mirror of
https://gitlab.freedesktop.org/uchardet/uchardet.git
synced 2025-12-24 12:44:46 +08:00
1b4c62ac21
I disable only ISO-8859-15 which is similar to ISO-8859-1 for all Spanish letters. Unfortunately illegal codepoints are similar too. Difference should likely be done on symbols (like the euro symbol) but our current algorithm does nothing about this for charset comparison. Text from https://es.wikipedia.org/wiki/España
50 lines
1.3 KiB
CMake
50 lines
1.3 KiB
CMake
set(
|
|
UCHARDET_TEST_SOURCES
|
|
uchardet-tests.c
|
|
)
|
|
|
|
add_executable(
|
|
uchardet-tests
|
|
${UCHARDET_TEST_SOURCES}
|
|
)
|
|
|
|
target_link_libraries(
|
|
uchardet-tests
|
|
libuchardet
|
|
)
|
|
|
|
set_target_properties(
|
|
uchardet-tests
|
|
PROPERTIES
|
|
LINKER_LANGUAGE
|
|
C
|
|
OUTPUT_NAME
|
|
uchardet-tests
|
|
)
|
|
|
|
# Iterate through all langs.
|
|
file(GLOB dirs "[a-z][a-z]")
|
|
foreach(dir ${dirs})
|
|
get_filename_component(lang ${dir} NAME)
|
|
file(GLOB files "${dir}/*")
|
|
# Iterate through all files.
|
|
foreach(file ${files})
|
|
get_filename_component(charset ${file} NAME_WE)
|
|
# These are tests known to fail (not supported or not efficient
|
|
# enough). We will have to take a closer look and fix these, but
|
|
# there is no need to break the whole `make test` right now,
|
|
# which may make actual regressions harder to notice.
|
|
if ("${lang}:${charset}" STREQUAL "el:windows-1253" OR
|
|
"${lang}:${charset}" STREQUAL "ja:utf-16le" OR
|
|
"${lang}:${charset}" STREQUAL "ja:utf-16be" OR
|
|
"${lang}:${charset}" STREQUAL "es:iso-8859-15" OR
|
|
"${lang}:${charset}" STREQUAL "he:iso-8859-8")
|
|
message(STATUS "Skipping test ${lang}:${charset} (known broken)")
|
|
else()
|
|
add_test(NAME "${lang}:${charset}"
|
|
COMMAND uchardet-tests ${file})
|
|
endif()
|
|
endforeach()
|
|
endforeach()
|
|
|