Introduces Bytes class (#1019)

This introduces a new `Bytes` standard library class, for working with
binary data.

* Add Bytes class to the standard library
* Change CLI to eval `output.bytes`
* Change code generators to map Bytes to respective underlying type
* Add subscript and concat operator support
* Add binary encoding for Bytes
* Add PCF and Plist rendering for Bytes

Co-authored-by: Kushal Pisavadia <kushi.p@gmail.com>
This commit is contained in:
Daniel Chao
2025-06-11 16:23:55 -07:00
committed by GitHub
parent 3bd8a88506
commit e9320557b7
104 changed files with 2210 additions and 545 deletions

View File

@@ -3684,6 +3684,41 @@ res5 = map.getOrNull("Falcon") // <5>
<4> result: `2`
<5> result: `null`
[[bytes]]
=== Bytes
A value of type `Bytes` is a sequence of `UInt8` elements.
`Bytes` can be constructed by passing byte values into the constructor.
[source,pkl]
----
bytes1 = Bytes(0xff, 0x00, 0x3f) // <1>
bytes2 = Bytes() // <2>
----
<1> Result: a `Bytes` with 3 elements
<2> Result: an empty `Bytes`
`Bytes` can also be constructed from a base64-encoded string, via `base64DecodedBytes`:
[source,pkl]
----
bytes3 = "cGFycm90".base64DecodedBytes // <1>
----
<1> Result: `Bytes(112, 97, 114, 114, 111, 116)`
==== `Bytes` vs `List<UInt8>`
`Bytes` is similar to `List<UInt8>` in that they are both sequences of `UInt8` elements.
However, they are semantically distinct.
`Bytes` represent binary data, and is typically rendered differently.
For example, they are rendered as `<data>` tags when using `PListRenderer`.
`Bytes` also have different performance characteristics; a value of type `Bytes` tends to be managed as a contiguous memory block.
Thus, they are more compact and consume less memory.
However, they are not optimized for transformations.
For example, given two values of size `M` and `N`, concatenating two `Bytes` values allocates O(M + N) space, whereas concatenating two `List` values allocates O(1) space.
[[regular-expressions]]
=== Regular Expressions
@@ -4671,6 +4706,10 @@ The following types are iterable:
|entry key (`Key`)
|entry value (`Value`)
|`Bytes`
|element index (`Int`)
|element value (`UInt8`)
|`Listing<Element>`
|element index (`Int`)
|element value (`Element`)
@@ -4751,6 +4790,9 @@ The following table describes how different iterables turn into object members:
| `IntSeq`
| Element
| `Bytes`
| Element
|===
These types can only be spread into enclosing objects that support that member type.