Add more unicode APIs

This commit is contained in:
2026-04-07 07:59:46 +10:00
parent 6d792ee4c0
commit cb3e856a6e
5 changed files with 195 additions and 1 deletions

View File

@@ -112,6 +112,23 @@ namespace ANSCENTER {
// Input: ASCII string with \uXXXX escapes (e.g., "\\u6c5f\\u6771 599")
// Output: UTF-8 encoded Unicode string.
static std::string ConvertUnicodeEscapesToUTF8(const std::string& escapedStr);
// Convert a UTF-8 string to Unicode escape sequences (\uXXXX).
// ASCII chars (0x20-0x7E) are preserved, all others become \uXXXX.
// Useful for encoding Unicode text into JSON-safe ASCII.
static std::string ConvertUTF8ToUnicodeEscapes(const std::string& utf8Str);
// Double-escape \uXXXX sequences: \u1ee7 becomes \\u1ee7.
// Useful when the string will be embedded in JSON and the literal \uXXXX must survive parsing.
static std::string DoubleEscapeUnicode(const std::string& str);
// Convert UTF-8 to double-escaped Unicode: "Thủy" becomes "Th\\u1ee7y".
// Combines ConvertUTF8ToUnicodeEscapes + DoubleEscapeUnicode in one call.
static std::string ConvertUTF8ToDoubleEscapedUnicode(const std::string& utf8Str);
// Unescape double-escaped Unicode: "Th\\u1ee7y" becomes "Th\u1ee7y".
// Reverses DoubleEscapeUnicode so other languages can parse \uXXXX correctly.
static std::string UnescapeDoubleEscapedUnicode(const std::string& str);
};
// Connection bundle for pool
struct S3Connection {
@@ -259,6 +276,10 @@ extern "C" ANSULT_API int ANSDecodeJsonUnicodeToUTF16LE(const char* escapedStr,
extern "C" ANSULT_API int ANSConvertUTF16LEToUTF8(const unsigned char* utf16leBytes, int byteLen, LStrHandle result);
extern "C" ANSULT_API int ANSConvertUTF16LEToUnicodeEscapes(const unsigned char* utf16leBytes, int byteLen, LStrHandle result);
extern "C" ANSULT_API int ANSConvertUnicodeEscapesToUTF8(const char* escapedStr, LStrHandle result);
extern "C" ANSULT_API int ANSConvertUTF8ToUnicodeEscapes(const char* utf8Str, LStrHandle result);
extern "C" ANSULT_API int ANSDoubleEscapeUnicode(const char* str, LStrHandle result);
extern "C" ANSULT_API int ANSConvertUTF8ToDoubleEscapedUnicode(const char* utf8Str, LStrHandle result);
extern "C" ANSULT_API int ANSUnescapeDoubleEscapedUnicode(const char* str, LStrHandle result);
// AWS S3 class