ICU-64260.0.1.tar.gz

[apple/icu.git] / icuSources / i18n / name2uni.cpp

diff --git a/icuSources/i18n/name2uni.cpp b/icuSources/i18n/name2uni.cpp
index 255a4bc63e6e736f5b27e35332b1e0cfcb2fb420..69e92b8c17e44974351253aa0e23c6e4cfe181a2 100644 (file)
--- a/icuSources/i18n/name2uni.cpp
+++ b/icuSources/i18n/name2uni.cpp
@@ -1,6 +1,8 @@
+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
 /*
 **********************************************************************
-*   Copyright (C) 2001-2004, International Business Machines
+*   Copyright (C) 2001-2011, International Business Machines
 *   Corporation and others.  All Rights Reserved.
 **********************************************************************
 *   Date        Name        Description
@@ -15,9 +17,12 @@
 #include "unicode/unifilt.h"
 #include "unicode/uchar.h"
 #include "unicode/uniset.h"
-#include "name2uni.h"
+#include "unicode/utf16.h"
 #include "cmemory.h"
+#include "name2uni.h"
+#include "patternprops.h"
 #include "uprops.h"
+#include "uinvchar.h"
 #include "util.h"
 
 U_NAMESPACE_BEGIN
@@ -35,10 +40,11 @@ U_CDECL_BEGIN
 // Does not use uset.h to reduce code dependencies
 static void U_CALLCONV
 _set_add(USet *set, UChar32 c) {
-    ((UnicodeSet *)set)->add(c);
+    uset_add(set, c);
 }
 
-static void U_CALLCONV
+// These functions aren't used.
+/*static void U_CALLCONV
 _set_addRange(USet *set, UChar32 start, UChar32 end) {
     ((UnicodeSet *)set)->add(start, end);
 }
@@ -46,7 +52,7 @@ _set_addRange(USet *set, UChar32 start, UChar32 end) {
 static void U_CALLCONV
 _set_addString(USet *set, const UChar *str, int32_t length) {
     ((UnicodeSet *)set)->add(UnicodeString((UBool)(length<0), str, length));
-}
+}*/
 
 U_CDECL_END
 
@@ -57,12 +63,15 @@ U_CDECL_END
 NameUnicodeTransliterator::NameUnicodeTransliterator(UnicodeFilter* adoptedFilter) :
     Transliterator(UNICODE_STRING("Name-Any", 8), adoptedFilter) {
 
+    UnicodeSet *legalPtr = &legal;
     // Get the legal character set
     USetAdder sa = {
-        (USet *)&legal, // USet* == UnicodeSet*
+        (USet *)legalPtr, // USet* == UnicodeSet*
         _set_add,
-        _set_addRange,
-        _set_addString
+        NULL, // Don't need _set_addRange
+        NULL, // Don't need _set_addString
+        NULL, // Don't need remove()
+        NULL
     };
     uprv_getCharNameCharacters(&sa);
 }
@@ -81,12 +90,12 @@ NameUnicodeTransliterator::NameUnicodeTransliterator(const NameUnicodeTransliter
 /**
  * Assignment operator.
  */
-NameUnicodeTransliterator& NameUnicodeTransliterator::operator=(
+/*NameUnicodeTransliterator& NameUnicodeTransliterator::operator=(
                              const NameUnicodeTransliterator& o) {
     Transliterator::operator=(o);
     // not necessary: the legal sets should all be the same -- legal=o.legal;
     return *this;
-}
+}*/
 
 /**
  * Transliterator API.
@@ -157,7 +166,7 @@ void NameUnicodeTransliterator::handleTransliterate(Replaceable& text, UTransPos
 
             // Convert \s+ => SPACE.  This assumes there are no
             // runs of >1 space characters in names.
-            if (uprv_isRuleWhiteSpace(c)) {
+            if (PatternProps::isWhiteSpace(c)) {
                 // Ignore leading whitespace
                 if (name.length() > 0 &&
                     name.charAt(name.length()-1) != SPACE) {
@@ -172,7 +181,6 @@ void NameUnicodeTransliterator::handleTransliterate(Replaceable& text, UTransPos
             }
 
             if (c == CLOSE_DELIM) {
-
                 int32_t len = name.length();
 
                 // Delete trailing space, if any
@@ -181,27 +189,29 @@ void NameUnicodeTransliterator::handleTransliterate(Replaceable& text, UTransPos
                     --len;
                 }
 
-                name.extract(0, len, cbuf, "");
+                if (uprv_isInvariantUString(name.getBuffer(), len)) {
+                    name.extract(0, len, cbuf, maxLen, US_INV);
 
-                UErrorCode status = U_ZERO_ERROR;
-                c = u_charFromName(U_EXTENDED_CHAR_NAME, cbuf, &status);
-                if (U_SUCCESS(status)) {
-                    // Lookup succeeded
+                    UErrorCode status = U_ZERO_ERROR;
+                    c = u_charFromName(U_EXTENDED_CHAR_NAME, cbuf, &status);
+                    if (U_SUCCESS(status)) {
+                        // Lookup succeeded
 
-                    // assert(UTF_CHAR_LENGTH(CLOSE_DELIM) == 1);
-                    cursor++; // advance over CLOSE_DELIM
+                        // assert(U16_LENGTH(CLOSE_DELIM) == 1);
+                        cursor++; // advance over CLOSE_DELIM
 
-                    str.truncate(0);
-                    str.append(c);
-                    text.handleReplaceBetween(openPos, cursor, str);
+                        str.truncate(0);
+                        str.append(c);
+                        text.handleReplaceBetween(openPos, cursor, str);
 
-                    // Adjust indices for the change in the length of
-                    // the string.  Do not assume that str.length() ==
-                    // 1, in case of surrogates.
-                    int32_t delta = cursor - openPos - str.length();
-                    cursor -= delta;
-                    limit -= delta;
-                    // assert(cursor == openPos + str.length());
+                        // Adjust indices for the change in the length of
+                        // the string.  Do not assume that str.length() ==
+                        // 1, in case of surrogates.
+                        int32_t delta = cursor - openPos - str.length();
+                        cursor -= delta;
+                        limit -= delta;
+                        // assert(cursor == openPos + str.length());
+                    }
                 }
                 // If the lookup failed, we leave things as-is and
                 // still switch to mode 0 and continue.
@@ -231,7 +241,7 @@ void NameUnicodeTransliterator::handleTransliterate(Replaceable& text, UTransPos
             break;
         }
 
-        cursor += UTF_CHAR_LENGTH(c);
+        cursor += U16_LENGTH(c);
     }
         
     offsets.contextLimit += limit - offsets.limit;