compatstd: minor optimization

Author: thislight

src/compatstd.zig

@@ -9,15 +9,15 @@ const nativeEndian = @import("builtin").cpu.arch.endian();
 const isZig0d14AndLater = builtin.zig_version.order(.{ .major = 0, .minor = 13, .patch = 0 }).compare(.gt);
 
 pub const mem = struct {
-    pub fn writeIntBig(T: type, dst: *[@divExact(meta.bitsOfType(T), 8)]u8, value: T) void {
+    pub inline fn writeIntBig(T: type, dst: *[@divExact(meta.bitsOfType(T), 8)]u8, value: T) void {
         return std.mem.writeInt(T, dst, value, .big);
     }
 
-    pub fn readIntBig(T: type, src: *const [@divExact(meta.bitsOfType(T), 8)]u8) T {
+    pub inline fn readIntBig(T: type, src: *const [@divExact(meta.bitsOfType(T), 8)]u8) T {
         return std.mem.readInt(T, src, .big);
     }
 
-    pub fn readFloatBig(comptime T: type, src: *const [@divExact(meta.bitsOfType(T), 8)]u8) T {
+    pub inline fn readFloatBig(comptime T: type, src: *const [@divExact(meta.bitsOfType(T), 8)]u8) T {
         if (meta.bitsOfType(T) > 64) {
             @compileError("readFloatBig does not support float types have more than 64 bits.");
         }
@@ -35,11 +35,11 @@ pub const mem = struct {
         return bytesAsValue(T, src).*;
     }
 
-    pub fn writeFloatBig(comptime T: type, dest: *[@divExact(meta.bitsOfType(T), 8)]u8, val: T) void {
+    pub inline fn writeFloatBig(comptime T: type, noalias dest: *[@divExact(meta.bitsOfType(T), 8)]u8, noalias val: T) void {
         if (meta.bitsOfType(T) > 64) {
             @compileError("writeFloatBig does not support float types have more than 64 bits.");
         }
-        std.mem.copyForwards(u8, dest, asBytes(&val));
+        @memcpy(dest, asBytes(&val));
         if (nativeEndian == .little) {
             for (0..@divExact(dest.len, 2)) |i| {
                 std.mem.swap(u8, &dest[i], &dest[dest.len - i - 1]);
@@ -49,7 +49,7 @@ pub const mem = struct {
 };
 
 pub const math = struct {
-    pub fn absCast(value: anytype) @TypeOf(@abs(value)) {
+    pub inline fn absCast(value: anytype) @TypeOf(@abs(value)) {
         return @abs(value);
     }
 };

src/root.zig

@@ -30,8 +30,8 @@
 //!
 //! var buf: [content.len + zigpak.PREFIX_BUFSIZE]u8 = undefined;
 //! const prefix = zigpak.AnyStr.prefix(@intCast(content.len));
-//! std.mem.copyForward(u8, &buf, prefix.constSlice());
-//! std.mem.copyForward(u8, buf[prefix.len..], content);
+//! @memcpy(u8, &buf, prefix.constSlice());
+//! @memcpy(u8, buf[prefix.len..], content);
 //!
 //! const result = buf[0..prefix.len + content.len]; // the constructed value
 //! ```
@@ -59,15 +59,15 @@
 //! AnyArray.pipe(buf.writer(), 3) catch unreachable;
 //!
 //! { // The first element: nil
-//!     Nil.pipe(buf.writer()) catch unreachable;
+//!     _ = Nil.pipe(buf.writer()) catch unreachable;
 //! }
 //!
 //! { // The second element: int 1
-//!     Int(i8).pipe(buf.writer(), 1) catch unreachable;
+//!     _ = Int(i8).pipe(buf.writer(), 1) catch unreachable;
 //! }
 //!
 //! { // The third element: a string
-//!     AnyStr.pipe(buf.writer(), @intCast(strContent.len)) catch unreachable;
+//!     _ = AnyStr.pipe(buf.writer(), @intCast(strContent.len)) catch unreachable;
 //!     buf.appendSliceAssumeCapacity(strContent);
 //! }
 //!