Eremey Valetov
38c0898bc2
New library (uc2_preprocess.h / uc2_preprocess.c) for Phase 4:
BCJ (Branch/Call/Jump) filter:
- E8/E9 x86 address normalization (relative → absolute)
- Makes calls to the same function from different locations produce
identical byte sequences, improving LZ77 matching
- Round-trip verified; address normalization confirmed
BWT (Burrows-Wheeler Transform):
- Suffix-array-based forward transform
- LF-mapping inverse with reverse reconstruction
- Groups similar contexts for better entropy coding
- Round-trip verified for text ("banana") and binary data
Delta filter:
- Byte-wise delta encoding with configurable stride
- Stride 1 for sequential data, stride 2+ for interleaved channels
- Constant-delta sequences (arithmetic progressions) reduce to
repeated single values
Content detection:
- Automatic content type identification (text/x86/structured/binary)
- MZ/PE and ELF header recognition for x86
- Printable ASCII ratio for text detection
11 unit tests covering all filters and detection.
107 lines
4.2 KiB
CMake
107 lines
4.2 KiB
CMake
# UC2 tests
|
|
|
|
add_executable(test_identify src/test_identify.c)
|
|
target_link_libraries(test_identify PRIVATE uc2)
|
|
target_include_directories(test_identify PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_identify PRIVATE c_std_99)
|
|
add_test(NAME identify COMMAND test_identify)
|
|
|
|
add_executable(test_extract src/test_extract.c)
|
|
target_link_libraries(test_extract PRIVATE uc2)
|
|
target_include_directories(test_extract PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_extract PRIVATE c_std_99)
|
|
add_test(NAME extract COMMAND test_extract
|
|
"${CMAKE_CURRENT_SOURCE_DIR}/archives"
|
|
"${CMAKE_CURRENT_SOURCE_DIR}/corpus"
|
|
)
|
|
|
|
add_executable(test_roundtrip src/test_roundtrip.c)
|
|
target_link_libraries(test_roundtrip PRIVATE uc2)
|
|
target_include_directories(test_roundtrip PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_roundtrip PRIVATE c_std_99)
|
|
add_test(NAME roundtrip COMMAND test_roundtrip)
|
|
|
|
# CLI create/extract round-trip test
|
|
add_test(NAME cli_create
|
|
COMMAND ${CMAKE_COMMAND}
|
|
-DUC2_CLI=$<TARGET_FILE:uc2-cli>
|
|
-DTEST_DIR=${CMAKE_CURRENT_BINARY_DIR}/cli_test
|
|
-P ${CMAKE_CURRENT_SOURCE_DIR}/test_cli_create.cmake
|
|
)
|
|
|
|
# CLI master-block deduplication round-trip test
|
|
add_test(NAME cli_master
|
|
COMMAND ${CMAKE_COMMAND}
|
|
-DUC2_CLI=$<TARGET_FILE:uc2-cli>
|
|
-DTEST_DIR=${CMAKE_CURRENT_BINARY_DIR}/cli_master_test
|
|
-P ${CMAKE_CURRENT_SOURCE_DIR}/test_cli_master.cmake
|
|
)
|
|
|
|
# CLI directory archival round-trip test
|
|
add_test(NAME cli_dirs
|
|
COMMAND ${CMAKE_COMMAND}
|
|
-DUC2_CLI=$<TARGET_FILE:uc2-cli>
|
|
-DTEST_DIR=${CMAKE_CURRENT_BINARY_DIR}/cli_dirs_test
|
|
-P ${CMAKE_CURRENT_SOURCE_DIR}/test_cli_dirs.cmake
|
|
)
|
|
|
|
add_executable(test_cdc src/test_cdc.c)
|
|
target_link_libraries(test_cdc PRIVATE uc2)
|
|
target_include_directories(test_cdc PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_cdc PRIVATE c_std_99)
|
|
add_test(NAME cdc COMMAND test_cdc)
|
|
|
|
add_executable(test_merkle src/test_merkle.c)
|
|
target_link_libraries(test_merkle PRIVATE uc2 m)
|
|
target_include_directories(test_merkle PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_merkle PRIVATE c_std_99)
|
|
add_test(NAME merkle COMMAND test_merkle)
|
|
|
|
add_executable(test_blockstore src/test_blockstore.c)
|
|
target_link_libraries(test_blockstore PRIVATE uc2)
|
|
target_include_directories(test_blockstore PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_blockstore PRIVATE c_std_99)
|
|
add_test(NAME blockstore COMMAND test_blockstore)
|
|
|
|
add_executable(test_simhash src/test_simhash.c)
|
|
target_link_libraries(test_simhash PRIVATE uc2)
|
|
target_include_directories(test_simhash PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_simhash PRIVATE c_std_99)
|
|
add_test(NAME simhash COMMAND test_simhash)
|
|
|
|
add_executable(test_delta src/test_delta.c)
|
|
target_link_libraries(test_delta PRIVATE uc2)
|
|
target_include_directories(test_delta PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_delta PRIVATE c_std_99)
|
|
add_test(NAME delta COMMAND test_delta)
|
|
|
|
add_executable(test_rans src/test_rans.c)
|
|
target_link_libraries(test_rans PRIVATE uc2 m)
|
|
target_include_directories(test_rans PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_rans PRIVATE c_std_99)
|
|
add_test(NAME rans COMMAND test_rans)
|
|
|
|
add_executable(test_dict src/test_dict.c)
|
|
target_link_libraries(test_dict PRIVATE uc2)
|
|
target_include_directories(test_dict PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_dict PRIVATE c_std_99)
|
|
add_test(NAME dict COMMAND test_dict)
|
|
|
|
add_executable(test_preprocess src/test_preprocess.c)
|
|
target_link_libraries(test_preprocess PRIVATE uc2)
|
|
target_include_directories(test_preprocess PRIVATE "${PROJECT_BINARY_DIR}/lib")
|
|
target_compile_features(test_preprocess PRIVATE c_std_99)
|
|
add_test(NAME preprocess COMMAND test_preprocess)
|
|
|
|
# Cross-tool round-trip: UC2 v3 <-> original uc2pro.exe via DOSBox-X
|
|
add_test(NAME roundtrip_dosbox
|
|
COMMAND bash ${CMAKE_CURRENT_SOURCE_DIR}/scripts/roundtrip_dosbox.sh
|
|
$<TARGET_FILE:uc2-cli>
|
|
${CMAKE_CURRENT_SOURCE_DIR}/../original/UC2_source/uc2pro.exe
|
|
${CMAKE_CURRENT_SOURCE_DIR}/corpus
|
|
)
|
|
set_tests_properties(roundtrip_dosbox PROPERTIES
|
|
LABELS "dosbox"
|
|
TIMEOUT 1200
|
|
)
|