452 lines
14 KiB
C++
452 lines
14 KiB
C++
//////////////////////////////////////////////////////////////////////////////
|
|
// regress.ipp
|
|
//
|
|
// (C) Copyright Eric Niebler 2004.
|
|
// Use, modification and distribution are subject to the
|
|
// Boost Software License, Version 1.0. (See accompanying file
|
|
// LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
|
|
|
|
/*
|
|
Revision history:
|
|
7 March 2004 : Initial version.
|
|
*/
|
|
|
|
#include <locale>
|
|
#include <vector>
|
|
#include <string>
|
|
#include <fstream>
|
|
#include <iostream>
|
|
#include <boost/lexical_cast.hpp>
|
|
#include <boost/xpressive/xpressive.hpp>
|
|
#include <boost/test/unit_test.hpp>
|
|
|
|
#if defined(_MSC_VER) && defined(_DEBUG)
|
|
# define _CRTDBG_MAP_ALLOC
|
|
# include <crtdbg.h>
|
|
#endif
|
|
|
|
#if defined(BOOST_XPRESSIVE_TEST_WREGEX) && !defined(BOOST_XPRESSIVE_NO_WREGEX)
|
|
namespace std
|
|
{
|
|
inline std::ostream &operator <<(std::ostream &sout, std::wstring const &wstr)
|
|
{
|
|
for(std::size_t n = 0; n < wstr.size(); ++n)
|
|
sout.put(BOOST_USE_FACET(std::ctype<wchar_t>, std::locale()).narrow(wstr[n], '?'));
|
|
return sout;
|
|
}
|
|
}
|
|
#endif
|
|
|
|
#define BOOST_XPR_CHECK(pred) \
|
|
if(pred) {} else { BOOST_ERROR(case_ << #pred); }
|
|
|
|
using namespace boost::unit_test;
|
|
using namespace boost::xpressive;
|
|
|
|
//////////////////////////////////////////////////////////////////////////////
|
|
// xpr_test_case
|
|
template<typename Char>
|
|
struct xpr_test_case
|
|
{
|
|
typedef std::basic_string<Char> string_type;
|
|
std::string section;
|
|
string_type str;
|
|
string_type pat;
|
|
string_type sub;
|
|
string_type res;
|
|
regex_constants::syntax_option_type syntax_flags;
|
|
regex_constants::match_flag_type match_flags;
|
|
std::vector<string_type> br;
|
|
|
|
xpr_test_case()
|
|
{
|
|
this->reset();
|
|
}
|
|
|
|
void reset()
|
|
{
|
|
this->section.clear();
|
|
this->str.clear();
|
|
this->pat.clear();
|
|
this->sub.clear();
|
|
this->res.clear();
|
|
this->br.clear();
|
|
this->syntax_flags = regex_constants::ECMAScript;
|
|
this->match_flags = regex_constants::match_default | regex_constants::format_first_only;
|
|
}
|
|
};
|
|
|
|
//////////////////////////////////////////////////////////////////////////////
|
|
// globals
|
|
std::ifstream in;
|
|
unsigned int test_count = 0;
|
|
|
|
// The global object that contains the current test case
|
|
xpr_test_case<char> test;
|
|
|
|
struct test_case_formatter
|
|
{
|
|
friend std::ostream &operator <<(std::ostream &sout, test_case_formatter)
|
|
{
|
|
sout << test.section << " /" << test.pat << "/ : ";
|
|
return sout;
|
|
}
|
|
};
|
|
|
|
test_case_formatter const case_ = {};
|
|
|
|
#if defined(BOOST_XPRESSIVE_TEST_WREGEX) && !defined(BOOST_XPRESSIVE_NO_WREGEX)
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// widen
|
|
// make a std::wstring from a std::string by widening according to the
|
|
// current ctype<char> facet
|
|
inline std::wstring widen(std::string const &str)
|
|
{
|
|
std::ctype<char> const &ct = BOOST_USE_FACET(std::ctype<char>, std::locale());
|
|
std::wstring res;
|
|
for(size_t i=0; i<str.size(); ++i)
|
|
{
|
|
res += ct.widen(str[i]);
|
|
}
|
|
return res;
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// widen
|
|
// widens an entire test case
|
|
xpr_test_case<wchar_t> widen(xpr_test_case<char> const &test)
|
|
{
|
|
xpr_test_case<wchar_t> wtest;
|
|
wtest.section = test.section;
|
|
wtest.str = ::widen(test.str);
|
|
wtest.pat = ::widen(test.pat);
|
|
wtest.sub = ::widen(test.sub);
|
|
wtest.res = ::widen(test.res);
|
|
wtest.syntax_flags = test.syntax_flags;
|
|
wtest.match_flags = test.match_flags;
|
|
wtest.br.reserve(test.br.size());
|
|
for(std::size_t i = 0; i < test.br.size(); ++i)
|
|
{
|
|
wtest.br.push_back(::widen(test.br[i]));
|
|
}
|
|
return wtest;
|
|
}
|
|
#endif // BOOST_XPRESSIVE_NO_WREGEX
|
|
|
|
std::string escape(std::string str)
|
|
{
|
|
for(std::string::size_type pos = 0; std::string::npos != (pos = str.find('\\', pos)); ++pos)
|
|
{
|
|
if(pos + 1 == str.size())
|
|
break;
|
|
|
|
switch(str[pos + 1])
|
|
{
|
|
case '\\': str.replace(pos, 2, "\\"); break;
|
|
case 'n': str.replace(pos, 2, "\n"); break;
|
|
case 'r': str.replace(pos, 2, "\r"); break;
|
|
}
|
|
}
|
|
return str;
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// get_test
|
|
// read the next section out of the input file, and fill out
|
|
// the global variables
|
|
bool get_test()
|
|
{
|
|
test.reset();
|
|
bool first = true;
|
|
std::string line;
|
|
smatch what;
|
|
|
|
sregex const rx_sec = '[' >> (s1= +_) >> ']';
|
|
sregex const rx_str = "str=" >> (s1= *_);
|
|
sregex const rx_pat = "pat=" >> (s1= *_);
|
|
sregex const rx_flg = "flg=" >> (s1= *_);
|
|
sregex const rx_sub = "sub=" >> (s1= *_);
|
|
sregex const rx_res = "res=" >> (s1= *_);
|
|
sregex const rx_br = "br" >> (s1= +digit) >> '=' >> (s2= *_);
|
|
|
|
while(in.good())
|
|
{
|
|
std::getline(in, line);
|
|
|
|
if(!line.empty() && '\r' == line[line.size()-1])
|
|
{
|
|
line.erase(line.size()-1);
|
|
}
|
|
|
|
if(regex_match(line, what, rx_sec))
|
|
{
|
|
if(!first)
|
|
{
|
|
if(what[1] != "end")
|
|
{
|
|
BOOST_FAIL(("invalid input : " + line).c_str());
|
|
}
|
|
break;
|
|
}
|
|
|
|
first = false;
|
|
test.section = what[1].str();
|
|
}
|
|
else if(regex_match(line, what, rx_str))
|
|
{
|
|
test.str = escape(what[1].str());
|
|
}
|
|
else if(regex_match(line, what, rx_pat))
|
|
{
|
|
test.pat = what[1].str();
|
|
}
|
|
else if(regex_match(line, what, rx_sub))
|
|
{
|
|
test.sub = what[1].str();
|
|
}
|
|
else if(regex_match(line, what, rx_res))
|
|
{
|
|
test.res = escape(what[1].str());
|
|
}
|
|
else if(regex_match(line, what, rx_flg))
|
|
{
|
|
std::string flg = what[1].str();
|
|
|
|
if(std::string::npos != flg.find('i'))
|
|
{
|
|
test.syntax_flags = test.syntax_flags | regex_constants::icase;
|
|
}
|
|
if(std::string::npos == flg.find('m'))
|
|
{
|
|
test.syntax_flags = test.syntax_flags | regex_constants::single_line;
|
|
}
|
|
if(std::string::npos == flg.find('s'))
|
|
{
|
|
test.syntax_flags = test.syntax_flags | regex_constants::not_dot_newline;
|
|
}
|
|
if(std::string::npos != flg.find('x'))
|
|
{
|
|
test.syntax_flags = test.syntax_flags | regex_constants::ignore_white_space;
|
|
}
|
|
if(std::string::npos != flg.find('g'))
|
|
{
|
|
test.match_flags = test.match_flags & ~regex_constants::format_first_only;
|
|
}
|
|
if(std::string::npos != flg.find('a'))
|
|
{
|
|
test.match_flags = test.match_flags | regex_constants::format_all;
|
|
}
|
|
if(std::string::npos != flg.find('p'))
|
|
{
|
|
test.match_flags = test.match_flags | regex_constants::format_perl;
|
|
}
|
|
if(std::string::npos != flg.find('d'))
|
|
{
|
|
test.match_flags = test.match_flags | regex_constants::format_sed;
|
|
}
|
|
}
|
|
else if(regex_match(line, what, rx_br))
|
|
{
|
|
std::size_t nbr = boost::lexical_cast<std::size_t>(what[1].str());
|
|
|
|
if(nbr >= test.br.size())
|
|
{
|
|
test.br.resize(nbr + 1);
|
|
}
|
|
|
|
test.br[nbr] = escape(what[2].str());
|
|
}
|
|
else if(!line.empty() && ';' != line[0])
|
|
{
|
|
BOOST_FAIL((std::string("invalid input : ") + line).c_str());
|
|
}
|
|
}
|
|
|
|
return !first;
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// run_test_impl
|
|
// run the test
|
|
template<typename Char>
|
|
void run_test_impl(xpr_test_case<Char> const &test)
|
|
{
|
|
try
|
|
{
|
|
Char const empty[] = {0};
|
|
typedef typename std::basic_string<Char>::const_iterator iterator;
|
|
basic_regex<iterator> rx = basic_regex<iterator>::compile(test.pat, test.syntax_flags);
|
|
|
|
// Build the same regex for use with C strings
|
|
basic_regex<Char const *> c_rx = basic_regex<Char const *>::compile(test.pat, test.syntax_flags);
|
|
|
|
if(!test.res.empty())
|
|
{
|
|
// test regex_replace
|
|
std::basic_string<Char> res = regex_replace(test.str, rx, test.sub, test.match_flags);
|
|
BOOST_CHECK_MESSAGE(res == test.res, case_ << res << " != " << test.res );
|
|
|
|
// test regex_replace with NTBS format string
|
|
std::basic_string<Char> res2 = regex_replace(test.str, rx, test.sub.c_str(), test.match_flags);
|
|
BOOST_CHECK_MESSAGE(res2 == test.res, case_ << res2 << " != " << test.res );
|
|
|
|
// test regex_replace with NTBS input string
|
|
std::basic_string<Char> res3 = regex_replace(test.str.c_str(), c_rx, test.sub, test.match_flags);
|
|
BOOST_CHECK_MESSAGE(res3 == test.res, case_ << res3 << " != " << test.res );
|
|
|
|
// test regex_replace with NTBS input string and NTBS format string
|
|
std::basic_string<Char> res4 = regex_replace(test.str.c_str(), c_rx, test.sub.c_str(), test.match_flags);
|
|
BOOST_CHECK_MESSAGE(res4 == test.res, case_ << res4 << " != " << test.res );
|
|
}
|
|
|
|
if(0 == (test.match_flags & regex_constants::format_first_only))
|
|
{
|
|
{
|
|
// global search, use regex_iterator
|
|
std::vector<sub_match<iterator> > br;
|
|
regex_iterator<iterator> begin(test.str.begin(), test.str.end(), rx, test.match_flags), end;
|
|
for(; begin != end; ++begin)
|
|
{
|
|
match_results<iterator> const &what = *begin;
|
|
br.insert(br.end(), what.begin(), what.end());
|
|
}
|
|
|
|
// match succeeded: was it expected to succeed?
|
|
BOOST_XPR_CHECK(br.size() == test.br.size());
|
|
|
|
for(std::size_t i = 0; i < br.size() && i < test.br.size(); ++i)
|
|
{
|
|
BOOST_XPR_CHECK((!br[i].matched && test.br[i] == empty) || test.br[i] == br[i].str());
|
|
}
|
|
}
|
|
|
|
{
|
|
// global search, use regex_token_iterator
|
|
std::vector<typename sub_match<iterator>::string_type> br2;
|
|
std::vector<int> subs(rx.mark_count() + 1, 0);
|
|
// regex_token_iterator will extract all sub_matches, in order:
|
|
for(std::size_t i = 0; i < subs.size(); ++i)
|
|
{
|
|
subs[i] = static_cast<int>(i);
|
|
}
|
|
regex_token_iterator<iterator> begin2(test.str.begin(), test.str.end(), rx, subs, test.match_flags), end2;
|
|
for(; begin2 != end2; ++begin2)
|
|
{
|
|
br2.push_back(*begin2);
|
|
}
|
|
|
|
// match succeeded: was it expected to succeed?
|
|
BOOST_XPR_CHECK(br2.size() == test.br.size());
|
|
|
|
for(std::size_t i = 0; i < br2.size() && i < test.br.size(); ++i)
|
|
{
|
|
BOOST_XPR_CHECK(test.br[i] == br2[i]);
|
|
}
|
|
}
|
|
}
|
|
else
|
|
{
|
|
// test regex_search
|
|
match_results<iterator> what;
|
|
if(regex_search(test.str, what, rx, test.match_flags))
|
|
{
|
|
// match succeeded: was it expected to succeed?
|
|
BOOST_XPR_CHECK(what.size() == test.br.size());
|
|
|
|
for(std::size_t i = 0; i < what.size() && i < test.br.size(); ++i)
|
|
{
|
|
BOOST_XPR_CHECK((!what[i].matched && test.br[i] == empty) || test.br[i] == what[i].str());
|
|
}
|
|
}
|
|
else
|
|
{
|
|
// match failed: was it expected to fail?
|
|
BOOST_XPR_CHECK(0 == test.br.size());
|
|
}
|
|
}
|
|
}
|
|
catch(regex_error const &e)
|
|
{
|
|
BOOST_ERROR(case_ << e.what());
|
|
}
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// run_test_impl
|
|
// run the current test
|
|
void run_test()
|
|
{
|
|
#ifdef BOOST_XPRESSIVE_TEST_WREGEX
|
|
xpr_test_case<wchar_t> wtest = ::widen(test);
|
|
run_test_impl(wtest);
|
|
#else
|
|
run_test_impl(test);
|
|
#endif
|
|
}
|
|
|
|
static char const * s_argv1;
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// open_test
|
|
bool open_test()
|
|
{
|
|
in.open( s_argv1? s_argv1: "regress.txt" );
|
|
return in.good();
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// test_main
|
|
// read the tests from the input file and execute them
|
|
void test_main()
|
|
{
|
|
#if !defined(BOOST_XPRESSIVE_TEST_WREGEX) || !defined(BOOST_XPRESSIVE_NO_WREGEX)
|
|
if(!open_test())
|
|
{
|
|
BOOST_ERROR("Error: unable to open input file.");
|
|
}
|
|
|
|
while(get_test())
|
|
{
|
|
run_test();
|
|
++test_count;
|
|
}
|
|
#endif
|
|
|
|
std::cout << test_count << " tests completed." << std::endl;
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// init_unit_test_suite
|
|
//
|
|
test_suite* init_unit_test_suite( int argc, char* argv[] )
|
|
{
|
|
s_argv1 = argv[1];
|
|
|
|
test_suite *test = BOOST_TEST_SUITE("basic regression test");
|
|
test->add(BOOST_TEST_CASE(&test_main));
|
|
return test;
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////////////
|
|
// debug_init
|
|
static const struct debug_init
|
|
{
|
|
debug_init()
|
|
{
|
|
#if defined(_MSC_VER) && defined(_DEBUG)
|
|
// Send warnings, errors and asserts to STDERR
|
|
_CrtSetReportMode(_CRT_WARN, _CRTDBG_MODE_FILE | _CRTDBG_MODE_DEBUG);
|
|
_CrtSetReportFile(_CRT_WARN, _CRTDBG_FILE_STDERR);
|
|
_CrtSetReportMode(_CRT_ERROR, _CRTDBG_MODE_FILE | _CRTDBG_MODE_DEBUG);
|
|
_CrtSetReportFile(_CRT_ERROR, _CRTDBG_FILE_STDERR);
|
|
_CrtSetReportMode(_CRT_ASSERT, _CRTDBG_MODE_FILE | _CRTDBG_MODE_DEBUG);
|
|
_CrtSetReportFile(_CRT_ASSERT, _CRTDBG_FILE_STDERR);
|
|
|
|
// Check for leaks at program termination
|
|
_CrtSetDbgFlag(_CRTDBG_LEAK_CHECK_DF | _CrtSetDbgFlag(_CRTDBG_REPORT_FLAG));
|
|
|
|
//_CrtSetBreakAlloc(221);
|
|
#endif
|
|
}
|
|
} g_debug_init;
|