ICU-64260.0.1.tar.gz

[apple/icu.git] / icuSources / common / ustr_imp.h

diff --git a/icuSources/common/ustr_imp.h b/icuSources/common/ustr_imp.h
index 3b162e8ab27d5bf315bbd3329f91fc7b0333acfe..34ca2f7428283685565b7be1fc85980ed804904f 100644 (file)
--- a/icuSources/common/ustr_imp.h
+++ b/icuSources/common/ustr_imp.h
@@ -1,10 +1,12 @@
+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html

 /*  
 **********************************************************************
 /*  
 **********************************************************************

-*   Copyright (C) 1999-2004, International Business Machines
+*   Copyright (C) 1999-2015, International Business Machines

 *   Corporation and others.  All Rights Reserved.
 **********************************************************************
 *   file name:  ustr_imp.h
 *   Corporation and others.  All Rights Reserved.
 **********************************************************************
 *   file name:  ustr_imp.h

-*   encoding:   US-ASCII
+*   encoding:   UTF-8

 *   tab size:   8 (not used)
 *   indentation:4
 *
 *   tab size:   8 (not used)
 *   indentation:4
 *


@@ -16,133 +18,33 @@
 #define __USTR_IMP_H__
 
 #include "unicode/utypes.h"
 #define __USTR_IMP_H__
 
 #include "unicode/utypes.h"

-#include "unicode/uiter.h"
-#include "ucase.h"
+#include "unicode/utf8.h"

 
 

-/** Simple declaration for u_strToTitle() to avoid including unicode/ubrk.h. */
-#ifndef UBRK_TYPEDEF_UBREAK_ITERATOR
-#   define UBRK_TYPEDEF_UBREAK_ITERATOR
-    typedef void UBreakIterator;
-#endif
+/**
+ * Internal option for unorm_cmpEquivFold() for strncmp style.
+ * If set, checks for both string length and terminating NUL.
+ */
+#define _STRNCMP_STYLE 0x1000

 
 /**
  * Compare two strings in code point order or code unit order.
  * Works in strcmp style (both lengths -1),
  * strncmp style (lengths equal and >=0, flag TRUE),
  * and memcmp/UnicodeString style (at least one length >=0).
 
 /**
  * Compare two strings in code point order or code unit order.
  * Works in strcmp style (both lengths -1),
  * strncmp style (lengths equal and >=0, flag TRUE),
  * and memcmp/UnicodeString style (at least one length >=0).

- * @internal

  */
  */

-U_CAPI int32_t U_EXPORT2
+U_CFUNC int32_t U_EXPORT2

 uprv_strCompare(const UChar *s1, int32_t length1,
                 const UChar *s2, int32_t length2,
                 UBool strncmpStyle, UBool codePointOrder);
 
 uprv_strCompare(const UChar *s1, int32_t length1,
                 const UChar *s2, int32_t length2,
                 UBool strncmpStyle, UBool codePointOrder);
 

-/**
- * Internal API, used by u_strcasecmp() etc.
- * Compare strings case-insensitively,
- * in code point order or code unit order.
- * @internal
- */
-U_CFUNC int32_t
-u_strcmpFold(const UChar *s1, int32_t length1,
-             const UChar *s2, int32_t length2,
-             uint32_t options,
-             UErrorCode *pErrorCode);
-
-/**
- * Are the Unicode properties loaded?
- * This must be used before internal functions are called that do
- * not perform this check.
- * Generate a debug assertion failure if data is not loaded, to flag the fact
- *   that u_init() wasn't called first, before trying to access character properties.
- * @internal
- */
-U_CFUNC UBool
-uprv_haveProperties(UErrorCode *pErrorCode);
-
-/**
-  * Load the Unicode property data.
-  * Intended primarily for use from u_init().
-  * Has no effect if property data is already loaded.
-  * NOT thread safe.
-  * @internal
-  */
-U_CFUNC int8_t
-uprv_loadPropsData(UErrorCode *errorCode);
-
-/**
- * Type of a function that may be passed to the internal case mapping functions
- * or similar for growing the destination buffer.
- * @internal
- */
-typedef UBool U_CALLCONV
-UGrowBuffer(void *context,      /* opaque pointer for this function */
-            UChar **pBuffer,    /* in/out destination buffer pointer */
-            int32_t *pCapacity, /* in/out buffer capacity in numbers of UChars */
-            int32_t reqCapacity,/* requested capacity */
-            int32_t length);    /* number of UChars to be copied to new buffer */
-
-/**
- * Default implementation of UGrowBuffer.
- * Takes a static buffer as context, allocates a new buffer,
- * and releases the old one if it is not the same as the one passed as context.
- * @internal
- */
-U_CAPI UBool /* U_CALLCONV U_EXPORT2 */
-u_growBufferFromStatic(void *context,
-                       UChar **pBuffer, int32_t *pCapacity, int32_t reqCapacity,
-                       int32_t length);
-
-/*
- * Internal string casing functions implementing
- * ustring.h/ustrcase.c and UnicodeString case mapping functions.
- */
-
-/**
- * @internal
- */
-U_CFUNC int32_t
-ustr_toLower(UCaseProps *csp,
-             UChar *dest, int32_t destCapacity,
-             const UChar *src, int32_t srcLength,
-             const char *locale,
-             UErrorCode *pErrorCode);
-
-/**
- * @internal
- */
-U_CFUNC int32_t
-ustr_toUpper(UCaseProps *csp,
-             UChar *dest, int32_t destCapacity,
-             const UChar *src, int32_t srcLength,
-             const char *locale,
-             UErrorCode *pErrorCode);
-
-#if !UCONFIG_NO_BREAK_ITERATION
-
-/**
- * @internal
- */
-U_CFUNC int32_t
-ustr_toTitle(UCaseProps *csp,
-             UChar *dest, int32_t destCapacity,
-             const UChar *src, int32_t srcLength,
-             UBreakIterator *titleIter,
-             const char *locale,
-             UErrorCode *pErrorCode);
+U_CAPI int32_t U_EXPORT2 
+ustr_hashUCharsN(const UChar *str, int32_t length);

 
 

-#endif
+U_CAPI int32_t U_EXPORT2 
+ustr_hashCharsN(const char *str, int32_t length);

 
 

-/**
- * Internal case folding function.
- * @internal
- */
-U_CFUNC int32_t
-ustr_foldCase(UCaseProps *csp,
-              UChar *dest, int32_t destCapacity,
-              const UChar *src, int32_t srcLength,
-              uint32_t options,
-              UErrorCode *pErrorCode);
+U_CAPI int32_t U_EXPORT2
+ustr_hashICharsN(const char *str, int32_t length);

 
 /**
  * NUL-terminate a UChar * string if possible.
 
 /**
  * NUL-terminate a UChar * string if possible.


@@ -155,7 +57,6 @@ ustr_foldCase(UCaseProps *csp,
  * @param length Number of UChars that were (to be) written to dest.
  * @param pErrorCode ICU error code.
  * @return length
  * @param length Number of UChars that were (to be) written to dest.
  * @param pErrorCode ICU error code.
  * @return length

- * @internal

  */
 U_CAPI int32_t U_EXPORT2
 u_terminateUChars(UChar *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode);
  */
 U_CAPI int32_t U_EXPORT2
 u_terminateUChars(UChar *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode);


@@ -181,4 +82,75 @@ u_terminateUChar32s(UChar32 *dest, int32_t destCapacity, int32_t length, UErrorC
 U_CAPI int32_t U_EXPORT2
 u_terminateWChars(wchar_t *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode);
 
 U_CAPI int32_t U_EXPORT2
 u_terminateWChars(wchar_t *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode);
 

+/**
+ * Counts the bytes of any whole valid sequence for a UTF-8 lead byte.
+ * Returns 1 for ASCII 0..0x7f.
+ * Returns 0 for 0x80..0xc1 as well as for 0xf5..0xff.
+ * leadByte might be evaluated multiple times.
+ *
+ * @param leadByte The first byte of a UTF-8 sequence. Must be 0..0xff.
+ * @return 0..4
+ */
+#define U8_COUNT_BYTES(leadByte) \
+    (U8_IS_SINGLE(leadByte) ? 1 : U8_COUNT_BYTES_NON_ASCII(leadByte))
+
+/**
+ * Counts the bytes of any whole valid sequence for a UTF-8 lead byte.
+ * Returns 0 for 0x00..0xc1 as well as for 0xf5..0xff.
+ * leadByte might be evaluated multiple times.
+ *
+ * @param leadByte The first byte of a UTF-8 sequence. Must be 0..0xff.
+ * @return 0 or 2..4
+ */
+#define U8_COUNT_BYTES_NON_ASCII(leadByte) \
+    (U8_IS_LEAD(leadByte) ? ((uint8_t)(leadByte)>=0xe0)+((uint8_t)(leadByte)>=0xf0)+2 : 0)
+
+#ifdef __cplusplus
+
+U_NAMESPACE_BEGIN
+
+class UTF8 {
+public:
+    UTF8() = delete;  // all static
+
+    /**
+     * Is t a valid UTF-8 trail byte?
+     *
+     * @param prev Must be the preceding lead byte if i==1 and length>=3;
+     *             otherwise ignored.
+     * @param t The i-th byte following the lead byte.
+     * @param i The index (1..3) of byte t in the byte sequence. 0<i<length
+     * @param length The length (2..4) of the byte sequence according to the lead byte.
+     * @return TRUE if t is a valid trail byte in this context.
+     */
+    static inline UBool isValidTrail(int32_t prev, uint8_t t, int32_t i, int32_t length) {
+        // The first trail byte after a 3- or 4-byte lead byte
+        // needs to be validated together with its lead byte.
+        if (length <= 2 || i > 1) {
+            return U8_IS_TRAIL(t);
+        } else if (length == 3) {
+            return U8_IS_VALID_LEAD3_AND_T1(prev, t);
+        } else {  // length == 4
+            return U8_IS_VALID_LEAD4_AND_T1(prev, t);
+        }
+    }
+};
+
+U_NAMESPACE_END
+
+#endif  // __cplusplus
+
+/**
+ * Check whether a code point has the emoji property.
+ * (implemented in uchar.cpp as reduced version of u_hasBinaryProperty)
+ *
+ * @param c Code point to test.
+ * @return TRUE or FALSE according to whether c has the emoji property.
+ *
+ * @internal Apple only
+ */
+U_CAPI UBool U_EXPORT2
+u_isEmoji(UChar32 c);
+
+

 #endif
 #endif