| 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 | |
| 34 | #include "utf8.h" |
| 35 | #include "mem.h" |
| 36 | #include "log.h" |
| 37 | #include "vector.h" |
| 38 | #include "charset.h" |
| 39 | #include "mime.h" |
| 40 | #include "hex.h" |
| 41 | #include "heap.h" |
| 42 | #include "unicode.h" |
| 43 | #include "inputline.h" |
| 44 | #include "wstat.h" |
| 45 | |
| 46 | static int tests, errors; |
| 47 | static int fail_first; |
| 48 | |
| 49 | static void count_error() { |
| 50 | ++errors; |
| 51 | if(fail_first) |
| 52 | abort(); |
| 53 | } |
| 54 | |
| 55 | /** @brief Checks that @p expr is nonzero */ |
| 56 | #define insist(expr) do { \ |
| 57 | if(!(expr)) { \ |
| 58 | count_error(); \ |
| 59 | fprintf(stderr, "%s:%d: error checking %s\n", \ |
| 60 | __FILE__, __LINE__, #expr); \ |
| 61 | } \ |
| 62 | ++tests; \ |
| 63 | } while(0) |
| 64 | |
| 65 | static const char *format(const char *s) { |
| 66 | struct dynstr d; |
| 67 | int c; |
| 68 | char buf[10]; |
| 69 | |
| 70 | dynstr_init(&d); |
| 71 | while((c = (unsigned char)*s++)) { |
| 72 | if(c >= ' ' && c <= '~') |
| 73 | dynstr_append(&d, c); |
| 74 | else { |
| 75 | sprintf(buf, "\\x%02X", (unsigned)c); |
| 76 | dynstr_append_string(&d, buf); |
| 77 | } |
| 78 | } |
| 79 | dynstr_terminate(&d); |
| 80 | return d.vec; |
| 81 | } |
| 82 | |
| 83 | static const char *format_utf32(const uint32_t *s) { |
| 84 | struct dynstr d; |
| 85 | uint32_t c; |
| 86 | char buf[64]; |
| 87 | |
| 88 | dynstr_init(&d); |
| 89 | while((c = *s++)) { |
| 90 | sprintf(buf, " %04lX", (long)c); |
| 91 | dynstr_append_string(&d, buf); |
| 92 | } |
| 93 | dynstr_terminate(&d); |
| 94 | return d.vec; |
| 95 | } |
| 96 | |
| 97 | #define check_string(GOT, WANT) do { \ |
| 98 | const char *g = GOT; \ |
| 99 | const char *w = WANT; \ |
| 100 | \ |
| 101 | if(w == 0) { \ |
| 102 | fprintf(stderr, "%s:%d: %s returned 0\n", \ |
| 103 | __FILE__, __LINE__, #GOT); \ |
| 104 | count_error(); \ |
| 105 | } else if(strcmp(w, g)) { \ |
| 106 | fprintf(stderr, "%s:%d: %s returned:\n%s\nexpected:\n%s\n", \ |
| 107 | __FILE__, __LINE__, #GOT, format(g), format(w)); \ |
| 108 | count_error(); \ |
| 109 | } \ |
| 110 | ++tests; \ |
| 111 | } while(0) |
| 112 | |
| 113 | static uint32_t *ucs4parse(const char *s) { |
| 114 | struct dynstr_ucs4 d; |
| 115 | char *e; |
| 116 | |
| 117 | dynstr_ucs4_init(&d); |
| 118 | while(*s) { |
| 119 | errno = 0; |
| 120 | dynstr_ucs4_append(&d, strtoul(s, &e, 0)); |
| 121 | if(errno) fatal(errno, "strtoul (%s)", s); |
| 122 | s = e; |
| 123 | } |
| 124 | dynstr_ucs4_terminate(&d); |
| 125 | return d.vec; |
| 126 | } |
| 127 | |
| 128 | static void test_utf8(void) { |
| 129 | /* Test validutf8, convert to UCS-4, check the answer is right, |
| 130 | * convert back to UTF-8, check we got to where we started */ |
| 131 | #define U8(CHARS, WORDS) do { \ |
| 132 | uint32_t *w = ucs4parse(WORDS); \ |
| 133 | uint32_t *ucs; \ |
| 134 | char *u8; \ |
| 135 | \ |
| 136 | insist(validutf8(CHARS)); \ |
| 137 | ucs = utf8_to_utf32(CHARS, strlen(CHARS), 0); \ |
| 138 | insist(ucs != 0); \ |
| 139 | insist(!utf32_cmp(w, ucs)); \ |
| 140 | u8 = utf32_to_utf8(ucs, utf32_len(ucs), 0); \ |
| 141 | insist(u8 != 0); \ |
| 142 | insist(!strcmp(u8, CHARS)); \ |
| 143 | } while(0) |
| 144 | |
| 145 | fprintf(stderr, "test_utf8\n"); |
| 146 | #define validutf8(S) utf8_valid((S), strlen(S)) |
| 147 | |
| 148 | /* empty string */ |
| 149 | |
| 150 | U8("", ""); |
| 151 | |
| 152 | /* ASCII characters */ |
| 153 | |
| 154 | U8(" !\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{|}~", |
| 155 | "0x20 0x21 0x22 0x23 0x24 0x25 0x26 0x27 0x28 0x29 0x2a 0x2b 0x2c 0x2d " |
| 156 | "0x2e 0x2f 0x30 0x31 0x32 0x33 0x34 0x35 0x36 0x37 0x38 0x39 0x3a " |
| 157 | "0x3b 0x3c 0x3d 0x3e 0x3f 0x40 0x41 0x42 0x43 0x44 0x45 0x46 0x47 " |
| 158 | "0x48 0x49 0x4a 0x4b 0x4c 0x4d 0x4e 0x4f 0x50 0x51 0x52 0x53 0x54 " |
| 159 | "0x55 0x56 0x57 0x58 0x59 0x5a 0x5b 0x5c 0x5d 0x5e 0x5f 0x60 0x61 " |
| 160 | "0x62 0x63 0x64 0x65 0x66 0x67 0x68 0x69 0x6a 0x6b 0x6c 0x6d 0x6e " |
| 161 | "0x6f 0x70 0x71 0x72 0x73 0x74 0x75 0x76 0x77 0x78 0x79 0x7a 0x7b " |
| 162 | "0x7c 0x7d 0x7e"); |
| 163 | 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", |
| 164 | "0x1 0x2 0x3 0x4 0x5 0x6 0x7 0x8 0x9 0xa 0xb 0xc 0xd 0xe 0xf 0x10 " |
| 165 | "0x11 0x12 0x13 0x14 0x15 0x16 0x17 0x18 0x19 0x1a 0x1b 0x1c 0x1d " |
| 166 | "0x1e 0x1f 0x7f"); |
| 167 | |
| 168 | /* from RFC3629 */ |
| 169 | |
| 170 | /* UTF8-2 = %xC2-DF UTF8-tail */ |
| 171 | insist(!validutf8("\xC0\x80")); |
| 172 | insist(!validutf8("\xC1\x80")); |
| 173 | insist(!validutf8("\xC2\x7F")); |
| 174 | U8("\xC2\x80", "0x80"); |
| 175 | U8("\xDF\xBF", "0x7FF"); |
| 176 | insist(!validutf8("\xDF\xC0")); |
| 177 | |
| 178 | /* UTF8-3 = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) / |
| 179 | * %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail ) |
| 180 | */ |
| 181 | insist(!validutf8("\xE0\x9F\x80")); |
| 182 | U8("\xE0\xA0\x80", "0x800"); |
| 183 | U8("\xE0\xBF\xBF", "0xFFF"); |
| 184 | insist(!validutf8("\xE0\xC0\xBF")); |
| 185 | |
| 186 | insist(!validutf8("\xE1\x80\x7F")); |
| 187 | U8("\xE1\x80\x80", "0x1000"); |
| 188 | U8("\xEC\xBF\xBF", "0xCFFF"); |
| 189 | insist(!validutf8("\xEC\xC0\xBF")); |
| 190 | |
| 191 | U8("\xED\x80\x80", "0xD000"); |
| 192 | U8("\xED\x9F\xBF", "0xD7FF"); |
| 193 | insist(!validutf8("\xED\xA0\xBF")); |
| 194 | |
| 195 | insist(!validutf8("\xEE\x7f\x80")); |
| 196 | U8("\xEE\x80\x80", "0xE000"); |
| 197 | U8("\xEF\xBF\xBF", "0xFFFF"); |
| 198 | insist(!validutf8("\xEF\xC0\xBF")); |
| 199 | |
| 200 | /* UTF8-4 = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) / |
| 201 | * %xF4 %x80-8F 2( UTF8-tail ) |
| 202 | */ |
| 203 | insist(!validutf8("\xF0\x8F\x80\x80")); |
| 204 | U8("\xF0\x90\x80\x80", "0x10000"); |
| 205 | U8("\xF0\xBF\xBF\xBF", "0x3FFFF"); |
| 206 | insist(!validutf8("\xF0\xC0\x80\x80")); |
| 207 | |
| 208 | insist(!validutf8("\xF1\x80\x80\x7F")); |
| 209 | U8("\xF1\x80\x80\x80", "0x40000"); |
| 210 | U8("\xF3\xBF\xBF\xBF", "0xFFFFF"); |
| 211 | insist(!validutf8("\xF3\xC0\x80\x80")); |
| 212 | |
| 213 | insist(!validutf8("\xF4\x80\x80\x7F")); |
| 214 | U8("\xF4\x80\x80\x80", "0x100000"); |
| 215 | U8("\xF4\x8F\xBF\xBF", "0x10FFFF"); |
| 216 | insist(!validutf8("\xF4\x90\x80\x80")); |
| 217 | |
| 218 | /* miscellaneous non-UTF-8 rubbish */ |
| 219 | insist(!validutf8("\x80")); |
| 220 | insist(!validutf8("\xBF")); |
| 221 | insist(!validutf8("\xC0")); |
| 222 | insist(!validutf8("\xC0\x7F")); |
| 223 | insist(!validutf8("\xC0\xC0")); |
| 224 | insist(!validutf8("\xE0")); |
| 225 | insist(!validutf8("\xE0\x7F")); |
| 226 | insist(!validutf8("\xE0\xC0")); |
| 227 | insist(!validutf8("\xE0\x80")); |
| 228 | insist(!validutf8("\xE0\x80\x7f")); |
| 229 | insist(!validutf8("\xE0\x80\xC0")); |
| 230 | insist(!validutf8("\xF0")); |
| 231 | insist(!validutf8("\xF0\x7F")); |
| 232 | insist(!validutf8("\xF0\xC0")); |
| 233 | insist(!validutf8("\xF0\x80")); |
| 234 | insist(!validutf8("\xF0\x80\x7f")); |
| 235 | insist(!validutf8("\xF0\x80\xC0")); |
| 236 | insist(!validutf8("\xF0\x80\x80\x7f")); |
| 237 | insist(!validutf8("\xF0\x80\x80\xC0")); |
| 238 | insist(!validutf8("\xF5\x80\x80\x80")); |
| 239 | insist(!validutf8("\xF8")); |
| 240 | } |
| 241 | |
| 242 | static void test_mime(void) { |
| 243 | char *t, *n, *v; |
| 244 | |
| 245 | fprintf(stderr, "test_mime\n"); |
| 246 | |
| 247 | t = n = v = 0; |
| 248 | insist(!mime_content_type("text/plain", &t, &n, &v)); |
| 249 | insist(!strcmp(t, "text/plain")); |
| 250 | insist(n == 0); |
| 251 | insist(v == 0); |
| 252 | |
| 253 | t = n = v = 0; |
| 254 | insist(!mime_content_type("TEXT ((nested) comment) /plain", &t, &n, &v)); |
| 255 | insist(!strcmp(t, "text/plain")); |
| 256 | insist(n == 0); |
| 257 | insist(v == 0); |
| 258 | |
| 259 | t = n = v = 0; |
| 260 | insist(!mime_content_type(" text/plain ; Charset=utf-8", &t, &n, &v)); |
| 261 | insist(!strcmp(t, "text/plain")); |
| 262 | insist(!strcmp(n, "charset")); |
| 263 | insist(!strcmp(v, "utf-8")); |
| 264 | |
| 265 | t = n = v = 0; |
| 266 | insist(!mime_content_type("text/plain;charset = ISO-8859-1 ", &t, &n, &v)); |
| 267 | insist(!strcmp(t, "text/plain")); |
| 268 | insist(!strcmp(n, "charset")); |
| 269 | insist(!strcmp(v, "ISO-8859-1")); |
| 270 | |
| 271 | /* XXX mime_parse */ |
| 272 | /* XXX mime_multipart */ |
| 273 | /* XXX mime_rfc2388_content_disposition */ |
| 274 | |
| 275 | check_string(mime_qp(""), ""); |
| 276 | check_string(mime_qp("foobar"), "foobar"); |
| 277 | check_string(mime_qp("foo=20bar"), "foo bar"); |
| 278 | check_string(mime_qp("x \r\ny"), "x\r\ny"); |
| 279 | check_string(mime_qp("x=\r\ny"), "xy"); |
| 280 | check_string(mime_qp("x= \r\ny"), "xy"); |
| 281 | check_string(mime_qp("x =\r\ny"), "x y"); |
| 282 | check_string(mime_qp("x = \r\ny"), "x y"); |
| 283 | |
| 284 | /* from RFC2045 */ |
| 285 | check_string(mime_qp("Now's the time =\r\n" |
| 286 | "for all folk to come=\r\n" |
| 287 | " to the aid of their country."), |
| 288 | "Now's the time for all folk to come to the aid of their country."); |
| 289 | |
| 290 | check_string(mime_base64(""), ""); |
| 291 | check_string(mime_base64("BBBB"), "\x04\x10\x41"); |
| 292 | check_string(mime_base64("////"), "\xFF\xFF\xFF"); |
| 293 | check_string(mime_base64("//BB"), "\xFF\xF0\x41"); |
| 294 | check_string(mime_base64("BBBB//BB////"), |
| 295 | "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF"); |
| 296 | check_string(mime_base64("B B B B / / B B / / / /"), |
| 297 | "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF"); |
| 298 | check_string(mime_base64("B\r\nBBB.// B-B//~//"), |
| 299 | "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF"); |
| 300 | check_string(mime_base64("BBBB="), |
| 301 | "\x04\x10\x41"); |
| 302 | check_string(mime_base64("BBBBx="), /* not actually valid base64 */ |
| 303 | "\x04\x10\x41"); |
| 304 | check_string(mime_base64("BBBB BB=="), |
| 305 | "\x04\x10\x41" "\x04"); |
| 306 | check_string(mime_base64("BBBB BBB="), |
| 307 | "\x04\x10\x41" "\x04\x10"); |
| 308 | } |
| 309 | |
| 310 | static void test_cookies(void) { |
| 311 | struct cookiedata cd[1]; |
| 312 | |
| 313 | fprintf(stderr, "test_cookies\n"); |
| 314 | |
| 315 | /* These are the examples from RFC2109 */ |
| 316 | insist(!parse_cookie("$Version=\"1\"; Customer=\"WILE_E_COYOTE\"; $Path=\"/acme\"", cd)); |
| 317 | insist(!strcmp(cd->version, "1")); |
| 318 | insist(cd->ncookies = 1); |
| 319 | insist(find_cookie(cd, "Customer") == &cd->cookies[0]); |
| 320 | check_string(cd->cookies[0].value, "WILE_E_COYOTE"); |
| 321 | check_string(cd->cookies[0].path, "/acme"); |
| 322 | insist(cd->cookies[0].domain == 0); |
| 323 | insist(!parse_cookie("$Version=\"1\";\n" |
| 324 | "Customer=\"WILE_E_COYOTE\"; $Path=\"/acme\";\n" |
| 325 | "Part_Number=\"Rocket_Launcher_0001\"; $Path=\"/acme\"", |
| 326 | cd)); |
| 327 | insist(cd->ncookies = 2); |
| 328 | insist(find_cookie(cd, "Customer") == &cd->cookies[0]); |
| 329 | insist(find_cookie(cd, "Part_Number") == &cd->cookies[1]); |
| 330 | check_string(cd->cookies[0].value, "WILE_E_COYOTE"); |
| 331 | check_string(cd->cookies[0].path, "/acme"); |
| 332 | insist(cd->cookies[0].domain == 0); |
| 333 | check_string(cd->cookies[1].value, "Rocket_Launcher_0001"); |
| 334 | check_string(cd->cookies[1].path, "/acme"); |
| 335 | insist(cd->cookies[1].domain == 0); |
| 336 | insist(!parse_cookie("$Version=\"1\";\n" |
| 337 | "Customer=\"WILE_E_COYOTE\"; $Path=\"/acme\";\n" |
| 338 | "Part_Number=\"Rocket_Launcher_0001\"; $Path=\"/acme\";\n" |
| 339 | "Shipping=\"FedEx\"; $Path=\"/acme\"", |
| 340 | cd)); |
| 341 | insist(cd->ncookies = 3); |
| 342 | insist(find_cookie(cd, "Customer") == &cd->cookies[0]); |
| 343 | insist(find_cookie(cd, "Part_Number") == &cd->cookies[1]); |
| 344 | insist(find_cookie(cd, "Shipping") == &cd->cookies[2]); |
| 345 | check_string(cd->cookies[0].value, "WILE_E_COYOTE"); |
| 346 | check_string(cd->cookies[0].path, "/acme"); |
| 347 | insist(cd->cookies[0].domain == 0); |
| 348 | check_string(cd->cookies[1].value, "Rocket_Launcher_0001"); |
| 349 | check_string(cd->cookies[1].path, "/acme"); |
| 350 | insist(cd->cookies[1].domain == 0); |
| 351 | check_string(cd->cookies[2].value, "FedEx"); |
| 352 | check_string(cd->cookies[2].path, "/acme"); |
| 353 | insist(cd->cookies[2].domain == 0); |
| 354 | } |
| 355 | |
| 356 | static void test_hex(void) { |
| 357 | unsigned n; |
| 358 | static const unsigned char h[] = { 0x00, 0xFF, 0x80, 0x7F }; |
| 359 | uint8_t *u; |
| 360 | size_t ul; |
| 361 | |
| 362 | fprintf(stderr, "test_hex\n"); |
| 363 | |
| 364 | for(n = 0; n <= UCHAR_MAX; ++n) { |
| 365 | if(!isxdigit(n)) |
| 366 | insist(unhexdigitq(n) == -1); |
| 367 | } |
| 368 | insist(unhexdigitq('0') == 0); |
| 369 | insist(unhexdigitq('1') == 1); |
| 370 | insist(unhexdigitq('2') == 2); |
| 371 | insist(unhexdigitq('3') == 3); |
| 372 | insist(unhexdigitq('4') == 4); |
| 373 | insist(unhexdigitq('5') == 5); |
| 374 | insist(unhexdigitq('6') == 6); |
| 375 | insist(unhexdigitq('7') == 7); |
| 376 | insist(unhexdigitq('8') == 8); |
| 377 | insist(unhexdigitq('9') == 9); |
| 378 | insist(unhexdigitq('a') == 10); |
| 379 | insist(unhexdigitq('b') == 11); |
| 380 | insist(unhexdigitq('c') == 12); |
| 381 | insist(unhexdigitq('d') == 13); |
| 382 | insist(unhexdigitq('e') == 14); |
| 383 | insist(unhexdigitq('f') == 15); |
| 384 | insist(unhexdigitq('A') == 10); |
| 385 | insist(unhexdigitq('B') == 11); |
| 386 | insist(unhexdigitq('C') == 12); |
| 387 | insist(unhexdigitq('D') == 13); |
| 388 | insist(unhexdigitq('E') == 14); |
| 389 | insist(unhexdigitq('F') == 15); |
| 390 | check_string(hex(h, sizeof h), "00ff807f"); |
| 391 | check_string(hex(0, 0), ""); |
| 392 | u = unhex("00ff807f", &ul); |
| 393 | insist(ul == 4); |
| 394 | insist(memcmp(u, h, 4) == 0); |
| 395 | u = unhex("00FF807F", &ul); |
| 396 | insist(ul == 4); |
| 397 | insist(memcmp(u, h, 4) == 0); |
| 398 | u = unhex("", &ul); |
| 399 | insist(ul == 0); |
| 400 | fprintf(stderr, "2 ERROR reports expected {\n"); |
| 401 | insist(unhex("F", 0) == 0); |
| 402 | insist(unhex("az", 0) == 0); |
| 403 | fprintf(stderr, "}\n"); |
| 404 | } |
| 405 | |
| 406 | static void test_casefold(void) { |
| 407 | uint32_t c, l; |
| 408 | const char *input, *canon_folded, *compat_folded, *canon_expected, *compat_expected; |
| 409 | |
| 410 | fprintf(stderr, "test_casefold\n"); |
| 411 | |
| 412 | /* This isn't a very exhaustive test. Unlike for normalization, there don't |
| 413 | * seem to be any public test vectors for these algorithms. */ |
| 414 | |
| 415 | for(c = 1; c < 256; ++c) { |
| 416 | input = utf32_to_utf8(&c, 1, 0); |
| 417 | canon_folded = utf8_casefold_canon(input, strlen(input), 0); |
| 418 | compat_folded = utf8_casefold_compat(input, strlen(input), 0); |
| 419 | switch(c) { |
| 420 | default: |
| 421 | if((c >= 'A' && c <= 'Z') |
| 422 | || (c >= 0xC0 && c <= 0xDE && c != 0xD7)) |
| 423 | l = c ^ 0x20; |
| 424 | else |
| 425 | l = c; |
| 426 | break; |
| 427 | case 0xB5: /* MICRO SIGN */ |
| 428 | l = 0x3BC; /* GREEK SMALL LETTER MU */ |
| 429 | break; |
| 430 | case 0xDF: /* LATIN SMALL LETTER SHARP S */ |
| 431 | insist(!strcmp(canon_folded, "ss")); |
| 432 | insist(!strcmp(compat_folded, "ss")); |
| 433 | l = 0; |
| 434 | break; |
| 435 | } |
| 436 | if(l) { |
| 437 | uint32_t *d; |
| 438 | /* Case-folded data is now normalized */ |
| 439 | d = utf32_decompose_canon(&l, 1, 0); |
| 440 | canon_expected = utf32_to_utf8(d, utf32_len(d), 0); |
| 441 | if(strcmp(canon_folded, canon_expected)) { |
| 442 | fprintf(stderr, "%s:%d: canon-casefolding %#lx got '%s', expected '%s'\n", |
| 443 | __FILE__, __LINE__, (unsigned long)c, |
| 444 | format(canon_folded), format(canon_expected)); |
| 445 | count_error(); |
| 446 | } |
| 447 | ++tests; |
| 448 | d = utf32_decompose_compat(&l, 1, 0); |
| 449 | compat_expected = utf32_to_utf8(d, utf32_len(d), 0); |
| 450 | if(strcmp(compat_folded, compat_expected)) { |
| 451 | fprintf(stderr, "%s:%d: compat-casefolding %#lx got '%s', expected '%s'\n", |
| 452 | __FILE__, __LINE__, (unsigned long)c, |
| 453 | format(compat_folded), format(compat_expected)); |
| 454 | count_error(); |
| 455 | } |
| 456 | ++tests; |
| 457 | } |
| 458 | } |
| 459 | check_string(utf8_casefold_canon("", 0, 0), ""); |
| 460 | } |
| 461 | |
| 462 | struct { |
| 463 | const char *in; |
| 464 | const char *expect[10]; |
| 465 | } wtest[] = { |
| 466 | /* Empty string */ |
| 467 | { "", { 0 } }, |
| 468 | /* Only whitespace and punctuation */ |
| 469 | { " ", { 0 } }, |
| 470 | { " ' ", { 0 } }, |
| 471 | { " ! ", { 0 } }, |
| 472 | { " \"\" ", { 0 } }, |
| 473 | { " @ ", { 0 } }, |
| 474 | /* Basics */ |
| 475 | { "wibble", { "wibble", 0 } }, |
| 476 | { " wibble", { "wibble", 0 } }, |
| 477 | { " wibble ", { "wibble", 0 } }, |
| 478 | { "wibble ", { "wibble", 0 } }, |
| 479 | { "wibble spong", { "wibble", "spong", 0 } }, |
| 480 | { " wibble spong", { "wibble", "spong", 0 } }, |
| 481 | { " wibble spong ", { "wibble", "spong", 0 } }, |
| 482 | { "wibble spong ", { "wibble", "spong", 0 } }, |
| 483 | { "wibble spong splat foo zot ", { "wibble", "spong", "splat", "foo", "zot", 0 } }, |
| 484 | /* Apostrophes */ |
| 485 | { "wibble 'spong", { "wibble", "spong", 0 } }, |
| 486 | { " wibble's", { "wibble's", 0 } }, |
| 487 | { " wibblespong' ", { "wibblespong", 0 } }, |
| 488 | { "wibble sp''ong ", { "wibble", "sp", "ong", 0 } }, |
| 489 | }; |
| 490 | #define NWTEST (sizeof wtest / sizeof *wtest) |
| 491 | |
| 492 | static void test_words(void) { |
| 493 | size_t t, nexpect, ngot, i; |
| 494 | int right; |
| 495 | |
| 496 | fprintf(stderr, "test_words\n"); |
| 497 | for(t = 0; t < NWTEST; ++t) { |
| 498 | char **got = utf8_word_split(wtest[t].in, strlen(wtest[t].in), &ngot, 0); |
| 499 | |
| 500 | for(nexpect = 0; wtest[t].expect[nexpect]; ++nexpect) |
| 501 | ; |
| 502 | if(nexpect == ngot) { |
| 503 | for(i = 0; i < ngot; ++i) |
| 504 | if(strcmp(wtest[t].expect[i], got[i])) |
| 505 | break; |
| 506 | right = i == ngot; |
| 507 | } else |
| 508 | right = 0; |
| 509 | if(!right) { |
| 510 | fprintf(stderr, "word split %zu failed\n", t); |
| 511 | fprintf(stderr, "input: %s\n", wtest[t].in); |
| 512 | fprintf(stderr, " | %-30s | %-30s\n", |
| 513 | "expected", "got"); |
| 514 | for(i = 0; i < nexpect || i < ngot; ++i) { |
| 515 | const char *e = i < nexpect ? wtest[t].expect[i] : "<none>"; |
| 516 | const char *g = i < ngot ? got[i] : "<none>"; |
| 517 | fprintf(stderr, " %2zu | %-30s | %-30s\n", i, e, g); |
| 518 | } |
| 519 | count_error(); |
| 520 | } |
| 521 | ++tests; |
| 522 | } |
| 523 | } |
| 524 | |
| 525 | /** @brief Less-than comparison function for integer heap */ |
| 526 | static inline int int_lt(int a, int b) { return a < b; } |
| 527 | |
| 528 | /** @struct iheap |
| 529 | * @brief A heap with @c int elements */ |
| 530 | HEAP_TYPE(iheap, int, int_lt); |
| 531 | HEAP_DEFINE(iheap, int, int_lt); |
| 532 | |
| 533 | /** @brief Tests for @ref heap.h */ |
| 534 | static void test_heap(void) { |
| 535 | struct iheap h[1]; |
| 536 | int n; |
| 537 | int last = -1; |
| 538 | |
| 539 | fprintf(stderr, "test_heap\n"); |
| 540 | |
| 541 | iheap_init(h); |
| 542 | for(n = 0; n < 1000; ++n) |
| 543 | iheap_insert(h, random() % 100); |
| 544 | for(n = 0; n < 1000; ++n) { |
| 545 | const int latest = iheap_remove(h); |
| 546 | if(last > latest) |
| 547 | fprintf(stderr, "should have %d <= %d\n", last, latest); |
| 548 | insist(last <= latest); |
| 549 | last = latest; |
| 550 | } |
| 551 | putchar('\n'); |
| 552 | } |
| 553 | |
| 554 | /** @brief Open a Unicode test file */ |
| 555 | static FILE *open_unicode_test(const char *path) { |
| 556 | const char *base; |
| 557 | FILE *fp; |
| 558 | char buffer[1024]; |
| 559 | int w; |
| 560 | |
| 561 | if((base = strrchr(path, '/'))) |
| 562 | ++base; |
| 563 | else |
| 564 | base = path; |
| 565 | if(!(fp = fopen(base, "r"))) { |
| 566 | snprintf(buffer, sizeof buffer, |
| 567 | "wget http://www.unicode.org/Public/5.0.0/ucd/%s", path); |
| 568 | if((w = system(buffer))) |
| 569 | fatal(0, "%s: %s", buffer, wstat(w)); |
| 570 | if(chmod(base, 0444) < 0) |
| 571 | fatal(errno, "chmod %s", base); |
| 572 | if(!(fp = fopen(base, "r"))) |
| 573 | fatal(errno, "%s", base); |
| 574 | } |
| 575 | return fp; |
| 576 | } |
| 577 | |
| 578 | /** @brief Run breaking tests for utf32_grapheme_boundary() etc */ |
| 579 | static void breaktest(const char *path, |
| 580 | int (*breakfn)(const uint32_t *, size_t, size_t)) { |
| 581 | FILE *fp = open_unicode_test(path); |
| 582 | int lineno = 0; |
| 583 | char *l, *lp; |
| 584 | size_t bn, n; |
| 585 | char break_allowed[1024]; |
| 586 | uint32_t buffer[1024]; |
| 587 | |
| 588 | while(!inputline(path, fp, &l, '\n')) { |
| 589 | ++lineno; |
| 590 | if(l[0] == '#') continue; |
| 591 | bn = 0; |
| 592 | lp = l; |
| 593 | while(*lp) { |
| 594 | if(*lp == ' ' || *lp == '\t') { |
| 595 | ++lp; |
| 596 | continue; |
| 597 | } |
| 598 | if(*lp == '#') |
| 599 | break; |
| 600 | if((unsigned char)*lp == 0xC3 && (unsigned char)lp[1] == 0xB7) { |
| 601 | /* 00F7 DIVISION SIGN */ |
| 602 | break_allowed[bn] = 1; |
| 603 | lp += 2; |
| 604 | continue; |
| 605 | } |
| 606 | if((unsigned char)*lp == 0xC3 && (unsigned char)lp[1] == 0x97) { |
| 607 | /* 00D7 MULTIPLICATION SIGN */ |
| 608 | break_allowed[bn] = 0; |
| 609 | lp += 2; |
| 610 | continue; |
| 611 | } |
| 612 | if(isxdigit((unsigned char)*lp)) { |
| 613 | buffer[bn++] = strtoul(lp, &lp, 16); |
| 614 | continue; |
| 615 | } |
| 616 | fatal(0, "%s:%d: evil line: %s", path, lineno, l); |
| 617 | } |
| 618 | for(n = 0; n <= bn; ++n) { |
| 619 | if(breakfn(buffer, bn, n) != break_allowed[n]) { |
| 620 | fprintf(stderr, |
| 621 | "%s:%d: offset %zu: mismatch\n" |
| 622 | "%s\n" |
| 623 | "\n", |
| 624 | path, lineno, n, l); |
| 625 | count_error(); |
| 626 | } |
| 627 | ++tests; |
| 628 | } |
| 629 | xfree(l); |
| 630 | } |
| 631 | fclose(fp); |
| 632 | } |
| 633 | |
| 634 | /** @brief Tests for @ref lib/unicode.h */ |
| 635 | static void test_unicode(void) { |
| 636 | FILE *fp; |
| 637 | int lineno = 0; |
| 638 | char *l, *lp; |
| 639 | uint32_t buffer[1024]; |
| 640 | uint32_t *c[6], *NFD_c[6], *NFKD_c[6], *NFC_c[6], *NFKC_c[6]; /* 1-indexed */ |
| 641 | int cn, bn; |
| 642 | |
| 643 | fprintf(stderr, "test_unicode\n"); |
| 644 | fp = open_unicode_test("NormalizationTest.txt"); |
| 645 | while(!inputline("NormalizationTest.txt", fp, &l, '\n')) { |
| 646 | ++lineno; |
| 647 | if(*l == '#' || *l == '@') |
| 648 | continue; |
| 649 | bn = 0; |
| 650 | cn = 1; |
| 651 | lp = l; |
| 652 | c[cn++] = &buffer[bn]; |
| 653 | while(*lp && *lp != '#') { |
| 654 | if(*lp == ' ') { |
| 655 | ++lp; |
| 656 | continue; |
| 657 | } |
| 658 | if(*lp == ';') { |
| 659 | buffer[bn++] = 0; |
| 660 | if(cn == 6) |
| 661 | break; |
| 662 | c[cn++] = &buffer[bn]; |
| 663 | ++lp; |
| 664 | continue; |
| 665 | } |
| 666 | buffer[bn++] = strtoul(lp, &lp, 16); |
| 667 | } |
| 668 | buffer[bn] = 0; |
| 669 | assert(cn == 6); |
| 670 | for(cn = 1; cn <= 5; ++cn) { |
| 671 | NFD_c[cn] = utf32_decompose_canon(c[cn], utf32_len(c[cn]), 0); |
| 672 | NFKD_c[cn] = utf32_decompose_compat(c[cn], utf32_len(c[cn]), 0); |
| 673 | NFC_c[cn] = utf32_compose_canon(c[cn], utf32_len(c[cn]), 0); |
| 674 | NFKC_c[cn] = utf32_compose_compat(c[cn], utf32_len(c[cn]), 0); |
| 675 | } |
| 676 | #define unt_check(T, A, B) do { \ |
| 677 | ++tests; \ |
| 678 | if(utf32_cmp(c[A], T##_c[B])) { \ |
| 679 | fprintf(stderr, \ |
| 680 | "NormalizationTest.txt:%d: c%d != "#T"(c%d)\n", \ |
| 681 | lineno, A, B); \ |
| 682 | fprintf(stderr, " c%d:%s\n", \ |
| 683 | A, format_utf32(c[A])); \ |
| 684 | fprintf(stderr, " c%d:%s\n", \ |
| 685 | B, format_utf32(c[B])); \ |
| 686 | fprintf(stderr, "%4s(c%d):%s\n", \ |
| 687 | #T, B, format_utf32(T##_c[B])); \ |
| 688 | count_error(); \ |
| 689 | } \ |
| 690 | } while(0) |
| 691 | unt_check(NFD, 3, 1); |
| 692 | unt_check(NFD, 3, 2); |
| 693 | unt_check(NFD, 3, 3); |
| 694 | unt_check(NFD, 5, 4); |
| 695 | unt_check(NFD, 5, 5); |
| 696 | unt_check(NFKD, 5, 1); |
| 697 | unt_check(NFKD, 5, 2); |
| 698 | unt_check(NFKD, 5, 3); |
| 699 | unt_check(NFKD, 5, 4); |
| 700 | unt_check(NFKD, 5, 5); |
| 701 | unt_check(NFC, 2, 1); |
| 702 | unt_check(NFC, 2, 2); |
| 703 | unt_check(NFC, 2, 3); |
| 704 | unt_check(NFC, 4, 4); |
| 705 | unt_check(NFC, 4, 5); |
| 706 | unt_check(NFKC, 4, 1); |
| 707 | unt_check(NFKC, 4, 2); |
| 708 | unt_check(NFKC, 4, 3); |
| 709 | unt_check(NFKC, 4, 4); |
| 710 | unt_check(NFKC, 4, 5); |
| 711 | for(cn = 1; cn <= 5; ++cn) { |
| 712 | xfree(NFD_c[cn]); |
| 713 | xfree(NFKD_c[cn]); |
| 714 | } |
| 715 | xfree(l); |
| 716 | } |
| 717 | fclose(fp); |
| 718 | breaktest("auxiliary/GraphemeBreakTest.txt", utf32_is_grapheme_boundary); |
| 719 | breaktest("auxiliary/WordBreakTest.txt", utf32_is_word_boundary); |
| 720 | } |
| 721 | |
| 722 | int main(void) { |
| 723 | fail_first = !!getenv("FAIL_FIRST"); |
| 724 | insist('\n' == 0x0A); |
| 725 | insist('\r' == 0x0D); |
| 726 | insist(' ' == 0x20); |
| 727 | insist('0' == 0x30); |
| 728 | insist('9' == 0x39); |
| 729 | insist('A' == 0x41); |
| 730 | insist('Z' == 0x5A); |
| 731 | insist('a' == 0x61); |
| 732 | insist('z' == 0x7A); |
| 733 | /* addr.c */ |
| 734 | /* asprintf.c */ |
| 735 | /* authhash.c */ |
| 736 | /* basen.c */ |
| 737 | /* charset.c */ |
| 738 | /* client.c */ |
| 739 | /* configuration.c */ |
| 740 | /* event.c */ |
| 741 | /* fprintf.c */ |
| 742 | /* heap.c */ |
| 743 | test_heap(); |
| 744 | /* hex.c */ |
| 745 | test_hex(); |
| 746 | /* inputline.c */ |
| 747 | /* kvp.c */ |
| 748 | /* log.c */ |
| 749 | /* mem.c */ |
| 750 | /* mime.c */ |
| 751 | test_mime(); |
| 752 | test_cookies(); |
| 753 | /* mixer.c */ |
| 754 | /* plugin.c */ |
| 755 | /* printf.c */ |
| 756 | /* queue.c */ |
| 757 | /* sink.c */ |
| 758 | /* snprintf.c */ |
| 759 | /* split.c */ |
| 760 | /* syscalls.c */ |
| 761 | /* table.c */ |
| 762 | /* unicode.c */ |
| 763 | test_unicode(); |
| 764 | /* utf8.c */ |
| 765 | test_utf8(); |
| 766 | /* vector.c */ |
| 767 | /* words.c */ |
| 768 | test_casefold(); |
| 769 | test_words(); |
| 770 | /* XXX words() */ |
| 771 | /* wstat.c */ |
| 772 | fprintf(stderr, "%d errors out of %d tests\n", errors, tests); |
| 773 | return !!errors; |
| 774 | } |
| 775 | |
| 776 | /* |
| 777 | Local Variables: |
| 778 | c-basic-offset:2 |
| 779 | comment-column:40 |
| 780 | fill-column:79 |
| 781 | indent-tabs-mode:nil |
| 782 | End: |
| 783 | */ |