3 files changed, 108 insertions, 16 deletions

diff --git a/src/GenCatData.zig b/src/GenCatData.zig
new file mode 100644
index 0000000..5496e4e
--- /dev/null
+++ b/src/GenCatData.zig
@@ -0,0 +1,83 @@
+const std = @import("std");
+const builtin = @import("builtin");
+const compress = std.compress;
+const mem = std.mem;
+
+/// General Category
+pub const Gc = enum {
+    Cc,
+    Cf,
+    Cn,
+    Co,
+    Cs,
+    Ll,
+    Lm,
+    Lo,
+    Lt,
+    Lu,
+    Mc,
+    Me,
+    Mn,
+    Nd,
+    Nl,
+    No,
+    Pc,
+    Pd,
+    Pe,
+    Pf,
+    Pi,
+    Po,
+    Ps,
+    Sc,
+    Sk,
+    Sm,
+    So,
+    Zl,
+    Zp,
+    Zs,
+};
+
+allocator: mem.Allocator,
+s1: []u16 = undefined,
+s2: []u5 = undefined,
+s3: []u5 = undefined,
+
+const Self = @This();
+
+pub fn init(allocator: mem.Allocator) !Self {
+    const decompressor = compress.deflate.decompressor;
+    const in_bytes = @embedFile("gencat");
+    var in_fbs = std.io.fixedBufferStream(in_bytes);
+    var in_decomp = try decompressor(allocator, in_fbs.reader(), null);
+    defer in_decomp.deinit();
+    var reader = in_decomp.reader();
+
+    const endian = builtin.cpu.arch.endian();
+
+    var self = Self{ .allocator = allocator };
+
+    const s1_len: u16 = try reader.readInt(u16, endian);
+    self.s1 = try allocator.alloc(u16, s1_len);
+    for (0..s1_len) |i| self.s1[i] = try reader.readInt(u16, endian);
+
+    const s2_len: u16 = try reader.readInt(u16, endian);
+    self.s2 = try allocator.alloc(u5, s2_len);
+    for (0..s2_len) |i| self.s2[i] = @intCast(try reader.readInt(u8, endian));
+
+    const s3_len: u16 = try reader.readInt(u8, endian);
+    self.s3 = try allocator.alloc(u5, s3_len);
+    for (0..s3_len) |i| self.s3[i] = @intCast(try reader.readInt(u8, endian));
+
+    return self;
+}
+
+pub fn deinit(self: *Self) void {
+    self.allocator.free(self.s1);
+    self.allocator.free(self.s2);
+    self.allocator.free(self.s3);
+}
+
+/// Lookup the General Category for `cp`.
+pub inline fn gc(self: Self, cp: u21) Gc {
+    return @enumFromInt(self.s3[self.s2[self.s1[cp >> 8] + (cp & 0xff)]]);
+}
diff --git a/src/HangulData.zig b/src/HangulData.zig
index 4d80c99..b97424c 100644
--- a/src/HangulData.zig
+++ b/src/HangulData.zig
@@ -15,7 +15,7 @@ pub const Syllable = enum {
 
 allocator: mem.Allocator,
 s1: []u16 = undefined,
-s2: []Syllable = undefined,
+s2: []u3 = undefined,
 
 const Self = @This();
 
@@ -35,8 +35,8 @@ pub fn init(allocator: mem.Allocator) !Self {
     for (0..stage_1_len) |i| self.s1[i] = try reader.readInt(u16, endian);
 
     const stage_2_len: u16 = try reader.readInt(u16, endian);
-    self.s2 = try allocator.alloc(Syllable, stage_2_len);
-    for (0..stage_2_len) |i| self.s2[i] = @enumFromInt(try reader.readInt(u8, endian));
+    self.s2 = try allocator.alloc(u3, stage_2_len);
+    for (0..stage_2_len) |i| self.s2[i] = @intCast(try reader.readInt(u8, endian));
 
     return self;
 }
@@ -48,5 +48,5 @@ pub fn deinit(self: *Self) void {
 
 /// Returns the Hangul syllable type for `cp`.
 pub inline fn syllable(self: Self, cp: u21) Syllable {
-    return self.s2[self.s1[cp >> 8] + (cp & 0xff)];
+    return @enumFromInt(self.s2[self.s1[cp >> 8] + (cp & 0xff)]);
 }
diff --git a/src/main.zig b/src/main.zig
index 0f1aab5..c521c4f 100644
--- a/src/main.zig
+++ b/src/main.zig
@@ -11,14 +11,16 @@ const std = @import("std");
 // const strWidth = @import("display_width").strWidth;
 
 // const CodePointIterator = @import("ziglyph").CodePointIterator;
-// const CodePointIterator = @import("code_point").Iterator;
+const CodePointIterator = @import("code_point").Iterator;
 
 // const ascii = @import("ascii");
 // const ascii = std.ascii;
 
 // const Normalizer = @import("ziglyph").Normalizer;
-const NormData = @import("Normalizer").NormData;
-const Normalizer = @import("Normalizer");
+// const NormData = @import("Normalizer").NormData;
+// const Normalizer = @import("Normalizer");
+
+const GenCatData = @import("GenCatData");
 
 pub fn main() !void {
     var args_iter = std.process.args();
@@ -32,16 +34,19 @@ pub fn main() !void {
     const input = try std.fs.cwd().readFileAlloc(allocator, in_path, std.math.maxInt(u32));
     defer allocator.free(input);
 
-    var data = try NormData.init(allocator);
-    defer data.deinit();
-    var n = Normalizer{ .norm_data = &data };
+    // var data = try NormData.init(allocator);
+    // defer data.deinit();
+    // var n = Normalizer{ .norm_data = &data };
     // var n = try Normalizer.init(allocator);
     // defer n.deinit();
 
+    var gencat_data = try GenCatData.init(allocator);
+    defer gencat_data.deinit();
+
     // var iter = GraphemeIterator.init(input, &data);
     // defer iter.deinit();
-    // var iter = CodePointIterator{ .bytes = input };
-    var iter = std.mem.splitScalar(u8, input, '\n');
+    var iter = CodePointIterator{ .bytes = input };
+    // var iter = std.mem.splitScalar(u8, input, '\n');
 
     var result: usize = 0;
     // var result: isize = 0;
@@ -50,10 +55,14 @@ pub fn main() !void {
     // while (iter.next()) |cp| result += codePointWidth(@intCast(cp.code));
     // while (iter.next()) |_| result += 1;
     // while (iter.next()) |line| result += strWidth(line, &data);
-    while (iter.next()) |line| {
-        const nfc = try n.nfc(allocator, line);
-        result += nfc.slice.len;
-        // nfc.deinit();
+    // while (iter.next()) |line| {
+    //     const nfc = try n.nfc(allocator, line);
+    //     result += nfc.slice.len;
+    //     // nfc.deinit();
+    // }
+    while (iter.next()) |cp| {
+        if (cp.code == 'É') std.debug.print("`{u}` Gc: {s}\n", .{ cp.code, @tagName(gencat_data.gc(cp.code)) });
+        result += 1;
     }
 
     std.debug.print("result: {}, took: {}\n", .{ result, timer.lap() / std.time.ns_per_ms });