Files
c3c/lib/std/core/array.c3
Zack Puhl 702b63ddb7 Add array::zip and Related Macros (#2370)
* zip / zip_into
* Deprecate `add_array` in favour of `push_all` on lists.
* Add support for generic lists for zip.

---------

Co-authored-by: Christoffer Lerno <christoffer@aegik.com>
2025-08-16 13:30:24 +02:00

340 lines
12 KiB
Plaintext

module std::core::array;
import std::core::array::slice;
import std::collections::pair;
<*
Returns true if the array contains at least one element, else false
@param [in] array
@param [in] element
@require @typekind(array) == SLICE || @typekind(array) == ARRAY
@require @typeis(array[0], $typeof(element)) : "array and element must have the same type"
*>
macro bool contains(array, element)
{
foreach (&item : array)
{
if (*item == element) return true;
}
return false;
}
<*
Return the first index of element found in the array, searching from the start.
@param [in] array
@param [in] element
@return "the first index of the element"
@return? NOT_FOUND
*>
macro index_of(array, element)
{
foreach (i, &e : array)
{
if (*e == element) return i;
}
return NOT_FOUND?;
}
<*
Slice a 2d array and create a Slice2d from it.
@param array_ptr : "the pointer to create a slice from"
@param x : "The starting position of the slice x, optional"
@param y : "The starting position of the slice y, optional"
@param xlen : "The length of the slice in x, defaults to the length of the array"
@param ylen : "The length of the slice in y, defaults to the length of the array"
@return "A Slice2d from the array"
@require @typekind(array_ptr) == POINTER
@require @typekind(*array_ptr) == VECTOR || @typekind(*array_ptr) == ARRAY
@require @typekind((*array_ptr)[0]) == VECTOR || @typekind((*array_ptr)[0]) == ARRAY
*>
macro slice2d(array_ptr, x = 0, xlen = 0, y = 0, ylen = 0)
{
if (xlen < 1) xlen = $typeof((*array_ptr)[0]).len + xlen;
if (ylen < 1) ylen = $typeof((*array_ptr)).len + ylen;
var $ElementType = $typeof((*array_ptr)[0][0]);
return (Slice2d{$ElementType}) { ($ElementType*)array_ptr, $typeof((*array_ptr)[0]).len, y, ylen, x, xlen };
}
<*
Return the first index of element found in the array, searching in reverse from the end.
@param [in] array
@param [in] element
@return "the last index of the element"
@return? NOT_FOUND
*>
macro rindex_of(array, element)
{
foreach_r (i, &e : array)
{
if (*e == element) return i;
}
return NOT_FOUND?;
}
<*
Concatenate two arrays or slices, returning a slice containing the concatenation of them.
@param [in] arr1
@param [in] arr2
@param [&inout] allocator : "The allocator to use, default is the heap allocator"
@require @typekind(arr1) == SLICE || @typekind(arr1) == ARRAY
@require @typekind(arr2) == SLICE || @typekind(arr2) == ARRAY
@require @typeis(arr1[0], $typeof(arr2[0])) : "Arrays must have the same type"
@ensure result.len == arr1.len + arr2.len
*>
macro concat(Allocator allocator, arr1, arr2) @nodiscard
{
var $Type = $typeof(arr1[0]);
$Type[] result = allocator::alloc_array(allocator, $Type, arr1.len + arr2.len);
if (arr1.len > 0)
{
mem::copy(result.ptr, &arr1[0], arr1.len * $Type.sizeof, $Type.alignof, $Type.alignof);
}
if (arr2.len > 0)
{
mem::copy(&result[arr1.len], &arr2[0], arr2.len * $Type.sizeof, $Type.alignof, $Type.alignof);
}
return result;
}
<*
Concatenate two arrays or slices, returning a slice containing the concatenation of them,
allocated using the temp allocator.
@param [in] arr1
@param [in] arr2
@require @typekind(arr1) == SLICE || @typekind(arr1) == ARRAY
@require @typekind(arr2) == SLICE || @typekind(arr2) == ARRAY
@require @typeis(arr1[0], $typeof(arr2[0])) : "Arrays must have the same type"
@ensure return.len == arr1.len + arr2.len
*>
macro tconcat(arr1, arr2) @nodiscard => concat(tmem, arr1, arr2);
<*
Zip together two separate arrays/slices into a single array of Pairs or return values. Values will
be collected up to the length of the shorter array if `fill_with` is left undefined; otherwise, they
will be collected up to the length of the LONGER array, with missing values in the shorter array being
assigned to the value of `fill_with`. Return array elements do not have to be of the same type.
For example:
```c3
uint[] chosen_session_ids = server::get_random_sessions(instance)[:128];
String[200] refreshed_session_keys = prng::new_keys_batch();
Pair { uint, String }[] sessions_meta = array::zip(mem, chosen_session_ids, refreshed_session_keys);
// The resulting Pair{}[] slice is then length of the shortest of the two arrays, so 128.
foreach (i, &sess : sessions:meta) {
// distribute new session keys to associated instance IDs
}
```
Or:
```c3
String[] client_names = server::online_usernames(instance);
uint128[] session_ids = server::user_keys();
// in this example, we 'know' ahead of time that 'session_ids' can only ever be SHORTER
// than 'client_names', but never longer, because it's possible new users have logged
// in without getting whatever this 'session ID' is delegated to them.
Pair { String, uint128 }[] zipped = array::tzip(client_names, session_ids, fill_with: uint128.max);
server::refresh_session_keys_by_pair(zipped)!;
```
### When an `operation` is supplied...
Apply an operation to each element of two slices or arrays and return the results of
each operation into a newly allocated array.
This essentially combines Iterable1 with Iterable2 using the `operation` functor.
See the functional `zipWith` construct, which has a more appropriate name than, e.g., `map`;
a la: https://hackage.haskell.org/package/base-4.21.0.0/docs/Prelude.html#v:zipWith
Similar to "normal" `zip`, this macro pads the shorter input array with a given `fill_with`, or
an empty value if one isn't supplied. This `fill_with` is supplied to the `operation` functor
_BEFORE_ calculating its result while zipping.
For example: a functor of `fn char (char a, char b) => a + b` with a `fill_with` of 7,
where the `left` array is the shorter iterable, will put 7 into that lambda in each place
where `left` is being filled in during the zip operation.
@param [&inout] allocator : "The allocator to use; default is the heap allocator."
@param [in] left : "The left-side array. These items will be placed as the First in each Pair"
@param [in] right : "The right-side array. These items will be placed as the Second in each Pair"
@param #operation : "The function to apply. Must have a signature of `$typeof(a) (a, b)`, where the type of 'a' and 'b' is the element type of left/right respectively."
@param fill_with : "The value used to fill or pad the shorter iterable to the length of the longer one while zipping."
@require @is_valid_list(left) &&& @is_valid_list(right) : "Left and right sides must be integer indexable"
@require @is_valid_operation(#operation, left, right) : "The operator must take two parameters matching the elements of the left and right side"
@require @is_valid_fill(left, right, fill_with) : "The specified fill value does not match either the left or the right array's underlying type."
*>
macro @zip(Allocator allocator, left, right, #operation = EMPTY_MACRO_SLOT, fill_with = EMPTY_MACRO_SLOT) @nodiscard
{
var $LeftType = $typeof(left[0]);
var $RightType = $typeof(right[0]);
var $Type = Pair { $LeftType, $RightType };
bool $is_op = @is_valid_macro_slot(#operation);
$if $is_op:
$Type = $typeof(#operation).returns;
$endif
usz left_len = find_len(left);
usz right_len = find_len(right);
$LeftType left_fill;
$RightType right_fill;
usz result_len = min(left_len, right_len);
bool $has_fill = @is_valid_macro_slot(fill_with);
$if $has_fill:
switch
{
case left_len > right_len:
$if !$defined(($RightType)fill_with):
unreachable();
$else
right_fill = ($RightType)fill_with;
result_len = left_len;
$endif
case left_len < right_len:
$if !$defined(($LeftType)fill_with):
unreachable();
$else
left_fill = ($LeftType)fill_with;
result_len = right_len;
$endif
}
$endif
if (result_len == 0) return ($Type[]){};
$Type[] result = allocator::alloc_array(allocator, $Type, result_len);
foreach (idx, &item : result)
{
$if $is_op:
var $LambdaType = $typeof(fn $Type ($LeftType a, $RightType b) => ($Type){});
$LambdaType $operation = ($LambdaType)#operation;
$LeftType lval = idx >= left_len ? left_fill : left[idx];
$RightType rval = idx >= right_len ? right_fill : right[idx];
*item = $operation(lval, rval);
$else
*item = {
idx >= left_len ? left_fill : left[idx],
idx >= right_len ? right_fill : right[idx]
};
$endif
}
return result;
}
<*
Array 'zip' using the temp allocator.
@param [in] left : "The left-side array. These items will be placed as the First in each Pair"
@param [in] right : "The right-side array. These items will be placed as the Second in each Pair"
@param #operation : "The function to apply. Must have a signature of `$typeof(a) (a, b)`, where the type of 'a' and 'b' is the element type of left/right respectively."
@param fill_with : "The value used to fill or pad the shorter iterable to the length of the longer one while zipping."
@require @is_valid_list(left) &&& @is_valid_list(right) : "Left and right sides must be integer indexable"
@require @is_valid_operation(#operation, left, right) : "The operator must take two parameters matching the elements of the left and right side"
@require @is_valid_fill(left, right, fill_with) : "The specified fill value does not match either the left or the right array's underlying type."
*>
macro @tzip(left, right, #operation = EMPTY_MACRO_SLOT, fill_with = EMPTY_MACRO_SLOT) @nodiscard
{
return @zip(tmem, left, right, #operation, fill_with);
}
<*
Apply an operation to each element of two slices or arrays and store the results of
each operation into the 'left' value.
This is useful because no memory allocations are required in order to perform the operation.
A good example of using this might be using algorithmic transformations on data in-place:
```
char[] partial_cipher = get_next_plaintext_block();
array::@zip_into(
partial_cipher[ENCRYPT_OFFSET:BASE_KEY.len],
BASE_KEY,
fn char (char a, char b) => a ^ (b * 5) % 37
);
```
This parameterizes the lambda function with left (`partial_cipher`) and right (`BASE_KEY`) slice
elements and stores the end result in-place within the left slice. This is in contrast to a
regular `zip_with` which will create a cloned final result and return it.
@param [inout] left : `Slice to store results of applied functor/operation.`
@param [in] right : `Slice to apply in the functor/operation.`
@param #operation : "The function to apply. Must have a signature of `$typeof(a) (a, b)`, where the type of 'a' and 'b' is the element type of left/right respectively."
@require @is_valid_list(left) : "Expected a valid list"
@require @is_valid_list(right) : "Expected a valid list"
@require find_len(right) >= find_len(left) : `Right side length must be >= the destination (left) side length; consider using a sub-array of data for the assignment.`
@require @assignable_to(#operation, $typefrom(@zip_into_fn(left, right))) : "The functor must use the same types as the `left` and `right` inputs, and return a value of the `left` type."
*>
macro @zip_into(left, right, #operation)
{
$typefrom(@zip_into_fn(left, right)) $operation = #operation;
foreach (i, &v : left) *v = $operation(left[i], right[i]);
}
// --- helper functions
macro typeid @zip_into_fn(#left, #right) @private @const
{
return @typeid(fn $typeof(#left[0]) ($typeof(#left[0]) l, $typeof(#right[0]) r) => l);
}
macro bool @is_valid_operation(#operation, #left, #right) @const @private
{
$switch:
$case @is_empty_macro_slot(#operation):
return true;
$case @typekind(#operation) != FUNC:
return false;
$default:
return $defined(#operation(#left[0], #right[0]));
$endswitch
}
macro bool @is_valid_list(#expr) @const @private
{
return $defined(#expr[0]) &&& ($defined(#expr.len) ||| $defined(#expr.len()));
}
macro bool @is_valid_fill(left, right, fill_with) @private
{
if (@is_empty_macro_slot(fill_with)) return true;
usz left_len = @select($defined(left.len()), left.len(), left.len);
usz right_len = @select($defined(right.len()), right.len(), right.len);
if (left_len == right_len) return true;
return left_len > right_len ? $defined(($typeof(right[0]))fill_with) : $defined(($typeof(left[0]))fill_with);
}
macro usz find_len(list)
{
$if $defined(list.len()):
return list.len();
$else
return list.len;
$endif
}