Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Initial support for Big Endian #75

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions CMakeLists.txt
Original file line number Diff line number Diff line change
Expand Up @@ -90,20 +90,27 @@ if(STRINGZILLA_INSTALL)
endif()

if(${STRINGZILLA_BUILD_TEST} OR ${STRINGZILLA_BUILD_BENCHMARK})
find_package(GTest REQUIRED)
include(GoogleTest)

add_executable(stringzilla_test scripts/test.cpp)
add_executable(stringzilla_unittest scripts/ut_stringzilla.cpp)
target_link_libraries(stringzilla_unittest GTest::GTest GTest::Main)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -g")
set(CMAKE_CXX_FLAGS
"${CMAKE_CXX_FLAGS} -O3 -flto -march=native -finline-functions -funroll-loops"
)

target_include_directories(stringzilla_test PRIVATE stringzilla)
target_include_directories(stringzilla_unittest PRIVATE stringzilla)
set_target_properties(stringzilla_test PROPERTIES RUNTIME_OUTPUT_DIRECTORY
${CMAKE_BINARY_DIR})

if(${CMAKE_VERSION} VERSION_EQUAL 3.13 OR ${CMAKE_VERSION} VERSION_GREATER
3.13)
include(CTest)
enable_testing()
add_test(NAME stringzilla_unittest COMMAND stringzilla_unittest)
add_test(NAME stringzilla_test COMMAND stringzilla_test)
endif()
endif()
9 changes: 8 additions & 1 deletion scripts/test.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,11 @@ int hybrid_sort_c_compare_uint32_t(const void *a, const void *b) {
return (int_a < int_b) ? -1 : (int_a > int_b);
}

#ifdef __APPLE__
int hybrid_sort_c_compare_strings(void *arg, const void *a, const void *b) {
#else
int hybrid_sort_c_compare_strings(const void *arg, const void *a, void *b) {
#endif
sz_sequence_t *sequence = (sz_sequence_t *)arg;
sz_size_t idx_a = *(sz_size_t *)a;
sz_size_t idx_b = *(sz_size_t *)b;
Expand Down Expand Up @@ -123,8 +127,11 @@ sz_size_t hybrid_sort_c(sz_sequence_t *sequence) {
}

// Sort the full strings.
#ifdef __APPLE__
qsort_r(sequence->order, sequence->count, sizeof(sz_size_t), sequence, hybrid_sort_c_compare_strings);

#else
qsort_r(sequence->order, sequence->count, sizeof(sz_size_t), hybrid_sort_c_compare_strings, sequence);
#endif
return sequence->count;
}

Expand Down
93 changes: 93 additions & 0 deletions scripts/ut_stringzilla.cpp
Original file line number Diff line number Diff line change
@@ -0,0 +1,93 @@
#include <gtest/gtest.h>
#include <string>
#include <stringzilla.h>

TEST(StringzillaSWAR, test_sz_count_char_swar) {
std::string haystack = "daddadddaddddaddddda";
std::string needle1 = "a";
std::string needle2 = "d";
ASSERT_EQ(sz_count_char_swar(haystack.data(), haystack.size(), needle1.data()), 5);
ASSERT_EQ(sz_count_char_swar(haystack.data(), haystack.size(), needle2.data()), 15);
}

TEST(StringzillaSWAR, test_sz_find_Xchar_swar) {
std::string haystack = "myneedleinhaystack";

ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "m") - haystack.data(), 0);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "y") - haystack.data(), 1);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "n") - haystack.data(), 2);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "e") - haystack.data(), 3);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "d") - haystack.data(), 5);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "l") - haystack.data(), 6);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "i") - haystack.data(), 8);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "h") - haystack.data(), 10);
ASSERT_EQ(sz_find_1char_swar(haystack.data(), haystack.size(), "k") - haystack.data(), 17);

ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "my") - haystack.data(), 0);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "yn") - haystack.data(), 1);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "ne") - haystack.data(), 2);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "ee") - haystack.data(), 3);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "ed") - haystack.data(), 4);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "dl") - haystack.data(), 5);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "le") - haystack.data(), 6);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "ei") - haystack.data(), 7);
ASSERT_EQ(sz_find_2char_swar(haystack.data(), haystack.size(), "in") - haystack.data(), 8);


ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "myn") - haystack.data(), 0);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "yne") - haystack.data(), 1);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "nee") - haystack.data(), 2);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "eed") - haystack.data(), 3);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "edl") - haystack.data(), 4);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "dle") - haystack.data(), 5);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "lei") - haystack.data(), 6);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "ein") - haystack.data(), 7);
ASSERT_EQ(sz_find_3char_swar(haystack.data(), haystack.size(), "inh") - haystack.data(), 8);

ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "myne") - haystack.data(), 0);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "ynee") - haystack.data(), 1);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "need") - haystack.data(), 2);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "eedl") - haystack.data(), 3);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "edle") - haystack.data(), 4);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "dlei") - haystack.data(), 5);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "lein") - haystack.data(), 6);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "einh") - haystack.data(), 7);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "inha") - haystack.data(), 8);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "nhay") - haystack.data(), 9);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "hays") - haystack.data(), 10);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "ayst") - haystack.data(), 11);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "ysta") - haystack.data(), 12);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "stac") - haystack.data(), 13);
ASSERT_EQ(sz_find_4char_swar(haystack.data(), haystack.size(), "tack") - haystack.data(), 14);
}

TEST(StringzillaSWAR, test_sz_rfind_1char_swar) {
std::string haystack = "myneedleinhaystack";

ASSERT_EQ(sz_rfind_1char_swar(haystack.data(), haystack.size(), "m") - haystack.data(), 0);
ASSERT_EQ(sz_rfind_1char_swar(haystack.data(), haystack.size(), "d") - haystack.data(), 5);
ASSERT_EQ(sz_rfind_1char_swar(haystack.data(), haystack.size(), "l") - haystack.data(), 6);
ASSERT_EQ(sz_rfind_1char_swar(haystack.data(), haystack.size(), "e") - haystack.data(), 7);
ASSERT_EQ(sz_rfind_1char_swar(haystack.data(), haystack.size(), "h") - haystack.data(), 10);
ASSERT_EQ(sz_rfind_1char_swar(haystack.data(), haystack.size(), "a") - haystack.data(), 15);

std::string alphabet = "abcdefghijklmnopqrstuvwxyz";

for (int i = 24; i < alphabet.size(); i++) {
std::string needle = std::string(1, alphabet[i]);
EXPECT_EQ(sz_rfind_1char_swar(alphabet.data(), alphabet.size(), needle.data()) - alphabet.data(), i);
}
}

TEST(StringzillaSWAR, test_sz_find_substring_swar) {
std::string haystack = "myneedleinhaystack";
std::string needle = "needle";
sz_string_start_t ptr = sz_find_substring_swar(haystack.data(), haystack.size(), needle.data(), needle.size());
ASSERT_TRUE(ptr);
ASSERT_EQ(std::string(ptr, 6), "needle");
}

int main(int argc, char **argv) {
::testing::InitGoogleTest(&argc, argv);
return RUN_ALL_TESTS();
}
Loading