1 // © 2016 and later: Unicode, Inc. and others. 2 // License & terms of use: http://www.unicode.org/copyright.html 3 /* 4 ******************************************************************************* 5 * Copyright (C) 2001-2009, International Business Machines 6 * Corporation and others. All Rights Reserved. 7 ******************************************************************************* 8 */ 9 10 package com.ibm.icu.samples.shaping; 11 12 import com.ibm.icu.text.ArabicShaping; 13 import com.ibm.icu.text.ArabicShapingException; 14 15 /** 16 * Interactive test for Arabic shaping. 17 * Invoke from a command line passing args and strings. Use '-help' to see description of arguments. 18 */ 19 public class ArabicShapingSample{ 20 private static final int COPY = 0; 21 private static final int INPLACE = 1; 22 private static final int STRING = 2; 23 main(String[] args)24 public static final void main(String[] args) { 25 int testtype = COPY; 26 int options = 0; 27 int ss = 0; 28 int sl = -1; 29 int ds = 0; 30 int dl = -1; 31 String text = "$22.4 test 123 \ufef6\u0644\u0622 456 \u0664\u0665\u0666!"; 32 33 for (int i = 0; i < args.length; ++i) { 34 String arg = args[i]; 35 if (arg.charAt(0) == '-') { 36 String opt = arg.substring(1); 37 String val = opt; 38 int index = arg.indexOf(':'); 39 if (index != -1) { 40 opt = opt.substring(0, Math.min(index, 3)); 41 val = arg.substring(index + 1); 42 } 43 44 if (opt.equalsIgnoreCase("len")) { 45 options &= ~ArabicShaping.LENGTH_MASK; 46 if (val.equalsIgnoreCase("gs")) { 47 options |= ArabicShaping.LENGTH_GROW_SHRINK; 48 } else if (val.equalsIgnoreCase("sn")) { 49 options |= ArabicShaping.LENGTH_FIXED_SPACES_NEAR; 50 } else if (val.equalsIgnoreCase("se")) { 51 options |= ArabicShaping.LENGTH_FIXED_SPACES_AT_END; 52 } else if (val.equalsIgnoreCase("sb")) { 53 options |= ArabicShaping.LENGTH_FIXED_SPACES_AT_BEGINNING; 54 } else { 55 throwValError(opt, val); 56 } 57 } else if (opt.equalsIgnoreCase("dir")) { 58 options &= ~ArabicShaping.TEXT_DIRECTION_MASK; 59 if (val.equalsIgnoreCase("log")) { 60 options |= ArabicShaping.TEXT_DIRECTION_LOGICAL; 61 } else if (val.equalsIgnoreCase("vis")) { 62 options |= ArabicShaping.TEXT_DIRECTION_VISUAL_LTR; 63 } else { 64 throwValError(opt, val); 65 } 66 } else if (opt.equalsIgnoreCase("let")) { 67 options &= ~ArabicShaping.LETTERS_MASK; 68 if (val.equalsIgnoreCase("no")) { 69 options |= ArabicShaping.LETTERS_NOOP; 70 } else if (val.equalsIgnoreCase("sh")) { 71 options |= ArabicShaping.LETTERS_SHAPE; 72 } else if (val.equalsIgnoreCase("un")) { 73 options |= ArabicShaping.LETTERS_UNSHAPE; 74 } else if (val.equalsIgnoreCase("ta")) { 75 options |= ArabicShaping.LETTERS_SHAPE_TASHKEEL_ISOLATED; 76 } else { 77 throwValError(opt, val); 78 } 79 } else if (opt.equalsIgnoreCase("dig")) { 80 options &= ~ArabicShaping.DIGITS_MASK; 81 if (val.equalsIgnoreCase("no")) { 82 options |= ArabicShaping.DIGITS_NOOP; 83 } else if (val.equalsIgnoreCase("ea")) { 84 options |= ArabicShaping.DIGITS_EN2AN; 85 } else if (val.equalsIgnoreCase("ae")) { 86 options |= ArabicShaping.DIGITS_AN2EN; 87 } else if (val.equalsIgnoreCase("lr")) { 88 options |= ArabicShaping.DIGITS_EN2AN_INIT_LR; 89 } else if (val.equalsIgnoreCase("al")) { 90 options |= ArabicShaping.DIGITS_EN2AN_INIT_AL; 91 } else { 92 throwValError(opt, val); 93 } 94 } else if (opt.equalsIgnoreCase("typ")) { 95 options &= ~ArabicShaping.DIGIT_TYPE_MASK; 96 if (val.equalsIgnoreCase("an")) { 97 options |= ArabicShaping.DIGIT_TYPE_AN; 98 } else if (val.equalsIgnoreCase("ex")) { 99 options |= ArabicShaping.DIGIT_TYPE_AN_EXTENDED; 100 } else { 101 throwValError(opt, val); 102 } 103 } else if (opt.equalsIgnoreCase("dst")) { 104 try { 105 ds = Integer.parseInt(val); 106 } 107 catch (Exception e) { 108 throwValError(opt, val); 109 } 110 } else if (opt.equalsIgnoreCase("dln")) { 111 try { 112 dl = Integer.parseInt(val); 113 } 114 catch (Exception e) { 115 throwValError(opt, val); 116 } 117 } else if (opt.equalsIgnoreCase("sst")) { 118 try { 119 ss = Integer.parseInt(val); 120 } 121 catch (Exception e) { 122 throwValError(opt, val); 123 } 124 } else if (opt.equalsIgnoreCase("sln")) { 125 try { 126 sl = Integer.parseInt(val); 127 } 128 catch (Exception e) { 129 throwValError(opt, val); 130 } 131 } else if (opt.equalsIgnoreCase("tes")) { 132 if (val.equalsIgnoreCase("cp")) { 133 testtype = COPY; 134 } else if (val.equalsIgnoreCase("ip")) { 135 testtype = INPLACE; 136 } else if (val.equalsIgnoreCase("st")) { 137 testtype = STRING; 138 } else { 139 throwValError(opt, val); 140 } 141 } else if (opt.equalsIgnoreCase("help")) { 142 System.out.println(usage); 143 } else { 144 throwOptError(opt); 145 } 146 } else { 147 // assume text 148 text = parseText(arg); 149 } 150 } 151 152 if (sl < 0) { 153 sl = text.length() - ss; 154 System.out.println("sl defaulting to " + sl); 155 } 156 if (dl < 0) { 157 dl = 2 * sl; 158 System.out.println("dl defaulting to " + dl); 159 } 160 161 ArabicShaping shaper = new ArabicShaping(options); 162 System.out.println("shaper: " + shaper); 163 164 char[] src = text.toCharArray(); 165 System.out.println(" input: '" + escapedText(src, ss, sl) + "'"); 166 if (testtype != STRING) { 167 System.out.println("start: " + ss + " length: " + sl + " total length: " + src.length); 168 } 169 170 int result = -1; 171 char[] dest = null; 172 173 try { 174 switch (testtype) { 175 case COPY: 176 dest = new char[ds + dl]; 177 result = shaper.shape(src, ss, sl, dest, ds, dl); 178 break; 179 180 case INPLACE: 181 shaper.shape(src, ss, sl); 182 ds = ss; 183 result = sl; 184 dest = src; 185 break; 186 187 case STRING: 188 dest = shaper.shape(text).toCharArray(); 189 ds = 0; 190 result = dest.length; 191 break; 192 } 193 194 System.out.println("output: '" + escapedText(dest, ds, result) + "'"); 195 System.out.println("length: " + result); 196 if (ds != 0 || result != dest.length) { 197 System.out.println("full output: '" + escapedText(dest, 0, dest.length) + "'"); 198 } 199 } 200 catch (ArabicShapingException e) { 201 System.out.println("Caught ArabicShapingException"); 202 System.out.println(e); 203 } 204 catch (Exception e) { 205 System.out.println("Caught Exception"); 206 System.out.println(e); 207 } 208 } 209 throwOptError(String opt)210 private static void throwOptError(String opt) { 211 throwUsageError("unknown option: " + opt); 212 } 213 throwValError(String opt, String val)214 private static void throwValError(String opt, String val) { 215 throwUsageError("unknown value: " + val + " for option: " + opt); 216 } 217 throwUsageError(String message)218 private static void throwUsageError(String message) { 219 StringBuffer buf = new StringBuffer("*** usage error ***\n"); 220 buf.append(message); 221 buf.append("\n"); 222 buf.append(usage); 223 throw new Error(buf.toString()); 224 } 225 226 private static final String usage = 227 "Usage: [option]* [text]\n" + 228 " where option is in the format '-opt[:val]'\n" + 229 " options are:\n" + 230 " -len:[gs|sn|se|sb] (length: grow/shrink, spaces near, spaces end, spaces beginning)\n" + 231 " -dir:[log|vis] (direction: logical, visual)\n" + 232 " -let:[no|sh|un|ta] (letters: noop, shape, unshape, tashkeel)\n" + 233 // " -let:[no|sh|un] (letters: noop, shape, unshape)\n" + 234 " -dig:[no|ea|ae|lr|al] (digits: noop, en2an, an2en, en2an_lr, en2an_al)\n" + 235 " -typ:[an|ex] (digit type: arabic, arabic extended)\n" + 236 " -dst:# (dest start: [integer])\n" + 237 " -dln:# (dest length (max size): [integer])\n" + 238 " -sst:# (source start: [integer])\n" + 239 " -sln:# (source length: [integer])\n" + 240 " -tes:[cp|ip|st] (test type: copy, in place, string)\n" + 241 " -help (print this help message)\n" + 242 " text can contain unicode escape values in the format '\\uXXXX' only\n"; 243 escapedText(char[] text, int start, int length)244 private static String escapedText(char[] text, int start, int length) { 245 StringBuffer buf = new StringBuffer(); 246 for (int i = start, e = start + length; i < e; ++i) { 247 char ch = text[i]; 248 if (ch < 0x20 || ch > 0x7e) { 249 buf.append("\\u"); 250 if (ch < 0x1000) { 251 buf.append('0'); 252 } 253 if (ch < 0x100) { 254 buf.append('0'); 255 } 256 if (ch < 0x10) { 257 buf.append('0'); 258 } 259 buf.append(Integer.toHexString(ch)); 260 } else { 261 buf.append(ch); 262 } 263 } 264 return buf.toString(); 265 } 266 parseText(String text)267 private static String parseText(String text) { 268 // process unicode escapes (only) 269 StringBuffer buf = new StringBuffer(); 270 char[] chars = text.toCharArray(); 271 for (int i = 0; i < chars.length; ++i) { 272 char ch = chars[i]; 273 if (ch == '\\') { 274 if ((i < chars.length - 1) && 275 (chars[i+1] == 'u')) { 276 int val = Integer.parseInt(text.substring(i+2, i+6), 16); 277 buf.append((char)val); 278 i += 5; 279 } else { 280 buf.append('\\'); 281 } 282 } else { 283 buf.append(ch); 284 } 285 } 286 return buf.toString(); 287 } 288 } 289