1 /* Copyright (C) 2000-2002 Free Software Foundation, Inc.
2 This file is part of the GNU C Library.
3 Contributed by Bruno Haible <haible@clisp.cons.org>, 2000.
5 The GNU C Library is free software; you can redistribute it and/or
6 modify it under the terms of the GNU Lesser General Public
7 License as published by the Free Software Foundation; either
8 version 2.1 of the License, or (at your option) any later version.
10 The GNU C Library is distributed in the hope that it will be useful,
11 but WITHOUT ANY WARRANTY; without even the implied warranty of
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 Lesser General Public License for more details.
15 You should have received a copy of the GNU Lesser General Public
16 License along with the GNU C Library; if not, write to the Free
17 Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
20 /* Create a table from CHARSET to Unicode.
21 This is a good test for CHARSET's iconv() module, in particular the
22 FROM_LOOP BODY macro. */
31 /* If nonzero, ignore conversions outside Unicode plane 0. */
34 /* Converts a byte buffer to a hexadecimal string. */
36 hexbuf (unsigned char buf[], unsigned int buflen)
43 sprintf (msg, "0x%02X", buf[0]);
46 sprintf (msg, "0x%02X%02X", buf[0], buf[1]);
49 sprintf (msg, "0x%02X%02X%02X", buf[0], buf[1], buf[2]);
52 sprintf (msg, "0x%02X%02X%02X%02X", buf[0], buf[1], buf[2], buf[3]);
60 /* Attempts to convert a byte buffer BUF (BUFLEN bytes) to OUT (12 bytes)
61 using the conversion descriptor CD. Returns the number of written bytes,
62 or 0 if ambiguous, or -1 if invalid. */
64 try (iconv_t cd, unsigned char buf[], unsigned int buflen, unsigned char *out)
66 const char *inbuf = (const char *) buf;
67 size_t inbytesleft = buflen;
68 char *outbuf = (char *) out;
69 size_t outbytesleft = 12;
72 iconv (cd, NULL, NULL, NULL, NULL);
73 result = iconv (cd, (char **) &inbuf, &inbytesleft, &outbuf, &outbytesleft);
74 if (result != (size_t)(-1))
75 result = iconv (cd, NULL, NULL, &outbuf, &outbytesleft);
77 if (result == (size_t)(-1))
83 else if (errno == EINVAL)
89 int saved_errno = errno;
90 fprintf (stderr, "%s: iconv error: ", hexbuf (buf, buflen));
100 fprintf (stderr, "%s: inbytes = %ld, outbytes = %ld\n",
101 hexbuf (buf, buflen),
102 (long) (buflen - inbytesleft),
103 (long) (12 - outbytesleft));
106 return 12 - outbytesleft;
110 /* Returns the out[] buffer as a Unicode value, formatted as 0x%04X. */
112 utf8_decode (const unsigned char *out, unsigned int outlen)
114 static char hexbuf[84];
124 sprintf (p, "0x%04X", out[0]);
125 out += 1; outlen -= 1;
127 else if (out[0] >= 0xc0 && out[0] < 0xe0 && outlen >= 2)
129 sprintf (p, "0x%04X", ((out[0] & 0x1f) << 6) + (out[1] & 0x3f));
130 out += 2; outlen -= 2;
132 else if (out[0] >= 0xe0 && out[0] < 0xf0 && outlen >= 3)
134 sprintf (p, "0x%04X", ((out[0] & 0x0f) << 12)
135 + ((out[1] & 0x3f) << 6) + (out[2] & 0x3f));
136 out += 3; outlen -= 3;
138 else if (out[0] >= 0xf0 && out[0] < 0xf8 && outlen >= 4)
140 sprintf (p, "0x%04X", ((out[0] & 0x07) << 18)
141 + ((out[1] & 0x3f) << 12)
142 + ((out[2] & 0x3f) << 6) + (out[3] & 0x3f));
143 out += 4; outlen -= 4;
145 else if (out[0] >= 0xf8 && out[0] < 0xfc && outlen >= 5)
147 sprintf (p, "0x%04X", ((out[0] & 0x03) << 24)
148 + ((out[1] & 0x3f) << 18)
149 + ((out[2] & 0x3f) << 12)
150 + ((out[3] & 0x3f) << 6) + (out[4] & 0x3f));
151 out += 5; outlen -= 5;
153 else if (out[0] >= 0xfc && out[0] < 0xfe && outlen >= 6)
155 sprintf (p, "0x%04X", ((out[0] & 0x01) << 30)
156 + ((out[1] & 0x3f) << 24)
157 + ((out[2] & 0x3f) << 18)
158 + ((out[3] & 0x3f) << 12)
159 + ((out[4] & 0x3f) << 6) + (out[5] & 0x3f));
160 out += 6; outlen -= 6;
164 sprintf (p, "0x????");
165 out += 1; outlen -= 1;
168 if (bmp_only && strlen (p) > 6)
169 /* Ignore conversions outside Unicode plane 0. */
179 main (int argc, char *argv[])
187 fprintf (stderr, "Usage: tst-table-from charset\n");
192 cd = iconv_open ("UTF-8", charset);
193 if (cd == (iconv_t)(-1))
195 perror ("iconv_open");
199 /* When testing UTF-8 or GB18030, stop at 0x10000, otherwise the output
200 file gets too big. */
201 bmp_only = (strcmp (charset, "UTF-8") == 0
202 || strcmp (charset, "GB18030") == 0);
203 search_depth = (strcmp (charset, "UTF-8") == 0 ? 3 : 4);
206 unsigned char out[12];
207 unsigned char buf[4];
208 unsigned int i0, i1, i2, i3;
211 for (i0 = 0; i0 < 0x100; i0++)
214 result = try (cd, buf, 1, out);
220 const char *unicode = utf8_decode (out, result);
222 printf ("0x%02X\t%s\n", i0, unicode);
226 for (i1 = 0; i1 < 0x100; i1++)
229 result = try (cd, buf, 2, out);
235 const char *unicode = utf8_decode (out, result);
237 printf ("0x%02X%02X\t%s\n", i0, i1, unicode);
241 for (i2 = 0; i2 < 0x100; i2++)
244 result = try (cd, buf, 3, out);
250 const char *unicode = utf8_decode (out, result);
252 printf ("0x%02X%02X%02X\t%s\n",
253 i0, i1, i2, unicode);
255 else if (search_depth > 3)
257 for (i3 = 0; i3 < 0x100; i3++)
260 result = try (cd, buf, 4, out);
266 const char *unicode =
267 utf8_decode (out, result);
269 printf ("0x%02X%02X%02X%02X\t%s\n",
270 i0, i1, i2, i3, unicode);
275 "%s: incomplete byte sequence\n",
288 if (iconv_close (cd) < 0)
290 perror ("iconv_close");
294 if (ferror (stdin) || fflush (stdout) || ferror (stdout))
296 fprintf (stderr, "I/O error\n");