diff options
| author | Timothy Flynn <trflynn89@pm.me> | 2023-01-17 08:34:38 -0500 |
|---|---|---|
| committer | Linus Groh <mail@linusgroh.de> | 2023-01-18 14:43:40 +0000 |
| commit | 8f2589b3b07470ebe5773c6e719ede169d894a93 (patch) | |
| tree | b044d9ab374db0ff09d6357bca0c978822815ade /Meta/CMake | |
| parent | 9226cf7272c57d57379a9bfdd1a9a6c6426818be (diff) | |
| download | serenity-8f2589b3b07470ebe5773c6e719ede169d894a93.zip | |
LibUnicode: Parse and generate case folding code point data
Case folding rules have a similar mapping style as special casing rules,
where one code point may map to zero or more case folding rules. These
will be used for case-insensitive string comparisons. To see how case
folding can differ from other casing rules, consider "ß" (U+00DF):
>>> "ß".lower()
'ß'
>>> "ß".upper()
'SS'
>>> "ß".title()
'Ss'
>>> "ß".casefold()
'ss'
Diffstat (limited to 'Meta/CMake')
| -rw-r--r-- | Meta/CMake/unicode_data.cmake | 6 |
1 files changed, 5 insertions, 1 deletions
diff --git a/Meta/CMake/unicode_data.cmake b/Meta/CMake/unicode_data.cmake index 6914b99a11..dc97c6362b 100644 --- a/Meta/CMake/unicode_data.cmake +++ b/Meta/CMake/unicode_data.cmake @@ -13,6 +13,9 @@ set(UNICODE_DATA_PATH "${UCD_PATH}/${UNICODE_DATA_SOURCE}") set(SPECIAL_CASING_SOURCE "SpecialCasing.txt") set(SPECIAL_CASING_PATH "${UCD_PATH}/${SPECIAL_CASING_SOURCE}") +set(CASE_FOLDING_SOURCE "CaseFolding.txt") +set(CASE_FOLDING_PATH "${UCD_PATH}/${CASE_FOLDING_SOURCE}") + set(DERIVED_GENERAL_CATEGORY_SOURCE "extracted/DerivedGeneralCategory.txt") set(DERIVED_GENERAL_CATEGORY_PATH "${UCD_PATH}/${DERIVED_GENERAL_CATEGORY_SOURCE}") @@ -71,6 +74,7 @@ if (ENABLE_UNICODE_DATABASE_DOWNLOAD) download_file("${UCD_ZIP_URL}" "${UCD_ZIP_PATH}") extract_path("${UCD_PATH}" "${UCD_ZIP_PATH}" "${UNICODE_DATA_SOURCE}" "${UNICODE_DATA_PATH}") extract_path("${UCD_PATH}" "${UCD_ZIP_PATH}" "${SPECIAL_CASING_SOURCE}" "${SPECIAL_CASING_PATH}") + extract_path("${UCD_PATH}" "${UCD_ZIP_PATH}" "${CASE_FOLDING_SOURCE}" "${CASE_FOLDING_PATH}") extract_path("${UCD_PATH}" "${UCD_ZIP_PATH}" "${DERIVED_GENERAL_CATEGORY_SOURCE}" "${DERIVED_GENERAL_CATEGORY_PATH}") extract_path("${UCD_PATH}" "${UCD_ZIP_PATH}" "${PROP_LIST_SOURCE}" "${PROP_LIST_PATH}") extract_path("${UCD_PATH}" "${UCD_ZIP_PATH}" "${DERIVED_CORE_PROP_SOURCE}" "${DERIVED_CORE_PROP_PATH}") @@ -105,7 +109,7 @@ if (ENABLE_UNICODE_DATABASE_DOWNLOAD) "${UCD_VERSION_FILE}" "${UNICODE_DATA_HEADER}" "${UNICODE_DATA_IMPLEMENTATION}" - arguments -u "${UNICODE_DATA_PATH}" -s "${SPECIAL_CASING_PATH}" -g "${DERIVED_GENERAL_CATEGORY_PATH}" -p "${PROP_LIST_PATH}" -d "${DERIVED_CORE_PROP_PATH}" -b "${DERIVED_BINARY_PROP_PATH}" -a "${PROP_ALIAS_PATH}" -v "${PROP_VALUE_ALIAS_PATH}" -r "${SCRIPTS_PATH}" -x "${SCRIPT_EXTENSIONS_PATH}" -k "${BLOCKS_PATH}" -e "${EMOJI_DATA_PATH}" -m "${NAME_ALIAS_PATH}" -n "${NORM_PROPS_PATH}" -f "${GRAPHEME_BREAK_PROP_PATH}" -w "${WORD_BREAK_PROP_PATH}" -i "${SENTENCE_BREAK_PROP_PATH}" + arguments -u "${UNICODE_DATA_PATH}" -s "${SPECIAL_CASING_PATH}" -o "${CASE_FOLDING_PATH}" -g "${DERIVED_GENERAL_CATEGORY_PATH}" -p "${PROP_LIST_PATH}" -d "${DERIVED_CORE_PROP_PATH}" -b "${DERIVED_BINARY_PROP_PATH}" -a "${PROP_ALIAS_PATH}" -v "${PROP_VALUE_ALIAS_PATH}" -r "${SCRIPTS_PATH}" -x "${SCRIPT_EXTENSIONS_PATH}" -k "${BLOCKS_PATH}" -e "${EMOJI_DATA_PATH}" -m "${NAME_ALIAS_PATH}" -n "${NORM_PROPS_PATH}" -f "${GRAPHEME_BREAK_PROP_PATH}" -w "${WORD_BREAK_PROP_PATH}" -i "${SENTENCE_BREAK_PROP_PATH}" ) invoke_generator( "EmojiData" |