy);
Yarns are intended for storing text, either as UTF-8 or as
probably-UTF-8 bytes; Yarn<str>
and Yarn<u8>
serve these purposes,
and can be inter-converted with each other. The Yarn::utf8_chunks()
function can be used to iterate over definitely-valid-UTF-8 chunks within
a string.
Both kinds of yarns can be Debug
ed and Display
ed, and will print out as
strings would. In particular, invalid UTF-8 is converted into either \xNN
escapes or replacement characters (for Debug
and Display
respectively).
let invalid = ByteYarn::from_byte(0xff);
assert_eq!(format!("{invalid:?}"), r#""\xFF""#);
assert_eq!(format!("{invalid}"), "οΏ½");
Dependencies
~0.6β1MB
~12K SLoC