A possible solution:
withUnsafeMutablePointer(&myVarOfStructProduct.name) {
strlcpy(UnsafeMutablePointer($0), productName, UInt(sizeofValue(myVarOfStructProduct.name)))
}
Inside the block, $0
is a (mutable) pointer to the tuple. This pointer is
converted to an UnsafeMutablePointer<Int8>
as expected by the
BSD library function strlcpy()
.
It also uses the fact that the Swift string productName
is automatically
to UnsafePointer<UInt8>
as explained in String value to UnsafePointer<UInt8> function parameter behavior. As mentioned in the comments in that
thread, this is done by creating a temporary UInt8
array (or sequence?).
So alternatively you could enumerate the UTF-8 bytes explicitly and put them
into the destination:
withUnsafeMutablePointer(&myVarOfStructProduct.name) {
tuplePtr -> Void in
var uint8Ptr = UnsafeMutablePointer<UInt8>(tuplePtr)
let size = sizeofValue(myVarOfStructProduct.name)
var idx = 0
if size == 0 { return } // C array has zero length.
for u in productName.utf8 {
if idx == size - 1 { break }
uint8Ptr[idx++] = u
}
uint8Ptr[idx] = 0 // NUL-terminate the C string in the array.
}
Yet another possible solution (with an intermediate NSData
object):
withUnsafeMutablePointer(&myVarOfStructProduct.name) {
tuplePtr -> Void in
let tmp = productName + String(UnicodeScalar(0)) // Add NUL-termination
let data = tmp.dataUsingEncoding(NSUTF8StringEncoding, allowLossyConversion: true)!
data.getBytes(tuplePtr, length: sizeofValue(myVarOfStructProduct.name))
}
Update for Swift 3:
withUnsafeMutablePointer(to: &myVarOfStructProduct.name) {
$0.withMemoryRebound(to: Int8.self, capacity: MemoryLayout.size(ofValue: myVarOfStructProduct.name)) {
_ = strlcpy($0, productName, MemoryLayout.size(ofValue: myVarOfStructProduct.name))
}
}
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…