add utf8proc back for bench

pkg/utf8proc/build.zig

@@ -0,0 +1,37 @@
+const std = @import("std");
+
+pub fn build(b: *std.Build) !void {
+    const target = b.standardTargetOptions(.{});
+    const optimize = b.standardOptimizeOption(.{});
+
+    const module = b.addModule("utf8proc", .{ .root_source_file = .{ .path = "main.zig" } });
+
+    const upstream = b.dependency("utf8proc", .{});
+    const lib = b.addStaticLibrary(.{
+        .name = "utf8proc",
+        .target = target,
+        .optimize = optimize,
+    });
+    lib.linkLibC();
+
+    lib.addIncludePath(upstream.path(""));
+    module.addIncludePath(upstream.path(""));
+
+    var flags = std.ArrayList([]const u8).init(b.allocator);
+    try flags.append("-DUTF8PROC_EXPORTS");
+    defer flags.deinit();
+    lib.addCSourceFiles(.{
+        .dependency = upstream,
+        .files = &.{"utf8proc.c"},
+        .flags = flags.items,
+    });
+
+    lib.installHeadersDirectoryOptions(.{
+        .source_dir = upstream.path(""),
+        .install_dir = .header,
+        .install_subdir = "",
+        .include_extensions = &.{".h"},
+    });
+
+    b.installArtifact(lib);
+}

pkg/utf8proc/build.zig.zon

@@ -0,0 +1,11 @@
+.{
+    .name = "utf8proc",
+    .version = "2.8.0",
+    .paths = .{""},
+    .dependencies = .{
+        .utf8proc = .{
+            .url = "https://github.com/JuliaStrings/utf8proc/archive/refs/tags/v2.8.0.tar.gz",
+            .hash = "1220056ce228a8c58f1fa66ab778f5c8965e62f720c1d30603c7d534cb7d8a605ad7",
+        },
+    },
+}

pkg/utf8proc/c.zig

@@ -0,0 +1,3 @@
+pub usingnamespace @cImport({
+    @cInclude("utf8proc.h");
+});

pkg/utf8proc/main.zig

@@ -0,0 +1,20 @@
+pub const c = @import("c.zig");
+
+/// Given a codepoint, return a character width analogous to `wcwidth(codepoint)`,
+/// except that a width of 0 is returned for non-printable codepoints
+/// instead of -1 as in `wcwidth`.
+pub fn charwidth(codepoint: u21) u8 {
+    return @intCast(c.utf8proc_charwidth(@intCast(codepoint)));
+}
+
+/// Given a pair of consecutive codepoints, return whether a grapheme break is
+/// permitted between them (as defined by the extended grapheme clusters in UAX#29).
+pub fn graphemeBreakStateful(cp1: u21, cp2: u21, state: *i32) bool {
+    return c.utf8proc_grapheme_break_stateful(
+        @intCast(cp1),
+        @intCast(cp2),
+        state,
+    );
+}
+
+test {}