| 1 | #include "cpp/data_lang.h"
|
| 2 |
|
| 3 | #include <stdio.h>
|
| 4 |
|
| 5 | #include "_gen/core/value.asdl.h"
|
| 6 | #include "data_lang/j8_libc.h" // for comparison
|
| 7 | #include "data_lang/j8_test_lib.h"
|
| 8 | #include "vendor/greatest.h"
|
| 9 |
|
| 10 | TEST PartIsUtf8_test() {
|
| 11 | BigStr* s = StrFromC("hi");
|
| 12 |
|
| 13 | ASSERT(pyj8::PartIsUtf8(s, 0, 2));
|
| 14 |
|
| 15 | // empty string is trivially UTF-8
|
| 16 | ASSERT(pyj8::PartIsUtf8(s, 0, 0));
|
| 17 |
|
| 18 | BigStr* binary = StrFromC("h\xff");
|
| 19 | ASSERT(!pyj8::PartIsUtf8(binary, 0, len(binary)));
|
| 20 |
|
| 21 | // first byte is UTF-8
|
| 22 | ASSERT(pyj8::PartIsUtf8(binary, 0, 1));
|
| 23 | // second byte isn't
|
| 24 | ASSERT(!pyj8::PartIsUtf8(binary, 1, 2));
|
| 25 |
|
| 26 | PASS();
|
| 27 | }
|
| 28 |
|
| 29 | // TODO: remove duplication
|
| 30 | #define LOSSY_JSON (1 << 3)
|
| 31 |
|
| 32 | TEST WriteString_test() {
|
| 33 | auto buf = Alloc<mylib::BufWriter>();
|
| 34 |
|
| 35 | for (int i = 0; J8_TEST_CASES[i]; ++i) {
|
| 36 | const char* s = J8_TEST_CASES[i];
|
| 37 | BigStr* s2 = StrFromC(s);
|
| 38 |
|
| 39 | buf = Alloc<mylib::BufWriter>();
|
| 40 | pyj8::WriteString(s2, LOSSY_JSON, buf);
|
| 41 |
|
| 42 | BigStr* result = buf->getvalue();
|
| 43 | log("result = %s", result->data_);
|
| 44 |
|
| 45 | buf = Alloc<mylib::BufWriter>();
|
| 46 | pyj8::WriteString(s2, 0, buf);
|
| 47 |
|
| 48 | result = buf->getvalue();
|
| 49 | log("result = %s", result->data_);
|
| 50 | }
|
| 51 |
|
| 52 | PASS();
|
| 53 | }
|
| 54 |
|
| 55 | TEST compare_c_test() {
|
| 56 | // Compare two implementations
|
| 57 |
|
| 58 | auto buf = Alloc<mylib::BufWriter>();
|
| 59 |
|
| 60 | for (int i = 0; J8_TEST_CASES[i]; ++i) {
|
| 61 | const char* s = J8_TEST_CASES[i];
|
| 62 | int input_len = strlen(s);
|
| 63 | j8_buf_t in = {(unsigned char*)s, input_len};
|
| 64 |
|
| 65 | j8_buf_t c_result = {0};
|
| 66 | J8EncodeString(in, &c_result, 0);
|
| 67 |
|
| 68 | printf("c_result %s\n", c_result.data);
|
| 69 | printf("c_result.len %d\n", c_result.len);
|
| 70 |
|
| 71 | BigStr* s2 = StrFromC(s);
|
| 72 |
|
| 73 | buf = Alloc<mylib::BufWriter>();
|
| 74 | pyj8::WriteString(s2, LOSSY_JSON, buf);
|
| 75 |
|
| 76 | BigStr* cpp_result = buf->getvalue();
|
| 77 |
|
| 78 | // Equal lengths
|
| 79 | ASSERT_EQ_FMT(c_result.len, len(cpp_result), "%d");
|
| 80 | // Equal contents
|
| 81 | ASSERT(memcmp(c_result.data, cpp_result->data_, c_result.len) == 0);
|
| 82 |
|
| 83 | free(c_result.data);
|
| 84 |
|
| 85 | //
|
| 86 | // Encode again with J8 fallback
|
| 87 | //
|
| 88 |
|
| 89 | c_result = {0};
|
| 90 | J8EncodeString(in, &c_result, 1);
|
| 91 |
|
| 92 | printf("c_result %s\n", c_result.data);
|
| 93 | printf("c_result.len %d\n", c_result.len);
|
| 94 |
|
| 95 | buf = Alloc<mylib::BufWriter>();
|
| 96 | pyj8::WriteString(s2, 0, buf);
|
| 97 |
|
| 98 | cpp_result = buf->getvalue();
|
| 99 |
|
| 100 | // Equal lengths
|
| 101 | ASSERT_EQ_FMT(c_result.len, len(cpp_result), "%d");
|
| 102 | // Equal contents
|
| 103 | ASSERT(memcmp(c_result.data, cpp_result->data_, c_result.len) == 0);
|
| 104 |
|
| 105 | free(c_result.data);
|
| 106 |
|
| 107 | printf("\n");
|
| 108 | }
|
| 109 |
|
| 110 | PASS();
|
| 111 | }
|
| 112 |
|
| 113 | using value_asdl::value;
|
| 114 | using value_asdl::value_t;
|
| 115 |
|
| 116 | TEST heap_id_test() {
|
| 117 | value_t* val1 = Alloc<value::Str>(kEmptyString);
|
| 118 | value_t* val2 = Alloc<value::Str>(kEmptyString);
|
| 119 |
|
| 120 | int id1 = j8::HeapValueId(val1);
|
| 121 | int id2 = j8::HeapValueId(val2);
|
| 122 |
|
| 123 | log("id1 = %d, id2 = %d", id1, id2);
|
| 124 | ASSERT(id1 != id2);
|
| 125 |
|
| 126 | PASS();
|
| 127 | }
|
| 128 |
|
| 129 | TEST utf8_decode_one_test() {
|
| 130 | #define ASSERT_DECODE(codepoint, bytes_read, string, start) \
|
| 131 | do { \
|
| 132 | Tuple2<int, int> result = fastfunc::Utf8DecodeOne((string), (start)); \
|
| 133 | ASSERT_EQ(result.at0(), (codepoint)); \
|
| 134 | ASSERT_EQ(result.at1(), (bytes_read)); \
|
| 135 | } while (false)
|
| 136 |
|
| 137 | BigStr* s = StrFromC("h\xE2\xA1\x80\xC5\x81");
|
| 138 | ASSERT_DECODE('h', 1, s, 0);
|
| 139 | ASSERT_DECODE(0x2840, 3, s, 1);
|
| 140 | ASSERT_DECODE(0x141, 2, s, 4);
|
| 141 |
|
| 142 | // UTF8_ERR_OVERLONG = 1
|
| 143 | ASSERT_DECODE(-1, 2, StrFromC("\xC1\x81"), 0);
|
| 144 |
|
| 145 | // UTF8_ERR_SURROGATE = 2
|
| 146 | ASSERT_DECODE(-2, 3, StrFromC("\xED\xBF\x80"), 0);
|
| 147 |
|
| 148 | // UTF8_ERR_TOO_LARGE = 3
|
| 149 | ASSERT_DECODE(-3, 4, StrFromC("\xF4\xA0\x80\x80"), 0);
|
| 150 |
|
| 151 | // UTF8_ERR_BAD_ENCODING = 4
|
| 152 | ASSERT_DECODE(-4, 2, StrFromC("\xC2\xFF"), 0);
|
| 153 |
|
| 154 | // UTF8_ERR_TRUNCATED_BYTES = 5
|
| 155 | ASSERT_DECODE(-5, 1, StrFromC("\xC2"), 0);
|
| 156 |
|
| 157 | PASS();
|
| 158 | #undef ASSERT_DECODE
|
| 159 | }
|
| 160 |
|
| 161 | GREATEST_MAIN_DEFS();
|
| 162 |
|
| 163 | int main(int argc, char** argv) {
|
| 164 | gHeap.Init();
|
| 165 |
|
| 166 | GREATEST_MAIN_BEGIN();
|
| 167 |
|
| 168 | RUN_TEST(PartIsUtf8_test);
|
| 169 | RUN_TEST(WriteString_test);
|
| 170 | RUN_TEST(compare_c_test);
|
| 171 | RUN_TEST(heap_id_test);
|
| 172 | RUN_TEST(utf8_decode_one_test);
|
| 173 |
|
| 174 | gHeap.CleanProcessExit();
|
| 175 |
|
| 176 | GREATEST_MAIN_END();
|
| 177 | return 0;
|
| 178 | }
|