| 1 |
// -*- coding: utf-8 -*-
|
| 2 |
//
|
| 3 |
// Copyright (c) 2005 - 2006, Google Inc.
|
| 4 |
// All rights reserved.
|
| 5 |
//
|
| 6 |
// Redistribution and use in source and binary forms, with or without
|
| 7 |
// modification, are permitted provided that the following conditions are
|
| 8 |
// met:
|
| 9 |
//
|
| 10 |
// * Redistributions of source code must retain the above copyright
|
| 11 |
// notice, this list of conditions and the following disclaimer.
|
| 12 |
// * Redistributions in binary form must reproduce the above
|
| 13 |
// copyright notice, this list of conditions and the following disclaimer
|
| 14 |
// in the documentation and/or other materials provided with the
|
| 15 |
// distribution.
|
| 16 |
// * Neither the name of Google Inc. nor the names of its
|
| 17 |
// contributors may be used to endorse or promote products derived from
|
| 18 |
// this software without specific prior written permission.
|
| 19 |
//
|
| 20 |
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
| 21 |
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
| 22 |
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
| 23 |
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
| 24 |
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
| 25 |
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
| 26 |
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
| 27 |
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
| 28 |
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
| 29 |
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
| 30 |
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
| 31 |
//
|
| 32 |
// Author: Sanjay Ghemawat
|
| 33 |
//
|
| 34 |
// TODO: Test extractions for PartialMatch/Consume
|
| 35 |
|
| 36 |
#ifdef HAVE_CONFIG_H
|
| 37 |
#include "config.h"
|
| 38 |
#endif
|
| 39 |
|
| 40 |
#ifdef HAVE_WINDOWS_H
|
| 41 |
#define snprintf _snprintf
|
| 42 |
#endif
|
| 43 |
|
| 44 |
#include <stdio.h>
|
| 45 |
#include <cassert>
|
| 46 |
#include <vector>
|
| 47 |
#include "pcrecpp.h"
|
| 48 |
|
| 49 |
using pcrecpp::StringPiece;
|
| 50 |
using pcrecpp::RE;
|
| 51 |
using pcrecpp::RE_Options;
|
| 52 |
using pcrecpp::Hex;
|
| 53 |
using pcrecpp::Octal;
|
| 54 |
using pcrecpp::CRadix;
|
| 55 |
|
| 56 |
static bool VERBOSE_TEST = false;
|
| 57 |
|
| 58 |
// CHECK dies with a fatal error if condition is not true. It is *not*
|
| 59 |
// controlled by NDEBUG, so the check will be executed regardless of
|
| 60 |
// compilation mode. Therefore, it is safe to do things like:
|
| 61 |
// CHECK_EQ(fp->Write(x), 4)
|
| 62 |
#define CHECK(condition) do { \
|
| 63 |
if (!(condition)) { \
|
| 64 |
fprintf(stderr, "%s:%d: Check failed: %s\n", \
|
| 65 |
__FILE__, __LINE__, #condition); \
|
| 66 |
exit(1); \
|
| 67 |
} \
|
| 68 |
} while (0)
|
| 69 |
|
| 70 |
#define CHECK_EQ(a, b) CHECK(a == b)
|
| 71 |
|
| 72 |
static void Timing1(int num_iters) {
|
| 73 |
// Same pattern lots of times
|
| 74 |
RE pattern("ruby:\\d+");
|
| 75 |
StringPiece p("ruby:1234");
|
| 76 |
for (int j = num_iters; j > 0; j--) {
|
| 77 |
CHECK(pattern.FullMatch(p));
|
| 78 |
}
|
| 79 |
}
|
| 80 |
|
| 81 |
static void Timing2(int num_iters) {
|
| 82 |
// Same pattern lots of times
|
| 83 |
RE pattern("ruby:(\\d+)");
|
| 84 |
int i;
|
| 85 |
for (int j = num_iters; j > 0; j--) {
|
| 86 |
CHECK(pattern.FullMatch("ruby:1234", &i));
|
| 87 |
CHECK_EQ(i, 1234);
|
| 88 |
}
|
| 89 |
}
|
| 90 |
|
| 91 |
static void Timing3(int num_iters) {
|
| 92 |
string text_string;
|
| 93 |
for (int j = num_iters; j > 0; j--) {
|
| 94 |
text_string += "this is another line\n";
|
| 95 |
}
|
| 96 |
|
| 97 |
RE line_matcher(".*\n");
|
| 98 |
string line;
|
| 99 |
StringPiece text(text_string);
|
| 100 |
int counter = 0;
|
| 101 |
while (line_matcher.Consume(&text)) {
|
| 102 |
counter++;
|
| 103 |
}
|
| 104 |
printf("Matched %d lines\n", counter);
|
| 105 |
}
|
| 106 |
|
| 107 |
#if 0 // uncomment this if you have a way of defining VirtualProcessSize()
|
| 108 |
|
| 109 |
static void LeakTest() {
|
| 110 |
// Check for memory leaks
|
| 111 |
unsigned long long initial_size = 0;
|
| 112 |
for (int i = 0; i < 100000; i++) {
|
| 113 |
if (i == 50000) {
|
| 114 |
initial_size = VirtualProcessSize();
|
| 115 |
printf("Size after 50000: %llu\n", initial_size);
|
| 116 |
}
|
| 117 |
char buf[100];
|
| 118 |
snprintf(buf, sizeof(buf), "pat%09d", i);
|
| 119 |
RE newre(buf);
|
| 120 |
}
|
| 121 |
uint64 final_size = VirtualProcessSize();
|
| 122 |
printf("Size after 100000: %llu\n", final_size);
|
| 123 |
const double growth = double(final_size - initial_size) / final_size;
|
| 124 |
printf("Growth: %0.2f%%", growth * 100);
|
| 125 |
CHECK(growth < 0.02); // Allow < 2% growth
|
| 126 |
}
|
| 127 |
|
| 128 |
#endif
|
| 129 |
|
| 130 |
static void RadixTests() {
|
| 131 |
printf("Testing hex\n");
|
| 132 |
|
| 133 |
#define CHECK_HEX(type, value) \
|
| 134 |
do { \
|
| 135 |
type v; \
|
| 136 |
CHECK(RE("([0-9a-fA-F]+)[uUlL]*").FullMatch(#value, Hex(&v))); \
|
| 137 |
CHECK_EQ(v, 0x ## value); \
|
| 138 |
CHECK(RE("([0-9a-fA-FxX]+)[uUlL]*").FullMatch("0x" #value, CRadix(&v))); \
|
| 139 |
CHECK_EQ(v, 0x ## value); \
|
| 140 |
} while(0)
|
| 141 |
|
| 142 |
CHECK_HEX(short, 2bad);
|
| 143 |
CHECK_HEX(unsigned short, 2badU);
|
| 144 |
CHECK_HEX(int, dead);
|
| 145 |
CHECK_HEX(unsigned int, deadU);
|
| 146 |
CHECK_HEX(long, 7eadbeefL);
|
| 147 |
CHECK_HEX(unsigned long, deadbeefUL);
|
| 148 |
#ifdef HAVE_LONG_LONG
|
| 149 |
CHECK_HEX(long long, 12345678deadbeefLL);
|
| 150 |
#endif
|
| 151 |
#ifdef HAVE_UNSIGNED_LONG_LONG
|
| 152 |
CHECK_HEX(unsigned long long, cafebabedeadbeefULL);
|
| 153 |
#endif
|
| 154 |
|
| 155 |
#undef CHECK_HEX
|
| 156 |
|
| 157 |
printf("Testing octal\n");
|
| 158 |
|
| 159 |
#define CHECK_OCTAL(type, value) \
|
| 160 |
do { \
|
| 161 |
type v; \
|
| 162 |
CHECK(RE("([0-7]+)[uUlL]*").FullMatch(#value, Octal(&v))); \
|
| 163 |
CHECK_EQ(v, 0 ## value); \
|
| 164 |
CHECK(RE("([0-9a-fA-FxX]+)[uUlL]*").FullMatch("0" #value, CRadix(&v))); \
|
| 165 |
CHECK_EQ(v, 0 ## value); \
|
| 166 |
} while(0)
|
| 167 |
|
| 168 |
CHECK_OCTAL(short, 77777);
|
| 169 |
CHECK_OCTAL(unsigned short, 177777U);
|
| 170 |
CHECK_OCTAL(int, 17777777777);
|
| 171 |
CHECK_OCTAL(unsigned int, 37777777777U);
|
| 172 |
CHECK_OCTAL(long, 17777777777L);
|
| 173 |
CHECK_OCTAL(unsigned long, 37777777777UL);
|
| 174 |
#ifdef HAVE_LONG_LONG
|
| 175 |
CHECK_OCTAL(long long, 777777777777777777777LL);
|
| 176 |
#endif
|
| 177 |
#ifdef HAVE_UNSIGNED_LONG_LONG
|
| 178 |
CHECK_OCTAL(unsigned long long, 1777777777777777777777ULL);
|
| 179 |
#endif
|
| 180 |
|
| 181 |
#undef CHECK_OCTAL
|
| 182 |
|
| 183 |
printf("Testing decimal\n");
|
| 184 |
|
| 185 |
#define CHECK_DECIMAL(type, value) \
|
| 186 |
do { \
|
| 187 |
type v; \
|
| 188 |
CHECK(RE("(-?[0-9]+)[uUlL]*").FullMatch(#value, &v)); \
|
| 189 |
CHECK_EQ(v, value); \
|
| 190 |
CHECK(RE("(-?[0-9a-fA-FxX]+)[uUlL]*").FullMatch(#value, CRadix(&v))); \
|
| 191 |
CHECK_EQ(v, value); \
|
| 192 |
} while(0)
|
| 193 |
|
| 194 |
CHECK_DECIMAL(short, -1);
|
| 195 |
CHECK_DECIMAL(unsigned short, 9999);
|
| 196 |
CHECK_DECIMAL(int, -1000);
|
| 197 |
CHECK_DECIMAL(unsigned int, 12345U);
|
| 198 |
CHECK_DECIMAL(long, -10000000L);
|
| 199 |
CHECK_DECIMAL(unsigned long, 3083324652U);
|
| 200 |
#ifdef HAVE_LONG_LONG
|
| 201 |
CHECK_DECIMAL(long long, -100000000000000LL);
|
| 202 |
#endif
|
| 203 |
#ifdef HAVE_UNSIGNED_LONG_LONG
|
| 204 |
CHECK_DECIMAL(unsigned long long, 1234567890987654321ULL);
|
| 205 |
#endif
|
| 206 |
|
| 207 |
#undef CHECK_DECIMAL
|
| 208 |
|
| 209 |
}
|
| 210 |
|
| 211 |
static void TestReplace() {
|
| 212 |
printf("Testing Replace\n");
|
| 213 |
|
| 214 |
struct ReplaceTest {
|
| 215 |
const char *regexp;
|
| 216 |
const char *rewrite;
|
| 217 |
const char *original;
|
| 218 |
const char *single;
|
| 219 |
const char *global;
|
| 220 |
};
|
| 221 |
static const ReplaceTest tests[] = {
|
| 222 |
{ "(qu|[b-df-hj-np-tv-z]*)([a-z]+)",
|
| 223 |
"\\2\\1ay",
|
| 224 |
"the quick brown fox jumps over the lazy dogs.",
|
| 225 |
"ethay quick brown fox jumps over the lazy dogs.",
|
| 226 |
"ethay ickquay ownbray oxfay umpsjay overay ethay azylay ogsday." },
|
| 227 |
{ "\\w+",
|
| 228 |
"\\0-NOSPAM",
|
| 229 |
"paul.haahr@google.com",
|
| 230 |
"paul-NOSPAM.haahr@google.com",
|
| 231 |
"paul-NOSPAM.haahr-NOSPAM@google-NOSPAM.com-NOSPAM" },
|
| 232 |
{ "^",
|
| 233 |
"(START)",
|
| 234 |
"foo",
|
| 235 |
"(START)foo",
|
| 236 |
"(START)foo" },
|
| 237 |
{ "^",
|
| 238 |
"(START)",
|
| 239 |
"",
|
| 240 |
"(START)",
|
| 241 |
"(START)" },
|
| 242 |
{ "$",
|
| 243 |
"(END)",
|
| 244 |
"",
|
| 245 |
"(END)",
|
| 246 |
"(END)" },
|
| 247 |
{ "b",
|
| 248 |
"bb",
|
| 249 |
"ababababab",
|
| 250 |
"abbabababab",
|
| 251 |
"abbabbabbabbabb" },
|
| 252 |
{ "b",
|
| 253 |
"bb",
|
| 254 |
"bbbbbb",
|
| 255 |
"bbbbbbb",
|
| 256 |
"bbbbbbbbbbbb" },
|
| 257 |
{ "b+",
|
| 258 |
"bb",
|
| 259 |
"bbbbbb",
|
| 260 |
"bb",
|
| 261 |
"bb" },
|
| 262 |
{ "b*",
|
| 263 |
"bb",
|
| 264 |
"bbbbbb",
|
| 265 |
"bb",
|
| 266 |
"bb" },
|
| 267 |
{ "b*",
|
| 268 |
"bb",
|
| 269 |
"aaaaa",
|
| 270 |
"bbaaaaa",
|
| 271 |
"bbabbabbabbabbabb" },
|
| 272 |
{ "b*",
|
| 273 |
"bb",
|
| 274 |
"aa\naa\n",
|
| 275 |
"bbaa\naa\n",
|
| 276 |
"bbabbabb\nbbabbabb\nbb" },
|
| 277 |
{ "b*",
|
| 278 |
"bb",
|
| 279 |
"aa\raa\r",
|
| 280 |
"bbaa\raa\r",
|
| 281 |
"bbabbabb\rbbabbabb\rbb" },
|
| 282 |
{ "b*",
|
| 283 |
"bb",
|
| 284 |
"aa\r\naa\r\n",
|
| 285 |
"bbaa\r\naa\r\n",
|
| 286 |
"bbabbabb\r\nbbabbabb\r\nbb" },
|
| 287 |
#ifdef SUPPORT_UTF8
|
| 288 |
{ "b*",
|
| 289 |
"bb",
|
| 290 |
"\xE3\x83\x9B\xE3\x83\xBC\xE3\x83\xA0\xE3\x81\xB8", // utf8
|
| 291 |
"bb\xE3\x83\x9B\xE3\x83\xBC\xE3\x83\xA0\xE3\x81\xB8",
|
| 292 |
"bb\xE3\x83\x9B""bb""\xE3\x83\xBC""bb""\xE3\x83\xA0""bb""\xE3\x81\xB8""bb" },
|
| 293 |
{ "b*",
|
| 294 |
"bb",
|
| 295 |
"\xE3\x83\x9B\r\n\xE3\x83\xBC\r\xE3\x83\xA0\n\xE3\x81\xB8\r\n", // utf8
|
| 296 |
"bb\xE3\x83\x9B\r\n\xE3\x83\xBC\r\xE3\x83\xA0\n\xE3\x81\xB8\r\n",
|
| 297 |
("bb\xE3\x83\x9B""bb\r\nbb""\xE3\x83\xBC""bb\rbb""\xE3\x83\xA0"
|
| 298 |
"bb\nbb""\xE3\x81\xB8""bb\r\nbb") },
|
| 299 |
#endif
|
| 300 |
{ "", NULL, NULL, NULL, NULL }
|
| 301 |
};
|
| 302 |
|
| 303 |
#ifdef SUPPORT_UTF8
|
| 304 |
const bool support_utf8 = true;
|
| 305 |
#else
|
| 306 |
const bool support_utf8 = false;
|
| 307 |
#endif
|
| 308 |
|
| 309 |
for (const ReplaceTest *t = tests; t->original != NULL; ++t) {
|
| 310 |
RE re(t->regexp, RE_Options(PCRE_NEWLINE_CRLF).set_utf8(support_utf8));
|
| 311 |
assert(re.error().empty());
|
| 312 |
string one(t->original);
|
| 313 |
CHECK(re.Replace(t->rewrite, &one));
|
| 314 |
CHECK_EQ(one, t->single);
|
| 315 |
string all(t->original);
|
| 316 |
CHECK(re.GlobalReplace(t->rewrite, &all) > 0);
|
| 317 |
CHECK_EQ(all, t->global);
|
| 318 |
}
|
| 319 |
|
| 320 |
// One final test: test \r\n replacement when we're not in CRLF mode
|
| 321 |
{
|
| 322 |
RE re("b*", RE_Options(PCRE_NEWLINE_CR).set_utf8(support_utf8));
|
| 323 |
assert(re.error().empty());
|
| 324 |
string all("aa\r\naa\r\n");
|
| 325 |
CHECK(re.GlobalReplace("bb", &all) > 0);
|
| 326 |
CHECK_EQ(all, string("bbabbabb\rbb\nbbabbabb\rbb\nbb"));
|
| 327 |
}
|
| 328 |
{
|
| 329 |
RE re("b*", RE_Options(PCRE_NEWLINE_LF).set_utf8(support_utf8));
|
| 330 |
assert(re.error().empty());
|
| 331 |
string all("aa\r\naa\r\n");
|
| 332 |
CHECK(re.GlobalReplace("bb", &all) > 0);
|
| 333 |
CHECK_EQ(all, string("bbabbabb\rbb\nbbabbabb\rbb\nbb"));
|
| 334 |
}
|
| 335 |
// TODO: test what happens when no PCRE_NEWLINE_* flag is set.
|
| 336 |
// Alas, the answer depends on how pcre was compiled.
|
| 337 |
}
|
| 338 |
|
| 339 |
static void TestExtract() {
|
| 340 |
printf("Testing Extract\n");
|
| 341 |
|
| 342 |
string s;
|
| 343 |
|
| 344 |
CHECK(RE("(.*)@([^.]*)").Extract("\\2!\\1", "boris@kremvax.ru", &s));
|
| 345 |
CHECK_EQ(s, "kremvax!boris");
|
| 346 |
|
| 347 |
// check the RE interface as well
|
| 348 |
CHECK(RE(".*").Extract("'\\0'", "foo", &s));
|
| 349 |
CHECK_EQ(s, "'foo'");
|
| 350 |
CHECK(!RE("bar").Extract("'\\0'", "baz", &s));
|
| 351 |
CHECK_EQ(s, "'foo'");
|
| 352 |
}
|
| 353 |
|
| 354 |
static void TestConsume() {
|
| 355 |
printf("Testing Consume\n");
|
| 356 |
|
| 357 |
string word;
|
| 358 |
|
| 359 |
string s(" aaa b!@#$@#$cccc");
|
| 360 |
StringPiece input(s);
|
| 361 |
|
| 362 |
RE r("\\s*(\\w+)"); // matches a word, possibly proceeded by whitespace
|
| 363 |
CHECK(r.Consume(&input, &word));
|
| 364 |
CHECK_EQ(word, "aaa");
|
| 365 |
CHECK(r.Consume(&input, &word));
|
| 366 |
CHECK_EQ(word, "b");
|
| 367 |
CHECK(! r.Consume(&input, &word));
|
| 368 |
}
|
| 369 |
|
| 370 |
static void TestFindAndConsume() {
|
| 371 |
printf("Testing FindAndConsume\n");
|
| 372 |
|
| 373 |
string word;
|
| 374 |
|
| 375 |
string s(" aaa b!@#$@#$cccc");
|
| 376 |
StringPiece input(s);
|
| 377 |
|
| 378 |
RE r("(\\w+)"); // matches a word
|
| 379 |
CHECK(r.FindAndConsume(&input, &word));
|
| 380 |
CHECK_EQ(word, "aaa");
|
| 381 |
CHECK(r.FindAndConsume(&input, &word));
|
| 382 |
CHECK_EQ(word, "b");
|
| 383 |
CHECK(r.FindAndConsume(&input, &word));
|
| 384 |
CHECK_EQ(word, "cccc");
|
| 385 |
CHECK(! r.FindAndConsume(&input, &word));
|
| 386 |
}
|
| 387 |
|
| 388 |
static void TestMatchNumberPeculiarity() {
|
| 389 |
printf("Testing match-number peculiaraity\n");
|
| 390 |
|
| 391 |
string word1;
|
| 392 |
string word2;
|
| 393 |
string word3;
|
| 394 |
|
| 395 |
RE r("(foo)|(bar)|(baz)");
|
| 396 |
CHECK(r.PartialMatch("foo", &word1, &word2, &word3));
|
| 397 |
CHECK_EQ(word1, "foo");
|
| 398 |
CHECK_EQ(word2, "");
|
| 399 |
CHECK_EQ(word3, "");
|
| 400 |
CHECK(r.PartialMatch("bar", &word1, &word2, &word3));
|
| 401 |
CHECK_EQ(word1, "");
|
| 402 |
CHECK_EQ(word2, "bar");
|
| 403 |
CHECK_EQ(word3, "");
|
| 404 |
CHECK(r.PartialMatch("baz", &word1, &word2, &word3));
|
| 405 |
CHECK_EQ(word1, "");
|
| 406 |
CHECK_EQ(word2, "");
|
| 407 |
CHECK_EQ(word3, "baz");
|
| 408 |
CHECK(!r.PartialMatch("f", &word1, &word2, &word3));
|
| 409 |
|
| 410 |
string a;
|
| 411 |
CHECK(RE("(foo)|hello").FullMatch("hello", &a));
|
| 412 |
CHECK_EQ(a, "");
|
| 413 |
}
|
| 414 |
|
| 415 |
static void TestRecursion() {
|
| 416 |
printf("Testing recursion\n");
|
| 417 |
|
| 418 |
// Get one string that passes (sometimes), one that never does.
|
| 419 |
string text_good("abcdefghijk");
|
| 420 |
string text_bad("acdefghijkl");
|
| 421 |
|
| 422 |
// According to pcretest, matching text_good against (\w+)*b
|
| 423 |
// requires match_limit of at least 8192, and match_recursion_limit
|
| 424 |
// of at least 37.
|
| 425 |
|
| 426 |
RE_Options options_ml;
|
| 427 |
options_ml.set_match_limit(8192);
|
| 428 |
RE re("(\\w+)*b", options_ml);
|
| 429 |
CHECK(re.PartialMatch(text_good) == true);
|
| 430 |
CHECK(re.PartialMatch(text_bad) == false);
|
| 431 |
CHECK(re.FullMatch(text_good) == false);
|
| 432 |
CHECK(re.FullMatch(text_bad) == false);
|
| 433 |
|
| 434 |
options_ml.set_match_limit(1024);
|
| 435 |
RE re2("(\\w+)*b", options_ml);
|
| 436 |
CHECK(re2.PartialMatch(text_good) == false); // because of match_limit
|
| 437 |
CHECK(re2.PartialMatch(text_bad) == false);
|
| 438 |
CHECK(re2.FullMatch(text_good) == false);
|
| 439 |
CHECK(re2.FullMatch(text_bad) == false);
|
| 440 |
|
| 441 |
RE_Options options_mlr;
|
| 442 |
options_mlr.set_match_limit_recursion(50);
|
| 443 |
RE re3("(\\w+)*b", options_mlr);
|
| 444 |
CHECK(re3.PartialMatch(text_good) == true);
|
| 445 |
CHECK(re3.PartialMatch(text_bad) == false);
|
| 446 |
CHECK(re3.FullMatch(text_good) == false);
|
| 447 |
CHECK(re3.FullMatch(text_bad) == false);
|
| 448 |
|
| 449 |
options_mlr.set_match_limit_recursion(10);
|
| 450 |
RE re4("(\\w+)*b", options_mlr);
|
| 451 |
CHECK(re4.PartialMatch(text_good) == false);
|
| 452 |
CHECK(re4.PartialMatch(text_bad) == false);
|
| 453 |
CHECK(re4.FullMatch(text_good) == false);
|
| 454 |
CHECK(re4.FullMatch(text_bad) == false);
|
| 455 |
}
|
| 456 |
|
| 457 |
// A meta-quoted string, interpreted as a pattern, should always match
|
| 458 |
// the original unquoted string.
|
| 459 |
static void TestQuoteMeta(string unquoted, RE_Options options = RE_Options()) {
|
| 460 |
string quoted = RE::QuoteMeta(unquoted);
|
| 461 |
RE re(quoted, options);
|
| 462 |
CHECK(re.FullMatch(unquoted));
|
| 463 |
}
|
| 464 |
|
| 465 |
// A string containing meaningful regexp characters, which is then meta-
|
| 466 |
// quoted, should not generally match a string the unquoted string does.
|
| 467 |
static void NegativeTestQuoteMeta(string unquoted, string should_not_match,
|
| 468 |
RE_Options options = RE_Options()) {
|
| 469 |
string quoted = RE::QuoteMeta(unquoted);
|
| 470 |
RE re(quoted, options);
|
| 471 |
CHECK(!re.FullMatch(should_not_match));
|
| 472 |
}
|
| 473 |
|
| 474 |
// Tests that quoted meta characters match their original strings,
|
| 475 |
// and that a few things that shouldn't match indeed do not.
|
| 476 |
static void TestQuotaMetaSimple() {
|
| 477 |
TestQuoteMeta("foo");
|
| 478 |
TestQuoteMeta("foo.bar");
|
| 479 |
TestQuoteMeta("foo\\.bar");
|
| 480 |
TestQuoteMeta("[1-9]");
|
| 481 |
TestQuoteMeta("1.5-2.0?");
|
| 482 |
TestQuoteMeta("\\d");
|
| 483 |
TestQuoteMeta("Who doesn't like ice cream?");
|
| 484 |
TestQuoteMeta("((a|b)c?d*e+[f-h]i)");
|
| 485 |
TestQuoteMeta("((?!)xxx).*yyy");
|
| 486 |
TestQuoteMeta("([");
|
| 487 |
}
|
| 488 |
|
| 489 |
static void TestQuoteMetaSimpleNegative() {
|
| 490 |
NegativeTestQuoteMeta("foo", "bar");
|
| 491 |
NegativeTestQuoteMeta("...", "bar");
|
| 492 |
NegativeTestQuoteMeta("\\.", ".");
|
| 493 |
NegativeTestQuoteMeta("\\.", "..");
|
| 494 |
NegativeTestQuoteMeta("(a)", "a");
|
| 495 |
NegativeTestQuoteMeta("(a|b)", "a");
|
| 496 |
NegativeTestQuoteMeta("(a|b)", "(a)");
|
| 497 |
NegativeTestQuoteMeta("(a|b)", "a|b");
|
| 498 |
NegativeTestQuoteMeta("[0-9]", "0");
|
| 499 |
NegativeTestQuoteMeta("[0-9]", "0-9");
|
| 500 |
NegativeTestQuoteMeta("[0-9]", "[9]");
|
| 501 |
NegativeTestQuoteMeta("((?!)xxx)", "xxx");
|
| 502 |
}
|
| 503 |
|
| 504 |
static void TestQuoteMetaLatin1() {
|
| 505 |
TestQuoteMeta("3\xb2 = 9");
|
| 506 |
}
|
| 507 |
|
| 508 |
static void TestQuoteMetaUtf8() {
|
| 509 |
#ifdef SUPPORT_UTF8
|
| 510 |
TestQuoteMeta("Pl\xc3\xa1\x63ido Domingo", pcrecpp::UTF8());
|
| 511 |
TestQuoteMeta("xyz", pcrecpp::UTF8()); // No fancy utf8
|
| 512 |
TestQuoteMeta("\xc2\xb0", pcrecpp::UTF8()); // 2-byte utf8 (degree symbol)
|
| 513 |
TestQuoteMeta("27\xc2\xb0 degrees", pcrecpp::UTF8()); // As a middle character
|
| 514 |
TestQuoteMeta("\xe2\x80\xb3", pcrecpp::UTF8()); // 3-byte utf8 (double prime)
|
| 515 |
TestQuoteMeta("\xf0\x9d\x85\x9f", pcrecpp::UTF8()); // 4-byte utf8 (music note)
|
| 516 |
TestQuoteMeta("27\xc2\xb0"); // Interpreted as Latin-1, but should still work
|
| 517 |
NegativeTestQuoteMeta("27\xc2\xb0", // 2-byte utf (degree symbol)
|
| 518 |
"27\\\xc2\\\xb0",
|
| 519 |
pcrecpp::UTF8());
|
| 520 |
#endif
|
| 521 |
}
|
| 522 |
|
| 523 |
static void TestQuoteMetaAll() {
|
| 524 |
printf("Testing QuoteMeta\n");
|
| 525 |
TestQuotaMetaSimple();
|
| 526 |
TestQuoteMetaSimpleNegative();
|
| 527 |
TestQuoteMetaLatin1();
|
| 528 |
TestQuoteMetaUtf8();
|
| 529 |
}
|
| 530 |
|
| 531 |
//
|
| 532 |
// Options tests contributed by
|
| 533 |
// Giuseppe Maxia, CTO, Stardata s.r.l.
|
| 534 |
// July 2005
|
| 535 |
//
|
| 536 |
static void GetOneOptionResult(
|
| 537 |
const char *option_name,
|
| 538 |
const char *regex,
|
| 539 |
const char *str,
|
| 540 |
RE_Options options,
|
| 541 |
bool full,
|
| 542 |
string expected) {
|
| 543 |
|
| 544 |
printf("Testing Option <%s>\n", option_name);
|
| 545 |
if(VERBOSE_TEST)
|
| 546 |
printf("/%s/ finds \"%s\" within \"%s\" \n",
|
| 547 |
regex,
|
| 548 |
expected.c_str(),
|
| 549 |
str);
|
| 550 |
string captured("");
|
| 551 |
if (full)
|
| 552 |
RE(regex,options).FullMatch(str, &captured);
|
| 553 |
else
|
| 554 |
RE(regex,options).PartialMatch(str, &captured);
|
| 555 |
CHECK_EQ(captured, expected);
|
| 556 |
}
|
| 557 |
|
| 558 |
static void TestOneOption(
|
| 559 |
const char *option_name,
|
| 560 |
const char *regex,
|
| 561 |
const char *str,
|
| 562 |
RE_Options options,
|
| 563 |
bool full,
|
| 564 |
bool assertive = true) {
|
| 565 |
|
| 566 |
printf("Testing Option <%s>\n", option_name);
|
| 567 |
if (VERBOSE_TEST)
|
| 568 |
printf("'%s' %s /%s/ \n",
|
| 569 |
str,
|
| 570 |
(assertive? "matches" : "doesn't match"),
|
| 571 |
regex);
|
| 572 |
if (assertive) {
|
| 573 |
if (full)
|
| 574 |
CHECK(RE(regex,options).FullMatch(str));
|
| 575 |
else
|
| 576 |
CHECK(RE(regex,options).PartialMatch(str));
|
| 577 |
} else {
|
| 578 |
if (full)
|
| 579 |
CHECK(!RE(regex,options).FullMatch(str));
|
| 580 |
else
|
| 581 |
CHECK(!RE(regex,options).PartialMatch(str));
|
| 582 |
}
|
| 583 |
}
|
| 584 |
|
| 585 |
static void Test_CASELESS() {
|
| 586 |
RE_Options options;
|
| 587 |
RE_Options options2;
|
| 588 |
|
| 589 |
options.set_caseless(true);
|
| 590 |
TestOneOption("CASELESS (class)", "HELLO", "hello", options, false);
|
| 591 |
TestOneOption("CASELESS (class2)", "HELLO", "hello", options2.set_caseless(true), false);
|
| 592 |
TestOneOption("CASELESS (class)", "^[A-Z]+$", "Hello", options, false);
|
| 593 |
|
| 594 |
TestOneOption("CASELESS (function)", "HELLO", "hello", pcrecpp::CASELESS(), false);
|
| 595 |
TestOneOption("CASELESS (function)", "^[A-Z]+$", "Hello", pcrecpp::CASELESS(), false);
|
| 596 |
options.set_caseless(false);
|
| 597 |
TestOneOption("no CASELESS", "HELLO", "hello", options, false, false);
|
| 598 |
}
|
| 599 |
|
| 600 |
static void Test_MULTILINE() {
|
| 601 |
RE_Options options;
|
| 602 |
RE_Options options2;
|
| 603 |
const char *str = "HELLO\n" "cruel\n" "world\n";
|
| 604 |
|
| 605 |
options.set_multiline(true);
|
| 606 |
TestOneOption("MULTILINE (class)", "^cruel$", str, options, false);
|
| 607 |
TestOneOption("MULTILINE (class2)", "^cruel$", str, options2.set_multiline(true), false);
|
| 608 |
TestOneOption("MULTILINE (function)", "^cruel$", str, pcrecpp::MULTILINE(), false);
|
| 609 |
options.set_multiline(false);
|
| 610 |
TestOneOption("no MULTILINE", "^cruel$", str, options, false, false);
|
| 611 |
}
|
| 612 |
|
| 613 |
static void Test_DOTALL() {
|
| 614 |
RE_Options options;
|
| 615 |
RE_Options options2;
|
| 616 |
const char *str = "HELLO\n" "cruel\n" "world";
|
| 617 |
|
| 618 |
options.set_dotall(true);
|
| 619 |
TestOneOption("DOTALL (class)", "HELLO.*world", str, options, true);
|
| 620 |
TestOneOption("DOTALL (class2)", "HELLO.*world", str, options2.set_dotall(true), true);
|
| 621 |
TestOneOption("DOTALL (function)", "HELLO.*world", str, pcrecpp::DOTALL(), true);
|
| 622 |
options.set_dotall(false);
|
| 623 |
TestOneOption("no DOTALL", "HELLO.*world", str, options, true, false);
|
| 624 |
}
|
| 625 |
|
| 626 |
static void Test_DOLLAR_ENDONLY() {
|
| 627 |
RE_Options options;
|
| 628 |
RE_Options options2;
|
| 629 |
const char *str = "HELLO world\n";
|
| 630 |
|
| 631 |
TestOneOption("no DOLLAR_ENDONLY", "world$", str, options, false);
|
| 632 |
options.set_dollar_endonly(true);
|
| 633 |
TestOneOption("DOLLAR_ENDONLY 1", "world$", str, options, false, false);
|
| 634 |
TestOneOption("DOLLAR_ENDONLY 2", "world$", str, options2.set_dollar_endonly(true), false, false);
|
| 635 |
}
|
| 636 |
|
| 637 |
static void Test_EXTRA() {
|
| 638 |
RE_Options options;
|
| 639 |
const char *str = "HELLO";
|
| 640 |
|
| 641 |
options.set_extra(true);
|
| 642 |
TestOneOption("EXTRA 1", "\\HELL\\O", str, options, true, false );
|
| 643 |
TestOneOption("EXTRA 2", "\\HELL\\O", str, RE_Options().set_extra(true), true, false );
|
| 644 |
options.set_extra(false);
|
| 645 |
TestOneOption("no EXTRA", "\\HELL\\O", str, options, true );
|
| 646 |
}
|
| 647 |
|
| 648 |
static void Test_EXTENDED() {
|
| 649 |
RE_Options options;
|
| 650 |
RE_Options options2;
|
| 651 |
const char *str = "HELLO world";
|
| 652 |
|
| 653 |
options.set_extended(true);
|
| 654 |
TestOneOption("EXTENDED (class)", "HELLO world", str, options, false, false);
|
| 655 |
TestOneOption("EXTENDED (class2)", "HELLO world", str, options2.set_extended(true), false, false);
|
| 656 |
TestOneOption("EXTENDED (class)",
|
| 657 |
"^ HE L{2} O "
|
| 658 |
"\\s+ "
|
| 659 |
"\\w+ $ ",
|
| 660 |
str,
|
| 661 |
options,
|
| 662 |
false);
|
| 663 |
|
| 664 |
TestOneOption("EXTENDED (function)", "HELLO world", str, pcrecpp::EXTENDED(), false, false);
|
| 665 |
TestOneOption("EXTENDED (function)",
|
| 666 |
"^ HE L{2} O "
|
| 667 |
"\\s+ "
|
| 668 |
"\\w+ $ ",
|
| 669 |
str,
|
| 670 |
pcrecpp::EXTENDED(),
|
| 671 |
false);
|
| 672 |
|
| 673 |
options.set_extended(false);
|
| 674 |
TestOneOption("no EXTENDED", "HELLO world", str, options, false);
|
| 675 |
}
|
| 676 |
|
| 677 |
static void Test_NO_AUTO_CAPTURE() {
|
| 678 |
RE_Options options;
|
| 679 |
const char *str = "HELLO world";
|
| 680 |
string captured;
|
| 681 |
|
| 682 |
printf("Testing Option <no NO_AUTO_CAPTURE>\n");
|
| 683 |
if (VERBOSE_TEST)
|
| 684 |
printf("parentheses capture text\n");
|
| 685 |
RE re("(world|universe)$", options);
|
| 686 |
CHECK(re.Extract("\\1", str , &captured));
|
| 687 |
CHECK_EQ(captured, "world");
|
| 688 |
options.set_no_auto_capture(true);
|
| 689 |
printf("testing Option <NO_AUTO_CAPTURE>\n");
|
| 690 |
if (VERBOSE_TEST)
|
| 691 |
printf("parentheses do not capture text\n");
|
| 692 |
re.Extract("\\1",str, &captured );
|
| 693 |
CHECK_EQ(captured, "world");
|
| 694 |
}
|
| 695 |
|
| 696 |
static void Test_UNGREEDY() {
|
| 697 |
RE_Options options;
|
| 698 |
const char *str = "HELLO, 'this' is the 'world'";
|
| 699 |
|
| 700 |
options.set_ungreedy(true);
|
| 701 |
GetOneOptionResult("UNGREEDY 1", "('.*')", str, options, false, "'this'" );
|
| 702 |
GetOneOptionResult("UNGREEDY 2", "('.*')", str, RE_Options().set_ungreedy(true), false, "'this'" );
|
| 703 |
GetOneOptionResult("UNGREEDY", "('.*?')", str, options, false, "'this' is the 'world'" );
|
| 704 |
|
| 705 |
options.set_ungreedy(false);
|
| 706 |
GetOneOptionResult("no UNGREEDY", "('.*')", str, options, false, "'this' is the 'world'" );
|
| 707 |
GetOneOptionResult("no UNGREEDY", "('.*?')", str, options, false, "'this'" );
|
| 708 |
}
|
| 709 |
|
| 710 |
static void Test_all_options() {
|
| 711 |
const char *str = "HELLO\n" "cruel\n" "world";
|
| 712 |
RE_Options options;
|
| 713 |
options.set_all_options(PCRE_CASELESS | PCRE_DOTALL);
|
| 714 |
|
| 715 |
TestOneOption("all_options (CASELESS|DOTALL)", "^hello.*WORLD", str , options, false);
|
| 716 |
options.set_all_options(0);
|
| 717 |
TestOneOption("all_options (0)", "^hello.*WORLD", str , options, false, false);
|
| 718 |
options.set_all_options(PCRE_MULTILINE | PCRE_EXTENDED);
|
| 719 |
|
| 720 |
TestOneOption("all_options (MULTILINE|EXTENDED)", " ^ c r u e l $ ", str, options, false);
|
| 721 |
TestOneOption("all_options (MULTILINE|EXTENDED) with constructor",
|
| 722 |
" ^ c r u e l $ ",
|
| 723 |
str,
|
| 724 |
RE_Options(PCRE_MULTILINE | PCRE_EXTENDED),
|
| 725 |
false);
|
| 726 |
|
| 727 |
TestOneOption("all_options (MULTILINE|EXTENDED) with concatenation",
|
| 728 |
" ^ c r u e l $ ",
|
| 729 |
str,
|
| 730 |
RE_Options()
|
| 731 |
.set_multiline(true)
|
| 732 |
.set_extended(true),
|
| 733 |
false);
|
| 734 |
|
| 735 |
options.set_all_options(0);
|
| 736 |
TestOneOption("all_options (0)", "^ c r u e l $", str, options, false, false);
|
| 737 |
|
| 738 |
}
|
| 739 |
|
| 740 |
static void TestOptions() {
|
| 741 |
printf("Testing Options\n");
|
| 742 |
Test_CASELESS();
|
| 743 |
Test_MULTILINE();
|
| 744 |
Test_DOTALL();
|
| 745 |
Test_DOLLAR_ENDONLY();
|
| 746 |
Test_EXTENDED();
|
| 747 |
Test_NO_AUTO_CAPTURE();
|
| 748 |
Test_UNGREEDY();
|
| 749 |
Test_EXTRA();
|
| 750 |
Test_all_options();
|
| 751 |
}
|
| 752 |
|
| 753 |
static void TestConstructors() {
|
| 754 |
printf("Testing constructors\n");
|
| 755 |
|
| 756 |
RE_Options options;
|
| 757 |
options.set_dotall(true);
|
| 758 |
const char *str = "HELLO\n" "cruel\n" "world";
|
| 759 |
|
| 760 |
RE orig("HELLO.*world", options);
|
| 761 |
CHECK(orig.FullMatch(str));
|
| 762 |
|
| 763 |
RE copy1(orig);
|
| 764 |
CHECK(copy1.FullMatch(str));
|
| 765 |
|
| 766 |
RE copy2("not a match");
|
| 767 |
CHECK(!copy2.FullMatch(str));
|
| 768 |
copy2 = copy1;
|
| 769 |
CHECK(copy2.FullMatch(str));
|
| 770 |
copy2 = orig;
|
| 771 |
CHECK(copy2.FullMatch(str));
|
| 772 |
|
| 773 |
// Make sure when we assign to ourselves, nothing bad happens
|
| 774 |
orig = orig;
|
| 775 |
copy1 = copy1;
|
| 776 |
copy2 = copy2;
|
| 777 |
CHECK(orig.FullMatch(str));
|
| 778 |
CHECK(copy1.FullMatch(str));
|
| 779 |
CHECK(copy2.FullMatch(str));
|
| 780 |
}
|
| 781 |
|
| 782 |
int main(int argc, char** argv) {
|
| 783 |
// Treat any flag as --help
|
| 784 |
if (argc > 1 && argv[1][0] == '-') {
|
| 785 |
printf("Usage: %s [timing1|timing2|timing3 num-iters]\n"
|
| 786 |
" If 'timingX ###' is specified, run the given timing test\n"
|
| 787 |
" with the given number of iterations, rather than running\n"
|
| 788 |
" the default corectness test.\n", argv[0]);
|
| 789 |
return 0;
|
| 790 |
}
|
| 791 |
|
| 792 |
if (argc > 1) {
|
| 793 |
if ( argc == 2 || atoi(argv[2]) == 0) {
|
| 794 |
printf("timing mode needs a num-iters argument\n");
|
| 795 |
return 1;
|
| 796 |
}
|
| 797 |
if (!strcmp(argv[1], "timing1"))
|
| 798 |
Timing1(atoi(argv[2]));
|
| 799 |
else if (!strcmp(argv[1], "timing2"))
|
| 800 |
Timing2(atoi(argv[2]));
|
| 801 |
else if (!strcmp(argv[1], "timing3"))
|
| 802 |
Timing3(atoi(argv[2]));
|
| 803 |
else
|
| 804 |
printf("Unknown argument '%s'\n", argv[1]);
|
| 805 |
return 0;
|
| 806 |
}
|
| 807 |
|
| 808 |
printf("Testing FullMatch\n");
|
| 809 |
|
| 810 |
int i;
|
| 811 |
string s;
|
| 812 |
|
| 813 |
/***** FullMatch with no args *****/
|
| 814 |
|
| 815 |
CHECK(RE("h.*o").FullMatch("hello"));
|
| 816 |
CHECK(!RE("h.*o").FullMatch("othello")); // Must be anchored at front
|
| 817 |
CHECK(!RE("h.*o").FullMatch("hello!")); // Must be anchored at end
|
| 818 |
CHECK(RE("a*").FullMatch("aaaa")); // Fullmatch with normal op
|
| 819 |
CHECK(RE("a*?").FullMatch("aaaa")); // Fullmatch with nongreedy op
|
| 820 |
CHECK(RE("a*?\\z").FullMatch("aaaa")); // Two unusual ops
|
| 821 |
|
| 822 |
/***** FullMatch with args *****/
|
| 823 |
|
| 824 |
// Zero-arg
|
| 825 |
CHECK(RE("\\d+").FullMatch("1001"));
|
| 826 |
|
| 827 |
// Single-arg
|
| 828 |
CHECK(RE("(\\d+)").FullMatch("1001", &i));
|
| 829 |
CHECK_EQ(i, 1001);
|
| 830 |
CHECK(RE("(-?\\d+)").FullMatch("-123", &i));
|
| 831 |
CHECK_EQ(i, -123);
|
| 832 |
CHECK(!RE("()\\d+").FullMatch("10", &i));
|
| 833 |
CHECK(!RE("(\\d+)").FullMatch("1234567890123456789012345678901234567890",
|
| 834 |
&i));
|
| 835 |
|
| 836 |
// Digits surrounding integer-arg
|
| 837 |
CHECK(RE("1(\\d*)4").FullMatch("1234", &i));
|
| 838 |
CHECK_EQ(i, 23);
|
| 839 |
CHECK(RE("(\\d)\\d+").FullMatch("1234", &i));
|
| 840 |
CHECK_EQ(i, 1);
|
| 841 |
CHECK(RE("(-\\d)\\d+").FullMatch("-1234", &i));
|
| 842 |
CHECK_EQ(i, -1);
|
| 843 |
CHECK(RE("(\\d)").PartialMatch("1234", &i));
|
| 844 |
CHECK_EQ(i, 1);
|
| 845 |
CHECK(RE("(-\\d)").PartialMatch("-1234", &i));
|
| 846 |
CHECK_EQ(i, -1);
|
| 847 |
|
| 848 |
// String-arg
|
| 849 |
CHECK(RE("h(.*)o").FullMatch("hello", &s));
|
| 850 |
CHECK_EQ(s, string("ell"));
|
| 851 |
|
| 852 |
// StringPiece-arg
|
| 853 |
StringPiece sp;
|
| 854 |
CHECK(RE("(\\w+):(\\d+)").FullMatch("ruby:1234", &sp, &i));
|
| 855 |
CHECK_EQ(sp.size(), 4);
|
| 856 |
CHECK(memcmp(sp.data(), "ruby", 4) == 0);
|
| 857 |
CHECK_EQ(i, 1234);
|
| 858 |
|
| 859 |
// Multi-arg
|
| 860 |
CHECK(RE("(\\w+):(\\d+)").FullMatch("ruby:1234", &s, &i));
|
| 861 |
CHECK_EQ(s, string("ruby"));
|
| 862 |
CHECK_EQ(i, 1234);
|
| 863 |
|
| 864 |
// Ignored arg
|
| 865 |
CHECK(RE("(\\w+)(:)(\\d+)").FullMatch("ruby:1234", &s, (void*)NULL, &i));
|
| 866 |
CHECK_EQ(s, string("ruby"));
|
| 867 |
CHECK_EQ(i, 1234);
|
| 868 |
|
| 869 |
// Type tests
|
| 870 |
{
|
| 871 |
char c;
|
| 872 |
CHECK(RE("(H)ello").FullMatch("Hello", &c));
|
| 873 |
CHECK_EQ(c, 'H');
|
| 874 |
}
|
| 875 |
{
|
| 876 |
unsigned char c;
|
| 877 |
CHECK(RE("(H)ello").FullMatch("Hello", &c));
|
| 878 |
CHECK_EQ(c, static_cast<unsigned char>('H'));
|
| 879 |
}
|
| 880 |
{
|
| 881 |
short v;
|
| 882 |
CHECK(RE("(-?\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100);
|
| 883 |
CHECK(RE("(-?\\d+)").FullMatch("-100", &v)); CHECK_EQ(v, -100);
|
| 884 |
CHECK(RE("(-?\\d+)").FullMatch("32767", &v)); CHECK_EQ(v, 32767);
|
| 885 |
CHECK(RE("(-?\\d+)").FullMatch("-32768", &v)); CHECK_EQ(v, -32768);
|
| 886 |
CHECK(!RE("(-?\\d+)").FullMatch("-32769", &v));
|
| 887 |
CHECK(!RE("(-?\\d+)").FullMatch("32768", &v));
|
| 888 |
}
|
| 889 |
{
|
| 890 |
unsigned short v;
|
| 891 |
CHECK(RE("(\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100);
|
| 892 |
CHECK(RE("(\\d+)").FullMatch("32767", &v)); CHECK_EQ(v, 32767);
|
| 893 |
CHECK(RE("(\\d+)").FullMatch("65535", &v)); CHECK_EQ(v, 65535);
|
| 894 |
CHECK(!RE("(\\d+)").FullMatch("65536", &v));
|
| 895 |
}
|
| 896 |
{
|
| 897 |
int v;
|
| 898 |
static const int max_value = 0x7fffffff;
|
| 899 |
static const int min_value = -max_value - 1;
|
| 900 |
CHECK(RE("(-?\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100);
|
| 901 |
CHECK(RE("(-?\\d+)").FullMatch("-100", &v)); CHECK_EQ(v, -100);
|
| 902 |
CHECK(RE("(-?\\d+)").FullMatch("2147483647", &v)); CHECK_EQ(v, max_value);
|
| 903 |
CHECK(RE("(-?\\d+)").FullMatch("-2147483648", &v)); CHECK_EQ(v, min_value);
|
| 904 |
CHECK(!RE("(-?\\d+)").FullMatch("-2147483649", &v));
|
| 905 |
CHECK(!RE("(-?\\d+)").FullMatch("2147483648", &v));
|
| 906 |
}
|
| 907 |
{
|
| 908 |
unsigned int v;
|
| 909 |
static const unsigned int max_value = 0xfffffffful;
|
| 910 |
CHECK(RE("(\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100);
|
| 911 |
CHECK(RE("(\\d+)").FullMatch("4294967295", &v)); CHECK_EQ(v, max_value);
|
| 912 |
CHECK(!RE("(\\d+)").FullMatch("4294967296", &v));
|
| 913 |
}
|
| 914 |
#ifdef HAVE_LONG_LONG
|
| 915 |
# if defined(__MINGW__) || defined(__MINGW32__)
|
| 916 |
# define LLD "%I64d"
|
| 917 |
# define LLU "%I64u"
|
| 918 |
# else
|
| 919 |
# define LLD "%lld"
|
| 920 |
# define LLU "%llu"
|
| 921 |
# endif
|
| 922 |
{
|
| 923 |
long long v;
|
| 924 |
static const long long max_value = 0x7fffffffffffffffLL;
|
| 925 |
static const long long min_value = -max_value - 1;
|
| 926 |
char buf[32];
|
| 927 |
|
| 928 |
CHECK(RE("(-?\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100);
|
| 929 |
CHECK(RE("(-?\\d+)").FullMatch("-100",&v)); CHECK_EQ(v, -100);
|
| 930 |
|
| 931 |
snprintf(buf, sizeof(buf), LLD, max_value);
|
| 932 |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value);
|
| 933 |
|
| 934 |
snprintf(buf, sizeof(buf), LLD, min_value);
|
| 935 |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, min_value);
|
| 936 |
|
| 937 |
snprintf(buf, sizeof(buf), LLD, max_value);
|
| 938 |
assert(buf[strlen(buf)-1] != '9');
|
| 939 |
buf[strlen(buf)-1]++;
|
| 940 |
CHECK(!RE("(-?\\d+)").FullMatch(buf, &v));
|
| 941 |
|
| 942 |
snprintf(buf, sizeof(buf), LLD, min_value);
|
| 943 |
assert(buf[strlen(buf)-1] != '9');
|
| 944 |
buf[strlen(buf)-1]++;
|
| 945 |
CHECK(!RE("(-?\\d+)").FullMatch(buf, &v));
|
| 946 |
}
|
| 947 |
#endif
|
| 948 |
#if defined HAVE_UNSIGNED_LONG_LONG && defined HAVE_LONG_LONG
|
| 949 |
{
|
| 950 |
unsigned long long v;
|
| 951 |
long long v2;
|
| 952 |
static const unsigned long long max_value = 0xffffffffffffffffULL;
|
| 953 |
char buf[32];
|
| 954 |
|
| 955 |
CHECK(RE("(-?\\d+)").FullMatch("100",&v)); CHECK_EQ(v, 100);
|
| 956 |
CHECK(RE("(-?\\d+)").FullMatch("-100",&v2)); CHECK_EQ(v2, -100);
|
| 957 |
|
| 958 |
snprintf(buf, sizeof(buf), LLU, max_value);
|
| 959 |
CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value);
|
| 960 |
|
| 961 |
assert(buf[strlen(buf)-1] != '9');
|
| 962 |
buf[strlen(buf)-1]++;
|
| 963 |
CHECK(!RE("(-?\\d+)").FullMatch(buf, &v));
|
| 964 |
}
|
| 965 |
#endif
|
| 966 |
{
|
| 967 |
float v;
|
| 968 |
CHECK(RE("(.*)").FullMatch("100", &v));
|
| 969 |
CHECK(RE("(.*)").FullMatch("-100.", &v));
|
| 970 |
CHECK(RE("(.*)").FullMatch("1e23", &v));
|
| 971 |
}
|
| 972 |
{
|
| 973 |
double v;
|
| 974 |
CHECK(RE("(.*)").FullMatch("100", &v));
|
| 975 |
CHECK(RE("(.*)").FullMatch("-100.", &v));
|
| 976 |
CHECK(RE("(.*)").FullMatch("1e23", &v));
|
| 977 |
}
|
| 978 |
|
| 979 |
// Check that matching is fully anchored
|
| 980 |
CHECK(!RE("(\\d+)").FullMatch("x1001", &i));
|
| 981 |
CHECK(!RE("(\\d+)").FullMatch("1001x", &i));
|
| 982 |
CHECK(RE("x(\\d+)").FullMatch("x1001", &i)); CHECK_EQ(i, 1001);
|
| 983 |
CHECK(RE("(\\d+)x").FullMatch("1001x", &i)); CHECK_EQ(i, 1001);
|
| 984 |
|
| 985 |
// Braces
|
| 986 |
CHECK(RE("[0-9a-f+.-]{5,}").FullMatch("0abcd"));
|
| 987 |
CHECK(RE("[0-9a-f+.-]{5,}").FullMatch("0abcde"));
|
| 988 |
CHECK(!RE("[0-9a-f+.-]{5,}").FullMatch("0abc"));
|
| 989 |
|
| 990 |
// Complicated RE
|
| 991 |
CHECK(RE("foo|bar|[A-Z]").FullMatch("foo"));
|
| 992 |
CHECK(RE("foo|bar|[A-Z]").FullMatch("bar"));
|
| 993 |
CHECK(RE("foo|bar|[A-Z]").FullMatch("X"));
|
| 994 |
CHECK(!RE("foo|bar|[A-Z]").FullMatch("XY"));
|
| 995 |
|
| 996 |
// Check full-match handling (needs '$' tacked on internally)
|
| 997 |
CHECK(RE("fo|foo").FullMatch("fo"));
|
| 998 |
CHECK(RE("fo|foo").FullMatch("foo"));
|
| 999 |
CHECK(RE("fo|foo$").FullMatch("fo"));
|
| 1000 |
CHECK(RE("fo|foo$").FullMatch("foo"));
|
| 1001 |
CHECK(RE("foo$").FullMatch("foo"));
|
| 1002 |
CHECK(!RE("foo\\$").FullMatch("foo$bar"));
|
| 1003 |
CHECK(!RE("fo|bar").FullMatch("fox"));
|
| 1004 |
|
| 1005 |
// Uncomment the following if we change the handling of '$' to
|
| 1006 |
// prevent it from matching a trailing newline
|
| 1007 |
if (false) {
|
| 1008 |
// Check that we don't get bitten by pcre's special handling of a
|
| 1009 |
// '\n' at the end of the string matching '$'
|
| 1010 |
CHECK(!RE("foo$").PartialMatch("foo\n"));
|
| 1011 |
}
|
| 1012 |
|
| 1013 |
// Number of args
|
| 1014 |
int a[16];
|
| 1015 |
CHECK(RE("").FullMatch(""));
|
| 1016 |
|
| 1017 |
memset(a, 0, sizeof(0));
|
| 1018 |
CHECK(RE("(\\d){1}").FullMatch("1",
|
| 1019 |
&a[0]));
|
| 1020 |
CHECK_EQ(a[0], 1);
|
| 1021 |
|
| 1022 |
memset(a, 0, sizeof(0));
|
| 1023 |
CHECK(RE("(\\d)(\\d)").FullMatch("12",
|
| 1024 |
&a[0], &a[1]));
|
| 1025 |
CHECK_EQ(a[0], 1);
|
| 1026 |
CHECK_EQ(a[1], 2);
|
| 1027 |
|
| 1028 |
memset(a, 0, sizeof(0));
|
| 1029 |
CHECK(RE("(\\d)(\\d)(\\d)").FullMatch("123",
|
| 1030 |
&a[0], &a[1], &a[2]));
|
| 1031 |
CHECK_EQ(a[0], 1);
|
| 1032 |
CHECK_EQ(a[1], 2);
|
| 1033 |
CHECK_EQ(a[2], 3);
|
| 1034 |
|
| 1035 |
memset(a, 0, sizeof(0));
|
| 1036 |
CHECK(RE("(\\d)(\\d)(\\d)(\\d)").FullMatch("1234",
|
| 1037 |
&a[0], &a[1], &a[2], &a[3]));
|
| 1038 |
CHECK_EQ(a[0], 1);
|
| 1039 |
CHECK_EQ(a[1], 2);
|
| 1040 |
CHECK_EQ(a[2], 3);
|
| 1041 |
CHECK_EQ(a[3], 4);
|
| 1042 |
|
| 1043 |
memset(a, 0, sizeof(0));
|
| 1044 |
CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch("12345",
|
| 1045 |
&a[0], &a[1], &a[2],
|
| 1046 |
&a[3], &a[4]));
|
| 1047 |
CHECK_EQ(a[0], 1);
|
| 1048 |
CHECK_EQ(a[1], 2);
|
| 1049 |
CHECK_EQ(a[2], 3);
|
| 1050 |
CHECK_EQ(a[3], 4);
|
| 1051 |
CHECK_EQ(a[4], 5);
|
| 1052 |
|
| 1053 |
memset(a, 0, sizeof(0));
|
| 1054 |
CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch("123456",
|
| 1055 |
&a[0], &a[1], &a[2],
|
| 1056 |
&a[3], &a[4], &a[5]));
|
| 1057 |
CHECK_EQ(a[0], 1);
|
| 1058 |
CHECK_EQ(a[1], 2);
|
| 1059 |
CHECK_EQ(a[2], 3);
|
| 1060 |
CHECK_EQ(a[3], 4);
|
| 1061 |
CHECK_EQ(a[4], 5);
|
| 1062 |
CHECK_EQ(a[5], 6);
|
| 1063 |
|
| 1064 |
memset(a, 0, sizeof(0));
|
| 1065 |
CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch("1234567",
|
| 1066 |
&a[0], &a[1], &a[2], &a[3],
|
| 1067 |
&a[4], &a[5], &a[6]));
|
| 1068 |
CHECK_EQ(a[0], 1);
|
| 1069 |
CHECK_EQ(a[1], 2);
|
| 1070 |
CHECK_EQ(a[2], 3);
|
| 1071 |
CHECK_EQ(a[3], 4);
|
| 1072 |
CHECK_EQ(a[4], 5);
|
| 1073 |
CHECK_EQ(a[5], 6);
|
| 1074 |
CHECK_EQ(a[6], 7);
|
| 1075 |
|
| 1076 |
memset(a, 0, sizeof(0));
|
| 1077 |
CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)"
|
| 1078 |
"(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch(
|
| 1079 |
"1234567890123456",
|
| 1080 |
&a[0], &a[1], &a[2], &a[3],
|
| 1081 |
&a[4], &a[5], &a[6], &a[7],
|
| 1082 |
&a[8], &a[9], &a[10], &a[11],
|
| 1083 |
&a[12], &a[13], &a[14], &a[15]));
|
| 1084 |
CHECK_EQ(a[0], 1);
|
| 1085 |
CHECK_EQ(a[1], 2);
|
| 1086 |
CHECK_EQ(a[2], 3);
|
| 1087 |
CHECK_EQ(a[3], 4);
|
| 1088 |
CHECK_EQ(a[4], 5);
|
| 1089 |
CHECK_EQ(a[5], 6);
|
| 1090 |
CHECK_EQ(a[6], 7);
|
| 1091 |
CHECK_EQ(a[7], 8);
|
| 1092 |
CHECK_EQ(a[8], 9);
|
| 1093 |
CHECK_EQ(a[9], 0);
|
| 1094 |
CHECK_EQ(a[10], 1);
|
| 1095 |
CHECK_EQ(a[11], 2);
|
| 1096 |
CHECK_EQ(a[12], 3);
|
| 1097 |
CHECK_EQ(a[13], 4);
|
| 1098 |
CHECK_EQ(a[14], 5);
|
| 1099 |
CHECK_EQ(a[15], 6);
|
| 1100 |
|
| 1101 |
/***** PartialMatch *****/
|
| 1102 |
|
| 1103 |
printf("Testing PartialMatch\n");
|
| 1104 |
|
| 1105 |
CHECK(RE("h.*o").PartialMatch("hello"));
|
| 1106 |
CHECK(RE("h.*o").PartialMatch("othello"));
|
| 1107 |
CHECK(RE("h.*o").PartialMatch("hello!"));
|
| 1108 |
CHECK(RE("((((((((((((((((((((x))))))))))))))))))))").PartialMatch("x"));
|
| 1109 |
|
| 1110 |
/***** other tests *****/
|
| 1111 |
|
| 1112 |
RadixTests();
|
| 1113 |
TestReplace();
|
| 1114 |
TestExtract();
|
| 1115 |
TestConsume();
|
| 1116 |
TestFindAndConsume();
|
| 1117 |
TestQuoteMetaAll();
|
| 1118 |
TestMatchNumberPeculiarity();
|
| 1119 |
|
| 1120 |
// Check the pattern() accessor
|
| 1121 |
{
|
| 1122 |
const string kPattern = "http://([^/]+)/.*";
|
| 1123 |
const RE re(kPattern);
|
| 1124 |
CHECK_EQ(kPattern, re.pattern());
|
| 1125 |
}
|
| 1126 |
|
| 1127 |
// Check RE error field.
|
| 1128 |
{
|
| 1129 |
RE re("foo");
|
| 1130 |
CHECK(re.error().empty()); // Must have no error
|
| 1131 |
}
|
| 1132 |
|
| 1133 |
#ifdef SUPPORT_UTF8
|
| 1134 |
// Check UTF-8 handling
|
| 1135 |
{
|
| 1136 |
printf("Testing UTF-8 handling\n");
|
| 1137 |
|
| 1138 |
// Three Japanese characters (nihongo)
|
| 1139 |
const char utf8_string[] = {
|
| 1140 |
0xe6, 0x97, 0xa5, // 65e5
|
| 1141 |
0xe6, 0x9c, 0xac, // 627c
|
| 1142 |
0xe8, 0xaa, 0x9e, // 8a9e
|
| 1143 |
0
|
| 1144 |
};
|
| 1145 |
const char utf8_pattern[] = {
|
| 1146 |
'.',
|
| 1147 |
0xe6, 0x9c, 0xac, // 627c
|
| 1148 |
'.',
|
| 1149 |
0
|
| 1150 |
};
|
| 1151 |
|
| 1152 |
// Both should match in either mode, bytes or UTF-8
|
| 1153 |
RE re_test1(".........");
|
| 1154 |
CHECK(re_test1.FullMatch(utf8_string));
|
| 1155 |
RE re_test2("...", pcrecpp::UTF8());
|
| 1156 |
CHECK(re_test2.FullMatch(utf8_string));
|
| 1157 |
|
| 1158 |
// Check that '.' matches one byte or UTF-8 character
|
| 1159 |
// according to the mode.
|
| 1160 |
string ss;
|
| 1161 |
RE re_test3("(.)");
|
| 1162 |
CHECK(re_test3.PartialMatch(utf8_string, &ss));
|
| 1163 |
CHECK_EQ(ss, string("\xe6"));
|
| 1164 |
RE re_test4("(.)", pcrecpp::UTF8());
|
| 1165 |
CHECK(re_test4.PartialMatch(utf8_string, &ss));
|
| 1166 |
CHECK_EQ(ss, string("\xe6\x97\xa5"));
|
| 1167 |
|
| 1168 |
// Check that string matches itself in either mode
|
| 1169 |
RE re_test5(utf8_string);
|
| 1170 |
CHECK(re_test5.FullMatch(utf8_string));
|
| 1171 |
RE re_test6(utf8_string, pcrecpp::UTF8());
|
| 1172 |
CHECK(re_test6.FullMatch(utf8_string));
|
| 1173 |
|
| 1174 |
// Check that pattern matches string only in UTF8 mode
|
| 1175 |
RE re_test7(utf8_pattern);
|
| 1176 |
CHECK(!re_test7.FullMatch(utf8_string));
|
| 1177 |
RE re_test8(utf8_pattern, pcrecpp::UTF8());
|
| 1178 |
CHECK(re_test8.FullMatch(utf8_string));
|
| 1179 |
}
|
| 1180 |
|
| 1181 |
// Check that ungreedy, UTF8 regular expressions don't match when they
|
| 1182 |
// oughtn't -- see bug 82246.
|
| 1183 |
{
|
| 1184 |
// This code always worked.
|
| 1185 |
const char* pattern = "\\w+X";
|
| 1186 |
const string target = "a aX";
|
| 1187 |
RE match_sentence(pattern);
|
| 1188 |
RE match_sentence_re(pattern, pcrecpp::UTF8());
|
| 1189 |
|
| 1190 |
CHECK(!match_sentence.FullMatch(target));
|
| 1191 |
CHECK(!match_sentence_re.FullMatch(target));
|
| 1192 |
}
|
| 1193 |
|
| 1194 |
{
|
| 1195 |
const char* pattern = "(?U)\\w+X";
|
| 1196 |
const string target = "a aX";
|
| 1197 |
RE match_sentence(pattern);
|
| 1198 |
RE match_sentence_re(pattern, pcrecpp::UTF8());
|
| 1199 |
|
| 1200 |
CHECK(!match_sentence.FullMatch(target));
|
| 1201 |
CHECK(!match_sentence_re.FullMatch(target));
|
| 1202 |
}
|
| 1203 |
#endif /* def SUPPORT_UTF8 */
|
| 1204 |
|
| 1205 |
printf("Testing error reporting\n");
|
| 1206 |
|
| 1207 |
{ RE re("a\\1"); CHECK(!re.error().empty()); }
|
| 1208 |
{
|
| 1209 |
RE re("a[x");
|
| 1210 |
CHECK(!re.error().empty());
|
| 1211 |
}
|
| 1212 |
{
|
| 1213 |
RE re("a[z-a]");
|
| 1214 |
CHECK(!re.error().empty());
|
| 1215 |
}
|
| 1216 |
{
|
| 1217 |
RE re("a[[:foobar:]]");
|
| 1218 |
CHECK(!re.error().empty());
|
| 1219 |
}
|
| 1220 |
{
|
| 1221 |
RE re("a(b");
|
| 1222 |
CHECK(!re.error().empty());
|
| 1223 |
}
|
| 1224 |
{
|
| 1225 |
RE re("a\\");
|
| 1226 |
CHECK(!re.error().empty());
|
| 1227 |
}
|
| 1228 |
|
| 1229 |
// Test that recursion is stopped
|
| 1230 |
TestRecursion();
|
| 1231 |
|
| 1232 |
// Test Options
|
| 1233 |
if (getenv("VERBOSE_TEST") != NULL)
|
| 1234 |
VERBOSE_TEST = true;
|
| 1235 |
TestOptions();
|
| 1236 |
|
| 1237 |
// Test the constructors
|
| 1238 |
TestConstructors();
|
| 1239 |
|
| 1240 |
// Done
|
| 1241 |
printf("OK\n");
|
| 1242 |
|
| 1243 |
return 0;
|
| 1244 |
}
|
Parent Directory
| 
Revision Log