deva2mon.cs

using System;
using System.Collections;
using System.IO;
using System.Text;
using System.Text.RegularExpressions;

namespace VRI.CSCD.Conversion
{
    class Deva2Mon
    {
        static void Main(string[] args)
        {
            try
            {
                if (args.Length < 2 || args.Length > 3)
                {
                    PrintUsage();
                    return;
                }

                FileInfo fi = new FileInfo(args[0]);
                if (!fi.Exists)
                {
                    Console.WriteLine("Input file does not exist");
                    return;
                }

                DirectoryInfo di = new DirectoryInfo(args[1]);
                if (!di.Exists)
                {
                    Console.Write("Output directory does not exist. Create it?[y/n] ");
                    if (Console.ReadLine()?.ToLower().StartsWith("y") == true)
                        di.Create();
                    else
                        return;
                }

                Deva2Mon d2m = new Deva2Mon();
                d2m.InputFilePath = args[0];
                d2m.OutputFilePath = Path.Combine(di.FullName, fi.Name);
                d2m.Convert();
            }
            catch (Exception ex)
            {
                Console.WriteLine("Exception: " + ex);
            }
        }

        static void PrintUsage()
        {
            Console.WriteLine("Transliterates Unicode Devanagari to Unicode Mon script");
            Console.WriteLine("syntax: deva2mon input output_dir");
        }

        private Hashtable deva2Mon;

        public Deva2Mon()
        {
            deva2Mon = new Hashtable();

            // --- Mapping Devanagari to Mon ---
            // Consonants (พยัญชนะ)
            deva2Mon['\x0915'] = '\x1000'; // ka
            deva2Mon['\x0916'] = '\x1001'; // kha
            deva2Mon['\x0917'] = '\x1002'; // ga
            deva2Mon['\x0918'] = '\x1003'; // gha
            deva2Mon['\x0919'] = '\x1004'; // nga
            
            deva2Mon['\x091A'] = '\x1005'; // ca
            deva2Mon['\x091B'] = '\x1006'; // cha
            deva2Mon['\x091C'] = '\x1007'; // ja
            deva2Mon['\x091D'] = '\x1008'; // jha
            deva2Mon['\x091E'] = '\x1009'; // nya

            deva2Mon['\x091F'] = '\x100B'; // tta
            deva2Mon['\x0920'] = '\x100C'; // ttha
            deva2Mon['\x0921'] = '\x100D'; // dda
            deva2Mon['\x0922'] = '\x100E'; // ddha
            deva2Mon['\x0923'] = '\x100F'; // nna

            deva2Mon['\x0924'] = '\x1010'; // ta
            deva2Mon['\x0925'] = '\x1011'; // tha
            deva2Mon['\x0926'] = '\x1012'; // da
            deva2Mon['\x0927'] = '\x1013'; // dha
            deva2Mon['\x0928'] = '\x1014'; // na

            deva2Mon['\x092A'] = '\x1015'; // pa
            deva2Mon['\x092B'] = '\x1016'; // pha
            deva2Mon['\x092C'] = '\x1017'; // ba
            deva2Mon['\x092D'] = '\x1018'; // bha
            deva2Mon['\x092E'] = '\x1019'; // ma

            deva2Mon['\x092F'] = '\x101A'; // ya
            deva2Mon['\x0930'] = '\x101B'; // ra
            deva2Mon['\x0932'] = '\x101C'; // la
            deva2Mon['\x0935'] = '\x101D'; // va
            deva2Mon['\x0938'] = '\x101E'; // sa
            deva2Mon['\x0939'] = '\x101F'; // ha
            deva2Mon['\x0933'] = '\x1020'; // lla

            // Independent Vowels (สระลอย)
            deva2Mon['\x0905'] = '\x1021'; // a
            deva2Mon['\x0906'] = "\x1021\x102B"; // aa (มอญใช้ Tall AA)
            deva2Mon['\x0907'] = '\x1023'; // i
            deva2Mon['\x0908'] = '\x1024'; // ii
            deva2Mon['\x0909'] = '\x1025'; // u
            deva2Mon['\x090A'] = '\x1026'; // uu
            deva2Mon['\x090F'] = '\x1027'; // e
            deva2Mon['\x0913'] = '\x1029'; // o

            // Dependent Vowel Signs (สระจม)
            deva2Mon['\x093E'] = '\x102B'; // aa (Mon prefers Tall AA \x102B)
            deva2Mon['\x093F'] = '\x102D'; // i
            deva2Mon['\x0940'] = '\x102E'; // ii
            deva2Mon['\x0941'] = '\x102F'; // u
            deva2Mon['\x0942'] = '\x1030'; // uu
            deva2Mon['\x0947'] = '\x1031'; // e
            deva2Mon['\x094B'] = "\x1031\x102B"; // o (e + tall aa)

            // Numerals
            deva2Mon['\x0966'] = '\x1040';
            deva2Mon['\x0967'] = '\x1041';
            deva2Mon['\x0968'] = '\x1042';
            deva2Mon['\x0969'] = '\x1043';
            deva2Mon['\x096A'] = '\x1044';
            deva2Mon['\x096B'] = '\x1045';
            deva2Mon['\x096C'] = '\x1046';
            deva2Mon['\x096D'] = '\x1047';
            deva2Mon['\x096E'] = '\x1048';
            deva2Mon['\x096F'] = '\x1049';

            // Other
            deva2Mon['\x0902'] = '\x1036'; // Anusvara
            deva2Mon['\x094D'] = '\x1039'; // Virama (Stacking)
            deva2Mon['\x0970'] = '.'; 
            deva2Mon['\x200C'] = ""; // ZWNJ
            deva2Mon['\x200D'] = ""; // ZWJ
        }

        public string InputFilePath { get; set; }
        public string OutputFilePath { get; set; }

        public void Convert()
        {
            string devStr = File.ReadAllText(InputFilePath);

            // ปรับชื่อ Stylesheet สำหรับภาษามอญ
            devStr = devStr.Replace("tipitaka-deva.xsl", "tipitaka-mon.xsl");

            // มอญ Peyyala (ฯลฯ)
            devStr = devStr.Replace("\x2026\x092A\x094B\x0970\x2026", "\x104B\x1015\x1031\x104B");

            string str = Convert(devStr);
            str = ConvertDandas(str);
            str = CleanupPunctuation(str);

            File.WriteAllText(OutputFilePath, str, Encoding.Unicode);
        }

        public string Convert(string devStr)
        {
            StringBuilder sb = new StringBuilder();
            foreach (char c in devStr)
            {
                if (deva2Mon.ContainsKey(c))
                    sb.Append(deva2Mon[c]);
                else
                    sb.Append(c);
            }

            string mon = sb.ToString();

            // --- กฎเฉพาะของอักษรมอญ ---

            // 1. รวม ña + virama + ña เป็นตัว NNYA (U+100A)
            mon = mon.Replace("\x1009\x1039\x1009", "\x100A");

            // 2. ตัวควบกล้ำ (Medials)
            mon = mon.Replace("\x1039\x101A", "\x103B"); // Ya
            mon = mon.Replace("\x1039\x101B", "\x103C"); // Ra
            mon = mon.Replace("\x1039\x101D", "\x103D"); // Wa
            mon = mon.Replace("\x1039\x101F", "\x103E"); // Ha

            // 3. จัดการเรื่อง Nga (U+1004) เมื่อเป็นตัวสะกดต้องมี Asat/Killer
            mon = mon.Replace("\x1004\x1039", "\x1004\x103A\x1039");

            // 4. การใช้ "Great SA" (ဿ) สำหรับ ssa
            mon = mon.Replace("\x101E\x1039\x101E", "\x103F");

            return mon;
        }

        public string ConvertDandas(string str)
        {
            // จัดการเครื่องหมายวรรคตอน (Danda)
            str = Regex.Replace(str, "<p rend=\"gatha[a-z0-9]*\">.+?</p>", m => {
                string s = m.Value;
                s = s.Replace(",", "\x104A"); // Single line
                s = s.Replace("\x0964", "\x104B"); // Double line
                s = s.Replace("\x0965", "\x104B");
                return s;
            });

            str = str.Replace("\x0964", "\x104B");
            str = str.Replace("\x0965", "\x104B");
            str = str.Replace(",", "\x104A");
            str = str.Replace("\x2026", "\x104B");

            return str;
        }

        public string CleanupPunctuation(string str)
        {
            str = str.Replace(" ,", ",");
            str = str.Replace(" ?", "?");
            str = str.Replace(" !", "!");
            str = str.Replace(" ;", ";");
            str = str.Replace(" .", ".");
            return str;
        }
    }
}