rapidjson/test/unittest/readertest.cpp
Milo Yip 57b91300fb Merge remote-tracking branch 'origin/master' into issue120floatprecision_customstrtod
Conflicts:
	include/rapidjson/internal/dtoa.h
	test/unittest/readertest.cpp
2014-11-14 22:23:16 +08:00

1174 lines
44 KiB
C++
Raw Blame History

// Copyright (C) 2011 Milo Yip
//
// Permission is hereby granted, free of charge, to any person obtaining a copy
// of this software and associated documentation files (the "Software"), to deal
// in the Software without restriction, including without limitation the rights
// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
// copies of the Software, and to permit persons to whom the Software is
// furnished to do so, subject to the following conditions:
//
// The above copyright notice and this permission notice shall be included in
// all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
// THE SOFTWARE.
#include "unittest.h"
#include "rapidjson/reader.h"
#include "rapidjson/internal/dtoa.h"
#include "rapidjson/internal/itoa.h"
#include "rapidjson/memorystream.h"
using namespace rapidjson;
#ifdef __GNUC__
RAPIDJSON_DIAG_PUSH
RAPIDJSON_DIAG_OFF(effc++)
#endif
template<bool expect>
struct ParseBoolHandler : BaseReaderHandler<UTF8<>, ParseBoolHandler<expect> > {
ParseBoolHandler() : step_(0) {}
bool Default() { ADD_FAILURE(); return false; }
// gcc 4.8.x generates warning in EXPECT_EQ(bool, bool) on this gtest version.
// Workaround with EXPECT_TRUE().
bool Bool(bool b) { /*EXPECT_EQ(expect, b); */EXPECT_TRUE(expect == b); ++step_; return true; }
unsigned step_;
};
TEST(Reader, ParseTrue) {
StringStream s("true");
ParseBoolHandler<true> h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(1u, h.step_);
}
TEST(Reader, ParseFalse) {
StringStream s("false");
ParseBoolHandler<false> h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(1u, h.step_);
}
struct ParseIntHandler : BaseReaderHandler<UTF8<>, ParseIntHandler> {
ParseIntHandler() : step_(0), actual_() {}
bool Default() { ADD_FAILURE(); return false; }
bool Int(int i) { actual_ = i; step_++; return true; }
unsigned step_;
int actual_;
};
struct ParseUintHandler : BaseReaderHandler<UTF8<>, ParseUintHandler> {
ParseUintHandler() : step_(0), actual_() {}
bool Default() { ADD_FAILURE(); return false; }
bool Uint(unsigned i) { actual_ = i; step_++; return true; }
unsigned step_;
unsigned actual_;
};
struct ParseInt64Handler : BaseReaderHandler<UTF8<>, ParseInt64Handler> {
ParseInt64Handler() : step_(0), actual_() {}
bool Default() { ADD_FAILURE(); return false; }
bool Int64(int64_t i) { actual_ = i; step_++; return true; }
unsigned step_;
int64_t actual_;
};
struct ParseUint64Handler : BaseReaderHandler<UTF8<>, ParseUint64Handler> {
ParseUint64Handler() : step_(0), actual_() {}
bool Default() { ADD_FAILURE(); return false; }
bool Uint64(uint64_t i) { actual_ = i; step_++; return true; }
unsigned step_;
uint64_t actual_;
};
struct ParseDoubleHandler : BaseReaderHandler<UTF8<>, ParseDoubleHandler> {
ParseDoubleHandler() : step_(0), actual_() {}
bool Default() { ADD_FAILURE(); return false; }
bool Double(double d) { actual_ = d; step_++; return true; }
unsigned step_;
double actual_;
};
TEST(Reader, ParseNumber_Integer) {
#define TEST_INTEGER(Handler, str, x) \
{ \
StringStream s(str); \
Handler h; \
Reader reader; \
reader.Parse(s, h); \
EXPECT_EQ(1u, h.step_); \
EXPECT_EQ(x, h.actual_); \
}
TEST_INTEGER(ParseUintHandler, "0", 0u);
TEST_INTEGER(ParseUintHandler, "123", 123u);
TEST_INTEGER(ParseUintHandler, "2147483648", 2147483648u); // 2^31 - 1 (cannot be stored in int)
TEST_INTEGER(ParseUintHandler, "4294967295", 4294967295u);
TEST_INTEGER(ParseIntHandler, "-123", -123);
TEST_INTEGER(ParseIntHandler, "-2147483648", static_cast<int32_t>(0x80000000)); // -2^31 (min of int)
TEST_INTEGER(ParseUint64Handler, "4294967296", RAPIDJSON_UINT64_C2(1, 0)); // 2^32 (max of unsigned + 1, force to use uint64_t)
TEST_INTEGER(ParseUint64Handler, "18446744073709551615", RAPIDJSON_UINT64_C2(0xFFFFFFFF, 0xFFFFFFFF)); // 2^64 - 1 (max of uint64_t)
TEST_INTEGER(ParseInt64Handler, "-2147483649", static_cast<int64_t>(RAPIDJSON_UINT64_C2(0xFFFFFFFF, 0x7FFFFFFF))); // -2^31 -1 (min of int - 1, force to use int64_t)
TEST_INTEGER(ParseInt64Handler, "-9223372036854775808", static_cast<int64_t>(RAPIDJSON_UINT64_C2(0x80000000, 0x00000000))); // -2^63 (min of int64_t)
// Random test for uint32_t/int32_t
{
union {
uint32_t u;
int32_t i;
}u;
Random r;
for (unsigned i = 0; i < 100000; i++) {
u.u = r();
char buffer[32];
*internal::u32toa(u.u, buffer) = '\0';
TEST_INTEGER(ParseUintHandler, buffer, u.u);
if (u.i < 0) {
*internal::i32toa(u.i, buffer) = '\0';
TEST_INTEGER(ParseIntHandler, buffer, u.i);
}
}
}
// Random test for uint64_t/int64_t
{
union {
uint64_t u;
int64_t i;
}u;
Random r;
for (unsigned i = 0; i < 100000; i++) {
u.u = uint64_t(r()) << 32;
u.u |= r();
char buffer[32];
if (u.u >= 4294967296ULL) {
*internal::u64toa(u.u, buffer) = '\0';
TEST_INTEGER(ParseUint64Handler, buffer, u.u);
}
if (u.i <= -2147483649LL) {
*internal::i64toa(u.i, buffer) = '\0';
TEST_INTEGER(ParseInt64Handler, buffer, u.i);
}
}
}
#undef TEST_INTEGER
}
template<bool fullPrecision>
static void TestParseDouble() {
#define TEST_DOUBLE(fullPrecision, str, x) \
{ \
StringStream s(str); \
ParseDoubleHandler h; \
Reader reader; \
ASSERT_EQ(kParseErrorNone, reader.Parse<fullPrecision ? kParseFullPrecisionFlag : 0>(s, h).Code()); \
EXPECT_EQ(1u, h.step_); \
if (fullPrecision) { \
EXPECT_EQ(x, h.actual_); \
if (x != h.actual_) \
printf(" String: %s\n Actual: %.17g\nExpected: %.17g\n", str, h.actual_, x); \
} \
else \
EXPECT_DOUBLE_EQ(x, h.actual_); \
}
TEST_DOUBLE(fullPrecision, "0.0", 0.0);
TEST_DOUBLE(fullPrecision, "1.0", 1.0);
TEST_DOUBLE(fullPrecision, "-1.0", -1.0);
TEST_DOUBLE(fullPrecision, "1.5", 1.5);
TEST_DOUBLE(fullPrecision, "-1.5", -1.5);
TEST_DOUBLE(fullPrecision, "3.1416", 3.1416);
TEST_DOUBLE(fullPrecision, "1E10", 1E10);
TEST_DOUBLE(fullPrecision, "1e10", 1e10);
TEST_DOUBLE(fullPrecision, "1E+10", 1E+10);
TEST_DOUBLE(fullPrecision, "1E-10", 1E-10);
TEST_DOUBLE(fullPrecision, "-1E10", -1E10);
TEST_DOUBLE(fullPrecision, "-1e10", -1e10);
TEST_DOUBLE(fullPrecision, "-1E+10", -1E+10);
TEST_DOUBLE(fullPrecision, "-1E-10", -1E-10);
TEST_DOUBLE(fullPrecision, "1.234E+10", 1.234E+10);
TEST_DOUBLE(fullPrecision, "1.234E-10", 1.234E-10);
TEST_DOUBLE(fullPrecision, "1.79769e+308", 1.79769e+308);
TEST_DOUBLE(fullPrecision, "2.22507e-308", 2.22507e-308);
TEST_DOUBLE(fullPrecision, "-1.79769e+308", -1.79769e+308);
TEST_DOUBLE(fullPrecision, "-2.22507e-308", -2.22507e-308);
TEST_DOUBLE(fullPrecision, "4.9406564584124654e-324", 4.9406564584124654e-324); // minimum denormal
TEST_DOUBLE(fullPrecision, "2.2250738585072009e-308", 2.2250738585072009e-308); // Max subnormal double
TEST_DOUBLE(fullPrecision, "2.2250738585072014e-308", 2.2250738585072014e-308); // Min normal positive double
TEST_DOUBLE(fullPrecision, "1.7976931348623157e+308", 1.7976931348623157e+308); // Max double
TEST_DOUBLE(fullPrecision, "1e-10000", 0.0); // must underflow
TEST_DOUBLE(fullPrecision, "18446744073709551616", 18446744073709551616.0); // 2^64 (max of uint64_t + 1, force to use double)
TEST_DOUBLE(fullPrecision, "-9223372036854775809", -9223372036854775809.0); // -2^63 - 1(min of int64_t + 1, force to use double)
TEST_DOUBLE(fullPrecision, "0.9868011474609375", 0.9868011474609375); // https://github.com/miloyip/rapidjson/issues/120
TEST_DOUBLE(fullPrecision, "123e34", 123e34); // Fast Path Cases In Disguise
TEST_DOUBLE(fullPrecision, "45913141877270640000.0", 45913141877270640000.0);
TEST_DOUBLE(fullPrecision, "2.2250738585072011e-308", 2.2250738585072011e-308); // http://www.exploringbinary.com/php-hangs-on-numeric-value-2-2250738585072011e-308/
// Since
// abs((2^-1022 - 2^-1074) - 2.2250738585072012e-308) = 3.109754131239141401123495768877590405345064751974375599... <20><> 10^-324
// abs((2^-1022) - 2.2250738585072012e-308) = 1.830902327173324040642192159804623318305533274168872044... <20><> 10 ^ -324
// So 2.2250738585072012e-308 should round to 2^-1022 = 2.2250738585072014e-308
TEST_DOUBLE(fullPrecision, "2.2250738585072012e-308", 2.2250738585072014e-308); // http://www.exploringbinary.com/java-hangs-when-converting-2-2250738585072012e-308/
// More closer to normal/subnormal boundary
// boundary = 2^-1022 - 2^-1075 = 2.225073858507201136057409796709131975934819546351645648... <20><> 10^-308
TEST_DOUBLE(fullPrecision, "2.22507385850720113605740979670913197593481954635164564e-308", 2.2250738585072009e-308);
TEST_DOUBLE(fullPrecision, "2.22507385850720113605740979670913197593481954635164565e-308", 2.2250738585072014e-308);
// 1.0 is in (1.0 - 2^-54, 1.0 + 2^-53)
// 1.0 - 2^-54 = 0.999999999999999944488848768742172978818416595458984375
TEST_DOUBLE(fullPrecision, "0.999999999999999944488848768742172978818416595458984375", 1.0); // round to even
TEST_DOUBLE(fullPrecision, "0.999999999999999944488848768742172978818416595458984374", 0.99999999999999989); // previous double
TEST_DOUBLE(fullPrecision, "0.999999999999999944488848768742172978818416595458984376", 1.0); // next double
// 1.0 + 2^-53 = 1.00000000000000011102230246251565404236316680908203125
TEST_DOUBLE(fullPrecision, "1.00000000000000011102230246251565404236316680908203125", 1.0); // round to even
TEST_DOUBLE(fullPrecision, "1.00000000000000011102230246251565404236316680908203124", 1.0); // previous double
TEST_DOUBLE(fullPrecision, "1.00000000000000011102230246251565404236316680908203126", 1.00000000000000022); // next double
{
char n1e308[310]; // '1' followed by 308 '0'
n1e308[0] = '1';
for (int i = 1; i < 309; i++)
n1e308[i] = '0';
n1e308[309] = '\0';
TEST_DOUBLE(fullPrecision, n1e308, 1E308);
}
#if 0 // Very slow
static const unsigned count = 10000000;
// Random test for double
{
Random r;
for (unsigned i = 0; i < count; i++) {
internal::Double d;
do {
// Need to call r() in two statements for cross-platform coherent sequence.
uint64_t u = uint64_t(r()) << 32;
u |= uint64_t(r());
d = internal::Double(u);
} while (d.IsNan() || d.IsInf()/* || !d.IsNormal()*/); // Also work for subnormal now
char buffer[32];
*internal::dtoa(d.Value(), buffer) = '\0';
TEST_DOUBLE(fullPrecision, buffer, d.Value());
}
}
#endif
#undef TEST_DOUBLE
}
TEST(Reader, ParseNumber_NormalPrecisionDouble) {
TestParseDouble<false>();
}
TEST(Reader, ParseNumber_FullPrecisionDouble) {
TestParseDouble<true>();
}
TEST(Reader, ParseNumber_NormalPrecisionError) {
static unsigned count = 1000000;
Random r;
double ulpSum = 0.0;
double ulpMax = 0.0;
for (unsigned i = 0; i < count; i++) {
internal::Double e, a;
do {
// Need to call r() in two statements for cross-platform coherent sequence.
uint64_t u = uint64_t(r()) << 32;
u |= uint64_t(r());
e = u;
} while (e.IsNan() || e.IsInf() || !e.IsNormal());
char buffer[32];
*internal::dtoa(e.Value(), buffer) = '\0';
StringStream s(buffer);
ParseDoubleHandler h;
Reader reader;
ASSERT_EQ(kParseErrorNone, reader.Parse(s, h).Code());
EXPECT_EQ(1u, h.step_);
a = h.actual_;
uint64_t bias1 = e.ToBias();
uint64_t bias2 = a.ToBias();
double ulp = bias1 >= bias2 ? bias1 - bias2 : bias2 - bias1;
ulpMax = std::max(ulpMax, ulp);
ulpSum += ulp;
}
printf("ULP Average = %g, Max = %g \n", ulpSum / count, ulpMax);
}
TEST(Reader, ParseNumber_Error) {
#define TEST_NUMBER_ERROR(errorCode, str) \
{ \
char buffer[1001]; \
sprintf(buffer, "%s", str); \
InsituStringStream s(buffer); \
BaseReaderHandler<> h; \
Reader reader; \
EXPECT_FALSE(reader.Parse(s, h)); \
EXPECT_EQ(errorCode, reader.GetParseErrorCode());\
}
// Number too big to be stored in double.
{
char n1e309[311]; // '1' followed by 309 '0'
n1e309[0] = '1';
for (int i = 1; i < 310; i++)
n1e309[i] = '0';
n1e309[310] = '\0';
TEST_NUMBER_ERROR(kParseErrorNumberTooBig, n1e309);
}
TEST_NUMBER_ERROR(kParseErrorNumberTooBig, "1e309");
// Miss fraction part in number.
TEST_NUMBER_ERROR(kParseErrorNumberMissFraction, "1.");
TEST_NUMBER_ERROR(kParseErrorNumberMissFraction, "1.a");
// Miss exponent in number.
TEST_NUMBER_ERROR(kParseErrorNumberMissExponent, "1e");
TEST_NUMBER_ERROR(kParseErrorNumberMissExponent, "1e_");
#undef TEST_NUMBER_ERROR
}
template <typename Encoding>
struct ParseStringHandler : BaseReaderHandler<Encoding, ParseStringHandler<Encoding> > {
ParseStringHandler() : str_(0), length_(0), copy_() {}
~ParseStringHandler() { EXPECT_TRUE(str_ != 0); if (copy_) free(const_cast<typename Encoding::Ch*>(str_)); }
ParseStringHandler(const ParseStringHandler&);
ParseStringHandler& operator=(const ParseStringHandler&);
bool Default() { ADD_FAILURE(); return false; }
bool String(const typename Encoding::Ch* str, size_t length, bool copy) {
EXPECT_EQ(0, str_);
if (copy) {
str_ = (typename Encoding::Ch*)malloc((length + 1) * sizeof(typename Encoding::Ch));
memcpy(const_cast<typename Encoding::Ch*>(str_), str, (length + 1) * sizeof(typename Encoding::Ch));
}
else
str_ = str;
length_ = length;
copy_ = copy;
return true;
}
const typename Encoding::Ch* str_;
size_t length_;
bool copy_;
};
TEST(Reader, ParseString) {
#define TEST_STRING(Encoding, e, x) \
{ \
Encoding::Ch* buffer = StrDup(x); \
GenericInsituStringStream<Encoding> is(buffer); \
ParseStringHandler<Encoding> h; \
GenericReader<Encoding, Encoding> reader; \
reader.Parse<kParseInsituFlag | kParseValidateEncodingFlag>(is, h); \
EXPECT_EQ(0, StrCmp<Encoding::Ch>(e, h.str_)); \
EXPECT_EQ(StrLen(e), h.length_); \
free(buffer); \
GenericStringStream<Encoding> s(x); \
ParseStringHandler<Encoding> h2; \
GenericReader<Encoding, Encoding> reader2; \
reader2.Parse(s, h2); \
EXPECT_EQ(0, StrCmp<Encoding::Ch>(e, h2.str_)); \
EXPECT_EQ(StrLen(e), h2.length_); \
}
// String constant L"\xXX" can only specify character code in bytes, which is not endianness-neutral.
// And old compiler does not support u"" and U"" string literal. So here specify string literal by array of Ch.
// In addition, GCC 4.8 generates -Wnarrowing warnings when character code >= 128 are assigned to signed integer types.
// Therefore, utype is added for declaring unsigned array, and then cast it to Encoding::Ch.
#define ARRAY(...) { __VA_ARGS__ }
#define TEST_STRINGARRAY(Encoding, utype, array, x) \
{ \
static const utype ue[] = array; \
static const Encoding::Ch* e = reinterpret_cast<const Encoding::Ch *>(&ue[0]); \
TEST_STRING(Encoding, e, x); \
}
#define TEST_STRINGARRAY2(Encoding, utype, earray, xarray) \
{ \
static const utype ue[] = earray; \
static const utype xe[] = xarray; \
static const Encoding::Ch* e = reinterpret_cast<const Encoding::Ch *>(&ue[0]); \
static const Encoding::Ch* x = reinterpret_cast<const Encoding::Ch *>(&xe[0]); \
TEST_STRING(Encoding, e, x); \
}
TEST_STRING(UTF8<>, "", "\"\"");
TEST_STRING(UTF8<>, "Hello", "\"Hello\"");
TEST_STRING(UTF8<>, "Hello\nWorld", "\"Hello\\nWorld\"");
TEST_STRING(UTF8<>, "\"\\/\b\f\n\r\t", "\"\\\"\\\\/\\b\\f\\n\\r\\t\"");
TEST_STRING(UTF8<>, "\x24", "\"\\u0024\""); // Dollar sign U+0024
TEST_STRING(UTF8<>, "\xC2\xA2", "\"\\u00A2\""); // Cents sign U+00A2
TEST_STRING(UTF8<>, "\xE2\x82\xAC", "\"\\u20AC\""); // Euro sign U+20AC
TEST_STRING(UTF8<>, "\xF0\x9D\x84\x9E", "\"\\uD834\\uDD1E\""); // G clef sign U+1D11E
// UTF16
TEST_STRING(UTF16<>, L"", L"\"\"");
TEST_STRING(UTF16<>, L"Hello", L"\"Hello\"");
TEST_STRING(UTF16<>, L"Hello\nWorld", L"\"Hello\\nWorld\"");
TEST_STRING(UTF16<>, L"\"\\/\b\f\n\r\t", L"\"\\\"\\\\/\\b\\f\\n\\r\\t\"");
TEST_STRINGARRAY(UTF16<>, wchar_t, ARRAY(0x0024, 0x0000), L"\"\\u0024\"");
TEST_STRINGARRAY(UTF16<>, wchar_t, ARRAY(0x00A2, 0x0000), L"\"\\u00A2\""); // Cents sign U+00A2
TEST_STRINGARRAY(UTF16<>, wchar_t, ARRAY(0x20AC, 0x0000), L"\"\\u20AC\""); // Euro sign U+20AC
TEST_STRINGARRAY(UTF16<>, wchar_t, ARRAY(0xD834, 0xDD1E, 0x0000), L"\"\\uD834\\uDD1E\""); // G clef sign U+1D11E
// UTF32
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY('\0'), ARRAY('\"', '\"', '\0'));
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY('H', 'e', 'l', 'l', 'o', '\0'), ARRAY('\"', 'H', 'e', 'l', 'l', 'o', '\"', '\0'));
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY('H', 'e', 'l', 'l', 'o', '\n', 'W', 'o', 'r', 'l', 'd', '\0'), ARRAY('\"', 'H', 'e', 'l', 'l', 'o', '\\', 'n', 'W', 'o', 'r', 'l', 'd', '\"', '\0'));
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY('\"', '\\', '/', '\b', '\f', '\n', '\r', '\t', '\0'), ARRAY('\"', '\\', '\"', '\\', '\\', '/', '\\', 'b', '\\', 'f', '\\', 'n', '\\', 'r', '\\', 't', '\"', '\0'));
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY(0x00024, 0x0000), ARRAY('\"', '\\', 'u', '0', '0', '2', '4', '\"', '\0'));
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY(0x000A2, 0x0000), ARRAY('\"', '\\', 'u', '0', '0', 'A', '2', '\"', '\0')); // Cents sign U+00A2
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY(0x020AC, 0x0000), ARRAY('\"', '\\', 'u', '2', '0', 'A', 'C', '\"', '\0')); // Euro sign U+20AC
TEST_STRINGARRAY2(UTF32<>, unsigned, ARRAY(0x1D11E, 0x0000), ARRAY('\"', '\\', 'u', 'D', '8', '3', '4', '\\', 'u', 'D', 'D', '1', 'E', '\"', '\0')); // G clef sign U+1D11E
#undef TEST_STRINGARRAY
#undef ARRAY
#undef TEST_STRING
// Support of null character in string
{
StringStream s("\"Hello\\u0000World\"");
const char e[] = "Hello\0World";
ParseStringHandler<UTF8<> > h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(0, memcmp(e, h.str_, h.length_ + 1));
EXPECT_EQ(11u, h.length_);
}
}
TEST(Reader, ParseString_Transcoding) {
const char* x = "\"Hello\"";
const wchar_t* e = L"Hello";
GenericStringStream<UTF8<> > is(x);
GenericReader<UTF8<>, UTF16<> > reader;
ParseStringHandler<UTF16<> > h;
reader.Parse(is, h);
EXPECT_EQ(0, StrCmp<UTF16<>::Ch>(e, h.str_));
EXPECT_EQ(StrLen(e), h.length_);
}
TEST(Reader, ParseString_NonDestructive) {
StringStream s("\"Hello\\nWorld\"");
ParseStringHandler<UTF8<> > h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(0, StrCmp("Hello\nWorld", h.str_));
EXPECT_EQ(11u, h.length_);
}
ParseErrorCode TestString(const char* str) {
StringStream s(str);
BaseReaderHandler<> h;
Reader reader;
reader.Parse<kParseValidateEncodingFlag>(s, h);
return reader.GetParseErrorCode();
}
TEST(Reader, ParseString_Error) {
#define TEST_STRING_ERROR(errorCode, str)\
EXPECT_EQ(errorCode, TestString(str))
#define ARRAY(...) { __VA_ARGS__ }
#define TEST_STRINGENCODING_ERROR(Encoding, utype, array) \
{ \
static const utype ue[] = array; \
static const Encoding::Ch* e = reinterpret_cast<const Encoding::Ch *>(&ue[0]); \
EXPECT_EQ(kParseErrorStringInvalidEncoding, TestString(e));\
}
// Invalid escape character in string.
TEST_STRING_ERROR(kParseErrorStringEscapeInvalid, "[\"\\a\"]");
// Incorrect hex digit after \\u escape in string.
TEST_STRING_ERROR(kParseErrorStringUnicodeEscapeInvalidHex, "[\"\\uABCG\"]");
// The surrogate pair in string is invalid.
TEST_STRING_ERROR(kParseErrorStringUnicodeSurrogateInvalid, "[\"\\uD800X\"]");
TEST_STRING_ERROR(kParseErrorStringUnicodeSurrogateInvalid, "[\"\\uD800\\uFFFF\"]");
// Missing a closing quotation mark in string.
TEST_STRING_ERROR(kParseErrorStringMissQuotationMark, "[\"Test]");
// http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt
// 3 Malformed sequences
// 3.1 Unexpected continuation bytes
{
char e[] = { '[', '\"', 0, '\"', ']', '\0' };
for (unsigned char c = 0x80u; c <= 0xBFu; c++) {
e[2] = c;
ParseErrorCode error = TestString(e);
EXPECT_EQ(kParseErrorStringInvalidEncoding, error);
if (error != kParseErrorStringInvalidEncoding)
std::cout << (unsigned)(unsigned char)c << std::endl;
}
}
// 3.2 Lonely start characters, 3.5 Impossible bytes
{
char e[] = { '[', '\"', 0, ' ', '\"', ']', '\0' };
for (unsigned c = 0xC0u; c <= 0xFFu; c++) {
e[2] = (char)c;
TEST_STRING_ERROR(kParseErrorStringInvalidEncoding, e);
}
}
// 4 Overlong sequences
// 4.1 Examples of an overlong ASCII character
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xC0u, 0xAFu, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xE0u, 0x80u, 0xAFu, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xF0u, 0x80u, 0x80u, 0xAFu, '\"', ']', '\0'));
// 4.2 Maximum overlong sequences
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xC1u, 0xBFu, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xE0u, 0x9Fu, 0xBFu, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xF0u, 0x8Fu, 0xBFu, 0xBFu, '\"', ']', '\0'));
// 4.3 Overlong representation of the NUL character
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xC0u, 0x80u, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xE0u, 0x80u, 0x80u, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xF0u, 0x80u, 0x80u, 0x80u, '\"', ']', '\0'));
// 5 Illegal code positions
// 5.1 Single UTF-16 surrogates
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xEDu, 0xA0u, 0x80u, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xEDu, 0xADu, 0xBFu, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xEDu, 0xAEu, 0x80u, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xEDu, 0xAFu, 0xBFu, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xEDu, 0xB0u, 0x80u, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xEDu, 0xBEu, 0x80u, '\"', ']', '\0'));
TEST_STRINGENCODING_ERROR(UTF8<>, unsigned char, ARRAY('[', '\"', 0xEDu, 0xBFu, 0xBFu, '\"', ']', '\0'));
#undef ARRAY
#undef TEST_STRINGARRAY_ERROR
}
template <unsigned count>
struct ParseArrayHandler : BaseReaderHandler<UTF8<>, ParseArrayHandler<count> > {
ParseArrayHandler() : step_(0) {}
bool Default() { ADD_FAILURE(); return false; }
bool Uint(unsigned i) { EXPECT_EQ(step_, i); step_++; return true; }
bool StartArray() { EXPECT_EQ(0u, step_); step_++; return true; }
bool EndArray(SizeType) { step_++; return true; }
unsigned step_;
};
TEST(Reader, ParseEmptyArray) {
char *json = StrDup("[ ] ");
InsituStringStream s(json);
ParseArrayHandler<0> h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(2u, h.step_);
free(json);
}
TEST(Reader, ParseArray) {
char *json = StrDup("[1, 2, 3, 4]");
InsituStringStream s(json);
ParseArrayHandler<4> h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(6u, h.step_);
free(json);
}
TEST(Reader, ParseArray_Error) {
#define TEST_ARRAY_ERROR(errorCode, str) \
{ \
char buffer[1001]; \
strncpy(buffer, str, 1000); \
InsituStringStream s(buffer); \
BaseReaderHandler<> h; \
GenericReader<UTF8<>, UTF8<>, CrtAllocator> reader; \
EXPECT_FALSE(reader.Parse(s, h)); \
EXPECT_EQ(errorCode, reader.GetParseErrorCode());\
}
// Missing a comma or ']' after an array element.
TEST_ARRAY_ERROR(kParseErrorArrayMissCommaOrSquareBracket, "[1");
TEST_ARRAY_ERROR(kParseErrorArrayMissCommaOrSquareBracket, "[1}");
TEST_ARRAY_ERROR(kParseErrorArrayMissCommaOrSquareBracket, "[1 2]");
#undef TEST_ARRAY_ERROR
}
struct ParseObjectHandler : BaseReaderHandler<UTF8<>, ParseObjectHandler> {
ParseObjectHandler() : step_(0) {}
bool Default() { ADD_FAILURE(); return false; }
bool Null() { EXPECT_EQ(8u, step_); step_++; return true; }
bool Bool(bool b) {
switch(step_) {
case 4: EXPECT_TRUE(b); step_++; return true;
case 6: EXPECT_FALSE(b); step_++; return true;
default: ADD_FAILURE(); return false;
}
}
bool Int(int i) {
switch(step_) {
case 10: EXPECT_EQ(123, i); step_++; return true;
case 15: EXPECT_EQ(1, i); step_++; return true;
case 16: EXPECT_EQ(2, i); step_++; return true;
case 17: EXPECT_EQ(3, i); step_++; return true;
default: ADD_FAILURE(); return false;
}
}
bool Uint(unsigned i) { return Int(i); }
bool Double(double d) { EXPECT_EQ(12u, step_); EXPECT_EQ(3.1416, d); step_++; return true; }
bool String(const char* str, size_t, bool) {
switch(step_) {
case 1: EXPECT_STREQ("hello", str); step_++; return true;
case 2: EXPECT_STREQ("world", str); step_++; return true;
case 3: EXPECT_STREQ("t", str); step_++; return true;
case 5: EXPECT_STREQ("f", str); step_++; return true;
case 7: EXPECT_STREQ("n", str); step_++; return true;
case 9: EXPECT_STREQ("i", str); step_++; return true;
case 11: EXPECT_STREQ("pi", str); step_++; return true;
case 13: EXPECT_STREQ("a", str); step_++; return true;
default: ADD_FAILURE(); return false;
}
}
bool StartObject() { EXPECT_EQ(0u, step_); step_++; return true; }
bool EndObject(SizeType memberCount) { EXPECT_EQ(19u, step_); EXPECT_EQ(7u, memberCount); step_++; return true; }
bool StartArray() { EXPECT_EQ(14u, step_); step_++; return true; }
bool EndArray(SizeType elementCount) { EXPECT_EQ(18u, step_); EXPECT_EQ(3u, elementCount); step_++; return true; }
unsigned step_;
};
TEST(Reader, ParseObject) {
const char* json = "{ \"hello\" : \"world\", \"t\" : true , \"f\" : false, \"n\": null, \"i\":123, \"pi\": 3.1416, \"a\":[1, 2, 3] } ";
// Insitu
{
char* json2 = StrDup(json);
InsituStringStream s(json2);
ParseObjectHandler h;
Reader reader;
reader.Parse<kParseInsituFlag>(s, h);
EXPECT_EQ(20u, h.step_);
free(json2);
}
// Normal
{
StringStream s(json);
ParseObjectHandler h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(20u, h.step_);
}
}
struct ParseEmptyObjectHandler : BaseReaderHandler<UTF8<>, ParseEmptyObjectHandler> {
ParseEmptyObjectHandler() : step_(0) {}
bool Default() { ADD_FAILURE(); return false; }
bool StartObject() { EXPECT_EQ(0u, step_); step_++; return true; }
bool EndObject(SizeType) { EXPECT_EQ(1u, step_); step_++; return true; }
unsigned step_;
};
TEST(Reader, Parse_EmptyObject) {
StringStream s("{ } ");
ParseEmptyObjectHandler h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(2u, h.step_);
}
struct ParseMultipleRootHandler : BaseReaderHandler<UTF8<>, ParseMultipleRootHandler> {
ParseMultipleRootHandler() : step_(0) {}
bool Default() { ADD_FAILURE(); return false; }
bool StartObject() { EXPECT_EQ(0u, step_); step_++; return true; }
bool EndObject(SizeType) { EXPECT_EQ(1u, step_); step_++; return true; }
bool StartArray() { EXPECT_EQ(2u, step_); step_++; return true; }
bool EndArray(SizeType) { EXPECT_EQ(3u, step_); step_++; return true; }
unsigned step_;
};
template <unsigned parseFlags>
void TestMultipleRoot() {
StringStream s("{}[] a");
ParseMultipleRootHandler h;
Reader reader;
EXPECT_TRUE(reader.Parse<parseFlags>(s, h));
EXPECT_EQ(2u, h.step_);
EXPECT_TRUE(reader.Parse<parseFlags>(s, h));
EXPECT_EQ(4u, h.step_);
EXPECT_EQ(' ', s.Take());
EXPECT_EQ('a', s.Take());
}
TEST(Reader, Parse_MultipleRoot) {
TestMultipleRoot<kParseStopWhenDoneFlag>();
}
TEST(Reader, ParseIterative_MultipleRoot) {
TestMultipleRoot<kParseIterativeFlag | kParseStopWhenDoneFlag>();
}
template <unsigned parseFlags>
void TestInsituMultipleRoot() {
char* buffer = strdup("{}[] a");
InsituStringStream s(buffer);
ParseMultipleRootHandler h;
Reader reader;
EXPECT_TRUE(reader.Parse<kParseInsituFlag | parseFlags>(s, h));
EXPECT_EQ(2u, h.step_);
EXPECT_TRUE(reader.Parse<kParseInsituFlag | parseFlags>(s, h));
EXPECT_EQ(4u, h.step_);
EXPECT_EQ(' ', s.Take());
EXPECT_EQ('a', s.Take());
free(buffer);
}
TEST(Reader, ParseInsitu_MultipleRoot) {
TestInsituMultipleRoot<kParseStopWhenDoneFlag>();
}
TEST(Reader, ParseInsituIterative_MultipleRoot) {
TestInsituMultipleRoot<kParseIterativeFlag | kParseStopWhenDoneFlag>();
}
#define TEST_ERROR(errorCode, str) \
{ \
char buffer[1001]; \
strncpy(buffer, str, 1000); \
InsituStringStream s(buffer); \
BaseReaderHandler<> h; \
Reader reader; \
EXPECT_FALSE(reader.Parse(s, h)); \
EXPECT_EQ(errorCode, reader.GetParseErrorCode());\
}
TEST(Reader, ParseDocument_Error) {
// The document is empty.
TEST_ERROR(kParseErrorDocumentEmpty, "");
TEST_ERROR(kParseErrorDocumentEmpty, " ");
TEST_ERROR(kParseErrorDocumentEmpty, " \n");
// The document root must not follow by other values.
TEST_ERROR(kParseErrorDocumentRootNotSingular, "[] 0");
TEST_ERROR(kParseErrorDocumentRootNotSingular, "{} 0");
TEST_ERROR(kParseErrorDocumentRootNotSingular, "null []");
TEST_ERROR(kParseErrorDocumentRootNotSingular, "0 {}");
}
TEST(Reader, ParseValue_Error) {
// Invalid value.
TEST_ERROR(kParseErrorValueInvalid, "nulL");
TEST_ERROR(kParseErrorValueInvalid, "truE");
TEST_ERROR(kParseErrorValueInvalid, "falsE");
TEST_ERROR(kParseErrorValueInvalid, "a]");
TEST_ERROR(kParseErrorValueInvalid, ".1");
}
TEST(Reader, ParseObject_Error) {
// Missing a name for object member.
TEST_ERROR(kParseErrorObjectMissName, "{1}");
TEST_ERROR(kParseErrorObjectMissName, "{:1}");
TEST_ERROR(kParseErrorObjectMissName, "{null:1}");
TEST_ERROR(kParseErrorObjectMissName, "{true:1}");
TEST_ERROR(kParseErrorObjectMissName, "{false:1}");
TEST_ERROR(kParseErrorObjectMissName, "{1:1}");
TEST_ERROR(kParseErrorObjectMissName, "{[]:1}");
TEST_ERROR(kParseErrorObjectMissName, "{{}:1}");
TEST_ERROR(kParseErrorObjectMissName, "{xyz:1}");
// Missing a colon after a name of object member.
TEST_ERROR(kParseErrorObjectMissColon, "{\"a\" 1}");
TEST_ERROR(kParseErrorObjectMissColon, "{\"a\",1}");
// Must be a comma or '}' after an object member
TEST_ERROR(kParseErrorObjectMissCommaOrCurlyBracket, "{\"a\":1]");
// This tests that MemoryStream is checking the length in Peek().
{
MemoryStream ms("{\"a\"", 1);
BaseReaderHandler<> h;
Reader reader;
EXPECT_FALSE(reader.Parse<kParseStopWhenDoneFlag>(ms, h));
EXPECT_EQ(kParseErrorObjectMissName, reader.GetParseErrorCode());
}
}
#undef TEST_ERROR
TEST(Reader, SkipWhitespace) {
StringStream ss(" A \t\tB\n \n\nC\r\r \rD \t\n\r E");
const char* expected = "ABCDE";
for (size_t i = 0; i < 5; i++) {
SkipWhitespace(ss);
EXPECT_EQ(expected[i], ss.Take());
}
}
// Test implementing a stream without copy stream optimization.
// Clone from GenericStringStream except that copy constructor is disabled.
template <typename Encoding>
class CustomStringStream {
public:
typedef typename Encoding::Ch Ch;
CustomStringStream(const Ch *src) : src_(src), head_(src) {}
Ch Peek() const { return *src_; }
Ch Take() { return *src_++; }
size_t Tell() const { return static_cast<size_t>(src_ - head_); }
Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; }
void Put(Ch) { RAPIDJSON_ASSERT(false); }
void Flush() { RAPIDJSON_ASSERT(false); }
size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; }
private:
// Prohibit copy constructor & assignment operator.
CustomStringStream(const CustomStringStream&);
CustomStringStream& operator=(const CustomStringStream&);
const Ch* src_; //!< Current read position.
const Ch* head_; //!< Original head of the string.
};
// If the following code is compiled, it should generate compilation error as predicted.
// Because CustomStringStream<> is not copyable via making copy constructor private.
#if 0
namespace rapidjson {
template <typename Encoding>
struct StreamTraits<CustomStringStream<Encoding> > {
enum { copyOptimization = 1 };
};
} // namespace rapidjson
#endif
TEST(Reader, CustomStringStream) {
const char* json = "{ \"hello\" : \"world\", \"t\" : true , \"f\" : false, \"n\": null, \"i\":123, \"pi\": 3.1416, \"a\":[1, 2, 3] } ";
CustomStringStream<UTF8<char> > s(json);
ParseObjectHandler h;
Reader reader;
reader.Parse(s, h);
EXPECT_EQ(20u, h.step_);
}
#include <sstream>
class IStreamWrapper {
public:
typedef char Ch;
IStreamWrapper(std::istream& is) : is_(is) {}
Ch Peek() const {
int c = is_.peek();
return c == std::char_traits<char>::eof() ? '\0' : (Ch)c;
}
Ch Take() {
int c = is_.get();
return c == std::char_traits<char>::eof() ? '\0' : (Ch)c;
}
size_t Tell() const { return (size_t)is_.tellg(); }
Ch* PutBegin() { assert(false); return 0; }
void Put(Ch) { assert(false); }
void Flush() { assert(false); }
size_t PutEnd(Ch*) { assert(false); return 0; }
private:
IStreamWrapper(const IStreamWrapper&);
IStreamWrapper& operator=(const IStreamWrapper&);
std::istream& is_;
};
TEST(Reader, Parse_IStreamWrapper_StringStream) {
const char* json = "[1,2,3,4]";
std::stringstream ss(json);
IStreamWrapper is(ss);
Reader reader;
ParseArrayHandler<4> h;
reader.Parse(is, h);
EXPECT_FALSE(reader.HasParseError());
}
// Test iterative parsing.
#define TESTERRORHANDLING(text, errorCode, offset)\
{\
StringStream json(text); \
BaseReaderHandler<> handler; \
Reader reader; \
reader.Parse<kParseIterativeFlag>(json, handler); \
EXPECT_TRUE(reader.HasParseError()); \
EXPECT_EQ(errorCode, reader.GetParseErrorCode()); \
EXPECT_EQ(offset, reader.GetErrorOffset()); \
}
TEST(Reader, IterativeParsing_ErrorHandling) {
TESTERRORHANDLING("{\"a\": a}", kParseErrorValueInvalid, 6u);
TESTERRORHANDLING("", kParseErrorDocumentEmpty, 0u);
TESTERRORHANDLING("{}{}", kParseErrorDocumentRootNotSingular, 2u);
TESTERRORHANDLING("{1}", kParseErrorObjectMissName, 1u);
TESTERRORHANDLING("{\"a\", 1}", kParseErrorObjectMissColon, 4u);
TESTERRORHANDLING("{\"a\"}", kParseErrorObjectMissColon, 4u);
TESTERRORHANDLING("{\"a\": 1", kParseErrorObjectMissCommaOrCurlyBracket, 7u);
TESTERRORHANDLING("[1 2 3]", kParseErrorArrayMissCommaOrSquareBracket, 3u);
}
template<typename Encoding = UTF8<> >
struct IterativeParsingReaderHandler {
typedef typename Encoding::Ch Ch;
const static int LOG_NULL = -1;
const static int LOG_BOOL = -2;
const static int LOG_INT = -3;
const static int LOG_UINT = -4;
const static int LOG_INT64 = -5;
const static int LOG_UINT64 = -6;
const static int LOG_DOUBLE = -7;
const static int LOG_STRING = -8;
const static int LOG_STARTOBJECT = -9;
const static int LOG_KEY = -10;
const static int LOG_ENDOBJECT = -11;
const static int LOG_STARTARRAY = -12;
const static int LOG_ENDARRAY = -13;
const static size_t LogCapacity = 256;
int Logs[LogCapacity];
size_t LogCount;
IterativeParsingReaderHandler() : LogCount(0) {
}
bool Null() { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_NULL; return true; }
bool Bool(bool) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_BOOL; return true; }
bool Int(int) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_INT; return true; }
bool Uint(unsigned) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_INT; return true; }
bool Int64(int64_t) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_INT64; return true; }
bool Uint64(uint64_t) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_UINT64; return true; }
bool Double(double) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_DOUBLE; return true; }
bool String(const Ch*, SizeType, bool) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_STRING; return true; }
bool StartObject() { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_STARTOBJECT; return true; }
bool Key (const Ch*, SizeType, bool) { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_KEY; return true; }
bool EndObject(SizeType c) {
RAPIDJSON_ASSERT(LogCount < LogCapacity);
Logs[LogCount++] = LOG_ENDOBJECT;
Logs[LogCount++] = (int)c;
return true;
}
bool StartArray() { RAPIDJSON_ASSERT(LogCount < LogCapacity); Logs[LogCount++] = LOG_STARTARRAY; return true; }
bool EndArray(SizeType c) {
RAPIDJSON_ASSERT(LogCount < LogCapacity);
Logs[LogCount++] = LOG_ENDARRAY;
Logs[LogCount++] = (int)c;
return true;
}
};
TEST(Reader, IterativeParsing_General) {
{
StringStream is("[1, {\"k\": [1, 2]}, null, false, true, \"string\", 1.2]");
Reader reader;
IterativeParsingReaderHandler<> handler;
ParseResult r = reader.Parse<kParseIterativeFlag>(is, handler);
EXPECT_FALSE(r.IsError());
EXPECT_FALSE(reader.HasParseError());
int e[] = {
handler.LOG_STARTARRAY,
handler.LOG_INT,
handler.LOG_STARTOBJECT,
handler.LOG_KEY,
handler.LOG_STARTARRAY,
handler.LOG_INT,
handler.LOG_INT,
handler.LOG_ENDARRAY, 2,
handler.LOG_ENDOBJECT, 1,
handler.LOG_NULL,
handler.LOG_BOOL,
handler.LOG_BOOL,
handler.LOG_STRING,
handler.LOG_DOUBLE,
handler.LOG_ENDARRAY, 7
};
EXPECT_EQ(sizeof(e) / sizeof(int), handler.LogCount);
for (size_t i = 0; i < handler.LogCount; ++i) {
EXPECT_EQ(e[i], handler.Logs[i]) << "i = " << i;
}
}
}
TEST(Reader, IterativeParsing_Count) {
{
StringStream is("[{}, {\"k\": 1}, [1], []]");
Reader reader;
IterativeParsingReaderHandler<> handler;
ParseResult r = reader.Parse<kParseIterativeFlag>(is, handler);
EXPECT_FALSE(r.IsError());
EXPECT_FALSE(reader.HasParseError());
int e[] = {
handler.LOG_STARTARRAY,
handler.LOG_STARTOBJECT,
handler.LOG_ENDOBJECT, 0,
handler.LOG_STARTOBJECT,
handler.LOG_KEY,
handler.LOG_INT,
handler.LOG_ENDOBJECT, 1,
handler.LOG_STARTARRAY,
handler.LOG_INT,
handler.LOG_ENDARRAY, 1,
handler.LOG_STARTARRAY,
handler.LOG_ENDARRAY, 0,
handler.LOG_ENDARRAY, 4
};
EXPECT_EQ(sizeof(e) / sizeof(int), handler.LogCount);
for (size_t i = 0; i < handler.LogCount; ++i) {
EXPECT_EQ(e[i], handler.Logs[i]) << "i = " << i;
}
}
}
// Test iterative parsing on kParseErrorTermination.
struct HandlerTerminateAtStartObject : public IterativeParsingReaderHandler<> {
bool StartObject() { return false; }
};
struct HandlerTerminateAtStartArray : public IterativeParsingReaderHandler<> {
bool StartArray() { return false; }
};
struct HandlerTerminateAtEndObject : public IterativeParsingReaderHandler<> {
bool EndObject(SizeType) { return false; }
};
struct HandlerTerminateAtEndArray : public IterativeParsingReaderHandler<> {
bool EndArray(SizeType) { return false; }
};
TEST(Reader, IterativeParsing_ShortCircuit) {
{
HandlerTerminateAtStartObject handler;
Reader reader;
StringStream is("[1, {}]");
ParseResult r = reader.Parse<kParseIterativeFlag>(is, handler);
EXPECT_TRUE(reader.HasParseError());
EXPECT_EQ(kParseErrorTermination, r.Code());
EXPECT_EQ(4u, r.Offset());
}
{
HandlerTerminateAtStartArray handler;
Reader reader;
StringStream is("{\"a\": []}");
ParseResult r = reader.Parse<kParseIterativeFlag>(is, handler);
EXPECT_TRUE(reader.HasParseError());
EXPECT_EQ(kParseErrorTermination, r.Code());
EXPECT_EQ(6u, r.Offset());
}
{
HandlerTerminateAtEndObject handler;
Reader reader;
StringStream is("[1, {}]");
ParseResult r = reader.Parse<kParseIterativeFlag>(is, handler);
EXPECT_TRUE(reader.HasParseError());
EXPECT_EQ(kParseErrorTermination, r.Code());
EXPECT_EQ(5u, r.Offset());
}
{
HandlerTerminateAtEndArray handler;
Reader reader;
StringStream is("{\"a\": []}");
ParseResult r = reader.Parse<kParseIterativeFlag>(is, handler);
EXPECT_TRUE(reader.HasParseError());
EXPECT_EQ(kParseErrorTermination, r.Code());
EXPECT_EQ(7u, r.Offset());
}
}
#ifdef __GNUC__
RAPIDJSON_DIAG_POP
#endif