jdk/src/share/npt/utf.c
changeset 24991 e10a57d6e9d0
parent 24990 c9ca62c6a329
parent 24989 7eefcdab7678
child 24998 0de5415b1d2a
equal deleted inserted replaced
24990:c9ca62c6a329 24991:e10a57d6e9d0
     1 /*
       
     2  * Copyright (c) 2004, 2013, Oracle and/or its affiliates. All rights reserved.
       
     3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
       
     4  *
       
     5  * This code is free software; you can redistribute it and/or modify it
       
     6  * under the terms of the GNU General Public License version 2 only, as
       
     7  * published by the Free Software Foundation.  Oracle designates this
       
     8  * particular file as subject to the "Classpath" exception as provided
       
     9  * by Oracle in the LICENSE file that accompanied this code.
       
    10  *
       
    11  * This code is distributed in the hope that it will be useful, but WITHOUT
       
    12  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
       
    13  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
       
    14  * version 2 for more details (a copy is included in the LICENSE file that
       
    15  * accompanied this code).
       
    16  *
       
    17  * You should have received a copy of the GNU General Public License version
       
    18  * 2 along with this work; if not, write to the Free Software Foundation,
       
    19  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
       
    20  *
       
    21  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
       
    22  * or visit www.oracle.com if you need additional information or have any
       
    23  * questions.
       
    24  */
       
    25 
       
    26 /* Misc functions for conversion of Unicode and UTF-8 and platform encoding */
       
    27 
       
    28 #include <stdio.h>
       
    29 #include <stddef.h>
       
    30 #include <stdlib.h>
       
    31 #include <stdarg.h>
       
    32 #include <string.h>
       
    33 #include <ctype.h>
       
    34 
       
    35 #include "jni.h"
       
    36 
       
    37 #include "utf.h"
       
    38 
       
    39 /*
       
    40  * Error handler
       
    41  */
       
    42 void
       
    43 utfError(char *file, int line, char *message)
       
    44 {
       
    45     (void)fprintf(stderr, "UTF ERROR [\"%s\":%d]: %s\n", file, line, message);
       
    46     abort();
       
    47 }
       
    48 
       
    49 /*
       
    50  * Convert UTF-8 to UTF-16
       
    51  *    Returns length or -1 if output overflows.
       
    52  */
       
    53 int JNICALL
       
    54 utf8ToUtf16(struct UtfInst *ui, jbyte *utf8, int len, unsigned short *output, int outputMaxLen)
       
    55 {
       
    56     int outputLen;
       
    57     int i;
       
    58 
       
    59     UTF_ASSERT(utf8);
       
    60     UTF_ASSERT(len>=0);
       
    61     UTF_ASSERT(output);
       
    62     UTF_ASSERT(outputMaxLen>0);
       
    63 
       
    64     i = 0;
       
    65     outputLen = 0;
       
    66     while ( i<len ) {
       
    67         unsigned code, x, y, z;
       
    68 
       
    69         if ( outputLen >= outputMaxLen ) {
       
    70             return -1;
       
    71         }
       
    72         x = (unsigned char)utf8[i++];
       
    73         code = x;
       
    74         if ( (x & 0xE0)==0xE0 ) {
       
    75             y = (unsigned char)utf8[i++];
       
    76             z = (unsigned char)utf8[i++];
       
    77             code = ((x & 0xF)<<12) + ((y & 0x3F)<<6) + (z & 0x3F);
       
    78         } else if ( (x & 0xC0)==0xC0 ) {
       
    79             y = (unsigned char)utf8[i++];
       
    80             code = ((x & 0x1F)<<6) + (y & 0x3F);
       
    81         }
       
    82         output[outputLen++] = code;
       
    83     }
       
    84     return outputLen;
       
    85 }
       
    86 
       
    87 /*
       
    88  * Convert UTF-16 to UTF-8 Modified
       
    89  *    Returns length or -1 if output overflows.
       
    90  */
       
    91 int JNICALL
       
    92 utf16ToUtf8m(struct UtfInst *ui, unsigned short *utf16, int len, jbyte *output, int outputMaxLen)
       
    93 {
       
    94     int i;
       
    95     int outputLen;
       
    96 
       
    97     UTF_ASSERT(utf16);
       
    98     UTF_ASSERT(len>=0);
       
    99     UTF_ASSERT(output);
       
   100     UTF_ASSERT(outputMaxLen>0);
       
   101 
       
   102     outputLen = 0;
       
   103     for (i = 0; i < len; i++) {
       
   104         unsigned code;
       
   105 
       
   106         code = utf16[i];
       
   107         if ( code >= 0x0001 && code <= 0x007F ) {
       
   108             if ( outputLen + 1 >= outputMaxLen ) {
       
   109                 return -1;
       
   110             }
       
   111             output[outputLen++] = code;
       
   112         } else if ( code == 0 || ( code >= 0x0080 && code <= 0x07FF ) ) {
       
   113             if ( outputLen + 2 >= outputMaxLen ) {
       
   114                 return -1;
       
   115             }
       
   116             output[outputLen++] = ((code>>6) & 0x1F) | 0xC0;
       
   117             output[outputLen++] = (code & 0x3F) | 0x80;
       
   118         } else if ( code >= 0x0800 && code <= 0xFFFF ) {
       
   119             if ( outputLen + 3 >= outputMaxLen ) {
       
   120                 return -1;
       
   121             }
       
   122             output[outputLen++] = ((code>>12) & 0x0F) | 0xE0;
       
   123             output[outputLen++] = ((code>>6) & 0x3F) | 0x80;
       
   124             output[outputLen++] = (code & 0x3F) | 0x80;
       
   125         }
       
   126     }
       
   127     output[outputLen] = 0;
       
   128     return outputLen;
       
   129 }
       
   130 
       
   131 int JNICALL
       
   132 utf16ToUtf8s(struct UtfInst *ui, unsigned short *utf16, int len, jbyte *output, int outputMaxLen)
       
   133 {
       
   134     return -1; /* FIXUP */
       
   135 }
       
   136 
       
   137 /* Determine length of this Standard UTF-8 in Modified UTF-8.
       
   138  *    Validation is done of the basic UTF encoding rules, returns
       
   139  *    length (no change) when errors are detected in the UTF encoding.
       
   140  *
       
   141  *    Note: Accepts Modified UTF-8 also, no verification on the
       
   142  *          correctness of Standard UTF-8 is done. e,g, 0xC080 input is ok.
       
   143  */
       
   144 int JNICALL
       
   145 utf8sToUtf8mLength(struct UtfInst *ui, jbyte *string, int length)
       
   146 {
       
   147     int newLength;
       
   148     int i;
       
   149 
       
   150     newLength = 0;
       
   151     for ( i = 0 ; i < length ; i++ ) {
       
   152         unsigned byte;
       
   153 
       
   154         byte = (unsigned char)string[i];
       
   155         if ( (byte & 0x80) == 0 ) { /* 1byte encoding */
       
   156             newLength++;
       
   157             if ( byte == 0 ) {
       
   158                 newLength++; /* We gain one byte in length on NULL bytes */
       
   159             }
       
   160         } else if ( (byte & 0xE0) == 0xC0 ) { /* 2byte encoding */
       
   161             /* Check encoding of following bytes */
       
   162             if ( (i+1) >= length || (string[i+1] & 0xC0) != 0x80 ) {
       
   163                 break; /* Error condition */
       
   164             }
       
   165             i++; /* Skip next byte */
       
   166             newLength += 2;
       
   167         } else if ( (byte & 0xF0) == 0xE0 ) { /* 3byte encoding */
       
   168             /* Check encoding of following bytes */
       
   169             if ( (i+2) >= length || (string[i+1] & 0xC0) != 0x80
       
   170                                  || (string[i+2] & 0xC0) != 0x80 ) {
       
   171                 break; /* Error condition */
       
   172             }
       
   173             i += 2; /* Skip next two bytes */
       
   174             newLength += 3;
       
   175         } else if ( (byte & 0xF8) == 0xF0 ) { /* 4byte encoding */
       
   176             /* Check encoding of following bytes */
       
   177             if ( (i+3) >= length || (string[i+1] & 0xC0) != 0x80
       
   178                                  || (string[i+2] & 0xC0) != 0x80
       
   179                                  || (string[i+3] & 0xC0) != 0x80 ) {
       
   180                 break; /* Error condition */
       
   181             }
       
   182             i += 3; /* Skip next 3 bytes */
       
   183             newLength += 6; /* 4byte encoding turns into 2 3byte ones */
       
   184         } else {
       
   185             break; /* Error condition */
       
   186         }
       
   187     }
       
   188     if ( i != length ) {
       
   189         /* Error in finding new length, return old length so no conversion */
       
   190         /* FIXUP: ERROR_MESSAGE? */
       
   191         return length;
       
   192     }
       
   193     return newLength;
       
   194 }
       
   195 
       
   196 /* Convert Standard UTF-8 to Modified UTF-8.
       
   197  *    Assumes the UTF-8 encoding was validated by utf8mLength() above.
       
   198  *
       
   199  *    Note: Accepts Modified UTF-8 also, no verification on the
       
   200  *          correctness of Standard UTF-8 is done. e,g, 0xC080 input is ok.
       
   201  */
       
   202 void JNICALL
       
   203 utf8sToUtf8m(struct UtfInst *ui, jbyte *string, int length, jbyte *newString, int newLength)
       
   204 {
       
   205     int i;
       
   206     int j;
       
   207 
       
   208     j = 0;
       
   209     for ( i = 0 ; i < length ; i++ ) {
       
   210         unsigned byte1;
       
   211 
       
   212         byte1 = (unsigned char)string[i];
       
   213 
       
   214         /* NULL bytes and bytes starting with 11110xxx are special */
       
   215         if ( (byte1 & 0x80) == 0 ) { /* 1byte encoding */
       
   216             if ( byte1 == 0 ) {
       
   217                 /* Bits out: 11000000 10000000 */
       
   218                 newString[j++] = (jbyte)0xC0;
       
   219                 newString[j++] = (jbyte)0x80;
       
   220             } else {
       
   221                 /* Single byte */
       
   222                 newString[j++] = byte1;
       
   223             }
       
   224         } else if ( (byte1 & 0xE0) == 0xC0 ) { /* 2byte encoding */
       
   225             newString[j++] = byte1;
       
   226             newString[j++] = string[++i];
       
   227         } else if ( (byte1 & 0xF0) == 0xE0 ) { /* 3byte encoding */
       
   228             newString[j++] = byte1;
       
   229             newString[j++] = string[++i];
       
   230             newString[j++] = string[++i];
       
   231         } else if ( (byte1 & 0xF8) == 0xF0 ) { /* 4byte encoding */
       
   232             /* Beginning of 4byte encoding, turn into 2 3byte encodings */
       
   233             unsigned byte2, byte3, byte4, u21;
       
   234 
       
   235             /* Bits in: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
       
   236             byte2 = (unsigned char)string[++i];
       
   237             byte3 = (unsigned char)string[++i];
       
   238             byte4 = (unsigned char)string[++i];
       
   239             /* Reconstruct full 21bit value */
       
   240             u21  = (byte1 & 0x07) << 18;
       
   241             u21 += (byte2 & 0x3F) << 12;
       
   242             u21 += (byte3 & 0x3F) << 6;
       
   243             u21 += (byte4 & 0x3F);
       
   244             /* Bits out: 11101101 1010xxxx 10xxxxxx */
       
   245             newString[j++] = (jbyte)0xED;
       
   246             newString[j++] = (jbyte)(0xA0 + (((u21 >> 16) - 1) & 0x0F));
       
   247             newString[j++] = (jbyte)(0x80 + ((u21 >> 10) & 0x3F));
       
   248             /* Bits out: 11101101 1011xxxx 10xxxxxx */
       
   249             newString[j++] = (jbyte)0xED;
       
   250             newString[j++] = (jbyte)(0xB0 + ((u21 >>  6) & 0x0F));
       
   251             newString[j++] = byte4;
       
   252         }
       
   253     }
       
   254     UTF_ASSERT(i==length);
       
   255     UTF_ASSERT(j==newLength);
       
   256     newString[j] = (jbyte)0;
       
   257 }
       
   258 
       
   259 /* Given a Modified UTF-8 string, calculate the Standard UTF-8 length.
       
   260  *   Basic validation of the UTF encoding rules is done, and length is
       
   261  *   returned (no change) when errors are detected.
       
   262  *
       
   263  *   Note: No validation is made that this is indeed Modified UTF-8 coming in.
       
   264  *
       
   265  */
       
   266 int JNICALL
       
   267 utf8mToUtf8sLength(struct UtfInst *ui, jbyte *string, int length)
       
   268 {
       
   269     int newLength;
       
   270     int i;
       
   271 
       
   272     newLength = 0;
       
   273     for ( i = 0 ; i < length ; i++ ) {
       
   274         unsigned byte1, byte2, byte3, byte4, byte5, byte6;
       
   275 
       
   276         byte1 = (unsigned char)string[i];
       
   277         if ( (byte1 & 0x80) == 0 ) { /* 1byte encoding */
       
   278             newLength++;
       
   279         } else if ( (byte1 & 0xE0) == 0xC0 ) { /* 2byte encoding */
       
   280             /* Check encoding of following bytes */
       
   281             if ( (i+1) >= length || (string[i+1] & 0xC0) != 0x80 ) {
       
   282                 break; /* Error condition */
       
   283             }
       
   284             byte2 = (unsigned char)string[++i];
       
   285             if ( byte1 != 0xC0 || byte2 != 0x80 ) {
       
   286                 newLength += 2; /* Normal 2byte encoding, not 0xC080 */
       
   287             } else {
       
   288                 newLength++;    /* We will turn 0xC080 into 0 */
       
   289             }
       
   290         } else if ( (byte1 & 0xF0) == 0xE0 ) { /* 3byte encoding */
       
   291             /* Check encoding of following bytes */
       
   292             if ( (i+2) >= length || (string[i+1] & 0xC0) != 0x80
       
   293                                  || (string[i+2] & 0xC0) != 0x80 ) {
       
   294                 break; /* Error condition */
       
   295             }
       
   296             byte2 = (unsigned char)string[++i];
       
   297             byte3 = (unsigned char)string[++i];
       
   298             newLength += 3;
       
   299             /* Possible process a second 3byte encoding */
       
   300             if ( (i+3) < length && byte1 == 0xED && (byte2 & 0xF0) == 0xA0 ) {
       
   301                 /* See if this is a pair of 3byte encodings */
       
   302                 byte4 = (unsigned char)string[i+1];
       
   303                 byte5 = (unsigned char)string[i+2];
       
   304                 byte6 = (unsigned char)string[i+3];
       
   305                 if ( byte4 == 0xED && (byte5 & 0xF0) == 0xB0 ) {
       
   306                     /* Check encoding of 3rd byte */
       
   307                     if ( (byte6 & 0xC0) != 0x80 ) {
       
   308                         break; /* Error condition */
       
   309                     }
       
   310                     newLength++; /* New string will have 4byte encoding */
       
   311                     i += 3;       /* Skip next 3 bytes */
       
   312                 }
       
   313             }
       
   314         } else {
       
   315             break; /* Error condition */
       
   316         }
       
   317     }
       
   318     if ( i != length ) {
       
   319         /* Error in UTF encoding */
       
   320         /*  FIXUP: ERROR_MESSAGE()? */
       
   321         return length;
       
   322     }
       
   323     return newLength;
       
   324 }
       
   325 
       
   326 /* Convert a Modified UTF-8 string into a Standard UTF-8 string
       
   327  *   It is assumed that this string has been validated in terms of the
       
   328  *   basic UTF encoding rules by utf8Length() above.
       
   329  *
       
   330  *   Note: No validation is made that this is indeed Modified UTF-8 coming in.
       
   331  *
       
   332  */
       
   333 void JNICALL
       
   334 utf8mToUtf8s(struct UtfInst *ui, jbyte *string, int length, jbyte *newString, int newLength)
       
   335 {
       
   336     int i;
       
   337     int j;
       
   338 
       
   339     j = 0;
       
   340     for ( i = 0 ; i < length ; i++ ) {
       
   341         unsigned byte1, byte2, byte3, byte4, byte5, byte6;
       
   342 
       
   343         byte1 = (unsigned char)string[i];
       
   344         if ( (byte1 & 0x80) == 0 ) { /* 1byte encoding */
       
   345             /* Single byte */
       
   346             newString[j++] = byte1;
       
   347         } else if ( (byte1 & 0xE0) == 0xC0 ) { /* 2byte encoding */
       
   348             byte2 = (unsigned char)string[++i];
       
   349             if ( byte1 != 0xC0 || byte2 != 0x80 ) {
       
   350                 newString[j++] = byte1;
       
   351                 newString[j++] = byte2;
       
   352             } else {
       
   353                 newString[j++] = 0;
       
   354             }
       
   355         } else if ( (byte1 & 0xF0) == 0xE0 ) { /* 3byte encoding */
       
   356             byte2 = (unsigned char)string[++i];
       
   357             byte3 = (unsigned char)string[++i];
       
   358             if ( i+3 < length && byte1 == 0xED && (byte2 & 0xF0) == 0xA0 ) {
       
   359                 /* See if this is a pair of 3byte encodings */
       
   360                 byte4 = (unsigned char)string[i+1];
       
   361                 byte5 = (unsigned char)string[i+2];
       
   362                 byte6 = (unsigned char)string[i+3];
       
   363                 if ( byte4 == 0xED && (byte5 & 0xF0) == 0xB0 ) {
       
   364                     unsigned u21;
       
   365 
       
   366                     /* Bits in: 11101101 1010xxxx 10xxxxxx */
       
   367                     /* Bits in: 11101101 1011xxxx 10xxxxxx */
       
   368                     i += 3;
       
   369 
       
   370                     /* Reconstruct 21 bit code */
       
   371                     u21  = ((byte2 & 0x0F) + 1) << 16;
       
   372                     u21 += (byte3 & 0x3F) << 10;
       
   373                     u21 += (byte5 & 0x0F) << 6;
       
   374                     u21 += (byte6 & 0x3F);
       
   375 
       
   376                     /* Bits out: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
       
   377 
       
   378                     /* Convert to 4byte encoding */
       
   379                     newString[j++] = 0xF0 + ((u21 >> 18) & 0x07);
       
   380                     newString[j++] = 0x80 + ((u21 >> 12) & 0x3F);
       
   381                     newString[j++] = 0x80 + ((u21 >>  6) & 0x3F);
       
   382                     newString[j++] = 0x80 + (u21 & 0x3F);
       
   383                     continue;
       
   384                 }
       
   385             }
       
   386             /* Normal 3byte encoding */
       
   387             newString[j++] = byte1;
       
   388             newString[j++] = byte2;
       
   389             newString[j++] = byte3;
       
   390         }
       
   391     }
       
   392     UTF_ASSERT(i==length);
       
   393     UTF_ASSERT(j==newLength);
       
   394     newString[j] = 0;
       
   395 }
       
   396 
       
   397 /* ================================================================= */
       
   398 
       
   399 #ifdef COMPILE_WITH_UTF_TEST  /* Test program */
       
   400 
       
   401 /*
       
   402  * Convert any byte array into a printable string.
       
   403  *    Returns length or -1 if output overflows.
       
   404  */
       
   405 static int
       
   406 bytesToPrintable(struct UtfInst *ui, char *bytes, int len, char *output, int outputMaxLen)
       
   407 {
       
   408     int outputLen;
       
   409     int i;
       
   410 
       
   411     UTF_ASSERT(bytes);
       
   412     UTF_ASSERT(len>=0);
       
   413     UTF_ASSERT(output);
       
   414     UTF_ASSERT(outputMaxLen>=0);
       
   415 
       
   416     outputLen = 0;
       
   417     for ( i=0; i<len ; i++ ) {
       
   418         unsigned byte;
       
   419 
       
   420         byte = bytes[i];
       
   421         if ( byte <= 0x7f && isprint(byte) && !iscntrl(byte) ) {
       
   422             if ( outputLen + 1 >= outputMaxLen ) {
       
   423                 return -1;
       
   424             }
       
   425             output[outputLen++] = (char)byte;
       
   426         } else {
       
   427             if ( outputLen + 4 >= outputMaxLen ) {
       
   428                 return -1;
       
   429             }
       
   430             (void)sprintf(output+outputLen,"\\x%02x",byte);
       
   431             outputLen += 4;
       
   432         }
       
   433     }
       
   434     output[outputLen] = 0;
       
   435     return outputLen;
       
   436 }
       
   437 
       
   438 static void
       
   439 test(void)
       
   440 {
       
   441     static char *strings[] = {
       
   442                 "characters",
       
   443                 "abcdefghijklmnopqrstuvwxyz",
       
   444                 "0123456789",
       
   445                 "!@#$%^&*()_+=-{}[]:;",
       
   446                 NULL };
       
   447     int i;
       
   448     struct UtfInst *ui;
       
   449 
       
   450     ui = utfInitialize(NULL);
       
   451 
       
   452     i = 0;
       
   453     while ( strings[i] != NULL ) {
       
   454         char *str;
       
   455         #define MAX 1024
       
   456         char buf0[MAX];
       
   457         char buf1[MAX];
       
   458         char buf2[MAX];
       
   459         unsigned short buf3[MAX];
       
   460         int len1;
       
   461         int len2;
       
   462         int len3;
       
   463 
       
   464         str = strings[i];
       
   465 
       
   466         (void)bytesToPrintable(ui, str, (int)strlen(str), buf0, 1024);
       
   467 
       
   468         len1 = utf8FromPlatform(ui, str, (int)strlen(str), (jbyte*)buf1, 1024);
       
   469 
       
   470         UTF_ASSERT(len1==(int)strlen(str));
       
   471 
       
   472         len3 = utf8ToUtf16(ui, (jbyte*)buf1, len1, (jchar*)buf3, 1024);
       
   473 
       
   474         UTF_ASSERT(len3==len1);
       
   475 
       
   476         len1 = utf16ToUtf8m(ui, (jchar*)buf3, len3, (jbyte*)buf1, 1024);
       
   477 
       
   478         UTF_ASSERT(len1==len3);
       
   479         UTF_ASSERT(strcmp(str, buf1) == 0);
       
   480 
       
   481         len2 = utf8ToPlatform(ui, (jbyte*)buf1, len1, buf2, 1024);
       
   482 
       
   483         UTF_ASSERT(len2==len1);
       
   484         UTF_ASSERT(strcmp(str, buf2) == 0);
       
   485 
       
   486         i++;
       
   487     }
       
   488 
       
   489     utfTerminate(ui, NULL);
       
   490 
       
   491 }
       
   492 
       
   493 int
       
   494 main(int argc, char **argv)
       
   495 {
       
   496     test();
       
   497     return 0;
       
   498 }
       
   499 
       
   500 #endif