macro_rules! bitvec {
($order:ty, $store:ty; $val:expr; $len:expr) => { ... };
($val:expr; $len:expr) => { ... };
($($arg:tt)*) => { ... };
}
Expand description
Constructs a new BitVec
from a bit-pattern description.
This macro takes a superset of the vec!
argument syntax: it may be invoked
with either a sequence of bit expressions, or a single bit expression and a
repetition counter. Additionally, you may provide the names of a BitOrder
and a BitStore
implementor as the BitVec
’s type arguments.
Argument Rules
Bit expressions must be integer literals. Ambiguity restrictions in the macro
syntax forbid the use of identifiers to existing variables, even const
values.
These are converted to bool
through the expression $val != 0
. Any non-zero
enteger becomes true
, and 0
becomes false
.
You may use any name or path to a BitOrder
implementation. However, the
identifier tokens Lsb0
, Msb0
, and LocalBits
are matched directly and
specialized to have compile-time constructions, whereäs any other name or path
will not be known to the macro, and will execute at runtime.
The BitStore
argument must be the name of an unsigned integer
fundamental, an atomic, or a Cell<>
wrapper of that unsigned integer. These
are matched by token, not by type, and no other identifier as accepted. Using
any other token will cause the macro to fail.
Examples
use bitvec::prelude::*;
use core::cell::Cell;
radium::if_atomic! { if atomic(32) {
use core::sync::atomic::AtomicU32;
} }
let a: BitVec = bitvec![0, 1, 0, 1, 2];
assert_eq!(a.count_ones(), 3);
let b: BitVec = bitvec![2; 5];
assert!(b.all());
assert_eq!(b.len(), 5);
let c = bitvec![Lsb0, Cell<u16>; 0, 1, 0, 0, 1];
let d = bitvec![Msb0, AtomicU32; 0, 0, 1, 0, 1];