79113
Ma
mini-coding-agent / files
mini-coding-agent / src / text.zig
3.7 KB
1const std = @import("std"); 2 3const binary_sniff = 8192; 4 5pub fn isBinary(bytes: []const u8) bool { 6 return std.mem.indexOfScalar(u8, bytes[0..@min(bytes.len, binary_sniff)], 0) != null; 7} 8 9pub fn countLines(content: []const u8) usize { 10 if (content.len == 0) return 0; 11 return std.mem.count(u8, content, "\n") + @intFromBool(content[content.len - 1] != '\n'); 12} 13 14pub fn plural(a: std.mem.Allocator, n: usize, one: []const u8, many: []const u8) []const u8 { 15 return std.fmt.allocPrint(a, "{d} {s}", .{ n, if (n == 1) one else many }) catch many; 16} 17 18pub fn join(a: std.mem.Allocator, parts: []const []const u8, sep: []const u8) ![]const u8 { 19 var out: std.ArrayList(u8) = .empty; 20 for (parts, 0..) |part, i| { 21 if (i > 0) try out.appendSlice(a, sep); 22 try out.appendSlice(a, part); 23 } 24 return out.items; 25} 26 27fn utf8SeqLen(p: []const u8) ?usize { 28 if (p.len == 0) return null; 29 const c = p[0]; 30 if (c >= 0xC2 and c <= 0xDF) { 31 if (p.len >= 2 and p[1] & 0xC0 == 0x80) return 2; 32 } else if (c == 0xE0) { 33 if (p.len >= 3 and p[1] >= 0xA0 and p[1] <= 0xBF and p[2] & 0xC0 == 0x80) return 3; 34 } else if (c >= 0xE1 and c <= 0xEC) { 35 if (p.len >= 3 and p[1] & 0xC0 == 0x80 and p[2] & 0xC0 == 0x80) return 3; 36 } else if (c == 0xED) { 37 if (p.len >= 3 and p[1] >= 0x80 and p[1] <= 0x9F and p[2] & 0xC0 == 0x80) return 3; 38 } else if (c >= 0xEE and c <= 0xEF) { 39 if (p.len >= 3 and p[1] & 0xC0 == 0x80 and p[2] & 0xC0 == 0x80) return 3; 40 } else if (c == 0xF0) { 41 if (p.len >= 4 and p[1] >= 0x90 and p[1] <= 0xBF and p[2] & 0xC0 == 0x80 and p[3] & 0xC0 == 0x80) return 4; 42 } else if (c >= 0xF1 and c <= 0xF3) { 43 if (p.len >= 4 and p[1] & 0xC0 == 0x80 and p[2] & 0xC0 == 0x80 and p[3] & 0xC0 == 0x80) return 4; 44 } else if (c == 0xF4) { 45 if (p.len >= 4 and p[1] >= 0x80 and p[1] <= 0x8F and p[2] & 0xC0 == 0x80 and p[3] & 0xC0 == 0x80) return 4; 46 } 47 return null; 48} 49 50pub const Cp = struct { cp: u21, len: usize }; 51 52pub fn decodeAt(bytes: []const u8, i: usize) ?Cp { 53 if (i >= bytes.len) return null; 54 const b = bytes[i]; 55 const n = std.unicode.utf8ByteSequenceLength(b) catch return .{ .cp = b, .len = 1 }; 56 if (i + n > bytes.len) return .{ .cp = b, .len = 1 }; 57 return .{ .cp = std.unicode.utf8Decode(bytes[i .. i + n]) catch b, .len = n }; 58} 59 60pub fn cpLen(bytes: []const u8) usize { 61 var i: usize = 0; 62 var n: usize = 0; 63 while (decodeAt(bytes, i)) |d| { 64 i += d.len; 65 n += 1; 66 } 67 return n; 68} 69 70pub fn utf8Clean(a: std.mem.Allocator, text: []const u8) ![]const u8 { 71 if (std.unicode.utf8ValidateSlice(text)) return text; 72 var out: std.ArrayList(u8) = .empty; 73 var i: usize = 0; 74 while (i < text.len) { 75 if (text[i] < 0x80) { 76 try out.append(a, text[i]); 77 i += 1; 78 } else if (utf8SeqLen(text[i..])) |len| { 79 try out.appendSlice(a, text[i .. i + len]); 80 i += len; 81 } else { 82 try out.appendSlice(a, "\u{FFFD}"); 83 i += 1; 84 } 85 } 86 return out.items; 87} 88 89pub fn slugify(a: std.mem.Allocator, text: []const u8) []const u8 { 90 var out: std.ArrayList(u8) = .empty; 91 var pending_dash = false; 92 for (text) |c| { 93 if (out.items.len >= 40) break; 94 const lower = std.ascii.toLower(c); 95 if (std.ascii.isAlphanumeric(lower)) { 96 if (pending_dash and out.items.len > 0 and out.items.len < 40) out.append(a, '-') catch {}; 97 pending_dash = false; 98 out.append(a, lower) catch {}; 99 } else { 100 pending_dash = true; 101 } 102 } 103 if (out.items.len == 0) return a.dupe(u8, "session") catch "session"; 104 return out.items; 105}