• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /*
2  * Copyright (C) 2006 The Android Open Source Project
3  *
4  * Licensed under the Apache License, Version 2.0 (the "License");
5  * you may not use this file except in compliance with the License.
6  * You may obtain a copy of the License at
7  *
8  *      http://www.apache.org/licenses/LICENSE-2.0
9  *
10  * Unless required by applicable law or agreed to in writing, software
11  * distributed under the License is distributed on an "AS IS" BASIS,
12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13  * See the License for the specific language governing permissions and
14  * limitations under the License.
15  */
16 
17 package com.android.internal.telephony;
18 
19 import android.annotation.UnsupportedAppUsage;
20 import android.content.res.Resources;
21 import android.os.Build;
22 import android.telephony.Rlog;
23 import android.text.TextUtils;
24 import android.util.SparseIntArray;
25 
26 import com.android.internal.R;
27 
28 import java.nio.ByteBuffer;
29 import java.nio.charset.Charset;
30 import java.util.ArrayList;
31 import java.util.List;
32 
33 /**
34  * This class implements the character set mapping between
35  * the GSM SMS 7-bit alphabet specified in TS 23.038 6.2.1
36  * and UTF-16
37  *
38  * {@hide}
39  */
40 public class GsmAlphabet {
41     private static final String TAG = "GSM";
42 
GsmAlphabet()43     private GsmAlphabet() { }
44 
45     /**
46      * This escapes extended characters, and when present indicates that the
47      * following character should be looked up in the "extended" table.
48      *
49      * gsmToChar(GSM_EXTENDED_ESCAPE) returns 0xffff
50      */
51     public static final byte GSM_EXTENDED_ESCAPE = 0x1B;
52 
53     /**
54      * User data header requires one octet for length. Count as one septet, because
55      * all combinations of header elements below will have at least one free bit
56      * when padding to the nearest septet boundary.
57      */
58     public static final int UDH_SEPTET_COST_LENGTH = 1;
59 
60     /**
61      * Using a non-default language locking shift table OR single shift table
62      * requires a user data header of 3 octets, or 4 septets, plus UDH length.
63      */
64     public static final int UDH_SEPTET_COST_ONE_SHIFT_TABLE = 4;
65 
66     /**
67      * Using a non-default language locking shift table AND single shift table
68      * requires a user data header of 6 octets, or 7 septets, plus UDH length.
69      */
70     public static final int UDH_SEPTET_COST_TWO_SHIFT_TABLES = 7;
71 
72     /**
73      * Multi-part messages require a user data header of 5 octets, or 6 septets,
74      * plus UDH length.
75      */
76     public static final int UDH_SEPTET_COST_CONCATENATED_MESSAGE = 6;
77 
78     /**
79      * For a specific text string, this object describes protocol
80      * properties of encoding it for transmission as message user
81      * data.
82      */
83     public static class TextEncodingDetails {
84         /**
85          *The number of SMS's required to encode the text.
86          */
87         @UnsupportedAppUsage
88         public int msgCount;
89 
90         /**
91          * The number of code units consumed so far, where code units
92          * are basically characters in the encoding -- for example,
93          * septets for the standard ASCII and GSM encodings, and 16
94          * bits for Unicode.
95          */
96         @UnsupportedAppUsage
97         public int codeUnitCount;
98 
99         /**
100          * How many code units are still available without spilling
101          * into an additional message.
102          */
103         @UnsupportedAppUsage
104         public int codeUnitsRemaining;
105 
106         /**
107          * The encoding code unit size (specified using
108          * android.telephony.SmsMessage ENCODING_*).
109          */
110         @UnsupportedAppUsage
111         public int codeUnitSize;
112 
113         /**
114          * The GSM national language table to use, or 0 for the default 7-bit alphabet.
115          */
116         @UnsupportedAppUsage
117         public int languageTable;
118 
119         /**
120          * The GSM national language shift table to use, or 0 for the default 7-bit extension table.
121          */
122         @UnsupportedAppUsage
123         public int languageShiftTable;
124 
125         @Override
toString()126         public String toString() {
127             return "TextEncodingDetails " +
128                     "{ msgCount=" + msgCount +
129                     ", codeUnitCount=" + codeUnitCount +
130                     ", codeUnitsRemaining=" + codeUnitsRemaining +
131                     ", codeUnitSize=" + codeUnitSize +
132                     ", languageTable=" + languageTable +
133                     ", languageShiftTable=" + languageShiftTable +
134                     " }";
135         }
136     }
137 
138     /**
139      * Converts a char to a GSM 7 bit table index.
140      * Returns ' ' in GSM alphabet if there's no possible match. Returns
141      * GSM_EXTENDED_ESCAPE if this character is in the extended table.
142      * In this case, you must call charToGsmExtended() for the value
143      * that should follow GSM_EXTENDED_ESCAPE in the GSM alphabet string.
144      * @param c the character to convert
145      * @return the GSM 7 bit table index for the specified character
146      */
147     @UnsupportedAppUsage
148     public static int
charToGsm(char c)149     charToGsm(char c) {
150         try {
151             return charToGsm(c, false);
152         } catch (EncodeException ex) {
153             // this should never happen
154             return sCharsToGsmTables[0].get(' ', ' ');
155         }
156     }
157 
158     /**
159      * Converts a char to a GSM 7 bit table index.
160      * Returns GSM_EXTENDED_ESCAPE if this character is in the extended table.
161      * In this case, you must call charToGsmExtended() for the value that
162      * should follow GSM_EXTENDED_ESCAPE in the GSM alphabet string.
163      *
164      * @param c the character to convert
165      * @param throwException If true, throws EncodeException on invalid char.
166      *   If false, returns GSM alphabet ' ' char.
167      * @throws EncodeException encode error when throwException is true
168      * @return the GSM 7 bit table index for the specified character
169      */
170     @UnsupportedAppUsage
171     public static int
charToGsm(char c, boolean throwException)172     charToGsm(char c, boolean throwException) throws EncodeException {
173         int ret;
174 
175         ret = sCharsToGsmTables[0].get(c, -1);
176 
177         if (ret == -1) {
178             ret = sCharsToShiftTables[0].get(c, -1);
179 
180             if (ret == -1) {
181                 if (throwException) {
182                     throw new EncodeException(c);
183                 } else {
184                     return sCharsToGsmTables[0].get(' ', ' ');
185                 }
186             } else {
187                 return GSM_EXTENDED_ESCAPE;
188             }
189         }
190 
191         return ret;
192     }
193 
194     /**
195      * Converts a char to an extended GSM 7 bit table index.
196      * Extended chars should be escaped with GSM_EXTENDED_ESCAPE.
197      * Returns ' ' in GSM alphabet if there's no possible match.
198      * @param c the character to convert
199      * @return the GSM 7 bit extended table index for the specified character
200      */
201     public static int
charToGsmExtended(char c)202     charToGsmExtended(char c) {
203         int ret;
204 
205         ret = sCharsToShiftTables[0].get(c, -1);
206 
207         if (ret == -1) {
208             return sCharsToGsmTables[0].get(' ', ' ');
209         }
210 
211         return ret;
212     }
213 
214     /**
215      * Converts a character in the GSM alphabet into a char.
216      *
217      * If GSM_EXTENDED_ESCAPE is passed, 0xffff is returned. In this case,
218      * the following character in the stream should be decoded with
219      * gsmExtendedToChar().
220      *
221      * If an unmappable value is passed (one greater than 127), ' ' is returned.
222      *
223      * @param gsmChar the GSM 7 bit table index to convert
224      * @return the decoded character
225      */
226     @UnsupportedAppUsage
227     public static char
gsmToChar(int gsmChar)228     gsmToChar(int gsmChar) {
229         if (gsmChar >= 0 && gsmChar < 128) {
230             return sLanguageTables[0].charAt(gsmChar);
231         } else {
232             return ' ';
233         }
234     }
235 
236     /**
237      * Converts a character in the extended GSM alphabet into a char
238      *
239      * if GSM_EXTENDED_ESCAPE is passed, ' ' is returned since no second
240      * extension page has yet been defined (see Note 1 in table 6.2.1.1 of
241      * TS 23.038 v7.00)
242      *
243      * If an unmappable value is passed, the character from the GSM 7 bit
244      * default table will be used (table 6.2.1.1 of TS 23.038).
245      *
246      * @param gsmChar the GSM 7 bit extended table index to convert
247      * @return the decoded character
248      */
249     public static char
gsmExtendedToChar(int gsmChar)250     gsmExtendedToChar(int gsmChar) {
251         if (gsmChar == GSM_EXTENDED_ESCAPE) {
252             return ' ';
253         } else if (gsmChar >= 0 && gsmChar < 128) {
254             char c = sLanguageShiftTables[0].charAt(gsmChar);
255             if (c == ' ') {
256                 return sLanguageTables[0].charAt(gsmChar);
257             } else {
258                 return c;
259             }
260         } else {
261             return ' ';     // out of range
262         }
263     }
264 
265     /**
266      * Converts a String into a byte array containing the 7-bit packed
267      * GSM Alphabet representation of the string. If a header is provided,
268      * this is included in the returned byte array and padded to a septet
269      * boundary. This method is used by OEM code.
270      *
271      * @param data The text string to encode.
272      * @param header Optional header (including length byte) that precedes
273      * the encoded data, padded to septet boundary.
274      * @return Byte array containing header and encoded data.
275      * @throws EncodeException if String is too large to encode
276      * @see #stringToGsm7BitPackedWithHeader(String, byte[], int, int)
277      */
stringToGsm7BitPackedWithHeader(String data, byte[] header)278     public static byte[] stringToGsm7BitPackedWithHeader(String data, byte[] header)
279             throws EncodeException {
280         return stringToGsm7BitPackedWithHeader(data, header, 0, 0);
281     }
282 
283     /**
284      * Converts a String into a byte array containing the 7-bit packed
285      * GSM Alphabet representation of the string. If a header is provided,
286      * this is included in the returned byte array and padded to a septet
287      * boundary.
288      *
289      * Unencodable chars are encoded as spaces
290      *
291      * Byte 0 in the returned byte array is the count of septets used,
292      * including the header and header padding. The returned byte array is
293      * the minimum size required to store the packed septets. The returned
294      * array cannot contain more than 255 septets.
295      *
296      * @param data The text string to encode.
297      * @param header Optional header (including length byte) that precedes
298      * the encoded data, padded to septet boundary.
299      * @param languageTable the 7 bit language table, or 0 for the default GSM alphabet
300      * @param languageShiftTable the 7 bit single shift language table, or 0 for the default
301      *     GSM extension table
302      * @return Byte array containing header and encoded data.
303      * @throws EncodeException if String is too large to encode
304      */
305     @UnsupportedAppUsage
stringToGsm7BitPackedWithHeader(String data, byte[] header, int languageTable, int languageShiftTable)306     public static byte[] stringToGsm7BitPackedWithHeader(String data, byte[] header,
307             int languageTable, int languageShiftTable)
308             throws EncodeException {
309         if (header == null || header.length == 0) {
310             return stringToGsm7BitPacked(data, languageTable, languageShiftTable);
311         }
312 
313         int headerBits = (header.length + 1) * 8;
314         int headerSeptets = (headerBits + 6) / 7;
315 
316         byte[] ret = stringToGsm7BitPacked(data, headerSeptets, true, languageTable,
317                 languageShiftTable);
318 
319         // Paste in the header
320         ret[1] = (byte)header.length;
321         System.arraycopy(header, 0, ret, 2, header.length);
322         return ret;
323     }
324 
325     /**
326      * Converts a String into a byte array containing
327      * the 7-bit packed GSM Alphabet representation of the string.
328      *
329      * Unencodable chars are encoded as spaces
330      *
331      * Byte 0 in the returned byte array is the count of septets used
332      * The returned byte array is the minimum size required to store
333      * the packed septets. The returned array cannot contain more than 255
334      * septets.
335      *
336      * @param data the data string to encode
337      * @return the encoded string
338      * @throws EncodeException if String is too large to encode
339      */
340     @UnsupportedAppUsage
stringToGsm7BitPacked(String data)341     public static byte[] stringToGsm7BitPacked(String data)
342             throws EncodeException {
343         return stringToGsm7BitPacked(data, 0, true, 0, 0);
344     }
345 
346     /**
347      * Converts a String into a byte array containing
348      * the 7-bit packed GSM Alphabet representation of the string.
349      *
350      * Unencodable chars are encoded as spaces
351      *
352      * Byte 0 in the returned byte array is the count of septets used
353      * The returned byte array is the minimum size required to store
354      * the packed septets. The returned array cannot contain more than 255
355      * septets.
356      *
357      * @param data the data string to encode
358      * @param languageTable the 7 bit language table, or 0 for the default GSM alphabet
359      * @param languageShiftTable the 7 bit single shift language table, or 0 for the default
360      *     GSM extension table
361      * @return the encoded string
362      * @throws EncodeException if String is too large to encode
363      */
stringToGsm7BitPacked(String data, int languageTable, int languageShiftTable)364     public static byte[] stringToGsm7BitPacked(String data, int languageTable,
365             int languageShiftTable)
366             throws EncodeException {
367         return stringToGsm7BitPacked(data, 0, true, languageTable, languageShiftTable);
368     }
369 
370     /**
371      * Converts a String into a byte array containing
372      * the 7-bit packed GSM Alphabet representation of the string.
373      *
374      * Byte 0 in the returned byte array is the count of septets used
375      * The returned byte array is the minimum size required to store
376      * the packed septets. The returned array cannot contain more than 255
377      * septets.
378      *
379      * @param data the text to convert to septets
380      * @param startingSeptetOffset the number of padding septets to put before
381      *  the character data at the beginning of the array
382      * @param throwException If true, throws EncodeException on invalid char.
383      *   If false, replaces unencodable char with GSM alphabet space char.
384      * @param languageTable the 7 bit language table, or 0 for the default GSM alphabet
385      * @param languageShiftTable the 7 bit single shift language table, or 0 for the default
386      *     GSM extension table
387      * @return the encoded message
388      *
389      * @throws EncodeException if String is too large to encode or any characters are unencodable
390      */
391     @UnsupportedAppUsage
stringToGsm7BitPacked(String data, int startingSeptetOffset, boolean throwException, int languageTable, int languageShiftTable)392     public static byte[] stringToGsm7BitPacked(String data, int startingSeptetOffset,
393             boolean throwException, int languageTable, int languageShiftTable)
394             throws EncodeException {
395         int dataLen = data.length();
396         int septetCount = countGsmSeptetsUsingTables(data, !throwException,
397                 languageTable, languageShiftTable);
398         if (septetCount == -1) {
399             throw new EncodeException("countGsmSeptetsUsingTables(): unencodable char");
400         }
401         septetCount += startingSeptetOffset;
402         if (septetCount > 255) {
403             throw new EncodeException(
404                     "Payload cannot exceed 255 septets", EncodeException.ERROR_EXCEED_SIZE);
405         }
406         int byteCount = ((septetCount * 7) + 7) / 8;
407         byte[] ret = new byte[byteCount + 1];  // Include space for one byte length prefix.
408         SparseIntArray charToLanguageTable = sCharsToGsmTables[languageTable];
409         SparseIntArray charToShiftTable = sCharsToShiftTables[languageShiftTable];
410         for (int i = 0, septets = startingSeptetOffset, bitOffset = startingSeptetOffset * 7;
411                  i < dataLen && septets < septetCount;
412                  i++, bitOffset += 7) {
413             char c = data.charAt(i);
414             int v = charToLanguageTable.get(c, -1);
415             if (v == -1) {
416                 v = charToShiftTable.get(c, -1);  // Lookup the extended char.
417                 if (v == -1) {
418                     if (throwException) {
419                         throw new EncodeException("stringToGsm7BitPacked(): unencodable char");
420                     } else {
421                         v = charToLanguageTable.get(' ', ' ');   // should return ASCII space
422                     }
423                 } else {
424                     packSmsChar(ret, bitOffset, GSM_EXTENDED_ESCAPE);
425                     bitOffset += 7;
426                     septets++;
427                 }
428             }
429             packSmsChar(ret, bitOffset, v);
430             septets++;
431         }
432         ret[0] = (byte) (septetCount);  // Validated by check above.
433         return ret;
434     }
435 
436     /**
437      * Pack a 7-bit char into its appropriate place in a byte array
438      *
439      * @param packedChars the destination byte array
440      * @param bitOffset the bit offset that the septet should be packed at
441      *                  (septet index * 7)
442      * @param value the 7-bit character to store
443      */
444     @UnsupportedAppUsage
445     private static void
packSmsChar(byte[] packedChars, int bitOffset, int value)446     packSmsChar(byte[] packedChars, int bitOffset, int value) {
447         int byteOffset = bitOffset / 8;
448         int shift = bitOffset % 8;
449 
450         packedChars[++byteOffset] |= value << shift;
451 
452         if (shift > 1) {
453             packedChars[++byteOffset] = (byte)(value >> (8 - shift));
454         }
455     }
456 
457     /**
458      * Convert a GSM alphabet 7 bit packed string (SMS string) into a
459      * {@link java.lang.String}.
460      *
461      * See TS 23.038 6.1.2.1 for SMS Character Packing
462      *
463      * @param pdu the raw data from the pdu
464      * @param offset the byte offset of
465      * @param lengthSeptets string length in septets, not bytes
466      * @return String representation or null on decoding exception
467      */
468     @UnsupportedAppUsage
gsm7BitPackedToString(byte[] pdu, int offset, int lengthSeptets)469     public static String gsm7BitPackedToString(byte[] pdu, int offset,
470             int lengthSeptets) {
471         return gsm7BitPackedToString(pdu, offset, lengthSeptets, 0, 0, 0);
472     }
473 
474     /**
475      * Convert a GSM alphabet 7 bit packed string (SMS string) into a
476      * {@link java.lang.String}.
477      *
478      * See TS 23.038 6.1.2.1 for SMS Character Packing
479      *
480      * @param pdu the raw data from the pdu
481      * @param offset the byte offset of
482      * @param lengthSeptets string length in septets, not bytes
483      * @param numPaddingBits the number of padding bits before the start of the
484      *  string in the first byte
485      * @param languageTable the 7 bit language table, or 0 for the default GSM alphabet
486      * @param shiftTable the 7 bit single shift language table, or 0 for the default
487      *     GSM extension table
488      * @return String representation or null on decoding exception
489      */
490     @UnsupportedAppUsage
gsm7BitPackedToString(byte[] pdu, int offset, int lengthSeptets, int numPaddingBits, int languageTable, int shiftTable)491     public static String gsm7BitPackedToString(byte[] pdu, int offset,
492             int lengthSeptets, int numPaddingBits, int languageTable, int shiftTable) {
493         StringBuilder ret = new StringBuilder(lengthSeptets);
494 
495         if (languageTable < 0 || languageTable > sLanguageTables.length) {
496             Rlog.w(TAG, "unknown language table " + languageTable + ", using default");
497             languageTable = 0;
498         }
499         if (shiftTable < 0 || shiftTable > sLanguageShiftTables.length) {
500             Rlog.w(TAG, "unknown single shift table " + shiftTable + ", using default");
501             shiftTable = 0;
502         }
503 
504         try {
505             boolean prevCharWasEscape = false;
506             String languageTableToChar = sLanguageTables[languageTable];
507             String shiftTableToChar = sLanguageShiftTables[shiftTable];
508 
509             if (languageTableToChar.isEmpty()) {
510                 Rlog.w(TAG, "no language table for code " + languageTable + ", using default");
511                 languageTableToChar = sLanguageTables[0];
512             }
513             if (shiftTableToChar.isEmpty()) {
514                 Rlog.w(TAG, "no single shift table for code " + shiftTable + ", using default");
515                 shiftTableToChar = sLanguageShiftTables[0];
516             }
517 
518             for (int i = 0 ; i < lengthSeptets ; i++) {
519                 int bitOffset = (7 * i) + numPaddingBits;
520 
521                 int byteOffset = bitOffset / 8;
522                 int shift = bitOffset % 8;
523                 int gsmVal;
524 
525                 gsmVal = (0x7f & (pdu[offset + byteOffset] >> shift));
526 
527                 // if it crosses a byte boundary
528                 if (shift > 1) {
529                     // set msb bits to 0
530                     gsmVal &= 0x7f >> (shift - 1);
531 
532                     gsmVal |= 0x7f & (pdu[offset + byteOffset + 1] << (8 - shift));
533                 }
534 
535                 if (prevCharWasEscape) {
536                     if (gsmVal == GSM_EXTENDED_ESCAPE) {
537                         ret.append(' ');    // display ' ' for reserved double escape sequence
538                     } else {
539                         char c = shiftTableToChar.charAt(gsmVal);
540                         if (c == ' ') {
541                             ret.append(languageTableToChar.charAt(gsmVal));
542                         } else {
543                             ret.append(c);
544                         }
545                     }
546                     prevCharWasEscape = false;
547                 } else if (gsmVal == GSM_EXTENDED_ESCAPE) {
548                     prevCharWasEscape = true;
549                 } else {
550                     ret.append(languageTableToChar.charAt(gsmVal));
551                 }
552             }
553         } catch (RuntimeException ex) {
554             Rlog.e(TAG, "Error GSM 7 bit packed: ", ex);
555             return null;
556         }
557 
558         return ret.toString();
559     }
560 
561 
562     /**
563      * Convert a GSM alphabet string that's stored in 8-bit unpacked
564      * format (as it often appears in SIM records) into a String
565      *
566      * Field may be padded with trailing 0xff's. The decode stops
567      * at the first 0xff encountered.
568      *
569      * @param data the byte array to decode
570      * @param offset array offset for the first character to decode
571      * @param length the number of bytes to decode
572      * @return the decoded string
573      */
574     @UnsupportedAppUsage
575     public static String
gsm8BitUnpackedToString(byte[] data, int offset, int length)576     gsm8BitUnpackedToString(byte[] data, int offset, int length) {
577         return gsm8BitUnpackedToString(data, offset, length, "");
578     }
579 
580     /**
581      * Convert a GSM alphabet string that's stored in 8-bit unpacked
582      * format (as it often appears in SIM records) into a String
583      *
584      * Field may be padded with trailing 0xff's. The decode stops
585      * at the first 0xff encountered.
586      *
587      * Additionally, in some country(ex. Korea), there are non-ASCII or MBCS characters.
588      * If a character set is given, characters in data are treat as MBCS.
589      */
590     @UnsupportedAppUsage(maxTargetSdk = Build.VERSION_CODES.P, trackingBug = 115609023)
591     public static String
gsm8BitUnpackedToString(byte[] data, int offset, int length, String characterset)592     gsm8BitUnpackedToString(byte[] data, int offset, int length, String characterset) {
593         boolean isMbcs = false;
594         Charset charset = null;
595         ByteBuffer mbcsBuffer = null;
596 
597         if (!TextUtils.isEmpty(characterset)
598                 && !characterset.equalsIgnoreCase("us-ascii")
599                 && Charset.isSupported(characterset)) {
600             isMbcs = true;
601             charset = Charset.forName(characterset);
602             mbcsBuffer = ByteBuffer.allocate(2);
603         }
604 
605         // Always use GSM 7 bit default alphabet table for this method
606         String languageTableToChar = sLanguageTables[0];
607         String shiftTableToChar = sLanguageShiftTables[0];
608 
609         StringBuilder ret = new StringBuilder(length);
610         boolean prevWasEscape = false;
611         for (int i = offset ; i < offset + length ; i++) {
612             // Never underestimate the pain that can be caused
613             // by signed bytes
614             int c = data[i] & 0xff;
615 
616             if (c == 0xff) {
617                 break;
618             } else if (c == GSM_EXTENDED_ESCAPE) {
619                 if (prevWasEscape) {
620                     // Two escape chars in a row
621                     // We treat this as a space
622                     // See Note 1 in table 6.2.1.1 of TS 23.038 v7.00
623                     ret.append(' ');
624                     prevWasEscape = false;
625                 } else {
626                     prevWasEscape = true;
627                 }
628             } else {
629                 if (prevWasEscape) {
630                     char shiftChar =
631                             c < shiftTableToChar.length() ? shiftTableToChar.charAt(c) : ' ';
632                     if (shiftChar == ' ') {
633                         // display character from main table if not present in shift table
634                         if (c < languageTableToChar.length()) {
635                             ret.append(languageTableToChar.charAt(c));
636                         } else {
637                             ret.append(' ');
638                         }
639                     } else {
640                         ret.append(shiftChar);
641                     }
642                 } else {
643                     if (!isMbcs || c < 0x80 || i + 1 >= offset + length) {
644                         if (c < languageTableToChar.length()) {
645                             ret.append(languageTableToChar.charAt(c));
646                         } else {
647                             ret.append(' ');
648                         }
649                     } else {
650                         // isMbcs must be true. So both mbcsBuffer and charset are initialized.
651                         mbcsBuffer.clear();
652                         mbcsBuffer.put(data, i++, 2);
653                         mbcsBuffer.flip();
654                         ret.append(charset.decode(mbcsBuffer).toString());
655                     }
656                 }
657                 prevWasEscape = false;
658             }
659         }
660 
661         return ret.toString();
662     }
663 
664     /**
665      * Convert a string into an 8-bit unpacked GSM alphabet byte array.
666      * Always uses GSM default 7-bit alphabet and extension table.
667      * @param s the string to encode
668      * @return the 8-bit GSM encoded byte array for the string
669      */
670     @UnsupportedAppUsage
671     public static byte[]
672     stringToGsm8BitPacked(String s) {
673         byte[] ret;
674 
675         int septets = countGsmSeptetsUsingTables(s, true, 0, 0);
676 
677         // Enough for all the septets and the length byte prefix
678         ret = new byte[septets];
679 
680         stringToGsm8BitUnpackedField(s, ret, 0, ret.length);
681 
682         return ret;
683     }
684 
685 
686     /**
687      * Write a String into a GSM 8-bit unpacked field of
688      * Field is padded with 0xff's, string is truncated if necessary
689      *
690      * @param s the string to encode
691      * @param dest the destination byte array
692      * @param offset the starting offset for the encoded string
693      * @param length the maximum number of bytes to write
694      */
695     public static void
696     stringToGsm8BitUnpackedField(String s, byte dest[], int offset, int length) {
697         int outByteIndex = offset;
698         SparseIntArray charToLanguageTable = sCharsToGsmTables[0];
699         SparseIntArray charToShiftTable = sCharsToShiftTables[0];
700 
701         // Septets are stored in byte-aligned octets
702         for (int i = 0, sz = s.length()
703                 ; i < sz && (outByteIndex - offset) < length
704                 ; i++
705         ) {
706             char c = s.charAt(i);
707 
708             int v = charToLanguageTable.get(c, -1);
709 
710             if (v == -1) {
711                 v = charToShiftTable.get(c, -1);
712                 if (v == -1) {
713                     v = charToLanguageTable.get(' ', ' ');  // fall back to ASCII space
714                 } else {
715                     // make sure we can fit an escaped char
716                     if (! (outByteIndex + 1 - offset < length)) {
717                         break;
718                     }
719 
720                     dest[outByteIndex++] = GSM_EXTENDED_ESCAPE;
721                 }
722             }
723 
724             dest[outByteIndex++] = (byte)v;
725         }
726 
727         // pad with 0xff's
728         while((outByteIndex - offset) < length) {
729             dest[outByteIndex++] = (byte)0xff;
730         }
731     }
732 
733     /**
734      * Returns the count of 7-bit GSM alphabet characters
735      * needed to represent this character. Counts unencodable char as 1 septet.
736      * @param c the character to examine
737      * @return the number of septets for this character
738      */
739     public static int
740     countGsmSeptets(char c) {
741         try {
742             return countGsmSeptets(c, false);
743         } catch (EncodeException ex) {
744             // This should never happen.
745             return 0;
746         }
747     }
748 
749     /**
750      * Returns the count of 7-bit GSM alphabet characters
751      * needed to represent this character using the default 7 bit GSM alphabet.
752      * @param c the character to examine
753      * @param throwsException If true, throws EncodeException if unencodable
754      * char. Otherwise, counts invalid char as 1 septet.
755      * @return the number of septets for this character
756      * @throws EncodeException the character can't be encoded and throwsException is true
757      */
758     @UnsupportedAppUsage
759     public static int
760     countGsmSeptets(char c, boolean throwsException) throws EncodeException {
761         if (sCharsToGsmTables[0].get(c, -1) != -1) {
762             return 1;
763         }
764 
765         if (sCharsToShiftTables[0].get(c, -1) != -1) {
766             return 2;
767         }
768 
769         if (throwsException) {
770             throw new EncodeException(c);
771         } else {
772             // count as a space char
773             return 1;
774         }
775     }
776 
777     public static boolean isGsmSeptets(char c) {
778         if (sCharsToGsmTables[0].get(c, -1) != -1) {
779             return true;
780         }
781 
782         if (sCharsToShiftTables[0].get(c, -1) != -1) {
783             return true;
784         }
785 
786         return false;
787     }
788 
789     /**
790      * Returns the count of 7-bit GSM alphabet characters needed
791      * to represent this string, using the specified 7-bit language table
792      * and extension table (0 for GSM default tables).
793      * @param s the Unicode string that will be encoded
794      * @param use7bitOnly allow using space in place of unencodable character if true,
795      *     otherwise, return -1 if any characters are unencodable
796      * @param languageTable the 7 bit language table, or 0 for the default GSM alphabet
797      * @param languageShiftTable the 7 bit single shift language table, or 0 for the default
798      *     GSM extension table
799      * @return the septet count for s using the specified language tables, or -1 if any
800      *     characters are unencodable and use7bitOnly is false
801      */
802     public static int countGsmSeptetsUsingTables(CharSequence s, boolean use7bitOnly,
803             int languageTable, int languageShiftTable) {
804         int count = 0;
805         int sz = s.length();
806         SparseIntArray charToLanguageTable = sCharsToGsmTables[languageTable];
807         SparseIntArray charToShiftTable = sCharsToShiftTables[languageShiftTable];
808         for (int i = 0; i < sz; i++) {
809             char c = s.charAt(i);
810             if (c == GSM_EXTENDED_ESCAPE) {
811                 Rlog.w(TAG, "countGsmSeptets() string contains Escape character, skipping.");
812                 continue;
813             }
814             if (charToLanguageTable.get(c, -1) != -1) {
815                 count++;
816             } else if (charToShiftTable.get(c, -1) != -1) {
817                 count += 2; // escape + shift table index
818             } else if (use7bitOnly) {
819                 count++;    // encode as space
820             } else {
821                 return -1;  // caller must check for this case
822             }
823         }
824         return count;
825     }
826 
827     /**
828      * Returns the count of 7-bit GSM alphabet characters
829      * needed to represent this string, and the language table and
830      * language shift table used to achieve this result.
831      * For multi-part text messages, each message part may use its
832      * own language table encoding as specified in the message header
833      * for that message. However, this method will only return the
834      * optimal encoding for the message as a whole. When the individual
835      * pieces are encoded, a more optimal encoding may be chosen for each
836      * piece of the message, but the message will be split into pieces
837      * based on the encoding chosen for the message as a whole.
838      * @param s the Unicode string that will be encoded
839      * @param use7bitOnly allow using space in place of unencodable character if true,
840      *     using the language table pair with the fewest unencodable characters
841      * @return a TextEncodingDetails object containing the message and
842      *     character counts for the most efficient 7-bit encoding,
843      *     or null if there are no suitable language tables to encode the string.
844      */
845     public static TextEncodingDetails
846     countGsmSeptets(CharSequence s, boolean use7bitOnly) {
847         // Load enabled language tables from config.xml, including any MCC overlays
848         if (!sDisableCountryEncodingCheck) {
849             enableCountrySpecificEncodings();
850         }
851         // fast path for common case where no national language shift tables are enabled
852         if (sEnabledSingleShiftTables.length + sEnabledLockingShiftTables.length == 0) {
853             TextEncodingDetails ted = new TextEncodingDetails();
854             int septets = GsmAlphabet.countGsmSeptetsUsingTables(s, use7bitOnly, 0, 0);
855             if (septets == -1) {
856                 return null;
857             }
858             ted.codeUnitSize = SmsConstants.ENCODING_7BIT;
859             ted.codeUnitCount = septets;
860             if (septets > SmsConstants.MAX_USER_DATA_SEPTETS) {
861                 ted.msgCount = (septets + (SmsConstants.MAX_USER_DATA_SEPTETS_WITH_HEADER - 1)) /
862                         SmsConstants.MAX_USER_DATA_SEPTETS_WITH_HEADER;
863                 ted.codeUnitsRemaining = (ted.msgCount *
864                         SmsConstants.MAX_USER_DATA_SEPTETS_WITH_HEADER) - septets;
865             } else {
866                 ted.msgCount = 1;
867                 ted.codeUnitsRemaining = SmsConstants.MAX_USER_DATA_SEPTETS - septets;
868             }
869             return ted;
870         }
871 
872         int maxSingleShiftCode = sHighestEnabledSingleShiftCode;
873         List<LanguagePairCount> lpcList = new ArrayList<LanguagePairCount>(
874                 sEnabledLockingShiftTables.length + 1);
875 
876         // Always add default GSM 7-bit alphabet table
877         lpcList.add(new LanguagePairCount(0));
878         for (int i : sEnabledLockingShiftTables) {
879             // Avoid adding default table twice in case 0 is in the list of allowed tables
880             if (i != 0 && !sLanguageTables[i].isEmpty()) {
881                 lpcList.add(new LanguagePairCount(i));
882             }
883         }
884 
885         int sz = s.length();
886         // calculate septet count for each valid table / shift table pair
887         for (int i = 0; i < sz && !lpcList.isEmpty(); i++) {
888             char c = s.charAt(i);
889             if (c == GSM_EXTENDED_ESCAPE) {
890                 Rlog.w(TAG, "countGsmSeptets() string contains Escape character, ignoring!");
891                 continue;
892             }
893             // iterate through enabled locking shift tables
894             for (LanguagePairCount lpc : lpcList) {
895                 int tableIndex = sCharsToGsmTables[lpc.languageCode].get(c, -1);
896                 if (tableIndex == -1) {
897                     // iterate through single shift tables for this locking table
898                     for (int table = 0; table <= maxSingleShiftCode; table++) {
899                         if (lpc.septetCounts[table] != -1) {
900                             int shiftTableIndex = sCharsToShiftTables[table].get(c, -1);
901                             if (shiftTableIndex == -1) {
902                                 if (use7bitOnly) {
903                                     // can't encode char, use space instead
904                                     lpc.septetCounts[table]++;
905                                     lpc.unencodableCounts[table]++;
906                                 } else {
907                                     // can't encode char, remove language pair from list
908                                     lpc.septetCounts[table] = -1;
909                                 }
910                             } else {
911                                 // encode as Escape + index into shift table
912                                 lpc.septetCounts[table] += 2;
913                             }
914                         }
915                     }
916                 } else {
917                     // encode as index into locking shift table for all pairs
918                     for (int table = 0; table <= maxSingleShiftCode; table++) {
919                         if (lpc.septetCounts[table] != -1) {
920                             lpc.septetCounts[table]++;
921                         }
922                     }
923                 }
924             }
925         }
926 
927         // find the least cost encoding (lowest message count and most code units remaining)
928         TextEncodingDetails ted = new TextEncodingDetails();
929         ted.msgCount = Integer.MAX_VALUE;
930         ted.codeUnitSize = SmsConstants.ENCODING_7BIT;
931         int minUnencodableCount = Integer.MAX_VALUE;
932         for (LanguagePairCount lpc : lpcList) {
933             for (int shiftTable = 0; shiftTable <= maxSingleShiftCode; shiftTable++) {
934                 int septets = lpc.septetCounts[shiftTable];
935                 if (septets == -1) {
936                     continue;
937                 }
938                 int udhLength;
939                 if (lpc.languageCode != 0 && shiftTable != 0) {
940                     udhLength = UDH_SEPTET_COST_LENGTH + UDH_SEPTET_COST_TWO_SHIFT_TABLES;
941                 } else if (lpc.languageCode != 0 || shiftTable != 0) {
942                     udhLength = UDH_SEPTET_COST_LENGTH + UDH_SEPTET_COST_ONE_SHIFT_TABLE;
943                 } else {
944                     udhLength = 0;
945                 }
946                 int msgCount;
947                 int septetsRemaining;
948                 if (septets + udhLength > SmsConstants.MAX_USER_DATA_SEPTETS) {
949                     if (udhLength == 0) {
950                         udhLength = UDH_SEPTET_COST_LENGTH;
951                     }
952                     udhLength += UDH_SEPTET_COST_CONCATENATED_MESSAGE;
953                     int septetsPerMessage = SmsConstants.MAX_USER_DATA_SEPTETS - udhLength;
954                     msgCount = (septets + septetsPerMessage - 1) / septetsPerMessage;
955                     septetsRemaining = (msgCount * septetsPerMessage) - septets;
956                 } else {
957                     msgCount = 1;
958                     septetsRemaining = SmsConstants.MAX_USER_DATA_SEPTETS - udhLength - septets;
959                 }
960                 // for 7-bit only mode, use language pair with the least unencodable chars
961                 int unencodableCount = lpc.unencodableCounts[shiftTable];
962                 if (use7bitOnly && unencodableCount > minUnencodableCount) {
963                     continue;
964                 }
965                 if ((use7bitOnly && unencodableCount < minUnencodableCount)
966                         || msgCount < ted.msgCount || (msgCount == ted.msgCount
967                         && septetsRemaining > ted.codeUnitsRemaining)) {
968                     minUnencodableCount = unencodableCount;
969                     ted.msgCount = msgCount;
970                     ted.codeUnitCount = septets;
971                     ted.codeUnitsRemaining = septetsRemaining;
972                     ted.languageTable = lpc.languageCode;
973                     ted.languageShiftTable = shiftTable;
974                 }
975             }
976         }
977 
978         if (ted.msgCount == Integer.MAX_VALUE) {
979             return null;
980         }
981 
982         return ted;
983     }
984 
985     /**
986      * Returns the index into <code>s</code> of the first character
987      * after <code>limit</code> septets have been reached, starting at
988      * index <code>start</code>.  This is used when dividing messages
989      * into units within the SMS message size limit.
990      *
991      * @param s source string
992      * @param start index of where to start counting septets
993      * @param limit maximum septets to include,
994      *   e.g. <code>MAX_USER_DATA_SEPTETS</code>
995      * @param langTable the 7 bit character table to use (0 for default GSM 7-bit alphabet)
996      * @param langShiftTable the 7 bit shift table to use (0 for default GSM extension table)
997      * @return index of first character that won't fit, or the length
998      *   of the entire string if everything fits
999      */
1000     @UnsupportedAppUsage
1001     public static int
1002     findGsmSeptetLimitIndex(String s, int start, int limit, int langTable, int langShiftTable) {
1003         int accumulator = 0;
1004         int size = s.length();
1005 
1006         SparseIntArray charToLangTable = sCharsToGsmTables[langTable];
1007         SparseIntArray charToLangShiftTable = sCharsToShiftTables[langShiftTable];
1008         for (int i = start; i < size; i++) {
1009             int encodedSeptet = charToLangTable.get(s.charAt(i), -1);
1010             if (encodedSeptet == -1) {
1011                 encodedSeptet = charToLangShiftTable.get(s.charAt(i), -1);
1012                 if (encodedSeptet == -1) {
1013                     // char not found, assume we're replacing with space
1014                     accumulator++;
1015                 } else {
1016                     accumulator += 2;  // escape character + shift table index
1017                 }
1018             } else {
1019                 accumulator++;
1020             }
1021             if (accumulator > limit) {
1022                 return i;
1023             }
1024         }
1025         return size;
1026     }
1027 
1028     /**
1029      * Modify the array of enabled national language single shift tables for SMS
1030      * encoding. This is used for unit testing, but could also be used to
1031      * modify the enabled encodings based on the active MCC/MNC, for example.
1032      *
1033      * @param tables the new list of enabled single shift tables
1034      */
1035     public static synchronized void setEnabledSingleShiftTables(int[] tables) {
1036         sEnabledSingleShiftTables = tables;
1037         sDisableCountryEncodingCheck = true;
1038 
1039         if (tables.length > 0) {
1040             sHighestEnabledSingleShiftCode = tables[tables.length - 1];
1041         } else {
1042             sHighestEnabledSingleShiftCode = 0;
1043         }
1044     }
1045 
1046     /**
1047      * Modify the array of enabled national language locking shift tables for SMS
1048      * encoding. This is used for unit testing, but could also be used to
1049      * modify the enabled encodings based on the active MCC/MNC, for example.
1050      *
1051      * @param tables the new list of enabled locking shift tables
1052      */
1053     public static synchronized void setEnabledLockingShiftTables(int[] tables) {
1054         sEnabledLockingShiftTables = tables;
1055         sDisableCountryEncodingCheck = true;
1056     }
1057 
1058     /**
1059      * Return the array of enabled national language single shift tables for SMS
1060      * encoding. This is used for unit testing. The returned array is not a copy, so
1061      * the caller should be careful not to modify it.
1062      *
1063      * @return the list of enabled single shift tables
1064      */
1065     public static synchronized int[] getEnabledSingleShiftTables() {
1066         return sEnabledSingleShiftTables;
1067     }
1068 
1069     /**
1070      * Return the array of enabled national language locking shift tables for SMS
1071      * encoding. This is used for unit testing. The returned array is not a copy, so
1072      * the caller should be careful not to modify it.
1073      *
1074      * @return the list of enabled locking shift tables
1075      */
1076     public static synchronized int[] getEnabledLockingShiftTables() {
1077         return sEnabledLockingShiftTables;
1078     }
1079 
1080     /**
1081      * Enable country-specific language tables from MCC-specific overlays.
1082      * @context the context to use to get the TelephonyManager
1083      */
1084     private static void enableCountrySpecificEncodings() {
1085         Resources r = Resources.getSystem();
1086         // See comments in frameworks/base/core/res/res/values/config.xml for allowed values
1087         sEnabledSingleShiftTables = r.getIntArray(R.array.config_sms_enabled_single_shift_tables);
1088         sEnabledLockingShiftTables = r.getIntArray(R.array.config_sms_enabled_locking_shift_tables);
1089 
1090         if (sEnabledSingleShiftTables.length > 0) {
1091             sHighestEnabledSingleShiftCode =
1092                     sEnabledSingleShiftTables[sEnabledSingleShiftTables.length-1];
1093         } else {
1094             sHighestEnabledSingleShiftCode = 0;
1095         }
1096     }
1097 
1098     /** Reverse mapping from Unicode characters to indexes into language tables. */
1099     @UnsupportedAppUsage
1100     private static final SparseIntArray[] sCharsToGsmTables;
1101 
1102     /** Reverse mapping from Unicode characters to indexes into language shift tables. */
1103     @UnsupportedAppUsage
1104     private static final SparseIntArray[] sCharsToShiftTables;
1105 
1106     /** OEM configured list of enabled national language single shift tables for encoding. */
1107     @UnsupportedAppUsage
1108     private static int[] sEnabledSingleShiftTables;
1109 
1110     /** OEM configured list of enabled national language locking shift tables for encoding. */
1111     @UnsupportedAppUsage
1112     private static int[] sEnabledLockingShiftTables;
1113 
1114     /** Highest language code to include in array of single shift counters. */
1115     @UnsupportedAppUsage
1116     private static int sHighestEnabledSingleShiftCode;
1117 
1118     /** Flag to bypass check for country-specific overlays (for test cases only). */
1119     private static boolean sDisableCountryEncodingCheck = false;
1120 
1121     /**
1122      * Septet counter for a specific locking shift table and all of
1123      * the single shift tables that it can be paired with.
1124      */
1125     private static class LanguagePairCount {
1126         @UnsupportedAppUsage
1127         final int languageCode;
1128         @UnsupportedAppUsage
1129         final int[] septetCounts;
1130         @UnsupportedAppUsage
1131         final int[] unencodableCounts;
1132         @UnsupportedAppUsage
1133         LanguagePairCount(int code) {
1134             this.languageCode = code;
1135             int maxSingleShiftCode = sHighestEnabledSingleShiftCode;
1136             septetCounts = new int[maxSingleShiftCode + 1];
1137             unencodableCounts = new int[maxSingleShiftCode + 1];
1138             // set counters for disabled single shift tables to -1
1139             // (GSM default extension table index 0 is always enabled)
1140             for (int i = 1, tableOffset = 0; i <= maxSingleShiftCode; i++) {
1141                 if (sEnabledSingleShiftTables[tableOffset] == i) {
1142                     tableOffset++;
1143                 } else {
1144                     septetCounts[i] = -1;   // disabled
1145                 }
1146             }
1147             // exclude Turkish locking + Turkish single shift table and
1148             // Portuguese locking + Spanish single shift table (these
1149             // combinations will never be optimal for any input).
1150             if (code == 1 && maxSingleShiftCode >= 1) {
1151                 septetCounts[1] = -1;   // Turkish + Turkish
1152             } else if (code == 3 && maxSingleShiftCode >= 2) {
1153                 septetCounts[2] = -1;   // Portuguese + Spanish
1154             }
1155         }
1156     }
1157 
1158     /**
1159      * GSM default 7 bit alphabet plus national language locking shift character tables.
1160      * Comment lines above strings indicate the lower four bits of the table position.
1161      */
1162     @UnsupportedAppUsage
1163     private static final String[] sLanguageTables = {
1164         /* 3GPP TS 23.038 V9.1.1 section 6.2.1 - GSM 7 bit Default Alphabet
1165          01.....23.....4.....5.....6.....7.....8.....9.....A.B.....C.....D.E.....F.....0.....1 */
1166         "@\u00a3$\u00a5\u00e8\u00e9\u00f9\u00ec\u00f2\u00c7\n\u00d8\u00f8\r\u00c5\u00e5\u0394_"
1167             // 2.....3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....
1168             + "\u03a6\u0393\u039b\u03a9\u03a0\u03a8\u03a3\u0398\u039e\uffff\u00c6\u00e6\u00df"
1169             // F.....012.34.....56789ABCDEF0123456789ABCDEF0.....123456789ABCDEF0123456789A
1170             + "\u00c9 !\"#\u00a4%&'()*+,-./0123456789:;<=>?\u00a1ABCDEFGHIJKLMNOPQRSTUVWXYZ"
1171             // B.....C.....D.....E.....F.....0.....123456789ABCDEF0123456789AB.....C.....D.....
1172             + "\u00c4\u00d6\u00d1\u00dc\u00a7\u00bfabcdefghijklmnopqrstuvwxyz\u00e4\u00f6\u00f1"
1173             // E.....F.....
1174             + "\u00fc\u00e0",
1175 
1176         /* A.3.1 Turkish National Language Locking Shift Table
1177          01.....23.....4.....5.....6.....7.....8.....9.....A.B.....C.....D.E.....F.....0.....1 */
1178         "@\u00a3$\u00a5\u20ac\u00e9\u00f9\u0131\u00f2\u00c7\n\u011e\u011f\r\u00c5\u00e5\u0394_"
1179             // 2.....3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....
1180             + "\u03a6\u0393\u039b\u03a9\u03a0\u03a8\u03a3\u0398\u039e\uffff\u015e\u015f\u00df"
1181             // F.....012.34.....56789ABCDEF0123456789ABCDEF0.....123456789ABCDEF0123456789A
1182             + "\u00c9 !\"#\u00a4%&'()*+,-./0123456789:;<=>?\u0130ABCDEFGHIJKLMNOPQRSTUVWXYZ"
1183             // B.....C.....D.....E.....F.....0.....123456789ABCDEF0123456789AB.....C.....D.....
1184             + "\u00c4\u00d6\u00d1\u00dc\u00a7\u00e7abcdefghijklmnopqrstuvwxyz\u00e4\u00f6\u00f1"
1185             // E.....F.....
1186             + "\u00fc\u00e0",
1187 
1188         /* A.3.2 Void (no locking shift table for Spanish) */
1189         "",
1190 
1191         /* A.3.3 Portuguese National Language Locking Shift Table
1192          01.....23.....4.....5.....6.....7.....8.....9.....A.B.....C.....D.E.....F.....0.....1 */
1193         "@\u00a3$\u00a5\u00ea\u00e9\u00fa\u00ed\u00f3\u00e7\n\u00d4\u00f4\r\u00c1\u00e1\u0394_"
1194             // 2.....3.....4.....5.....67.8.....9.....AB.....C.....D.....E.....F.....012.34.....
1195             + "\u00aa\u00c7\u00c0\u221e^\\\u20ac\u00d3|\uffff\u00c2\u00e2\u00ca\u00c9 !\"#\u00ba"
1196             // 56789ABCDEF0123456789ABCDEF0.....123456789ABCDEF0123456789AB.....C.....D.....E.....
1197             + "%&'()*+,-./0123456789:;<=>?\u00cdABCDEFGHIJKLMNOPQRSTUVWXYZ\u00c3\u00d5\u00da\u00dc"
1198             // F.....0123456789ABCDEF0123456789AB.....C.....DE.....F.....
1199             + "\u00a7~abcdefghijklmnopqrstuvwxyz\u00e3\u00f5`\u00fc\u00e0",
1200 
1201         /* A.3.4 Bengali National Language Locking Shift Table
1202          0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....CD.EF.....0..... */
1203         "\u0981\u0982\u0983\u0985\u0986\u0987\u0988\u0989\u098a\u098b\n\u098c \r \u098f\u0990"
1204             // 123.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....F.....
1205             + "  \u0993\u0994\u0995\u0996\u0997\u0998\u0999\u099a\uffff\u099b\u099c\u099d\u099e"
1206             // 012.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....0123456789ABC
1207             + " !\u099f\u09a0\u09a1\u09a2\u09a3\u09a4)(\u09a5\u09a6,\u09a7.\u09a80123456789:; "
1208             // D.....E.....F0.....1.....2.....3.....4.....56.....789A.....B.....C.....D.....
1209             + "\u09aa\u09ab?\u09ac\u09ad\u09ae\u09af\u09b0 \u09b2   \u09b6\u09b7\u09b8\u09b9"
1210             // E.....F.....0.....1.....2.....3.....4.....5.....6.....789.....A.....BCD.....E.....
1211             + "\u09bc\u09bd\u09be\u09bf\u09c0\u09c1\u09c2\u09c3\u09c4  \u09c7\u09c8  \u09cb\u09cc"
1212             // F.....0.....123456789ABCDEF0123456789AB.....C.....D.....E.....F.....
1213             + "\u09cd\u09ceabcdefghijklmnopqrstuvwxyz\u09d7\u09dc\u09dd\u09f0\u09f1",
1214 
1215         /* A.3.5 Gujarati National Language Locking Shift Table
1216          0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....C.....D.EF.....0.....*/
1217         "\u0a81\u0a82\u0a83\u0a85\u0a86\u0a87\u0a88\u0a89\u0a8a\u0a8b\n\u0a8c\u0a8d\r \u0a8f\u0a90"
1218             // 1.....23.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....
1219             + "\u0a91 \u0a93\u0a94\u0a95\u0a96\u0a97\u0a98\u0a99\u0a9a\uffff\u0a9b\u0a9c\u0a9d"
1220             // F.....012.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....0123456789AB
1221             + "\u0a9e !\u0a9f\u0aa0\u0aa1\u0aa2\u0aa3\u0aa4)(\u0aa5\u0aa6,\u0aa7.\u0aa80123456789:;"
1222             // CD.....E.....F0.....1.....2.....3.....4.....56.....7.....89.....A.....B.....C.....
1223             + " \u0aaa\u0aab?\u0aac\u0aad\u0aae\u0aaf\u0ab0 \u0ab2\u0ab3 \u0ab5\u0ab6\u0ab7\u0ab8"
1224             // D.....E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89.....A.....
1225             + "\u0ab9\u0abc\u0abd\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac4\u0ac5 \u0ac7\u0ac8"
1226             // B.....CD.....E.....F.....0.....123456789ABCDEF0123456789AB.....C.....D.....E.....
1227             + "\u0ac9 \u0acb\u0acc\u0acd\u0ad0abcdefghijklmnopqrstuvwxyz\u0ae0\u0ae1\u0ae2\u0ae3"
1228             // F.....
1229             + "\u0af1",
1230 
1231         /* A.3.6 Hindi National Language Locking Shift Table
1232          0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....C.....D.E.....F.....*/
1233         "\u0901\u0902\u0903\u0905\u0906\u0907\u0908\u0909\u090a\u090b\n\u090c\u090d\r\u090e\u090f"
1234             // 0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....
1235             + "\u0910\u0911\u0912\u0913\u0914\u0915\u0916\u0917\u0918\u0919\u091a\uffff\u091b\u091c"
1236             // E.....F.....012.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....012345
1237             + "\u091d\u091e !\u091f\u0920\u0921\u0922\u0923\u0924)(\u0925\u0926,\u0927.\u0928012345"
1238             // 6789ABC.....D.....E.....F0.....1.....2.....3.....4.....5.....6.....7.....8.....
1239             + "6789:;\u0929\u092a\u092b?\u092c\u092d\u092e\u092f\u0930\u0931\u0932\u0933\u0934"
1240             // 9.....A.....B.....C.....D.....E.....F.....0.....1.....2.....3.....4.....5.....6.....
1241             + "\u0935\u0936\u0937\u0938\u0939\u093c\u093d\u093e\u093f\u0940\u0941\u0942\u0943\u0944"
1242             // 7.....8.....9.....A.....B.....C.....D.....E.....F.....0.....123456789ABCDEF012345678
1243             + "\u0945\u0946\u0947\u0948\u0949\u094a\u094b\u094c\u094d\u0950abcdefghijklmnopqrstuvwx"
1244             // 9AB.....C.....D.....E.....F.....
1245             + "yz\u0972\u097b\u097c\u097e\u097f",
1246 
1247         /* A.3.7 Kannada National Language Locking Shift Table
1248            NOTE: TS 23.038 V9.1.1 shows code 0x24 as \u0caa, corrected to \u0ca1 (typo)
1249          01.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....CD.E.....F.....0.....1 */
1250         " \u0c82\u0c83\u0c85\u0c86\u0c87\u0c88\u0c89\u0c8a\u0c8b\n\u0c8c \r\u0c8e\u0c8f\u0c90 "
1251             // 2.....3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....F.....
1252             + "\u0c92\u0c93\u0c94\u0c95\u0c96\u0c97\u0c98\u0c99\u0c9a\uffff\u0c9b\u0c9c\u0c9d\u0c9e"
1253             // 012.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....0123456789ABC
1254             + " !\u0c9f\u0ca0\u0ca1\u0ca2\u0ca3\u0ca4)(\u0ca5\u0ca6,\u0ca7.\u0ca80123456789:; "
1255             // D.....E.....F0.....1.....2.....3.....4.....5.....6.....7.....89.....A.....B.....
1256             + "\u0caa\u0cab?\u0cac\u0cad\u0cae\u0caf\u0cb0\u0cb1\u0cb2\u0cb3 \u0cb5\u0cb6\u0cb7"
1257             // C.....D.....E.....F.....0.....1.....2.....3.....4.....5.....6.....78.....9.....
1258             + "\u0cb8\u0cb9\u0cbc\u0cbd\u0cbe\u0cbf\u0cc0\u0cc1\u0cc2\u0cc3\u0cc4 \u0cc6\u0cc7"
1259             // A.....BC.....D.....E.....F.....0.....123456789ABCDEF0123456789AB.....C.....D.....
1260             + "\u0cc8 \u0cca\u0ccb\u0ccc\u0ccd\u0cd5abcdefghijklmnopqrstuvwxyz\u0cd6\u0ce0\u0ce1"
1261             // E.....F.....
1262             + "\u0ce2\u0ce3",
1263 
1264         /* A.3.8 Malayalam National Language Locking Shift Table
1265          01.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....CD.E.....F.....0.....1 */
1266         " \u0d02\u0d03\u0d05\u0d06\u0d07\u0d08\u0d09\u0d0a\u0d0b\n\u0d0c \r\u0d0e\u0d0f\u0d10 "
1267             // 2.....3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....F.....
1268             + "\u0d12\u0d13\u0d14\u0d15\u0d16\u0d17\u0d18\u0d19\u0d1a\uffff\u0d1b\u0d1c\u0d1d\u0d1e"
1269             // 012.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....0123456789ABC
1270             + " !\u0d1f\u0d20\u0d21\u0d22\u0d23\u0d24)(\u0d25\u0d26,\u0d27.\u0d280123456789:; "
1271             // D.....E.....F0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.....
1272             + "\u0d2a\u0d2b?\u0d2c\u0d2d\u0d2e\u0d2f\u0d30\u0d31\u0d32\u0d33\u0d34\u0d35\u0d36"
1273             // B.....C.....D.....EF.....0.....1.....2.....3.....4.....5.....6.....78.....9.....
1274             + "\u0d37\u0d38\u0d39 \u0d3d\u0d3e\u0d3f\u0d40\u0d41\u0d42\u0d43\u0d44 \u0d46\u0d47"
1275             // A.....BC.....D.....E.....F.....0.....123456789ABCDEF0123456789AB.....C.....D.....
1276             + "\u0d48 \u0d4a\u0d4b\u0d4c\u0d4d\u0d57abcdefghijklmnopqrstuvwxyz\u0d60\u0d61\u0d62"
1277             // E.....F.....
1278             + "\u0d63\u0d79",
1279 
1280         /* A.3.9 Oriya National Language Locking Shift Table
1281          0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....CD.EF.....0.....12 */
1282         "\u0b01\u0b02\u0b03\u0b05\u0b06\u0b07\u0b08\u0b09\u0b0a\u0b0b\n\u0b0c \r \u0b0f\u0b10  "
1283             // 3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....F.....01
1284             + "\u0b13\u0b14\u0b15\u0b16\u0b17\u0b18\u0b19\u0b1a\uffff\u0b1b\u0b1c\u0b1d\u0b1e !"
1285             // 2.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....0123456789ABCD.....
1286             + "\u0b1f\u0b20\u0b21\u0b22\u0b23\u0b24)(\u0b25\u0b26,\u0b27.\u0b280123456789:; \u0b2a"
1287             // E.....F0.....1.....2.....3.....4.....56.....7.....89.....A.....B.....C.....D.....
1288             + "\u0b2b?\u0b2c\u0b2d\u0b2e\u0b2f\u0b30 \u0b32\u0b33 \u0b35\u0b36\u0b37\u0b38\u0b39"
1289             // E.....F.....0.....1.....2.....3.....4.....5.....6.....789.....A.....BCD.....E.....
1290             + "\u0b3c\u0b3d\u0b3e\u0b3f\u0b40\u0b41\u0b42\u0b43\u0b44  \u0b47\u0b48  \u0b4b\u0b4c"
1291             // F.....0.....123456789ABCDEF0123456789AB.....C.....D.....E.....F.....
1292             + "\u0b4d\u0b56abcdefghijklmnopqrstuvwxyz\u0b57\u0b60\u0b61\u0b62\u0b63",
1293 
1294         /* A.3.10 Punjabi National Language Locking Shift Table
1295          0.....1.....2.....3.....4.....5.....6.....7.....8.....9A.BCD.EF.....0.....123.....4.....*/
1296         "\u0a01\u0a02\u0a03\u0a05\u0a06\u0a07\u0a08\u0a09\u0a0a \n  \r \u0a0f\u0a10  \u0a13\u0a14"
1297             // 5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....F.....012.....3.....
1298             + "\u0a15\u0a16\u0a17\u0a18\u0a19\u0a1a\uffff\u0a1b\u0a1c\u0a1d\u0a1e !\u0a1f\u0a20"
1299             // 4.....5.....6.....7.....89A.....B.....CD.....EF.....0123456789ABCD.....E.....F0.....
1300             + "\u0a21\u0a22\u0a23\u0a24)(\u0a25\u0a26,\u0a27.\u0a280123456789:; \u0a2a\u0a2b?\u0a2c"
1301             // 1.....2.....3.....4.....56.....7.....89.....A.....BC.....D.....E.....F0.....1.....
1302             + "\u0a2d\u0a2e\u0a2f\u0a30 \u0a32\u0a33 \u0a35\u0a36 \u0a38\u0a39\u0a3c \u0a3e\u0a3f"
1303             // 2.....3.....4.....56789.....A.....BCD.....E.....F.....0.....123456789ABCDEF012345678
1304             + "\u0a40\u0a41\u0a42    \u0a47\u0a48  \u0a4b\u0a4c\u0a4d\u0a51abcdefghijklmnopqrstuvwx"
1305             // 9AB.....C.....D.....E.....F.....
1306             + "yz\u0a70\u0a71\u0a72\u0a73\u0a74",
1307 
1308         /* A.3.11 Tamil National Language Locking Shift Table
1309          01.....2.....3.....4.....5.....6.....7.....8.....9A.BCD.E.....F.....0.....12.....3..... */
1310         " \u0b82\u0b83\u0b85\u0b86\u0b87\u0b88\u0b89\u0b8a \n  \r\u0b8e\u0b8f\u0b90 \u0b92\u0b93"
1311             // 4.....5.....6789.....A.....B.....CD.....EF.....012.....3456.....7.....89ABCDEF.....
1312             + "\u0b94\u0b95   \u0b99\u0b9a\uffff \u0b9c \u0b9e !\u0b9f   \u0ba3\u0ba4)(  , .\u0ba8"
1313             // 0123456789ABC.....D.....EF012.....3.....4.....5.....6.....7.....8.....9.....A.....
1314             + "0123456789:;\u0ba9\u0baa ?  \u0bae\u0baf\u0bb0\u0bb1\u0bb2\u0bb3\u0bb4\u0bb5\u0bb6"
1315             // B.....C.....D.....EF0.....1.....2.....3.....4.....5678.....9.....A.....BC.....D.....
1316             + "\u0bb7\u0bb8\u0bb9  \u0bbe\u0bbf\u0bc0\u0bc1\u0bc2   \u0bc6\u0bc7\u0bc8 \u0bca\u0bcb"
1317             // E.....F.....0.....123456789ABCDEF0123456789AB.....C.....D.....E.....F.....
1318             + "\u0bcc\u0bcd\u0bd0abcdefghijklmnopqrstuvwxyz\u0bd7\u0bf0\u0bf1\u0bf2\u0bf9",
1319 
1320         /* A.3.12 Telugu National Language Locking Shift Table
1321          0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....CD.E.....F.....0.....*/
1322         "\u0c01\u0c02\u0c03\u0c05\u0c06\u0c07\u0c08\u0c09\u0c0a\u0c0b\n\u0c0c \r\u0c0e\u0c0f\u0c10"
1323             // 12.....3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....E.....
1324             + " \u0c12\u0c13\u0c14\u0c15\u0c16\u0c17\u0c18\u0c19\u0c1a\uffff\u0c1b\u0c1c\u0c1d"
1325             // F.....012.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....0123456789AB
1326             + "\u0c1e !\u0c1f\u0c20\u0c21\u0c22\u0c23\u0c24)(\u0c25\u0c26,\u0c27.\u0c280123456789:;"
1327             // CD.....E.....F0.....1.....2.....3.....4.....5.....6.....7.....89.....A.....B.....
1328             + " \u0c2a\u0c2b?\u0c2c\u0c2d\u0c2e\u0c2f\u0c30\u0c31\u0c32\u0c33 \u0c35\u0c36\u0c37"
1329             // C.....D.....EF.....0.....1.....2.....3.....4.....5.....6.....78.....9.....A.....B
1330             + "\u0c38\u0c39 \u0c3d\u0c3e\u0c3f\u0c40\u0c41\u0c42\u0c43\u0c44 \u0c46\u0c47\u0c48 "
1331             // C.....D.....E.....F.....0.....123456789ABCDEF0123456789AB.....C.....D.....E.....
1332             + "\u0c4a\u0c4b\u0c4c\u0c4d\u0c55abcdefghijklmnopqrstuvwxyz\u0c56\u0c60\u0c61\u0c62"
1333             // F.....
1334             + "\u0c63",
1335 
1336         /* A.3.13 Urdu National Language Locking Shift Table
1337          0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.B.....C.....D.E.....F.....*/
1338         "\u0627\u0622\u0628\u067b\u0680\u067e\u06a6\u062a\u06c2\u067f\n\u0679\u067d\r\u067a\u067c"
1339             // 0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.....B.....C.....D.....
1340             + "\u062b\u062c\u0681\u0684\u0683\u0685\u0686\u0687\u062d\u062e\u062f\uffff\u068c\u0688"
1341             // E.....F.....012.....3.....4.....5.....6.....7.....89A.....B.....CD.....EF.....012345
1342             + "\u0689\u068a !\u068f\u068d\u0630\u0631\u0691\u0693)(\u0699\u0632,\u0696.\u0698012345"
1343             // 6789ABC.....D.....E.....F0.....1.....2.....3.....4.....5.....6.....7.....8.....
1344             + "6789:;\u069a\u0633\u0634?\u0635\u0636\u0637\u0638\u0639\u0641\u0642\u06a9\u06aa"
1345             // 9.....A.....B.....C.....D.....E.....F.....0.....1.....2.....3.....4.....5.....6.....
1346             + "\u06ab\u06af\u06b3\u06b1\u0644\u0645\u0646\u06ba\u06bb\u06bc\u0648\u06c4\u06d5\u06c1"
1347             // 7.....8.....9.....A.....B.....C.....D.....E.....F.....0.....123456789ABCDEF012345678
1348             + "\u06be\u0621\u06cc\u06d0\u06d2\u064d\u0650\u064f\u0657\u0654abcdefghijklmnopqrstuvwx"
1349             // 9AB.....C.....D.....E.....F.....
1350             + "yz\u0655\u0651\u0653\u0656\u0670"
1351     };
1352 
1353     /**
1354      * GSM default extension table plus national language single shift character tables.
1355      */
1356     @UnsupportedAppUsage
1357     private static final String[] sLanguageShiftTables = new String[]{
1358         /* 6.2.1.1 GSM 7 bit Default Alphabet Extension Table
1359          0123456789A.....BCDEF0123456789ABCDEF0123456789ABCDEF.0123456789ABCDEF0123456789ABCDEF */
1360         "          \u000c         ^                   {}     \\            [~] |               "
1361             // 0123456789ABCDEF012345.....6789ABCDEF0123456789ABCDEF
1362             + "                     \u20ac                          ",
1363 
1364         /* A.2.1 Turkish National Language Single Shift Table
1365          0123456789A.....BCDEF0123456789ABCDEF0123456789ABCDEF.0123456789ABCDEF01234567.....8 */
1366         "          \u000c         ^                   {}     \\            [~] |      \u011e "
1367             // 9.....ABCDEF0123.....456789ABCDEF0123.....45.....67.....89.....ABCDEF0123.....
1368             + "\u0130         \u015e               \u00e7 \u20ac \u011f \u0131         \u015f"
1369             // 456789ABCDEF
1370             + "            ",
1371 
1372         /* A.2.2 Spanish National Language Single Shift Table
1373          0123456789.....A.....BCDEF0123456789ABCDEF0123456789ABCDEF.0123456789ABCDEF01.....23 */
1374         "         \u00e7\u000c         ^                   {}     \\            [~] |\u00c1  "
1375             // 456789.....ABCDEF.....012345.....6789ABCDEF01.....2345.....6789.....ABCDEF.....012
1376             + "     \u00cd     \u00d3     \u00da           \u00e1   \u20ac   \u00ed     \u00f3   "
1377             // 345.....6789ABCDEF
1378             + "  \u00fa          ",
1379 
1380         /* A.2.3 Portuguese National Language Single Shift Table
1381          012345.....6789.....A.....B.....C.....DE.....F.....012.....3.....45.....6.....7.....8....*/
1382         "     \u00ea   \u00e7\u000c\u00d4\u00f4 \u00c1\u00e1  \u03a6\u0393^\u03a9\u03a0\u03a8\u03a3"
1383             // 9.....ABCDEF.....0123456789ABCDEF.0123456789ABCDEF01.....23456789.....ABCDE
1384             + "\u0398     \u00ca        {}     \\            [~] |\u00c0       \u00cd     "
1385             // F.....012345.....6789AB.....C.....DEF01.....2345.....6789.....ABCDEF.....01234
1386             + "\u00d3     \u00da     \u00c3\u00d5    \u00c2   \u20ac   \u00ed     \u00f3     "
1387             // 5.....6789AB.....C.....DEF.....
1388             + "\u00fa     \u00e3\u00f5  \u00e2",
1389 
1390         /* A.2.4 Bengali National Language Single Shift Table
1391          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1392         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u09e6\u09e7 \u09e8\u09e9"
1393             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....
1394             + "\u09ea\u09eb\u09ec\u09ed\u09ee\u09ef\u09df\u09e0\u09e1\u09e2{}\u09e3\u09f2\u09f3"
1395             // D.....E.....F.0.....1.....2.....3.....4.....56789ABCDEF0123456789ABCDEF
1396             + "\u09f4\u09f5\\\u09f6\u09f7\u09f8\u09f9\u09fa       [~] |ABCDEFGHIJKLMNO"
1397             // 0123456789ABCDEF012345.....6789ABCDEF0123456789ABCDEF
1398             + "PQRSTUVWXYZ          \u20ac                          ",
1399 
1400         /* A.2.5 Gujarati National Language Single Shift Table
1401          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1402         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0964\u0965 \u0ae6\u0ae7"
1403             // E.....F.....0.....1.....2.....3.....4.....5.....6789ABCDEF.0123456789ABCDEF
1404             + "\u0ae8\u0ae9\u0aea\u0aeb\u0aec\u0aed\u0aee\u0aef  {}     \\            [~] "
1405             // 0123456789ABCDEF0123456789ABCDEF012345.....6789ABCDEF0123456789ABCDEF
1406             + "|ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac                          ",
1407 
1408         /* A.2.6 Hindi National Language Single Shift Table
1409          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1410         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0964\u0965 \u0966\u0967"
1411             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....
1412             + "\u0968\u0969\u096a\u096b\u096c\u096d\u096e\u096f\u0951\u0952{}\u0953\u0954\u0958"
1413             // D.....E.....F.0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.....
1414             + "\u0959\u095a\\\u095b\u095c\u095d\u095e\u095f\u0960\u0961\u0962\u0963\u0970\u0971"
1415             // BCDEF0123456789ABCDEF0123456789ABCDEF012345.....6789ABCDEF0123456789ABCDEF
1416             + " [~] |ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac                          ",
1417 
1418         /* A.2.7 Kannada National Language Single Shift Table
1419          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1420         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0964\u0965 \u0ce6\u0ce7"
1421             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....BCDEF.01234567
1422             + "\u0ce8\u0ce9\u0cea\u0ceb\u0cec\u0ced\u0cee\u0cef\u0cde\u0cf1{}\u0cf2    \\        "
1423             // 89ABCDEF0123456789ABCDEF0123456789ABCDEF012345.....6789ABCDEF0123456789ABCDEF
1424             + "    [~] |ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac                          ",
1425 
1426         /* A.2.8 Malayalam National Language Single Shift Table
1427          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1428         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0964\u0965 \u0d66\u0d67"
1429             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....
1430             + "\u0d68\u0d69\u0d6a\u0d6b\u0d6c\u0d6d\u0d6e\u0d6f\u0d70\u0d71{}\u0d72\u0d73\u0d74"
1431             // D.....E.....F.0.....1.....2.....3.....4.....56789ABCDEF0123456789ABCDEF0123456789A
1432             + "\u0d75\u0d7a\\\u0d7b\u0d7c\u0d7d\u0d7e\u0d7f       [~] |ABCDEFGHIJKLMNOPQRSTUVWXYZ"
1433             // BCDEF012345.....6789ABCDEF0123456789ABCDEF
1434             + "          \u20ac                          ",
1435 
1436         /* A.2.9 Oriya National Language Single Shift Table
1437          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1438         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0964\u0965 \u0b66\u0b67"
1439             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....DE
1440             + "\u0b68\u0b69\u0b6a\u0b6b\u0b6c\u0b6d\u0b6e\u0b6f\u0b5c\u0b5d{}\u0b5f\u0b70\u0b71  "
1441             // F.0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF012345.....6789ABCDEF0123456789A
1442             + "\\            [~] |ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac                     "
1443             // BCDEF
1444             + "     ",
1445 
1446         /* A.2.10 Punjabi National Language Single Shift Table
1447          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1448         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0964\u0965 \u0a66\u0a67"
1449             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....
1450             + "\u0a68\u0a69\u0a6a\u0a6b\u0a6c\u0a6d\u0a6e\u0a6f\u0a59\u0a5a{}\u0a5b\u0a5c\u0a5e"
1451             // D.....EF.0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF012345.....6789ABCDEF01
1452             + "\u0a75 \\            [~] |ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac            "
1453             // 23456789ABCDEF
1454             + "              ",
1455 
1456         /* A.2.11 Tamil National Language Single Shift Table
1457            NOTE: TS 23.038 V9.1.1 shows code 0x24 as \u0bef, corrected to \u0bee (typo)
1458          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1459         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0964\u0965 \u0be6\u0be7"
1460             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....
1461             + "\u0be8\u0be9\u0bea\u0beb\u0bec\u0bed\u0bee\u0bef\u0bf3\u0bf4{}\u0bf5\u0bf6\u0bf7"
1462             // D.....E.....F.0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF012345.....6789ABC
1463             + "\u0bf8\u0bfa\\            [~] |ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac       "
1464             // DEF0123456789ABCDEF
1465             + "                   ",
1466 
1467         /* A.2.12 Telugu National Language Single Shift Table
1468            NOTE: TS 23.038 V9.1.1 shows code 0x22-0x23 as \u06cc\u06cd, corrected to \u0c6c\u0c6d
1469          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789ABC.....D.....E.....F..... */
1470         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*   \u0c66\u0c67\u0c68\u0c69"
1471             // 0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....D.....E.....F.
1472             + "\u0c6a\u0c6b\u0c6c\u0c6d\u0c6e\u0c6f\u0c58\u0c59{}\u0c78\u0c79\u0c7a\u0c7b\u0c7c\\"
1473             // 0.....1.....2.....3456789ABCDEF0123456789ABCDEF0123456789ABCDEF012345.....6789ABCD
1474             + "\u0c7d\u0c7e\u0c7f         [~] |ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac        "
1475             // EF0123456789ABCDEF
1476             + "                  ",
1477 
1478         /* A.2.13 Urdu National Language Single Shift Table
1479          01.....23.....4.....5.6.....789A.....BCDEF0123.....45.....6789.....A.....BC.....D..... */
1480         "@\u00a3$\u00a5\u00bf\"\u00a4%&'\u000c*+ -/<=>\u00a1^\u00a1_#*\u0600\u0601 \u06f0\u06f1"
1481             // E.....F.....0.....1.....2.....3.....4.....5.....6.....7.....89A.....B.....C.....
1482             + "\u06f2\u06f3\u06f4\u06f5\u06f6\u06f7\u06f8\u06f9\u060c\u060d{}\u060e\u060f\u0610"
1483             // D.....E.....F.0.....1.....2.....3.....4.....5.....6.....7.....8.....9.....A.....
1484             + "\u0611\u0612\\\u0613\u0614\u061b\u061f\u0640\u0652\u0658\u066b\u066c\u0672\u0673"
1485             // B.....CDEF.....0123456789ABCDEF0123456789ABCDEF012345.....6789ABCDEF0123456789ABCDEF
1486             + "\u06cd[~]\u06d4|ABCDEFGHIJKLMNOPQRSTUVWXYZ          \u20ac                          "
1487     };
1488 
1489     static {
1490         enableCountrySpecificEncodings();
1491         int numTables = sLanguageTables.length;
1492         int numShiftTables = sLanguageShiftTables.length;
1493         if (numTables != numShiftTables) {
1494             Rlog.e(TAG, "Error: language tables array length " + numTables +
1495                     " != shift tables array length " + numShiftTables);
1496         }
1497 
1498         sCharsToGsmTables = new SparseIntArray[numTables];
1499         for (int i = 0; i < numTables; i++) {
1500             String table = sLanguageTables[i];
1501 
1502             int tableLen = table.length();
1503             if (tableLen != 0 && tableLen != 128) {
1504                 Rlog.e(TAG, "Error: language tables index " + i +
1505                         " length " + tableLen + " (expected 128 or 0)");
1506             }
1507 
1508             SparseIntArray charToGsmTable = new SparseIntArray(tableLen);
1509             sCharsToGsmTables[i] = charToGsmTable;
1510             for (int j = 0; j < tableLen; j++) {
1511                 char c = table.charAt(j);
1512                 charToGsmTable.put(c, j);
1513             }
1514         }
1515 
1516         sCharsToShiftTables = new SparseIntArray[numTables];
1517         for (int i = 0; i < numShiftTables; i++) {
1518             String shiftTable = sLanguageShiftTables[i];
1519 
1520             int shiftTableLen = shiftTable.length();
1521             if (shiftTableLen != 0 && shiftTableLen != 128) {
1522                 Rlog.e(TAG, "Error: language shift tables index " + i +
1523                         " length " + shiftTableLen + " (expected 128 or 0)");
1524             }
1525 
1526             SparseIntArray charToShiftTable = new SparseIntArray(shiftTableLen);
1527             sCharsToShiftTables[i] = charToShiftTable;
1528             for (int j = 0; j < shiftTableLen; j++) {
1529                 char c = shiftTable.charAt(j);
1530                 if (c != ' ') {
1531                     charToShiftTable.put(c, j);
1532                 }
1533             }
1534         }
1535     }
1536 }
1537