"Fossies" - the Fresh Open Source Software Archive

Member "file-5.35/src/encoding.c" (15 Oct 2018, 18219 Bytes) of package /linux/misc/file-5.35.tar.gz:


As a special service "Fossies" has tried to format the requested source page into HTML format using (guessed) C and C++ source code syntax highlighting (style: standard) with prefixed line numbers and code folding option. Alternatively you can here view or download the uninterpreted source code file. For more information about "encoding.c" see the Fossies "Dox" file reference documentation and the latest Fossies "Diffs" side-by-side code changes report: 5.34_vs_5.35.

    1 /*
    2  * Copyright (c) Ian F. Darwin 1986-1995.
    3  * Software written by Ian F. Darwin and others;
    4  * maintained 1995-present by Christos Zoulas and others.
    5  *
    6  * Redistribution and use in source and binary forms, with or without
    7  * modification, are permitted provided that the following conditions
    8  * are met:
    9  * 1. Redistributions of source code must retain the above copyright
   10  *    notice immediately at the beginning of the file, without modification,
   11  *    this list of conditions, and the following disclaimer.
   12  * 2. Redistributions in binary form must reproduce the above copyright
   13  *    notice, this list of conditions and the following disclaimer in the
   14  *    documentation and/or other materials provided with the distribution.
   15  *
   16  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
   17  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
   18  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
   19  * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
   20  * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
   21  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
   22  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
   23  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
   24  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
   25  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
   26  * SUCH DAMAGE.
   27  */
   28 /*
   29  * Encoding -- determine the character encoding of a text file.
   30  *
   31  * Joerg Wunsch <joerg@freebsd.org> wrote the original support for 8-bit
   32  * international characters.
   33  */
   34 
   35 #include "file.h"
   36 
   37 #ifndef lint
   38 FILE_RCSID("@(#)$File: encoding.c,v 1.15 2018/10/15 16:29:16 christos Exp $")
   39 #endif  /* lint */
   40 
   41 #include "magic.h"
   42 #include <string.h>
   43 #include <memory.h>
   44 #include <stdlib.h>
   45 
   46 
   47 private int looks_ascii(const unsigned char *, size_t, unichar *, size_t *);
   48 private int looks_utf8_with_BOM(const unsigned char *, size_t, unichar *,
   49     size_t *);
   50 private int looks_utf7(const unsigned char *, size_t, unichar *, size_t *);
   51 private int looks_ucs16(const unsigned char *, size_t, unichar *, size_t *);
   52 private int looks_latin1(const unsigned char *, size_t, unichar *, size_t *);
   53 private int looks_extended(const unsigned char *, size_t, unichar *, size_t *);
   54 private void from_ebcdic(const unsigned char *, size_t, unsigned char *);
   55 
   56 #ifdef DEBUG_ENCODING
   57 #define DPRINTF(a) printf a
   58 #else
   59 #define DPRINTF(a)
   60 #endif
   61 
   62 /*
   63  * Try to determine whether text is in some character code we can
   64  * identify.  Each of these tests, if it succeeds, will leave
   65  * the text converted into one-unichar-per-character Unicode in
   66  * ubuf, and the number of characters converted in ulen.
   67  */
   68 protected int
   69 file_encoding(struct magic_set *ms, const struct buffer *b, unichar **ubuf,
   70     size_t *ulen, const char **code, const char **code_mime, const char **type)
   71 {
   72     const unsigned char *buf = CAST(const unsigned char *, b->fbuf);
   73     size_t nbytes = b->flen;
   74     size_t mlen;
   75     int rv = 1, ucs_type;
   76     unsigned char *nbuf = NULL;
   77     unichar *udefbuf;
   78     size_t udeflen;
   79 
   80     if (ubuf == NULL)
   81         ubuf = &udefbuf;
   82     if (ulen == NULL)
   83         ulen = &udeflen;
   84 
   85     *type = "text";
   86     *ulen = 0;
   87     *code = "unknown";
   88     *code_mime = "binary";
   89 
   90     mlen = (nbytes + 1) * sizeof((*ubuf)[0]);
   91     if ((*ubuf = CAST(unichar *, calloc((size_t)1, mlen))) == NULL) {
   92         file_oomem(ms, mlen);
   93         goto done;
   94     }
   95     mlen = (nbytes + 1) * sizeof(nbuf[0]);
   96     if ((nbuf = CAST(unsigned char *, calloc((size_t)1, mlen))) == NULL) {
   97         file_oomem(ms, mlen);
   98         goto done;
   99     }
  100 
  101     if (looks_ascii(buf, nbytes, *ubuf, ulen)) {
  102         if (looks_utf7(buf, nbytes, *ubuf, ulen) > 0) {
  103             DPRINTF(("utf-7 %" SIZE_T_FORMAT "u\n", *ulen));
  104             *code = "UTF-7 Unicode";
  105             *code_mime = "utf-7";
  106         } else {
  107             DPRINTF(("ascii %" SIZE_T_FORMAT "u\n", *ulen));
  108             *code = "ASCII";
  109             *code_mime = "us-ascii";
  110         }
  111     } else if (looks_utf8_with_BOM(buf, nbytes, *ubuf, ulen) > 0) {
  112         DPRINTF(("utf8/bom %" SIZE_T_FORMAT "u\n", *ulen));
  113         *code = "UTF-8 Unicode (with BOM)";
  114         *code_mime = "utf-8";
  115     } else if (file_looks_utf8(buf, nbytes, *ubuf, ulen) > 1) {
  116         DPRINTF(("utf8 %" SIZE_T_FORMAT "u\n", *ulen));
  117         *code = "UTF-8 Unicode";
  118         *code_mime = "utf-8";
  119     } else if ((ucs_type = looks_ucs16(buf, nbytes, *ubuf, ulen)) != 0) {
  120         if (ucs_type == 1) {
  121             *code = "Little-endian UTF-16 Unicode";
  122             *code_mime = "utf-16le";
  123         } else {
  124             *code = "Big-endian UTF-16 Unicode";
  125             *code_mime = "utf-16be";
  126         }
  127         DPRINTF(("ucs16 %" SIZE_T_FORMAT "u\n", *ulen));
  128     } else if (looks_latin1(buf, nbytes, *ubuf, ulen)) {
  129         DPRINTF(("latin1 %" SIZE_T_FORMAT "u\n", *ulen));
  130         *code = "ISO-8859";
  131         *code_mime = "iso-8859-1";
  132     } else if (looks_extended(buf, nbytes, *ubuf, ulen)) {
  133         DPRINTF(("extended %" SIZE_T_FORMAT "u\n", *ulen));
  134         *code = "Non-ISO extended-ASCII";
  135         *code_mime = "unknown-8bit";
  136     } else {
  137         from_ebcdic(buf, nbytes, nbuf);
  138 
  139         if (looks_ascii(nbuf, nbytes, *ubuf, ulen)) {
  140             DPRINTF(("ebcdic %" SIZE_T_FORMAT "u\n", *ulen));
  141             *code = "EBCDIC";
  142             *code_mime = "ebcdic";
  143         } else if (looks_latin1(nbuf, nbytes, *ubuf, ulen)) {
  144             DPRINTF(("ebcdic/international %" SIZE_T_FORMAT "u\n",
  145                 *ulen));
  146             *code = "International EBCDIC";
  147             *code_mime = "ebcdic";
  148         } else { /* Doesn't look like text at all */
  149             DPRINTF(("binary\n"));
  150             rv = 0;
  151             *type = "binary";
  152         }
  153     }
  154 
  155  done:
  156     free(nbuf);
  157     if (ubuf == &udefbuf)
  158         free(udefbuf);
  159 
  160     return rv;
  161 }
  162 
  163 /*
  164  * This table reflects a particular philosophy about what constitutes
  165  * "text," and there is room for disagreement about it.
  166  *
  167  * Version 3.31 of the file command considered a file to be ASCII if
  168  * each of its characters was approved by either the isascii() or
  169  * isalpha() function.  On most systems, this would mean that any
  170  * file consisting only of characters in the range 0x00 ... 0x7F
  171  * would be called ASCII text, but many systems might reasonably
  172  * consider some characters outside this range to be alphabetic,
  173  * so the file command would call such characters ASCII.  It might
  174  * have been more accurate to call this "considered textual on the
  175  * local system" than "ASCII."
  176  *
  177  * It considered a file to be "International language text" if each
  178  * of its characters was either an ASCII printing character (according
  179  * to the real ASCII standard, not the above test), a character in
  180  * the range 0x80 ... 0xFF, or one of the following control characters:
  181  * backspace, tab, line feed, vertical tab, form feed, carriage return,
  182  * escape.  No attempt was made to determine the language in which files
  183  * of this type were written.
  184  *
  185  *
  186  * The table below considers a file to be ASCII if all of its characters
  187  * are either ASCII printing characters (again, according to the X3.4
  188  * standard, not isascii()) or any of the following controls: bell,
  189  * backspace, tab, line feed, form feed, carriage return, esc, nextline.
  190  *
  191  * I include bell because some programs (particularly shell scripts)
  192  * use it literally, even though it is rare in normal text.  I exclude
  193  * vertical tab because it never seems to be used in real text.  I also
  194  * include, with hesitation, the X3.64/ECMA-43 control nextline (0x85),
  195  * because that's what the dd EBCDIC->ASCII table maps the EBCDIC newline
  196  * character to.  It might be more appropriate to include it in the 8859
  197  * set instead of the ASCII set, but it's got to be included in *something*
  198  * we recognize or EBCDIC files aren't going to be considered textual.
  199  * Some old Unix source files use SO/SI (^N/^O) to shift between Greek
  200  * and Latin characters, so these should possibly be allowed.  But they
  201  * make a real mess on VT100-style displays if they're not paired properly,
  202  * so we are probably better off not calling them text.
  203  *
  204  * A file is considered to be ISO-8859 text if its characters are all
  205  * either ASCII, according to the above definition, or printing characters
  206  * from the ISO-8859 8-bit extension, characters 0xA0 ... 0xFF.
  207  *
  208  * Finally, a file is considered to be international text from some other
  209  * character code if its characters are all either ISO-8859 (according to
  210  * the above definition) or characters in the range 0x80 ... 0x9F, which
  211  * ISO-8859 considers to be control characters but the IBM PC and Macintosh
  212  * consider to be printing characters.
  213  */
  214 
  215 #define F 0   /* character never appears in text */
  216 #define T 1   /* character appears in plain ASCII text */
  217 #define I 2   /* character appears in ISO-8859 text */
  218 #define X 3   /* character appears in non-ISO extended ASCII (Mac, IBM PC) */
  219 
  220 private char text_chars[256] = {
  221     /*                  BEL BS HT LF VT FF CR    */
  222     F, F, F, F, F, F, F, T, T, T, T, T, T, T, F, F,  /* 0x0X */
  223     /*                              ESC          */
  224     F, F, F, F, F, F, F, F, F, F, F, T, F, F, F, F,  /* 0x1X */
  225     T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x2X */
  226     T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x3X */
  227     T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x4X */
  228     T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x5X */
  229     T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x6X */
  230     T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, F,  /* 0x7X */
  231     /*            NEL                            */
  232     X, X, X, X, X, T, X, X, X, X, X, X, X, X, X, X,  /* 0x8X */
  233     X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, X,  /* 0x9X */
  234     I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xaX */
  235     I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xbX */
  236     I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xcX */
  237     I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xdX */
  238     I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xeX */
  239     I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I   /* 0xfX */
  240 };
  241 
  242 private int
  243 looks_ascii(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  244     size_t *ulen)
  245 {
  246     size_t i;
  247 
  248     *ulen = 0;
  249 
  250     for (i = 0; i < nbytes; i++) {
  251         int t = text_chars[buf[i]];
  252 
  253         if (t != T)
  254             return 0;
  255 
  256         ubuf[(*ulen)++] = buf[i];
  257     }
  258 
  259     return 1;
  260 }
  261 
  262 private int
  263 looks_latin1(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
  264 {
  265     size_t i;
  266 
  267     *ulen = 0;
  268 
  269     for (i = 0; i < nbytes; i++) {
  270         int t = text_chars[buf[i]];
  271 
  272         if (t != T && t != I)
  273             return 0;
  274 
  275         ubuf[(*ulen)++] = buf[i];
  276     }
  277 
  278     return 1;
  279 }
  280 
  281 private int
  282 looks_extended(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  283     size_t *ulen)
  284 {
  285     size_t i;
  286 
  287     *ulen = 0;
  288 
  289     for (i = 0; i < nbytes; i++) {
  290         int t = text_chars[buf[i]];
  291 
  292         if (t != T && t != I && t != X)
  293             return 0;
  294 
  295         ubuf[(*ulen)++] = buf[i];
  296     }
  297 
  298     return 1;
  299 }
  300 
  301 /*
  302  * Decide whether some text looks like UTF-8. Returns:
  303  *
  304  *     -1: invalid UTF-8
  305  *      0: uses odd control characters, so doesn't look like text
  306  *      1: 7-bit text
  307  *      2: definitely UTF-8 text (valid high-bit set bytes)
  308  *
  309  * If ubuf is non-NULL on entry, text is decoded into ubuf, *ulen;
  310  * ubuf must be big enough!
  311  */
  312 protected int
  313 file_looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
  314 {
  315     size_t i;
  316     int n;
  317     unichar c;
  318     int gotone = 0, ctrl = 0;
  319 
  320     if (ubuf)
  321         *ulen = 0;
  322 
  323     for (i = 0; i < nbytes; i++) {
  324         if ((buf[i] & 0x80) == 0) {    /* 0xxxxxxx is plain ASCII */
  325             /*
  326              * Even if the whole file is valid UTF-8 sequences,
  327              * still reject it if it uses weird control characters.
  328              */
  329 
  330             if (text_chars[buf[i]] != T)
  331                 ctrl = 1;
  332 
  333             if (ubuf)
  334                 ubuf[(*ulen)++] = buf[i];
  335         } else if ((buf[i] & 0x40) == 0) { /* 10xxxxxx never 1st byte */
  336             return -1;
  337         } else {               /* 11xxxxxx begins UTF-8 */
  338             int following;
  339 
  340             if ((buf[i] & 0x20) == 0) {     /* 110xxxxx */
  341                 c = buf[i] & 0x1f;
  342                 following = 1;
  343             } else if ((buf[i] & 0x10) == 0) {  /* 1110xxxx */
  344                 c = buf[i] & 0x0f;
  345                 following = 2;
  346             } else if ((buf[i] & 0x08) == 0) {  /* 11110xxx */
  347                 c = buf[i] & 0x07;
  348                 following = 3;
  349             } else if ((buf[i] & 0x04) == 0) {  /* 111110xx */
  350                 c = buf[i] & 0x03;
  351                 following = 4;
  352             } else if ((buf[i] & 0x02) == 0) {  /* 1111110x */
  353                 c = buf[i] & 0x01;
  354                 following = 5;
  355             } else
  356                 return -1;
  357 
  358             for (n = 0; n < following; n++) {
  359                 i++;
  360                 if (i >= nbytes)
  361                     goto done;
  362 
  363                 if ((buf[i] & 0x80) == 0 || (buf[i] & 0x40))
  364                     return -1;
  365 
  366                 c = (c << 6) + (buf[i] & 0x3f);
  367             }
  368 
  369             if (ubuf)
  370                 ubuf[(*ulen)++] = c;
  371             gotone = 1;
  372         }
  373     }
  374 done:
  375     return ctrl ? 0 : (gotone ? 2 : 1);
  376 }
  377 
  378 /*
  379  * Decide whether some text looks like UTF-8 with BOM. If there is no
  380  * BOM, return -1; otherwise return the result of looks_utf8 on the
  381  * rest of the text.
  382  */
  383 private int
  384 looks_utf8_with_BOM(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  385     size_t *ulen)
  386 {
  387     if (nbytes > 3 && buf[0] == 0xef && buf[1] == 0xbb && buf[2] == 0xbf)
  388         return file_looks_utf8(buf + 3, nbytes - 3, ubuf, ulen);
  389     else
  390         return -1;
  391 }
  392 
  393 private int
  394 looks_utf7(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
  395 {
  396     if (nbytes > 4 && buf[0] == '+' && buf[1] == '/' && buf[2] == 'v')
  397         switch (buf[3]) {
  398         case '8':
  399         case '9':
  400         case '+':
  401         case '/':
  402             if (ubuf)
  403                 *ulen = 0;
  404             return 1;
  405         default:
  406             return -1;
  407         }
  408     else
  409         return -1;
  410 }
  411 
  412 private int
  413 looks_ucs16(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  414     size_t *ulen)
  415 {
  416     int bigend;
  417     size_t i;
  418 
  419     if (nbytes < 2)
  420         return 0;
  421 
  422     if (buf[0] == 0xff && buf[1] == 0xfe)
  423         bigend = 0;
  424     else if (buf[0] == 0xfe && buf[1] == 0xff)
  425         bigend = 1;
  426     else
  427         return 0;
  428 
  429     *ulen = 0;
  430 
  431     for (i = 2; i + 1 < nbytes; i += 2) {
  432         /* XXX fix to properly handle chars > 65536 */
  433 
  434         if (bigend)
  435             ubuf[(*ulen)++] = buf[i + 1] + 256 * buf[i];
  436         else
  437             ubuf[(*ulen)++] = buf[i] + 256 * buf[i + 1];
  438 
  439         if (ubuf[*ulen - 1] == 0xfffe)
  440             return 0;
  441         if (ubuf[*ulen - 1] < 128 &&
  442             text_chars[(size_t)ubuf[*ulen - 1]] != T)
  443             return 0;
  444     }
  445 
  446     return 1 + bigend;
  447 }
  448 
  449 #undef F
  450 #undef T
  451 #undef I
  452 #undef X
  453 
  454 /*
  455  * This table maps each EBCDIC character to an (8-bit extended) ASCII
  456  * character, as specified in the rationale for the dd(1) command in
  457  * draft 11.2 (September, 1991) of the POSIX P1003.2 standard.
  458  *
  459  * Unfortunately it does not seem to correspond exactly to any of the
  460  * five variants of EBCDIC documented in IBM's _Enterprise Systems
  461  * Architecture/390: Principles of Operation_, SA22-7201-06, Seventh
  462  * Edition, July, 1999, pp. I-1 - I-4.
  463  *
  464  * Fortunately, though, all versions of EBCDIC, including this one, agree
  465  * on most of the printing characters that also appear in (7-bit) ASCII.
  466  * Of these, only '|', '!', '~', '^', '[', and ']' are in question at all.
  467  *
  468  * Fortunately too, there is general agreement that codes 0x00 through
  469  * 0x3F represent control characters, 0x41 a nonbreaking space, and the
  470  * remainder printing characters.
  471  *
  472  * This is sufficient to allow us to identify EBCDIC text and to distinguish
  473  * between old-style and internationalized examples of text.
  474  */
  475 
  476 private unsigned char ebcdic_to_ascii[] = {
  477   0,   1,   2,   3, 156,   9, 134, 127, 151, 141, 142,  11,  12,  13,  14,  15,
  478  16,  17,  18,  19, 157, 133,   8, 135,  24,  25, 146, 143,  28,  29,  30,  31,
  479 128, 129, 130, 131, 132,  10,  23,  27, 136, 137, 138, 139, 140,   5,   6,   7,
  480 144, 145,  22, 147, 148, 149, 150,   4, 152, 153, 154, 155,  20,  21, 158,  26,
  481 ' ', 160, 161, 162, 163, 164, 165, 166, 167, 168, 213, '.', '<', '(', '+', '|',
  482 '&', 169, 170, 171, 172, 173, 174, 175, 176, 177, '!', '$', '*', ')', ';', '~',
  483 '-', '/', 178, 179, 180, 181, 182, 183, 184, 185, 203, ',', '%', '_', '>', '?',
  484 186, 187, 188, 189, 190, 191, 192, 193, 194, '`', ':', '#', '@', '\'','=', '"',
  485 195, 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 196, 197, 198, 199, 200, 201,
  486 202, 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', '^', 204, 205, 206, 207, 208,
  487 209, 229, 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 210, 211, 212, '[', 214, 215,
  488 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, ']', 230, 231,
  489 '{', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 232, 233, 234, 235, 236, 237,
  490 '}', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 238, 239, 240, 241, 242, 243,
  491 '\\',159, 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 244, 245, 246, 247, 248, 249,
  492 '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 250, 251, 252, 253, 254, 255
  493 };
  494 
  495 #ifdef notdef
  496 /*
  497  * The following EBCDIC-to-ASCII table may relate more closely to reality,
  498  * or at least to modern reality.  It comes from
  499  *
  500  *   http://ftp.s390.ibm.com/products/oe/bpxqp9.html
  501  *
  502  * and maps the characters of EBCDIC code page 1047 (the code used for
  503  * Unix-derived software on IBM's 390 systems) to the corresponding
  504  * characters from ISO 8859-1.
  505  *
  506  * If this table is used instead of the above one, some of the special
  507  * cases for the NEL character can be taken out of the code.
  508  */
  509 
  510 private unsigned char ebcdic_1047_to_8859[] = {
  511 0x00,0x01,0x02,0x03,0x9C,0x09,0x86,0x7F,0x97,0x8D,0x8E,0x0B,0x0C,0x0D,0x0E,0x0F,
  512 0x10,0x11,0x12,0x13,0x9D,0x0A,0x08,0x87,0x18,0x19,0x92,0x8F,0x1C,0x1D,0x1E,0x1F,
  513 0x80,0x81,0x82,0x83,0x84,0x85,0x17,0x1B,0x88,0x89,0x8A,0x8B,0x8C,0x05,0x06,0x07,
  514 0x90,0x91,0x16,0x93,0x94,0x95,0x96,0x04,0x98,0x99,0x9A,0x9B,0x14,0x15,0x9E,0x1A,
  515 0x20,0xA0,0xE2,0xE4,0xE0,0xE1,0xE3,0xE5,0xE7,0xF1,0xA2,0x2E,0x3C,0x28,0x2B,0x7C,
  516 0x26,0xE9,0xEA,0xEB,0xE8,0xED,0xEE,0xEF,0xEC,0xDF,0x21,0x24,0x2A,0x29,0x3B,0x5E,
  517 0x2D,0x2F,0xC2,0xC4,0xC0,0xC1,0xC3,0xC5,0xC7,0xD1,0xA6,0x2C,0x25,0x5F,0x3E,0x3F,
  518 0xF8,0xC9,0xCA,0xCB,0xC8,0xCD,0xCE,0xCF,0xCC,0x60,0x3A,0x23,0x40,0x27,0x3D,0x22,
  519 0xD8,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0xAB,0xBB,0xF0,0xFD,0xFE,0xB1,
  520 0xB0,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F,0x70,0x71,0x72,0xAA,0xBA,0xE6,0xB8,0xC6,0xA4,
  521 0xB5,0x7E,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0xA1,0xBF,0xD0,0x5B,0xDE,0xAE,
  522 0xAC,0xA3,0xA5,0xB7,0xA9,0xA7,0xB6,0xBC,0xBD,0xBE,0xDD,0xA8,0xAF,0x5D,0xB4,0xD7,
  523 0x7B,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0xAD,0xF4,0xF6,0xF2,0xF3,0xF5,
  524 0x7D,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F,0x50,0x51,0x52,0xB9,0xFB,0xFC,0xF9,0xFA,0xFF,
  525 0x5C,0xF7,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0xB2,0xD4,0xD6,0xD2,0xD3,0xD5,
  526 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0xB3,0xDB,0xDC,0xD9,0xDA,0x9F
  527 };
  528 #endif
  529 
  530 /*
  531  * Copy buf[0 ... nbytes-1] into out[], translating EBCDIC to ASCII.
  532  */
  533 private void
  534 from_ebcdic(const unsigned char *buf, size_t nbytes, unsigned char *out)
  535 {
  536     size_t i;
  537 
  538     for (i = 0; i < nbytes; i++) {
  539         out[i] = ebcdic_to_ascii[buf[i]];
  540     }
  541 }