scuffed-code/icu4c/source/test/intltest/regextst.cpp

577 lines
18 KiB
C++
Raw Normal View History

/********************************************************************
* COPYRIGHT:
* Copyright (c) 2002, International Business Machines Corporation and
* others. All Rights Reserved.
********************************************************************/
//
// regex.cpp
//
// ICU Regular Expressions test, part of intltest.
//
#include "unicode/utypes.h"
#include "intltest.h"
#include "regextst.h"
RegexTest::RegexTest()
{
};
RegexTest::~RegexTest()
{
};
void RegexTest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
{
if (exec) logln("TestSuite RegexTest: ");
switch (index) {
case 0: name = "API_Match";
if (exec) API_Match();
break;
case 1: name = "Basic";
if (exec) Basic();
break;
case 2: name = "API_Replace";
if (exec) API_Replace();
break;
case 3: name = "API_Pattern";
if (exec) API_Pattern();
break;
default: name = "";
break; //needed to end loop
}
}
//---------------------------------------------------------------------------
//
// REGEX_TESTLM Macro + invocation function to simplify writing quick tests
// for the LookingAt() and Match() functions.
//
// usage:
// REGEX_TESTLM("pattern", "input text", lookingAt expected, matches expected);
//
// The expected results are UBool - TRUE or FALSE.
// The input text is unescaped. The pattern is not.
//
//
//---------------------------------------------------------------------------
#define REGEX_CHECK_STATUS {if (U_FAILURE(status)) {errln("RegexTest failure at line %d. status=%d\n", \
__LINE__, status); return;}}
#define REGEX_ASSERT(expr) {if ((expr)==FALSE) {errln("RegexTest failure at line %d.\n", __LINE__);};}
#define REGEX_ASSERT_FAIL(expr, errcode) {UErrorCode status=U_ZERO_ERROR; (expr);\
if (status!=errcode) {errln("RegexTest failure at line %d.\n", __LINE__);};}
#define REGEX_TESTLM(pat, text, looking, match) doRegexLMTest(pat, text, looking, match, __LINE__);
UBool RegexTest::doRegexLMTest(char *pat, char *text, UBool looking, UBool match, int line) {
const UnicodeString pattern(pat);
const UnicodeString inputText(text);
UErrorCode status = U_ZERO_ERROR;
UParseError pe;
RegexPattern *REPattern = NULL;
RegexMatcher *REMatcher = NULL;
UBool retVal = TRUE;
UnicodeString patString(pat);
REPattern = RegexPattern::compile(patString, 0, pe, status);
if (U_FAILURE(status)) {
errln("RegexTest failure in RegexPattern::compile() at line %d. Status = %d\n", line, status);
return FALSE;
}
UnicodeString inputString(inputText);
UnicodeString unEscapedInput = inputString.unescape();
REMatcher = REPattern->matcher(unEscapedInput, status);
if (U_FAILURE(status)) {
errln("RegexTest failure in REPattern::matcher() at line %d. Status = %d\n", line, status);
return FALSE;
}
UBool actualmatch;
actualmatch = REMatcher->lookingAt(status);
if (U_FAILURE(status)) {
errln("RegexTest failure in lookingAt() at line %d. Status = %d\n", line, status);
retVal = FALSE;
}
if (actualmatch != looking) {
errln("RegexTest: wrong return from lookingAt() at line %d.\n", line);
retVal = FALSE;
}
status = U_ZERO_ERROR;
actualmatch = REMatcher->matches(status);
if (U_FAILURE(status)) {
errln("RegexTest failure in matches() at line %d. Status = %d\n", line, status);
retVal = FALSE;
}
if (actualmatch != match) {
errln("RegexTest: wrong return from matches() at line %d.\n", line);
retVal = FALSE;
}
if (retVal == FALSE) {
REPattern->dump();
}
delete REPattern;
delete REMatcher;
return retVal;
}
//---------------------------------------------------------------------------
//
// API_Match
//
//---------------------------------------------------------------------------
void RegexTest::API_Match() {
UParseError pe;
UErrorCode status=U_ZERO_ERROR;
int32_t flags = 0;
//
// Debug - slide failing test cases early
//
#if 0
{
}
return;
#endif
//
// Simple pattern compilation
//
{
UnicodeString re("abc");
RegexPattern *pat2;
pat2 = RegexPattern::compile(re, flags, pe, status);
REGEX_CHECK_STATUS;
UnicodeString inStr1 = "abcdef this is a test";
UnicodeString instr2 = "not abc";
UnicodeString empty = "";
//
// Matcher creation and reset.
//
RegexMatcher *m1 = pat2->matcher(inStr1, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(m1->lookingAt(status) == TRUE);
REGEX_ASSERT(m1->input() == inStr1);
m1->reset(instr2);
REGEX_ASSERT(m1->lookingAt(status) == FALSE);
REGEX_ASSERT(m1->input() == instr2);
m1->reset(inStr1);
REGEX_ASSERT(m1->input() == inStr1);
REGEX_ASSERT(m1->lookingAt(status) == TRUE);
m1->reset(empty);
REGEX_ASSERT(m1->lookingAt(status) == FALSE);
REGEX_ASSERT(m1->input() == empty);
REGEX_ASSERT(&m1->pattern() == pat2);
delete m1;
delete pat2;
}
//
// Capture Group.
// RegexMatcher::start();
// RegexMatcher::end();
// RegexMatcher::groupCount();
//
{
int32_t flags=0;
UParseError pe;
UErrorCode status=U_ZERO_ERROR;
UnicodeString re("01(23(45)67)(.*)");
RegexPattern *pat = RegexPattern::compile(re, flags, pe, status);
REGEX_CHECK_STATUS;
UnicodeString data = "0123456789";
RegexMatcher *matcher = pat->matcher(data, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(matcher->lookingAt(status) == TRUE);
int matchStarts[] = {0, 2, 4, 8};
int matchEnds[] = {10, 8, 6, 10};
int i;
for (i=0; i<4; i++) {
int32_t actualStart = matcher->start(i, status);
REGEX_CHECK_STATUS;
if (actualStart != matchStarts[i]) {
errln("RegexTest failure at line %d, index %d. Expected %d, got %d\n",
__LINE__, i, matchStarts[i], actualStart);
}
int32_t actualEnd = matcher->end(i, status);
REGEX_CHECK_STATUS;
if (actualEnd != matchEnds[i]) {
errln("RegexTest failure at line %d index %d. Expected %d, got %d\n",
__LINE__, i, matchEnds[i], actualEnd);
}
}
REGEX_ASSERT(matcher->start(0, status) == matcher->start(status));
REGEX_ASSERT(matcher->end(0, status) == matcher->end(status));
REGEX_ASSERT_FAIL(matcher->start(-1, status), U_INDEX_OUTOFBOUNDS_ERROR);
REGEX_ASSERT_FAIL(matcher->start( 4, status), U_INDEX_OUTOFBOUNDS_ERROR);
matcher->reset();
REGEX_ASSERT_FAIL(matcher->start( 0, status), U_REGEX_INVALID_STATE);
matcher->lookingAt(status);
REGEX_ASSERT(matcher->group(status) == "0123456789");
REGEX_ASSERT(matcher->group(0, status) == "0123456789");
REGEX_ASSERT(matcher->group(1, status) == "234567" );
REGEX_ASSERT(matcher->group(2, status) == "45" );
REGEX_ASSERT(matcher->group(3, status) == "89" );
REGEX_CHECK_STATUS;
REGEX_ASSERT_FAIL(matcher->group(-1, status), U_INDEX_OUTOFBOUNDS_ERROR);
REGEX_ASSERT_FAIL(matcher->group( 4, status), U_INDEX_OUTOFBOUNDS_ERROR);
matcher->reset();
REGEX_ASSERT_FAIL(matcher->group( 0, status), U_REGEX_INVALID_STATE);
delete matcher;
delete pat;
}
//
// find
//
{
int32_t flags=0;
UParseError pe;
UErrorCode status=U_ZERO_ERROR;
UnicodeString re("abc");
RegexPattern *pat = RegexPattern::compile(re, flags, pe, status);
REGEX_CHECK_STATUS;
UnicodeString data = ".abc..abc...abc..";
// 012345678901234567
RegexMatcher *matcher = pat->matcher(data, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(matcher->find());
REGEX_ASSERT(matcher->start(status) == 1);
REGEX_ASSERT(matcher->find());
REGEX_ASSERT(matcher->start(status) == 6);
REGEX_ASSERT(matcher->find());
REGEX_ASSERT(matcher->start(status) == 12);
REGEX_ASSERT(matcher->find() == FALSE);
REGEX_ASSERT(matcher->find() == FALSE);
matcher->reset();
REGEX_ASSERT(matcher->find());
REGEX_ASSERT(matcher->start(status) == 1);
REGEX_ASSERT(matcher->find(0, status));
REGEX_ASSERT(matcher->start(status) == 1);
REGEX_ASSERT(matcher->find(1, status));
REGEX_ASSERT(matcher->start(status) == 1);
REGEX_ASSERT(matcher->find(2, status));
REGEX_ASSERT(matcher->start(status) == 6);
REGEX_ASSERT(matcher->find(12, status));
REGEX_ASSERT(matcher->start(status) == 12);
REGEX_ASSERT(matcher->find(13, status) == FALSE);
REGEX_ASSERT(matcher->find(16, status) == FALSE);
REGEX_ASSERT_FAIL(matcher->start(status), U_REGEX_INVALID_STATE);
REGEX_CHECK_STATUS;
REGEX_ASSERT_FAIL(matcher->find(-1, status), U_INDEX_OUTOFBOUNDS_ERROR);
REGEX_ASSERT_FAIL(matcher->find(17, status), U_INDEX_OUTOFBOUNDS_ERROR);
REGEX_ASSERT(matcher->groupCount() == 0);
delete matcher;
delete pat;
}
}
//---------------------------------------------------------------------------
//
// Basic Check for basic functionality of
// regex pattern matching.
//
//---------------------------------------------------------------------------
void RegexTest::Basic() {
//
// Debug - slide failing test cases early
//
#if 0
{
}
return;
#endif
//
// Pattern with parentheses
//
REGEX_TESTLM("st(abc)ring", "stabcring thing", TRUE, FALSE);
REGEX_TESTLM("st(abc)ring", "stabcring", TRUE, TRUE);
REGEX_TESTLM("st(abc)ring", "stabcrung", FALSE, FALSE);
//
// Patterns with *
//
REGEX_TESTLM("st(abc)*ring", "string", TRUE, TRUE);
REGEX_TESTLM("st(abc)*ring", "stabcring", TRUE, TRUE);
REGEX_TESTLM("st(abc)*ring", "stabcabcring", TRUE, TRUE);
REGEX_TESTLM("st(abc)*ring", "stabcabcdring", FALSE, FALSE);
REGEX_TESTLM("st(abc)*ring", "stabcabcabcring etc.", TRUE, FALSE);
REGEX_TESTLM("a*", "", TRUE, TRUE);
REGEX_TESTLM("a*", "b", TRUE, FALSE);
//
// Patterns with "."
//
REGEX_TESTLM(".", "abc", TRUE, FALSE);
REGEX_TESTLM("...", "abc", TRUE, TRUE);
REGEX_TESTLM("....", "abc", FALSE, FALSE);
REGEX_TESTLM(".*", "abcxyz123", TRUE, TRUE);
REGEX_TESTLM("ab.*xyz", "abcdefghij", FALSE, FALSE);
REGEX_TESTLM("ab.*xyz", "abcdefg...wxyz", TRUE, TRUE);
REGEX_TESTLM("ab.*xyz", "abcde...wxyz...abc..xyz", TRUE, TRUE);
REGEX_TESTLM("ab.*xyz", "abcde...wxyz...abc..xyz...", TRUE, FALSE);
//
// Patterns with * applied to chars at end of literal string
//
REGEX_TESTLM("abc*", "ab", TRUE, TRUE);
REGEX_TESTLM("abc*", "abccccc", TRUE, TRUE);
//
// Supplemental chars match as single chars, not a pair of surrogates.
//
REGEX_TESTLM(".", "\\U00011000", TRUE, TRUE);
REGEX_TESTLM("...", "\\U00011000x\\U00012002", TRUE, TRUE);
REGEX_TESTLM("...", "\\U00011000x\\U00012002y", TRUE, FALSE);
//
// UnicodeSets in the pattern
//
REGEX_TESTLM("[1-6]", "1", TRUE, TRUE);
REGEX_TESTLM("[1-6]", "3", TRUE, TRUE);
REGEX_TESTLM("[1-6]", "7", FALSE, FALSE);
REGEX_TESTLM("a[1-6]", "a3", TRUE, TRUE);
REGEX_TESTLM("a[1-6]", "a3", TRUE, TRUE);
REGEX_TESTLM("a[1-6]b", "a3b", TRUE, TRUE);
REGEX_TESTLM("a[0-9]*b", "a123b", TRUE, TRUE);
REGEX_TESTLM("a[0-9]*b", "abc", TRUE, FALSE);
REGEX_TESTLM("[\\p{Nd}]*", "123456", TRUE, TRUE);
REGEX_TESTLM("[\\p{Nd}]*", "a123456", TRUE, FALSE); // note that * matches 0 occurences.
REGEX_TESTLM("[a][b][[:Zs:]]*", "ab ", TRUE, TRUE);
//
// OR operator in patterns
//
REGEX_TESTLM("(a|b)", "a", TRUE, TRUE);
REGEX_TESTLM("(a|b)", "b", TRUE, TRUE);
REGEX_TESTLM("(a|b)", "c", FALSE, FALSE);
REGEX_TESTLM("a|b", "b", TRUE, TRUE);
REGEX_TESTLM("(a|b|c)*", "aabcaaccbcabc", TRUE, TRUE);
REGEX_TESTLM("(a|b|c)*", "aabcaaccbcabdc", TRUE, FALSE);
REGEX_TESTLM("(a(b|c|d)(x|y|z)*|123)", "ac", TRUE, TRUE);
REGEX_TESTLM("(a(b|c|d)(x|y|z)*|123)", "123", TRUE, TRUE);
REGEX_TESTLM("(a|(1|2)*)(b|c|d)(x|y|z)*|123", "123", TRUE, TRUE);
REGEX_TESTLM("(a|(1|2)*)(b|c|d)(x|y|z)*|123", "222211111czzzzw", TRUE, FALSE);
//
// +
//
REGEX_TESTLM("ab+", "abbc", TRUE, FALSE);
REGEX_TESTLM("ab+c", "ac", FALSE, FALSE);
REGEX_TESTLM("b+", "", FALSE, FALSE);
REGEX_TESTLM("(abc|def)+", "defabc", TRUE, TRUE);
REGEX_TESTLM(".+y", "zippity dooy dah ", TRUE, FALSE);
REGEX_TESTLM(".+y", "zippity dooy", TRUE, TRUE);
//
// ?
//
REGEX_TESTLM("ab?", "ab", TRUE, TRUE);
REGEX_TESTLM("ab?", "a", TRUE, TRUE);
REGEX_TESTLM("ab?", "ac", TRUE, FALSE);
REGEX_TESTLM("ab?", "abb", TRUE, FALSE);
REGEX_TESTLM("a(b|c)?d", "abd", TRUE, TRUE);
REGEX_TESTLM("a(b|c)?d", "acd", TRUE, TRUE);
REGEX_TESTLM("a(b|c)?d", "ad", TRUE, TRUE);
REGEX_TESTLM("a(b|c)?d", "abcd", FALSE, FALSE);
REGEX_TESTLM("a(b|c)?d", "ab", FALSE, FALSE);
};
//---------------------------------------------------------------------------
//
// API_Replace
//
//---------------------------------------------------------------------------
void RegexTest::API_Replace() {
}
//---------------------------------------------------------------------------
//
// API_Pattern
//
//---------------------------------------------------------------------------
void RegexTest::API_Pattern() {
RegexPattern pata; // Test default constructor to not crash.
RegexPattern patb;
REGEX_ASSERT(pata == patb);
REGEX_ASSERT(pata == pata);
UnicodeString re1("abc[a-l][m-z]");
UnicodeString re2("def");
UErrorCode status = U_ZERO_ERROR;
UParseError pe;
RegexPattern *pat1 = RegexPattern::compile(re1, 0, pe, status);
RegexPattern *pat2 = RegexPattern::compile(re2, 0, pe, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(*pat1 == *pat1);
REGEX_ASSERT(*pat1 != pata);
// Assign
patb = *pat1;
REGEX_ASSERT(patb == *pat1);
// Copy Construct
RegexPattern patc(*pat1);
REGEX_ASSERT(patc == *pat1);
REGEX_ASSERT(patb == patc);
REGEX_ASSERT(pat1 != pat2);
patb = *pat2;
REGEX_ASSERT(patb != patc);
REGEX_ASSERT(patb == *pat2);
// Compile with no flags.
RegexPattern *pat1a = RegexPattern::compile(re1, pe, status);
REGEX_ASSERT(*pat1a == *pat1);
// Compile with different flags should be not equal
RegexPattern *pat1b = RegexPattern::compile(re1, UREGEX_CASE_INSENSITIVE, pe, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(*pat1b != *pat1a);
REGEX_ASSERT(pat1b->flags() == UREGEX_CASE_INSENSITIVE);
REGEX_ASSERT(pat1a->flags() == 0);
// clone
RegexPattern *pat1c = pat1b->clone();
REGEX_ASSERT(*pat1b == *pat1c);
REGEX_ASSERT(*pat1a != *pat1c);
// TODO: Actually do some matches with the cloned/copied/assigned patterns.
delete pat1c;
delete pat1b;
delete pat1a;
delete pat1;
delete pat2;
//
// matches convenience API
//
REGEX_ASSERT(RegexPattern::matches(".*", "random input", pe, status) == TRUE);
REGEX_CHECK_STATUS;
REGEX_ASSERT(RegexPattern::matches("abc", "random input", pe, status) == FALSE);
REGEX_CHECK_STATUS;
REGEX_ASSERT(RegexPattern::matches(".*nput", "random input", pe, status) == TRUE);
REGEX_CHECK_STATUS;
REGEX_ASSERT(RegexPattern::matches("random input", "random input", pe, status) == TRUE);
REGEX_CHECK_STATUS;
REGEX_ASSERT(RegexPattern::matches(".*u", "random input", pe, status) == FALSE);
REGEX_CHECK_STATUS;
status = U_INDEX_OUTOFBOUNDS_ERROR;
REGEX_ASSERT(RegexPattern::matches("abc", "abc", pe, status) == FALSE);
REGEX_ASSERT(status == U_INDEX_OUTOFBOUNDS_ERROR);
//
// Split()
//
status = U_ZERO_ERROR;
pat1 = RegexPattern::compile(" +", pe, status);
REGEX_CHECK_STATUS;
UnicodeString fields[10];
int32_t n;
n = pat1->split("Now is the time", fields, 10, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(n==4);
REGEX_ASSERT(fields[0]=="Now");
REGEX_ASSERT(fields[1]=="is");
REGEX_ASSERT(fields[2]=="the");
REGEX_ASSERT(fields[3]=="time");
REGEX_ASSERT(fields[4]=="");
n = pat1->split("Now is the time", fields, 2, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(n==2);
REGEX_ASSERT(fields[0]=="Now");
REGEX_ASSERT(fields[1]=="is the time");
REGEX_ASSERT(fields[2]=="the"); // left over from previous test
fields[1] = "*";
n = pat1->split("Now is the time", fields, 1, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(n==1);
REGEX_ASSERT(fields[0]=="Now is the time");
REGEX_ASSERT(fields[1]=="*");
n = pat1->split(" Now is the time ", fields, 10, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(n==5);
REGEX_ASSERT(fields[0]=="");
REGEX_ASSERT(fields[1]=="Now");
REGEX_ASSERT(fields[2]=="is");
REGEX_ASSERT(fields[3]=="the");
REGEX_ASSERT(fields[4]=="time");
REGEX_ASSERT(fields[5]=="");
n = pat1->split(" ", fields, 10, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(n==1);
REGEX_ASSERT(fields[0]=="");
fields[0] = "foo";
n = pat1->split("", fields, 10, status);
REGEX_CHECK_STATUS;
REGEX_ASSERT(n==0);
REGEX_ASSERT(fields[0]=="foo");
}