feat(tazjin/german-string): init initial representation r/8522

This adds an initial implementation of the so-called "German Strings" in Rust.

https://cedardb.com/blog/german_strings/

This implementation is *far from* complete, the only thing that can be done
right now is construct a string, and even that I'm not fully happy with.

Change-Id: I2697932a0ef373be76ffd14d59677493a5783b58
Reviewed-on: https://cl.tvl.fyi/c/depot/+/12234
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>

4 files changed, 82 insertions, 0 deletions

diff --git a/users/tazjin/german-string/.gitignore b/users/tazjin/german-string/.gitignore
new file mode 100644
index 000000000000..2f7896d1d136
--- /dev/null
+++ b/users/tazjin/german-string/.gitignore
@@ -0,0 +1 @@
+target/
diff --git a/users/tazjin/german-string/Cargo.lock b/users/tazjin/german-string/Cargo.lock
new file mode 100644
index 000000000000..14d31ce55628
--- /dev/null
+++ b/users/tazjin/german-string/Cargo.lock
@@ -0,0 +1,7 @@
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+version = 3
+
+[[package]]
+name = "german-string"
+version = "0.1.0"
diff --git a/users/tazjin/german-string/Cargo.toml b/users/tazjin/german-string/Cargo.toml
new file mode 100644
index 000000000000..90462e964839
--- /dev/null
+++ b/users/tazjin/german-string/Cargo.toml
@@ -0,0 +1,6 @@
+[package]
+name = "german-string"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
diff --git a/users/tazjin/german-string/src/lib.rs b/users/tazjin/german-string/src/lib.rs
new file mode 100644
index 000000000000..c35cb3ad4464
--- /dev/null
+++ b/users/tazjin/german-string/src/lib.rs
@@ -0,0 +1,68 @@
+use std::alloc::Layout;
+
+#[derive(Clone, Copy)]
+#[repr(C)]
+struct GSSmall {
+    len: u32,
+    data: [u8; 12],
+}
+
+#[derive(Clone, Copy)]
+#[repr(C)]
+struct GSLarge {
+    len: u32,
+    prefix: [u8; 4],
+    data: *mut u8,
+}
+
+const _ASSERT_VARIANTS_SIZE: () = assert!(
+    std::mem::size_of::<GSSmall>() == std::mem::size_of::<GSLarge>(),
+    "German String variants must have the same size"
+);
+
+union GSRepr {
+    small: GSSmall,
+    large: GSLarge,
+}
+
+#[repr(transparent)]
+pub struct GermanString(GSRepr);
+
+const _ASSERT_GSTRING_SIZE: () = assert!(
+    std::mem::size_of::<GermanString>() == 16,
+    "German String should be 16 bytes in size",
+);
+
+impl GermanString {
+    // Creates a new transient German String from the given bytes. Transient
+    // strings are destroyed when the object is destroyed. Persistent strings
+    // are not supported yet.
+    pub fn new_transient(bytes: &[u8]) -> GermanString {
+        if bytes.len() <= 12 {
+            let mut s = GSSmall {
+                len: bytes.len() as u32,
+                data: [0u8; 12],
+            };
+            s.data[..bytes.len()].copy_from_slice(bytes);
+            GermanString(GSRepr { small: s })
+        } else {
+            let layout = Layout::array::<u8>(bytes.len()).unwrap();
+            let mut large = GSLarge {
+                len: bytes.len() as u32,
+                prefix: [0u8; 4],
+                data: unsafe {
+                    let ptr = std::alloc::alloc(layout);
+                    if ptr.is_null() {
+                        std::alloc::handle_alloc_error(layout);
+                    }
+                    std::ptr::copy_nonoverlapping(bytes.as_ptr(), ptr, bytes.len());
+                    ptr
+                },
+            };
+
+            large.prefix.copy_from_slice(&bytes[..4]);
+
+            GermanString(GSRepr { large })
+        }
+    }
+}