| 1 | #ifndef NU_TOUPPER_H |
| 2 | #define NU_TOUPPER_H |
| 3 | |
| 4 | #include <stdint.h> |
| 5 | |
| 6 | #include <libnu/config.h> |
| 7 | #include <libnu/defines.h> |
| 8 | #include <libnu/strings.h> |
| 9 | #include <libnu/udb.h> |
| 10 | |
| 11 | #if defined (__cplusplus) || defined (c_plusplus) |
| 12 | extern "C" { |
| 13 | #endif |
| 14 | |
| 15 | /** |
| 16 | * @example folding.c |
| 17 | * @example special_casing.c |
| 18 | */ |
| 19 | |
| 20 | /** Synonim to nu_casemap_read. It is recommended to use |
| 21 | * nu_casemap_read instead. |
| 22 | */ |
| 23 | #define NU_CASEMAP_DECODING_FUNCTION NU_UDB_DECODING_FUNCTION |
| 24 | /** Read (decoding) function for use with transformation results of |
| 25 | * casemapping functions. E.g. nu_casemap_read(nu_tolower(0x0041)); |
| 26 | * will read first codepoint of 'A' transformed to lower case. |
| 27 | */ |
| 28 | #define nu_casemap_read (nu_udb_read) |
| 29 | |
| 30 | /** Casemap codepoint |
| 31 | * |
| 32 | * @ingroup transformations |
| 33 | */ |
| 34 | typedef nu_transformation_t nu_casemapping_t; |
| 35 | |
| 36 | #ifdef NU_WITH_TOUPPER |
| 37 | |
| 38 | /** Return uppercase value of codepoint. Uncoditional casemapping. |
| 39 | * |
| 40 | * @ingroup transformations |
| 41 | * @param codepoint unicode codepoint |
| 42 | * @return uppercase codepoint or 0 if mapping doesn't exist |
| 43 | */ |
| 44 | NU_EXPORT |
| 45 | const char* nu_toupper(uint32_t codepoint); |
| 46 | |
| 47 | /** Return uppercase value of codepoint. Context-sensitivity is not |
| 48 | * implemented internally, returned result is equal to calling nu_toupper() |
| 49 | * on corresponding codepoint. |
| 50 | * |
| 51 | * @ingroup transformations_internal |
| 52 | * @param encoded pointer to encoded string |
| 53 | * @param limit memory limit of encoded string or NU_UNLIMITED |
| 54 | * @param read read (decoding) function |
| 55 | * @param u (optional) codepoint which was (or wasn't) transformed |
| 56 | * @param transform output value of codepoint transformed into uppercase or 0 |
| 57 | * if mapping doesn't exist. Can't be NULL, supposed to be decoded with |
| 58 | * nu_casemap_read |
| 59 | * @param context not used |
| 60 | * @return pointer to the next codepoint in string |
| 61 | */ |
| 62 | NU_EXPORT |
| 63 | const char* _nu_toupper(const char *encoded, const char *limit, nu_read_iterator_t read, |
| 64 | uint32_t *u, const char **transform, |
| 65 | void *context); |
| 66 | |
| 67 | #endif /* NU_WITH_TOUPPER */ |
| 68 | |
| 69 | #ifdef NU_WITH_TOLOWER |
| 70 | |
| 71 | /** Return lowercase value of codepoint. Unconditional casemapping. |
| 72 | * |
| 73 | * @ingroup transformations |
| 74 | * @param codepoint unicode codepoint |
| 75 | * @return lowercase codepoint or 0 if mapping doesn't exist |
| 76 | */ |
| 77 | NU_EXPORT |
| 78 | const char* nu_tolower(uint32_t codepoint); |
| 79 | |
| 80 | /** Return lowercase value of codepoint. Will transform uppercase |
| 81 | * Sigma ('Σ') into final sigma ('ς') if it occurs at string boundary or |
| 82 | * followed by U+0000. Might require single read-ahead when |
| 83 | * encountering Sigma. |
| 84 | * |
| 85 | * @ingroup transformations_internal |
| 86 | * @param encoded pointer to encoded string |
| 87 | * @param limit memory limit of encoded string or NU_UNLIMITED |
| 88 | * @param read read (decoding) function |
| 89 | * @param u (optional) codepoint which was (or wasn't) transformed |
| 90 | * @param transform output value of codepoint transformed into lowercase or 0 |
| 91 | * if mapping doesn't exist. Can't be NULL, supposed to be decoded with |
| 92 | * nu_casemap_read |
| 93 | * @param context not used |
| 94 | * @return pointer to the next codepoint in string |
| 95 | */ |
| 96 | NU_EXPORT |
| 97 | const char* _nu_tolower(const char *encoded, const char *limit, nu_read_iterator_t read, |
| 98 | uint32_t *u, const char **transform, |
| 99 | void *context); |
| 100 | |
| 101 | #endif /* NU_WITH_TOLOWER */ |
| 102 | |
| 103 | #ifdef NU_WITH_TOFOLD |
| 104 | |
| 105 | /** Return value of codepoint with case differences eliminated |
| 106 | * |
| 107 | * @ingroup transformations |
| 108 | * @param codepoint unicode codepoint |
| 109 | * @return casefolded codepoint or 0 if mapping doesn't exist |
| 110 | */ |
| 111 | NU_EXPORT |
| 112 | const char* nu_tofold(uint32_t codepoint); |
| 113 | |
| 114 | /** Return value of codepoint with case differences eliminated. |
| 115 | * Context-sensitivity is not implemented internally, returned result is equal |
| 116 | * to calling nu_tofold() on corresponding codepoint. |
| 117 | * |
| 118 | * @ingroup transformations_internal |
| 119 | * @param encoded pointer to encoded string |
| 120 | * @param limit memory limit of encoded string or NU_UNLIMITED |
| 121 | * @param read read (decoding) function |
| 122 | * @param u (optional) codepoint which was (or wasn't) transformed |
| 123 | * @param transform output value of casefolded codepoint or 0 |
| 124 | * if mapping doesn't exist. Can't be NULL, supposed to be decoded with |
| 125 | * nu_casemap_read |
| 126 | * @param context not used |
| 127 | * @return pointer to the next codepoint in string |
| 128 | */ |
| 129 | NU_EXPORT |
| 130 | const char* _nu_tofold(const char *encoded, const char *limit, nu_read_iterator_t read, |
| 131 | uint32_t *u, const char **transform, |
| 132 | void *context); |
| 133 | |
| 134 | #endif /* NU_WITH_TOFOLD */ |
| 135 | |
| 136 | #if defined (__cplusplus) || defined (c_plusplus) |
| 137 | } |
| 138 | #endif |
| 139 | |
| 140 | #endif /* NU_TOUPPER_H */ |
| 141 | |