]> git.saurik.com Git - apple/icu.git/blobdiff - icuSources/common/unistr_case.cpp
ICU-6.2.4.tar.gz
[apple/icu.git] / icuSources / common / unistr_case.cpp
diff --git a/icuSources/common/unistr_case.cpp b/icuSources/common/unistr_case.cpp
new file mode 100644 (file)
index 0000000..dd4a081
--- /dev/null
@@ -0,0 +1,244 @@
+/*
+*******************************************************************************
+*
+*   Copyright (C) 1999-2004, International Business Machines
+*   Corporation and others.  All Rights Reserved.
+*
+*******************************************************************************
+*   file name:  unistr_case.cpp
+*   encoding:   US-ASCII
+*   tab size:   8 (not used)
+*   indentation:2
+*
+*   created on: 2004aug19
+*   created by: Markus W. Scherer
+*
+*   Case-mapping functions moved here from unistr.cpp
+*/
+
+#include "unicode/utypes.h"
+#include "unicode/putil.h"
+#include "unicode/locid.h"
+#include "cstring.h"
+#include "cmemory.h"
+#include "unicode/ustring.h"
+#include "unicode/unistr.h"
+#include "unicode/uchar.h"
+#include "unicode/ubrk.h"
+#include "ustr_imp.h"
+#include "unormimp.h"
+
+U_NAMESPACE_BEGIN
+
+//========================================
+// Read-only implementation
+//========================================
+
+int8_t
+UnicodeString::doCaseCompare(int32_t start,
+                             int32_t length,
+                             const UChar *srcChars,
+                             int32_t srcStart,
+                             int32_t srcLength,
+                             uint32_t options) const
+{
+  // compare illegal string values
+  // treat const UChar *srcChars==NULL as an empty string
+  if(isBogus()) {
+    return -1;
+  }
+
+  // pin indices to legal values
+  pinIndices(start, length);
+
+  if(srcChars == NULL) {
+    srcStart = srcLength = 0;
+  }
+
+  // get the correct pointer
+  const UChar *chars = getArrayStart();
+
+  chars += start;
+  srcChars += srcStart;
+
+  if(chars != srcChars) {
+    UErrorCode errorCode=U_ZERO_ERROR;
+    int32_t result=u_strcmpFold(chars, length, srcChars, srcLength,
+                                options|U_COMPARE_IGNORE_CASE, &errorCode);
+    if(result!=0) {
+      return (int8_t)(result >> 24 | 1);
+    }
+  } else {
+    // get the srcLength if necessary
+    if(srcLength < 0) {
+      srcLength = u_strlen(srcChars + srcStart);
+    }
+    if(length != srcLength) {
+      return (int8_t)((length - srcLength) >> 24 | 1);
+    }
+  }
+  return 0;
+}
+
+//========================================
+// Write implementation
+//========================================
+
+/*
+ * Implement argument checking and buffer handling
+ * for string case mapping as a common function.
+ */
+enum {
+    TO_LOWER,
+    TO_UPPER,
+    TO_TITLE,
+    FOLD_CASE
+};
+
+UnicodeString &
+UnicodeString::caseMap(BreakIterator *titleIter,
+                       const char *locale,
+                       uint32_t options,
+                       int32_t toWhichCase) {
+  if(fLength <= 0) {
+    // nothing to do
+    return *this;
+  }
+
+  UErrorCode errorCode;
+
+  errorCode = U_ZERO_ERROR;
+  UCaseProps *csp=ucase_getSingleton(&errorCode);
+  if(U_FAILURE(errorCode)) {
+    setToBogus();
+    return *this;
+  }
+
+  // We need to allocate a new buffer for the internal string case mapping function.
+  // This is very similar to how doReplace() below keeps the old array pointer
+  // and deletes the old array itself after it is done.
+  // In addition, we are forcing cloneArrayIfNeeded() to always allocate a new array.
+  UChar *oldArray = fArray;
+  int32_t oldLength = fLength;
+  int32_t *bufferToDelete = 0;
+
+  // Make sure that if the string is in fStackBuffer we do not overwrite it!
+  int32_t capacity;
+  if(fLength <= US_STACKBUF_SIZE) {
+    if(fArray == fStackBuffer) {
+      capacity = 2 * US_STACKBUF_SIZE; // make sure that cloneArrayIfNeeded() allocates a new buffer
+    } else {
+      capacity = US_STACKBUF_SIZE;
+    }
+  } else {
+    capacity = fLength + 20;
+  }
+  if(!cloneArrayIfNeeded(capacity, capacity, FALSE, &bufferToDelete, TRUE)) {
+    return *this;
+  }
+
+#if !UCONFIG_NO_BREAK_ITERATION
+  // set up the titlecasing break iterator
+  UBreakIterator *cTitleIter = 0;
+
+  if(toWhichCase == TO_TITLE) {
+    errorCode = U_ZERO_ERROR;
+    if(titleIter != 0) {
+      cTitleIter = (UBreakIterator *)titleIter;
+      ubrk_setText(cTitleIter, oldArray, oldLength, &errorCode);
+    } else {
+      cTitleIter = ubrk_open(UBRK_WORD, locale,
+                             oldArray, oldLength,
+                             &errorCode);
+    }
+    if(U_FAILURE(errorCode)) {
+      uprv_free(bufferToDelete);
+      setToBogus();
+      return *this;
+    }
+  }
+#endif
+
+  // Case-map, and if the result is too long, then reallocate and repeat.
+  do {
+    errorCode = U_ZERO_ERROR;
+    if(toWhichCase==TO_LOWER) {
+      fLength = ustr_toLower(csp, fArray, fCapacity,
+                             oldArray, oldLength,
+                             locale, &errorCode);
+    } else if(toWhichCase==TO_UPPER) {
+      fLength = ustr_toUpper(csp, fArray, fCapacity,
+                             oldArray, oldLength,
+                             locale, &errorCode);
+    } else if(toWhichCase==TO_TITLE) {
+#if UCONFIG_NO_BREAK_ITERATION
+        errorCode=U_UNSUPPORTED_ERROR;
+#else
+      fLength = ustr_toTitle(csp, fArray, fCapacity,
+                             oldArray, oldLength,
+                             cTitleIter, locale, &errorCode);
+#endif
+    } else {
+      fLength = ustr_foldCase(csp, fArray, fCapacity,
+                              oldArray, oldLength,
+                              options,
+                              &errorCode);
+    }
+  } while(errorCode==U_BUFFER_OVERFLOW_ERROR && cloneArrayIfNeeded(fLength, fLength, FALSE));
+
+#if !UCONFIG_NO_BREAK_ITERATION
+  if(cTitleIter != 0 && titleIter == 0) {
+    ubrk_close(cTitleIter);
+  }
+#endif
+
+  if (bufferToDelete) {
+    uprv_free(bufferToDelete);
+  }
+  if(U_FAILURE(errorCode)) {
+    setToBogus();
+  }
+  return *this;
+}
+
+UnicodeString &
+UnicodeString::toLower() {
+  return caseMap(0, Locale::getDefault().getName(), 0, TO_LOWER);
+}
+
+UnicodeString &
+UnicodeString::toLower(const Locale &locale) {
+  return caseMap(0, locale.getName(), 0, TO_LOWER);
+}
+
+UnicodeString &
+UnicodeString::toUpper() {
+  return caseMap(0, Locale::getDefault().getName(), 0, TO_UPPER);
+}
+
+UnicodeString &
+UnicodeString::toUpper(const Locale &locale) {
+  return caseMap(0, locale.getName(), 0, TO_UPPER);
+}
+
+#if !UCONFIG_NO_BREAK_ITERATION
+
+UnicodeString &
+UnicodeString::toTitle(BreakIterator *titleIter) {
+  return caseMap(titleIter, Locale::getDefault().getName(), 0, TO_TITLE);
+}
+
+UnicodeString &
+UnicodeString::toTitle(BreakIterator *titleIter, const Locale &locale) {
+  return caseMap(titleIter, locale.getName(), 0, TO_TITLE);
+}
+
+#endif
+
+UnicodeString &
+UnicodeString::foldCase(uint32_t options) {
+    /* The Locale parameter isn't used. Use "" instead. */
+    return caseMap(0, "", options, FOLD_CASE);
+}
+
+U_NAMESPACE_END