mdw@git.distorted.org.uk Git - disorder/blob - lib/test.c

   1 /*
   2  * This file is part of DisOrder.
   3  * Copyright (C) 2005, 2007 Richard Kettlewell
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful, but
  11  * WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  * General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License
  16  * along with this program; if not, write to the Free Software
  17  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
  18  * USA
  19  */
  20 /** @file lib/test.c @brief Library tests */
  21
  22 #include <config.h>
  23 #include "types.h"
  24
  25 #include <stdio.h>
  26 #include <string.h>
  27 #include <stdlib.h>
  28 #include <errno.h>
  29 #include <ctype.h>
  30 #include <assert.h>
  31 #include <sys/types.h>
  32 #include <sys/stat.h>
  33 #include <unistd.h>
  34
  35 #include "utf8.h"
  36 #include "mem.h"
  37 #include "log.h"
  38 #include "vector.h"
  39 #include "charset.h"
  40 #include "mime.h"
  41 #include "hex.h"
  42 #include "heap.h"
  43 #include "unicode.h"
  44 #include "inputline.h"
  45 #include "wstat.h"
  46 #include "signame.h"
  47 #include "cache.h"
  48
  49 static int tests, errors;
  50 static int fail_first;
  51
  52 static void count_error() {
  53   ++errors;
  54   if(fail_first)
  55     abort();
  56 }
  57
  58 /** @brief Checks that @p expr is nonzero */
  59 #define insist(expr) do {                               \
  60   if(!(expr)) {                                         \
  61     count_error();                                              \
  62     fprintf(stderr, "%s:%d: error checking %s\n",       \
  63             __FILE__, __LINE__, #expr);                 \
  64   }                                                     \
  65   ++tests;                                              \
  66 } while(0)
  67
  68 static const char *format(const char *s) {
  69   struct dynstr d;
  70   int c;
  71   char buf[10];
  72
  73   dynstr_init(&d);
  74   while((c = (unsigned char)*s++)) {
  75     if(c >= ' ' && c <= '~')
  76       dynstr_append(&d, c);
  77     else {
  78       sprintf(buf, "\\x%02X", (unsigned)c);
  79       dynstr_append_string(&d, buf);
  80     }
  81   }
  82   dynstr_terminate(&d);
  83   return d.vec;
  84 }
  85
  86 static const char *format_utf32(const uint32_t *s) {
  87   struct dynstr d;
  88   uint32_t c;
  89   char buf[64];
  90
  91   dynstr_init(&d);
  92   while((c = *s++)) {
  93     sprintf(buf, " %04lX", (long)c);
  94     dynstr_append_string(&d, buf);
  95   }
  96   dynstr_terminate(&d);
  97   return d.vec;
  98 }
  99
 100 #define check_string(GOT, WANT) do {                            \
 101   const char *g = GOT;                                          \
 102   const char *w = WANT;                                         \
 103                                                                 \
 104   if(w == 0) {                                                  \
 105     fprintf(stderr, "%s:%d: %s returned 0\n",                   \
 106             __FILE__, __LINE__, #GOT);                          \
 107     count_error();                                                      \
 108   } else if(strcmp(w, g)) {                                     \
 109     fprintf(stderr, "%s:%d: %s returned:\n%s\nexpected:\n%s\n", \
 110             __FILE__, __LINE__, #GOT, format(g), format(w));    \
 111     count_error();                                                      \
 112   }                                                             \
 113   ++tests;                                                      \
 114  } while(0)
 115
 116 static uint32_t *ucs4parse(const char *s) {
 117   struct dynstr_ucs4 d;
 118   char *e;
 119
 120   dynstr_ucs4_init(&d);
 121   while(*s) {
 122     errno = 0;
 123     dynstr_ucs4_append(&d, strtoul(s, &e, 0));
 124     if(errno) fatal(errno, "strtoul (%s)", s);
 125     s = e;
 126   }
 127   dynstr_ucs4_terminate(&d);
 128   return d.vec;
 129 }
 130
 131 static void test_utf8(void) {
 132   /* Test validutf8, convert to UCS-4, check the answer is right,
 133    * convert back to UTF-8, check we got to where we started */
 134 #define U8(CHARS, WORDS) do {                   \
 135   uint32_t *w = ucs4parse(WORDS);               \
 136   uint32_t *ucs;                                \
 137   char *u8;                                     \
 138                                                 \
 139   insist(validutf8(CHARS));                     \
 140   ucs = utf8_to_utf32(CHARS, strlen(CHARS), 0); \
 141   insist(ucs != 0);                             \
 142   insist(!utf32_cmp(w, ucs));                   \
 143   u8 = utf32_to_utf8(ucs, utf32_len(ucs), 0);   \
 144   insist(u8 != 0);                              \
 145   insist(!strcmp(u8, CHARS));                   \
 146 } while(0)
 147
 148   fprintf(stderr, "test_utf8\n");
 149 #define validutf8(S) utf8_valid((S), strlen(S))
 150
 151   /* empty string */
 152
 153   U8("", "");
 154
 155   /* ASCII characters */
 156
 157   U8(" !\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{|}~",
 158      "0x20 0x21 0x22 0x23 0x24 0x25 0x26 0x27 0x28 0x29 0x2a 0x2b 0x2c 0x2d "
 159      "0x2e 0x2f 0x30 0x31 0x32 0x33 0x34 0x35 0x36 0x37 0x38 0x39 0x3a "
 160      "0x3b 0x3c 0x3d 0x3e 0x3f 0x40 0x41 0x42 0x43 0x44 0x45 0x46 0x47 "
 161      "0x48 0x49 0x4a 0x4b 0x4c 0x4d 0x4e 0x4f 0x50 0x51 0x52 0x53 0x54 "
 162      "0x55 0x56 0x57 0x58 0x59 0x5a 0x5b 0x5c 0x5d 0x5e 0x5f 0x60 0x61 "
 163      "0x62 0x63 0x64 0x65 0x66 0x67 0x68 0x69 0x6a 0x6b 0x6c 0x6d 0x6e "
 164      "0x6f 0x70 0x71 0x72 0x73 0x74 0x75 0x76 0x77 0x78 0x79 0x7a 0x7b "
 165      "0x7c 0x7d 0x7e");
 166   U8("\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037\177",
 167      "0x1 0x2 0x3 0x4 0x5 0x6 0x7 0x8 0x9 0xa 0xb 0xc 0xd 0xe 0xf 0x10 "
 168      "0x11 0x12 0x13 0x14 0x15 0x16 0x17 0x18 0x19 0x1a 0x1b 0x1c 0x1d "
 169      "0x1e 0x1f 0x7f");
 170
 171   /* from RFC3629 */
 172
 173   /* UTF8-2      = %xC2-DF UTF8-tail */
 174   insist(!validutf8("\xC0\x80"));
 175   insist(!validutf8("\xC1\x80"));
 176   insist(!validutf8("\xC2\x7F"));
 177   U8("\xC2\x80", "0x80");
 178   U8("\xDF\xBF", "0x7FF");
 179   insist(!validutf8("\xDF\xC0"));
 180
 181   /*  UTF8-3      = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) /
 182    *                %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail )
 183    */
 184   insist(!validutf8("\xE0\x9F\x80"));
 185   U8("\xE0\xA0\x80", "0x800");
 186   U8("\xE0\xBF\xBF", "0xFFF");
 187   insist(!validutf8("\xE0\xC0\xBF"));
 188
 189   insist(!validutf8("\xE1\x80\x7F"));
 190   U8("\xE1\x80\x80", "0x1000");
 191   U8("\xEC\xBF\xBF", "0xCFFF");
 192   insist(!validutf8("\xEC\xC0\xBF"));
 193
 194   U8("\xED\x80\x80", "0xD000");
 195   U8("\xED\x9F\xBF", "0xD7FF");
 196   insist(!validutf8("\xED\xA0\xBF"));
 197
 198   insist(!validutf8("\xEE\x7f\x80"));
 199   U8("\xEE\x80\x80", "0xE000");
 200   U8("\xEF\xBF\xBF", "0xFFFF");
 201   insist(!validutf8("\xEF\xC0\xBF"));
 202
 203   /*  UTF8-4      = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) /
 204    *                %xF4 %x80-8F 2( UTF8-tail )
 205    */
 206   insist(!validutf8("\xF0\x8F\x80\x80"));
 207   U8("\xF0\x90\x80\x80", "0x10000");
 208   U8("\xF0\xBF\xBF\xBF", "0x3FFFF");
 209   insist(!validutf8("\xF0\xC0\x80\x80"));
 210
 211   insist(!validutf8("\xF1\x80\x80\x7F"));
 212   U8("\xF1\x80\x80\x80", "0x40000");
 213   U8("\xF3\xBF\xBF\xBF", "0xFFFFF");
 214   insist(!validutf8("\xF3\xC0\x80\x80"));
 215
 216   insist(!validutf8("\xF4\x80\x80\x7F"));
 217   U8("\xF4\x80\x80\x80", "0x100000");
 218   U8("\xF4\x8F\xBF\xBF", "0x10FFFF");
 219   insist(!validutf8("\xF4\x90\x80\x80"));
 220
 221   /* miscellaneous non-UTF-8 rubbish */
 222   insist(!validutf8("\x80"));
 223   insist(!validutf8("\xBF"));
 224   insist(!validutf8("\xC0"));
 225   insist(!validutf8("\xC0\x7F"));
 226   insist(!validutf8("\xC0\xC0"));
 227   insist(!validutf8("\xE0"));
 228   insist(!validutf8("\xE0\x7F"));
 229   insist(!validutf8("\xE0\xC0"));
 230   insist(!validutf8("\xE0\x80"));
 231   insist(!validutf8("\xE0\x80\x7f"));
 232   insist(!validutf8("\xE0\x80\xC0"));
 233   insist(!validutf8("\xF0"));
 234   insist(!validutf8("\xF0\x7F"));
 235   insist(!validutf8("\xF0\xC0"));
 236   insist(!validutf8("\xF0\x80"));
 237   insist(!validutf8("\xF0\x80\x7f"));
 238   insist(!validutf8("\xF0\x80\xC0"));
 239   insist(!validutf8("\xF0\x80\x80\x7f"));
 240   insist(!validutf8("\xF0\x80\x80\xC0"));
 241   insist(!validutf8("\xF5\x80\x80\x80"));
 242   insist(!validutf8("\xF8"));
 243 }
 244
 245 static void test_mime(void) {
 246   char *t, *n, *v;
 247
 248   fprintf(stderr, "test_mime\n");
 249
 250   t = n = v = 0;
 251   insist(!mime_content_type("text/plain", &t, &n, &v));
 252   insist(!strcmp(t, "text/plain"));
 253   insist(n == 0);
 254   insist(v == 0);
 255
 256   t = n = v = 0;
 257   insist(!mime_content_type("TEXT ((nested) comment) /plain", &t, &n, &v));
 258   insist(!strcmp(t, "text/plain"));
 259   insist(n == 0);
 260   insist(v == 0);
 261
 262   t = n = v = 0;
 263   insist(!mime_content_type(" text/plain ; Charset=utf-8", &t, &n, &v));
 264   insist(!strcmp(t, "text/plain"));
 265   insist(!strcmp(n, "charset"));
 266   insist(!strcmp(v, "utf-8"));
 267
 268   t = n = v = 0;
 269   insist(!mime_content_type("text/plain;charset = ISO-8859-1 ", &t, &n, &v));
 270   insist(!strcmp(t, "text/plain"));
 271   insist(!strcmp(n, "charset"));
 272   insist(!strcmp(v, "ISO-8859-1"));
 273
 274   /* XXX mime_parse */
 275   /* XXX mime_multipart */
 276   /* XXX mime_rfc2388_content_disposition */
 277
 278   check_string(mime_qp(""), "");
 279   check_string(mime_qp("foobar"), "foobar");
 280   check_string(mime_qp("foo=20bar"), "foo bar");
 281   check_string(mime_qp("x \r\ny"), "x\r\ny");
 282   check_string(mime_qp("x=\r\ny"), "xy");
 283   check_string(mime_qp("x= \r\ny"), "xy");
 284   check_string(mime_qp("x =\r\ny"), "x y");
 285   check_string(mime_qp("x = \r\ny"), "x y");
 286
 287   /* from RFC2045 */
 288   check_string(mime_qp("Now's the time =\r\n"
 289 "for all folk to come=\r\n"
 290 " to the aid of their country."),
 291                "Now's the time for all folk to come to the aid of their country.");
 292
 293   check_string(mime_base64(""),  "");
 294   check_string(mime_base64("BBBB"), "\x04\x10\x41");
 295   check_string(mime_base64("////"), "\xFF\xFF\xFF");
 296   check_string(mime_base64("//BB"), "\xFF\xF0\x41");
 297   check_string(mime_base64("BBBB//BB////"),
 298                "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
 299   check_string(mime_base64("B B B B  / / B B / / / /"),
 300                "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
 301   check_string(mime_base64("B\r\nBBB.// B-B//~//"),
 302                "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
 303   check_string(mime_base64("BBBB="),
 304                "\x04\x10\x41");
 305   check_string(mime_base64("BBBBx="),   /* not actually valid base64 */
 306                "\x04\x10\x41");
 307   check_string(mime_base64("BBBB BB=="),
 308                "\x04\x10\x41" "\x04");
 309   check_string(mime_base64("BBBB BBB="),
 310                "\x04\x10\x41" "\x04\x10");
 311 }
 312
 313 static void test_hex(void) {
 314   unsigned n;
 315   static const unsigned char h[] = { 0x00, 0xFF, 0x80, 0x7F };
 316   uint8_t *u;
 317   size_t ul;
 318
 319   fprintf(stderr, "test_hex\n");
 320
 321   for(n = 0; n <= UCHAR_MAX; ++n) {
 322     if(!isxdigit(n))
 323       insist(unhexdigitq(n) == -1);
 324   }
 325   insist(unhexdigitq('0') == 0);
 326   insist(unhexdigitq('1') == 1);
 327   insist(unhexdigitq('2') == 2);
 328   insist(unhexdigitq('3') == 3);
 329   insist(unhexdigitq('4') == 4);
 330   insist(unhexdigitq('5') == 5);
 331   insist(unhexdigitq('6') == 6);
 332   insist(unhexdigitq('7') == 7);
 333   insist(unhexdigitq('8') == 8);
 334   insist(unhexdigitq('9') == 9);
 335   insist(unhexdigitq('a') == 10);
 336   insist(unhexdigitq('b') == 11);
 337   insist(unhexdigitq('c') == 12);
 338   insist(unhexdigitq('d') == 13);
 339   insist(unhexdigitq('e') == 14);
 340   insist(unhexdigitq('f') == 15);
 341   insist(unhexdigitq('A') == 10);
 342   insist(unhexdigitq('B') == 11);
 343   insist(unhexdigitq('C') == 12);
 344   insist(unhexdigitq('D') == 13);
 345   insist(unhexdigitq('E') == 14);
 346   insist(unhexdigitq('F') == 15);
 347   check_string(hex(h, sizeof h), "00ff807f");
 348   check_string(hex(0, 0), "");
 349   u = unhex("00ff807f", &ul);
 350   insist(ul == 4);
 351   insist(memcmp(u, h, 4) == 0);
 352   u = unhex("00FF807F", &ul);
 353   insist(ul == 4);
 354   insist(memcmp(u, h, 4) == 0);
 355   u = unhex("", &ul);
 356   insist(ul == 0);
 357   fprintf(stderr, "2 ERROR reports expected {\n");
 358   insist(unhex("F", 0) == 0);
 359   insist(unhex("az", 0) == 0);
 360   fprintf(stderr, "}\n");
 361 }
 362
 363 static void test_casefold(void) {
 364   uint32_t c, l;
 365   const char *input, *canon_folded, *compat_folded, *canon_expected, *compat_expected;
 366
 367   fprintf(stderr, "test_casefold\n");
 368
 369   /* This isn't a very exhaustive test.  Unlike for normalization, there don't
 370    * seem to be any public test vectors for these algorithms. */
 371
 372   for(c = 1; c < 256; ++c) {
 373     input = utf32_to_utf8(&c, 1, 0);
 374     canon_folded = utf8_casefold_canon(input, strlen(input), 0);
 375     compat_folded = utf8_casefold_compat(input, strlen(input), 0);
 376     switch(c) {
 377     default:
 378       if((c >= 'A' && c <= 'Z')
 379          || (c >= 0xC0 && c <= 0xDE && c != 0xD7))
 380         l = c ^ 0x20;
 381       else
 382         l = c;
 383       break;
 384     case 0xB5:                          /* MICRO SIGN */
 385       l = 0x3BC;                        /* GREEK SMALL LETTER MU */
 386       break;
 387     case 0xDF:                          /* LATIN SMALL LETTER SHARP S */
 388       insist(!strcmp(canon_folded, "ss"));
 389       insist(!strcmp(compat_folded, "ss"));
 390       l = 0;
 391       break;
 392     }
 393     if(l) {
 394       uint32_t *d;
 395       /* Case-folded data is now normalized */
 396       d = utf32_decompose_canon(&l, 1, 0);
 397       canon_expected = utf32_to_utf8(d, utf32_len(d), 0);
 398       if(strcmp(canon_folded, canon_expected)) {
 399         fprintf(stderr, "%s:%d: canon-casefolding %#lx got '%s', expected '%s'\n",
 400                 __FILE__, __LINE__, (unsigned long)c,
 401                 format(canon_folded), format(canon_expected));
 402         count_error();
 403       }
 404       ++tests;
 405       d = utf32_decompose_compat(&l, 1, 0);
 406       compat_expected = utf32_to_utf8(d, utf32_len(d), 0);
 407       if(strcmp(compat_folded, compat_expected)) {
 408         fprintf(stderr, "%s:%d: compat-casefolding %#lx got '%s', expected '%s'\n",
 409                 __FILE__, __LINE__, (unsigned long)c,
 410                 format(compat_folded), format(compat_expected));
 411         count_error();
 412       }
 413       ++tests;
 414     }
 415   }
 416   check_string(utf8_casefold_canon("", 0, 0), "");
 417 }
 418
 419 struct {
 420   const char *in;
 421   const char *expect[10];
 422 } wtest[] = {
 423   /* Empty string */
 424   { "", { 0 } },
 425   /* Only whitespace and punctuation */
 426   { "    ", { 0 } },
 427   { " '   ", { 0 } },
 428   { " !  ", { 0 } },
 429   { " \"\"  ", { 0 } },
 430   { " @  ", { 0 } },
 431   /* Basics */
 432   { "wibble", { "wibble", 0 } },
 433   { " wibble", { "wibble", 0 } },
 434   { " wibble ", { "wibble", 0 } },
 435   { "wibble ", { "wibble", 0 } },
 436   { "wibble spong", { "wibble", "spong", 0 } },
 437   { " wibble  spong", { "wibble", "spong", 0 } },
 438   { " wibble  spong   ", { "wibble", "spong", 0 } },
 439   { "wibble   spong  ", { "wibble", "spong", 0 } },
 440   { "wibble   spong splat foo zot  ", { "wibble", "spong", "splat", "foo", "zot", 0 } },
 441   /* Apostrophes */
 442   { "wibble 'spong", { "wibble", "spong", 0 } },
 443   { " wibble's", { "wibble's", 0 } },
 444   { " wibblespong'   ", { "wibblespong", 0 } },
 445   { "wibble   sp''ong  ", { "wibble", "sp", "ong", 0 } },
 446 };
 447 #define NWTEST (sizeof wtest / sizeof *wtest)
 448
 449 static void test_words(void) {
 450   size_t t, nexpect, ngot, i;
 451   int right;
 452
 453   fprintf(stderr, "test_words\n");
 454   for(t = 0; t < NWTEST; ++t) {
 455     char **got = utf8_word_split(wtest[t].in, strlen(wtest[t].in), &ngot, 0);
 456
 457     for(nexpect = 0; wtest[t].expect[nexpect]; ++nexpect)
 458       ;
 459     if(nexpect == ngot) {
 460       for(i = 0; i < ngot; ++i)
 461         if(strcmp(wtest[t].expect[i], got[i]))
 462           break;
 463       right = i == ngot;
 464     } else
 465       right = 0;
 466     if(!right) {
 467       fprintf(stderr, "word split %zu failed\n", t);
 468       fprintf(stderr, "input: %s\n", wtest[t].in);
 469       fprintf(stderr, "    | %-30s | %-30s\n",
 470               "expected", "got");
 471       for(i = 0; i < nexpect || i < ngot; ++i) {
 472         const char *e = i < nexpect ? wtest[t].expect[i] : "<none>";
 473         const char *g = i < ngot ? got[i] : "<none>";
 474         fprintf(stderr, " %2zu | %-30s | %-30s\n", i, e, g);
 475       }
 476       count_error();
 477     }
 478     ++tests;
 479   }
 480 }
 481
 482 /** @brief Less-than comparison function for integer heap */
 483 static inline int int_lt(int a, int b) { return a < b; }
 484
 485 /** @struct iheap
 486  * @brief A heap with @c int elements */
 487 HEAP_TYPE(iheap, int, int_lt);
 488 HEAP_DEFINE(iheap, int, int_lt);
 489
 490 /** @brief Tests for @ref heap.h */
 491 static void test_heap(void) {
 492   struct iheap h[1];
 493   int n;
 494   int last = -1;
 495
 496   fprintf(stderr, "test_heap\n");
 497
 498   iheap_init(h);
 499   for(n = 0; n < 1000; ++n)
 500     iheap_insert(h, random() % 100);
 501   for(n = 0; n < 1000; ++n) {
 502     const int latest = iheap_remove(h);
 503     if(last > latest)
 504       fprintf(stderr, "should have %d <= %d\n", last, latest);
 505     insist(last <= latest);
 506     last = latest;
 507   }
 508   putchar('\n');
 509 }
 510
 511 /** @brief Open a Unicode test file */
 512 static FILE *open_unicode_test(const char *path) {
 513   const char *base;
 514   FILE *fp;
 515   char buffer[1024];
 516   int w;
 517
 518   if((base = strrchr(path, '/')))
 519     ++base;
 520   else
 521     base = path;
 522   if(!(fp = fopen(base, "r"))) {
 523     snprintf(buffer, sizeof buffer,
 524              "wget http://www.unicode.org/Public/5.0.0/ucd/%s", path);
 525     if((w = system(buffer)))
 526       fatal(0, "%s: %s", buffer, wstat(w));
 527     if(chmod(base, 0444) < 0)
 528       fatal(errno, "chmod %s", base);
 529     if(!(fp = fopen(base, "r")))
 530       fatal(errno, "%s", base);
 531   }
 532   return fp;
 533 }
 534
 535 /** @brief Run breaking tests for utf32_grapheme_boundary() etc */
 536 static void breaktest(const char *path,
 537                       int (*breakfn)(const uint32_t *, size_t, size_t)) {
 538   FILE *fp = open_unicode_test(path);
 539   int lineno = 0;
 540   char *l, *lp;
 541   size_t bn, n;
 542   char break_allowed[1024];
 543   uint32_t buffer[1024];
 544
 545   while(!inputline(path, fp, &l, '\n')) {
 546     ++lineno;
 547     if(l[0] == '#') continue;
 548     bn = 0;
 549     lp = l;
 550     while(*lp) {
 551       if(*lp == ' ' || *lp == '\t') {
 552         ++lp;
 553         continue;
 554       }
 555       if(*lp == '#')
 556         break;
 557       if((unsigned char)*lp == 0xC3 && (unsigned char)lp[1] == 0xB7) {
 558         /* 00F7 DIVISION SIGN */
 559         break_allowed[bn] = 1;
 560         lp += 2;
 561         continue;
 562       }
 563       if((unsigned char)*lp == 0xC3 && (unsigned char)lp[1] == 0x97) {
 564         /* 00D7 MULTIPLICATION SIGN */
 565         break_allowed[bn] = 0;
 566         lp += 2;
 567         continue;
 568       }
 569       if(isxdigit((unsigned char)*lp)) {
 570         buffer[bn++] = strtoul(lp, &lp, 16);
 571         continue;
 572       }
 573       fatal(0, "%s:%d: evil line: %s", path, lineno, l);
 574     }
 575     for(n = 0; n <= bn; ++n) {
 576       if(breakfn(buffer, bn, n) != break_allowed[n]) {
 577         fprintf(stderr,
 578                 "%s:%d: offset %zu: mismatch\n"
 579                 "%s\n"
 580                 "\n",
 581                 path, lineno, n, l);
 582         count_error();
 583       }
 584       ++tests;
 585     }
 586     xfree(l);
 587   }
 588   fclose(fp);
 589 }
 590
 591 /** @brief Tests for @ref lib/unicode.h */
 592 static void test_unicode(void) {
 593   FILE *fp;
 594   int lineno = 0;
 595   char *l, *lp;
 596   uint32_t buffer[1024];
 597   uint32_t *c[6], *NFD_c[6], *NFKD_c[6], *NFC_c[6], *NFKC_c[6]; /* 1-indexed */
 598   int cn, bn;
 599
 600   fprintf(stderr, "test_unicode\n");
 601   fp = open_unicode_test("NormalizationTest.txt");
 602   while(!inputline("NormalizationTest.txt", fp, &l, '\n')) {
 603     ++lineno;
 604     if(*l == '#' || *l == '@')
 605       continue;
 606     bn = 0;
 607     cn = 1;
 608     lp = l;
 609     c[cn++] = &buffer[bn];
 610     while(*lp && *lp != '#') {
 611       if(*lp == ' ') {
 612         ++lp;
 613         continue;
 614       }
 615       if(*lp == ';') {
 616         buffer[bn++] = 0;
 617         if(cn == 6)
 618           break;
 619         c[cn++] = &buffer[bn];
 620         ++lp;
 621         continue;
 622       }
 623       buffer[bn++] = strtoul(lp, &lp, 16);
 624     }
 625     buffer[bn] = 0;
 626     assert(cn == 6);
 627     for(cn = 1; cn <= 5; ++cn) {
 628       NFD_c[cn] = utf32_decompose_canon(c[cn], utf32_len(c[cn]), 0);
 629       NFKD_c[cn] = utf32_decompose_compat(c[cn], utf32_len(c[cn]), 0);
 630       NFC_c[cn] = utf32_compose_canon(c[cn], utf32_len(c[cn]), 0);
 631       NFKC_c[cn] = utf32_compose_compat(c[cn], utf32_len(c[cn]), 0);
 632     }
 633 #define unt_check(T, A, B) do {                                 \
 634     ++tests;                                                    \
 635     if(utf32_cmp(c[A], T##_c[B])) {                             \
 636       fprintf(stderr,                                           \
 637               "NormalizationTest.txt:%d: c%d != "#T"(c%d)\n",   \
 638               lineno, A, B);                                    \
 639       fprintf(stderr, "      c%d:%s\n",                         \
 640               A, format_utf32(c[A]));                           \
 641       fprintf(stderr, "      c%d:%s\n",                         \
 642               B, format_utf32(c[B]));                           \
 643       fprintf(stderr, "%4s(c%d):%s\n",                          \
 644               #T, B, format_utf32(T##_c[B]));                   \
 645       count_error();                                            \
 646     }                                                           \
 647   } while(0)
 648     unt_check(NFD, 3, 1);
 649     unt_check(NFD, 3, 2);
 650     unt_check(NFD, 3, 3);
 651     unt_check(NFD, 5, 4);
 652     unt_check(NFD, 5, 5);
 653     unt_check(NFKD, 5, 1);
 654     unt_check(NFKD, 5, 2);
 655     unt_check(NFKD, 5, 3);
 656     unt_check(NFKD, 5, 4);
 657     unt_check(NFKD, 5, 5);
 658     unt_check(NFC, 2, 1);
 659     unt_check(NFC, 2, 2);
 660     unt_check(NFC, 2, 3);
 661     unt_check(NFC, 4, 4);
 662     unt_check(NFC, 4, 5);
 663     unt_check(NFKC, 4, 1);
 664     unt_check(NFKC, 4, 2);
 665     unt_check(NFKC, 4, 3);
 666     unt_check(NFKC, 4, 4);
 667     unt_check(NFKC, 4, 5);
 668     for(cn = 1; cn <= 5; ++cn) {
 669       xfree(NFD_c[cn]);
 670       xfree(NFKD_c[cn]);
 671     }
 672     xfree(l);
 673   }
 674   fclose(fp);
 675   breaktest("auxiliary/GraphemeBreakTest.txt", utf32_is_grapheme_boundary);
 676   breaktest("auxiliary/WordBreakTest.txt", utf32_is_word_boundary);
 677 }
 678
 679 static void test_signame(void) {
 680   fprintf(stderr, "test_signame\n");
 681   insist(find_signal("SIGTERM") == SIGTERM);
 682   insist(find_signal("SIGHUP") == SIGHUP);
 683   insist(find_signal("SIGINT") == SIGINT);
 684   insist(find_signal("SIGQUIT") == SIGQUIT);
 685   insist(find_signal("SIGKILL") == SIGKILL);
 686   insist(find_signal("SIGYOURMUM") == -1);
 687 }
 688
 689 static void test_cache(void) {
 690   const struct cache_type t1 = { 1 }, t2 = { 10 };
 691   const char v11[] = "spong", v12[] = "wibble", v2[] = "blat";
 692   fprintf(stderr, "test_cache\n");
 693   cache_put(&t1, "1_1", v11);
 694   cache_put(&t1, "1_2", v12);
 695   cache_put(&t2, "2", v2);
 696   insist(cache_count() == 3);
 697   insist(cache_get(&t2, "2") == v2);
 698   insist(cache_get(&t1, "1_1") == v11);
 699   insist(cache_get(&t1, "1_2") == v12);
 700   insist(cache_get(&t1, "2") == 0);
 701   insist(cache_get(&t2, "1_1") == 0);
 702   insist(cache_get(&t2, "1_2") == 0);
 703   insist(cache_get(&t1, "2") == 0);
 704   insist(cache_get(&t2, "1_1") == 0);
 705   insist(cache_get(&t2, "1_2") == 0);
 706   sleep(2);
 707   cache_expire();
 708   insist(cache_count() == 1);
 709   insist(cache_get(&t1, "1_1") == 0);
 710   insist(cache_get(&t1, "1_2") == 0);
 711   insist(cache_get(&t2, "2") == v2);
 712   cache_clean(0);
 713   insist(cache_count() == 0);
 714   insist(cache_get(&t2, "2") == 0);
 715 }
 716
 717 int main(void) {
 718   fail_first = !!getenv("FAIL_FIRST");
 719   insist('\n' == 0x0A);
 720   insist('\r' == 0x0D);
 721   insist(' ' == 0x20);
 722   insist('0' == 0x30);
 723   insist('9' == 0x39);
 724   insist('A' == 0x41);
 725   insist('Z' == 0x5A);
 726   insist('a' == 0x61);
 727   insist('z' == 0x7A);
 728   /* addr.c */
 729   /* asprintf.c */
 730   /* authhash.c */
 731   /* basen.c */
 732   /* charset.c */
 733   /* client.c */
 734   /* configuration.c */
 735   /* event.c */
 736   /* fprintf.c */
 737   /* heap.c */
 738   test_heap();
 739   /* hex.c */
 740   test_hex();
 741   /* inputline.c */
 742   /* kvp.c */
 743   /* log.c */
 744   /* mem.c */
 745   /* mime.c */
 746   test_mime();
 747   /* mixer.c */
 748   /* plugin.c */
 749   /* printf.c */
 750   /* queue.c */
 751   /* sink.c */
 752   /* snprintf.c */
 753   /* split.c */
 754   /* syscalls.c */
 755   /* table.c */
 756   /* unicode.c */
 757   test_unicode();
 758   /* utf8.c */
 759   test_utf8();
 760   /* vector.c */
 761   /* words.c */
 762   test_casefold();
 763   test_words();
 764   /* wstat.c */
 765   /* signame.c */
 766   test_signame();
 767   /* cache.c */
 768   test_cache();
 769   fprintf(stderr,  "%d errors out of %d tests\n", errors, tests);
 770   return !!errors;
 771 }
 772
 773 /*
 774 Local Variables:
 775 c-basic-offset:2
 776 comment-column:40
 777 fill-column:79
 778 indent-tabs-mode:nil
 779 End:
 780 */