Merge pull request #104104 from bruvzg/icu771

ICU4C: Update to version 77.1

thirdparty/README.md

@@ -420,13 +420,13 @@ Files extracted from upstream source:
 ## icu4c
 
 - Upstream: https://github.com/unicode-org/icu
-- Version: 76.1 (8eca245c7484ac6cc179e3e5f7c1ea7680810f39, 2024)
+- Version: 77.1 (457157a92aa053e632cc7fcfd0e12f8a943b2d11, 2025)
 - License: Unicode
 
 Files extracted from upstream source:
 
 - The `common` folder
-- `scriptset.*`, `ucln_in.*`, `uspoof.cpp"` and `uspoof_impl.cpp` from the `i18n` folder
+- `scriptset.*`, `ucln_in.*`, `uspoof.cpp` and `uspoof_impl.*` from the `i18n` folder
 - `uspoof.h` from the `i18n/unicode` folder
 - `LICENSE`
 

thirdparty/icu4c/LICENSE

@@ -2,7 +2,7 @@ UNICODE LICENSE V3
 
 COPYRIGHT AND PERMISSION NOTICE
 
-Copyright © 2016-2024 Unicode, Inc.
+Copyright © 2016-2025 Unicode, Inc.
 
 NOTICE TO USER: Carefully read the following legal agreement. BY
 DOWNLOADING, INSTALLING, COPYING OR OTHERWISE USING DATA FILES, AND/OR

thirdparty/icu4c/common/brkiter.cpp

@@ -59,7 +59,7 @@ BreakIterator::buildInstance(const Locale& loc, const char *type, UErrorCode &st
 {
     char fnbuff[256];
     char ext[4]={'\0'};
-    CharString actualLocale;
+    CharString actual;
     int32_t size;
     const char16_t* brkfname = nullptr;
     UResourceBundle brkRulesStack;
@@ -94,7 +94,7 @@ BreakIterator::buildInstance(const Locale& loc, const char *type, UErrorCode &st
 
         // Use the string if we found it
         if (U_SUCCESS(status) && brkfname) {
-            actualLocale.append(ures_getLocaleInternal(brkName, &status), -1, status);
+            actual.append(ures_getLocaleInternal(brkName, &status), -1, status);
 
             char16_t* extStart=u_strchr(brkfname, 0x002e);
             int len = 0;
@@ -123,10 +123,9 @@ BreakIterator::buildInstance(const Locale& loc, const char *type, UErrorCode &st
     if (U_SUCCESS(status) && result != nullptr) {
         U_LOCALE_BASED(locBased, *(BreakIterator*)result);
 
-        locBased.setLocaleIDs(ures_getLocaleByType(b, ULOC_VALID_LOCALE, &status), 
-                              actualLocale.data());
-        uprv_strncpy(result->requestLocale, loc.getName(), ULOC_FULLNAME_CAPACITY);
-        result->requestLocale[ULOC_FULLNAME_CAPACITY-1] = 0; // always terminate
+        locBased.setLocaleIDs(ures_getLocaleByType(b, ULOC_VALID_LOCALE, &status),
+                              actual.data(), status);
+        LocaleBased::setLocaleID(loc.getName(), result->requestLocale, status);
     }
 
     ures_close(b);
@@ -206,26 +205,32 @@ BreakIterator::getAvailableLocales(int32_t& count)
 
 BreakIterator::BreakIterator()
 {
-    *validLocale = *actualLocale = *requestLocale = 0;
 }
 
 BreakIterator::BreakIterator(const BreakIterator &other) : UObject(other) {
-    uprv_strncpy(actualLocale, other.actualLocale, sizeof(actualLocale));
-    uprv_strncpy(validLocale, other.validLocale, sizeof(validLocale));
-    uprv_strncpy(requestLocale, other.requestLocale, sizeof(requestLocale));
+    UErrorCode status = U_ZERO_ERROR;
+    U_LOCALE_BASED(locBased, *this);
+    locBased.setLocaleIDs(other.validLocale, other.actualLocale, status);
+    LocaleBased::setLocaleID(other.requestLocale, requestLocale, status);
+    U_ASSERT(U_SUCCESS(status));
 }
 
 BreakIterator &BreakIterator::operator =(const BreakIterator &other) {
     if (this != &other) {
-        uprv_strncpy(actualLocale, other.actualLocale, sizeof(actualLocale));
-        uprv_strncpy(validLocale, other.validLocale, sizeof(validLocale));
-        uprv_strncpy(requestLocale, other.requestLocale, sizeof(requestLocale));
+        UErrorCode status = U_ZERO_ERROR;
+        U_LOCALE_BASED(locBased, *this);
+        locBased.setLocaleIDs(other.validLocale, other.actualLocale, status);
+        LocaleBased::setLocaleID(other.requestLocale, requestLocale, status);
+        U_ASSERT(U_SUCCESS(status));
     }
     return *this;
 }
 
 BreakIterator::~BreakIterator()
 {
+    delete validLocale;
+    delete actualLocale;
+    delete requestLocale;
 }
 
 // ------------------------------------------
@@ -394,7 +399,7 @@ BreakIterator::createInstance(const Locale& loc, int32_t kind, UErrorCode& statu
         // revisit this in ICU 3.0 and clean it up/fix it/remove it.
         if (U_SUCCESS(status) && (result != nullptr) && *actualLoc.getName() != 0) {
             U_LOCALE_BASED(locBased, *result);
-            locBased.setLocaleIDs(actualLoc.getName(), actualLoc.getName());
+            locBased.setLocaleIDs(actualLoc.getName(), actualLoc.getName(), status);
         }
         return result;
     }
@@ -488,6 +493,7 @@ BreakIterator::makeInstance(const Locale& loc, int32_t kind, UErrorCode& status)
     }
 
     if (U_FAILURE(status)) {
+        delete result;
         return nullptr;
     }
 
@@ -496,20 +502,25 @@ BreakIterator::makeInstance(const Locale& loc, int32_t kind, UErrorCode& status)
 
 Locale
 BreakIterator::getLocale(ULocDataLocaleType type, UErrorCode& status) const {
+    if (U_FAILURE(status)) {
+        return Locale::getRoot();
+    }
     if (type == ULOC_REQUESTED_LOCALE) {
-        return {requestLocale};
+        return requestLocale == nullptr ?
+            Locale::getRoot() : Locale(requestLocale->data());
     }
-    U_LOCALE_BASED(locBased, *this);
-    return locBased.getLocale(type, status);
+    return LocaleBased::getLocale(validLocale, actualLocale, type, status);
 }
 
 const char *
 BreakIterator::getLocaleID(ULocDataLocaleType type, UErrorCode& status) const {
+    if (U_FAILURE(status)) {
+        return nullptr;
+    }
     if (type == ULOC_REQUESTED_LOCALE) {
-        return requestLocale;
+        return requestLocale == nullptr ?  "" : requestLocale->data();
     }
-    U_LOCALE_BASED(locBased, *this);
-    return locBased.getLocaleID(type, status);
+    return LocaleBased::getLocaleID(validLocale, actualLocale, type, status);
 }
 
 
@@ -536,8 +547,10 @@ int32_t BreakIterator::getRuleStatusVec(int32_t *fillInVec, int32_t capacity, UE
 }
 
 BreakIterator::BreakIterator (const Locale& valid, const Locale& actual) {
+  UErrorCode status = U_ZERO_ERROR;
   U_LOCALE_BASED(locBased, (*this));
-  locBased.setLocaleIDs(valid, actual);
+  locBased.setLocaleIDs(valid.getName(), actual.getName(), status);
+  U_ASSERT(U_SUCCESS(status));
 }
 
 U_NAMESPACE_END

thirdparty/icu4c/common/charstr.cpp

@@ -70,6 +70,15 @@ CharString &CharString::copyFrom(const CharString &s, UErrorCode &errorCode) {
     return *this;
 }
 
+CharString &CharString::copyFrom(StringPiece s, UErrorCode &errorCode) {
+    if (U_FAILURE(errorCode)) {
+        return *this;
+    }
+    len = 0;
+    append(s, errorCode);
+    return *this;
+}
+
 int32_t CharString::lastIndexOf(char c) const {
     for(int32_t i=len; i>0;) {
         if(buffer[--i]==c) {
@@ -143,7 +152,7 @@ CharString &CharString::append(const char *s, int32_t sLength, UErrorCode &error
     return *this;
 }
 
-CharString &CharString::appendNumber(int32_t number, UErrorCode &status) {
+CharString &CharString::appendNumber(int64_t number, UErrorCode &status) {
     if (number < 0) {
         this->append('-', status);
         if (U_FAILURE(status)) {

thirdparty/icu4c/common/charstr.h

@@ -74,6 +74,7 @@ public:
      * use a UErrorCode where memory allocations might be needed.
      */
     CharString &copyFrom(const CharString &other, UErrorCode &errorCode);
+    CharString &copyFrom(StringPiece s, UErrorCode &errorCode);
 
     UBool isEmpty() const { return len==0; }
     int32_t length() const { return len; }
@@ -135,7 +136,7 @@ public:
     }
     CharString &append(const char *s, int32_t sLength, UErrorCode &status);
 
-    CharString &appendNumber(int32_t number, UErrorCode &status);
+    CharString &appendNumber(int64_t number, UErrorCode &status);
 
     /**
      * Returns a writable buffer for appending and writes the buffer's capacity to

thirdparty/icu4c/common/locbased.cpp

@@ -12,44 +12,84 @@
 */
 #include "locbased.h"
 #include "cstring.h"
+#include "charstr.h"
 
 U_NAMESPACE_BEGIN
 
-Locale LocaleBased::getLocale(ULocDataLocaleType type, UErrorCode& status) const {
-    const char* id = getLocaleID(type, status);
+Locale LocaleBased::getLocale(const CharString* valid, const CharString* actual,
+                              ULocDataLocaleType type, UErrorCode& status) {
+    const char* id = getLocaleID(valid, actual, type, status);
     return Locale(id != nullptr ? id : "");
 }
 
-const char* LocaleBased::getLocaleID(ULocDataLocaleType type, UErrorCode& status) const {
+const char* LocaleBased::getLocaleID(const CharString* valid, const CharString* actual,
+                                     ULocDataLocaleType type, UErrorCode& status) {
     if (U_FAILURE(status)) {
         return nullptr;
     }
 
     switch(type) {
     case ULOC_VALID_LOCALE:
-        return valid;
+        return valid == nullptr ? "" : valid->data();
     case ULOC_ACTUAL_LOCALE:
-        return actual;
+        return actual == nullptr ? "" : actual->data();
     default:
         status = U_ILLEGAL_ARGUMENT_ERROR;
         return nullptr;
     }
 }
 
-void LocaleBased::setLocaleIDs(const char* validID, const char* actualID) {
-    if (validID != nullptr) {
-      uprv_strncpy(valid, validID, ULOC_FULLNAME_CAPACITY);
-      valid[ULOC_FULLNAME_CAPACITY-1] = 0; // always terminate
+void LocaleBased::setLocaleIDs(const CharString* validID, const CharString* actualID, UErrorCode& status) {
+    setValidLocaleID(validID, status);
+    setActualLocaleID(actualID,status);
+}
+void LocaleBased::setLocaleIDs(const char* validID, const char* actualID, UErrorCode& status) {
+    setValidLocaleID(validID, status);
+    setActualLocaleID(actualID,status);
+}
+
+void LocaleBased::setLocaleID(const char* id, CharString*& dest, UErrorCode& status) {
+    if (U_FAILURE(status)) { return; }
+    if (id == nullptr || *id == 0) {
+        delete dest;
+        dest = nullptr;
+    } else {
+        if (dest == nullptr) {
+            dest = new CharString(id, status);
+            if (dest == nullptr) {
+                status = U_MEMORY_ALLOCATION_ERROR;
+                return;
+            }
+        } else {
+            dest->copyFrom(id, status);
+        }
     }
-    if (actualID != nullptr) {
-      uprv_strncpy(actual, actualID, ULOC_FULLNAME_CAPACITY);
-      actual[ULOC_FULLNAME_CAPACITY-1] = 0; // always terminate
+}
+
+void LocaleBased::setLocaleID(const CharString* id, CharString*& dest, UErrorCode& status) {
+    if (U_FAILURE(status)) { return; }
+    if (id == nullptr || id->isEmpty()) {
+        delete dest;
+        dest = nullptr;
+    } else {
+        if (dest == nullptr) {
+            dest = new CharString(*id, status);
+            if (dest == nullptr) {
+                status = U_MEMORY_ALLOCATION_ERROR;
+                return;
+            }
+        } else {
+            dest->copyFrom(*id, status);
+        }
     }
 }
 
-void LocaleBased::setLocaleIDs(const Locale& validID, const Locale& actualID) {
-  uprv_strcpy(valid, validID.getName());
-  uprv_strcpy(actual, actualID.getName());
+bool LocaleBased::equalIDs(const CharString* left, const CharString* right) {
+    // true if both are nullptr
+    if (left == nullptr && right == nullptr) return true;
+    // false if only one is nullptr
+    if (left == nullptr || right == nullptr) return false;
+    return *left == *right;
 }
 
 U_NAMESPACE_END

thirdparty/icu4c/common/locbased.h

@@ -19,13 +19,14 @@
 /**
  * Macro to declare a locale LocaleBased wrapper object for the given
  * object, which must have two members named `validLocale' and
- * `actualLocale' of size ULOC_FULLNAME_CAPACITY
+ * `actualLocale' of which are pointers to the internal icu::CharString.
  */
 #define U_LOCALE_BASED(varname, objname) \
   LocaleBased varname((objname).validLocale, (objname).actualLocale)
 
 U_NAMESPACE_BEGIN
 
+class CharString;
 /**
  * A utility class that unifies the implementation of getLocale() by
  * various ICU services.  This class is likely to be removed in the
@@ -41,33 +42,35 @@ class U_COMMON_API LocaleBased : public UMemory {
      * Construct a LocaleBased wrapper around the two pointers.  These
      * will be aliased for the lifetime of this object.
      */
-    inline LocaleBased(char* validAlias, char* actualAlias);
-
-    /**
-     * Construct a LocaleBased wrapper around the two const pointers.
-     * These will be aliased for the lifetime of this object.
-     */
-    inline LocaleBased(const char* validAlias, const char* actualAlias);
+    inline LocaleBased(CharString*& validAlias, CharString*& actualAlias);
 
     /**
      * Return locale meta-data for the service object wrapped by this
      * object.  Either the valid or the actual locale may be
      * retrieved.
+     * @param valid The valid locale.
+     * @param actual The actual locale.
      * @param type either ULOC_VALID_LOCALE or ULOC_ACTUAL_LOCALE
      * @param status input-output error code
      * @return the indicated locale
      */
-    Locale getLocale(ULocDataLocaleType type, UErrorCode& status) const;
+    static Locale getLocale(
+        const CharString* valid, const CharString* actual,
+        ULocDataLocaleType type, UErrorCode& status);
 
     /**
      * Return the locale ID for the service object wrapped by this
      * object.  Either the valid or the actual locale may be
      * retrieved.
+     * @param valid The valid locale.
+     * @param actual The actual locale.
      * @param type either ULOC_VALID_LOCALE or ULOC_ACTUAL_LOCALE
      * @param status input-output error code
      * @return the indicated locale ID
      */
-    const char* getLocaleID(ULocDataLocaleType type, UErrorCode& status) const;
+    static const char* getLocaleID(
+        const CharString* valid, const CharString* actual,
+        ULocDataLocaleType type, UErrorCode& status);
 
     /**
      * Set the locale meta-data for the service object wrapped by this
@@ -75,31 +78,40 @@ class U_COMMON_API LocaleBased : public UMemory {
      * @param valid the ID of the valid locale
      * @param actual the ID of the actual locale
      */
-    void setLocaleIDs(const char* valid, const char* actual);
+    void setLocaleIDs(const char* valid, const char* actual, UErrorCode& status);
+    void setLocaleIDs(const CharString* valid, const CharString* actual, UErrorCode& status);
 
-    /**
-     * Set the locale meta-data for the service object wrapped by this
-     * object.
-     * @param valid the ID of the valid locale
-     * @param actual the ID of the actual locale
-     */
-    void setLocaleIDs(const Locale& valid, const Locale& actual);
+    static void setLocaleID(const char* id, CharString*& dest, UErrorCode& status);
+    static void setLocaleID(const CharString* id, CharString*& dest, UErrorCode& status);
+
+    static bool equalIDs(const CharString* left, const CharString* right);
 
  private:
 
-    char* valid;
-    
-    char* actual;
+    void setValidLocaleID(const CharString* id, UErrorCode& status);
+    void setActualLocaleID(const CharString* id, UErrorCode& status);
+    void setValidLocaleID(const char* id, UErrorCode& status);
+    void setActualLocaleID(const char* id, UErrorCode& status);
+
+    CharString*& valid;
+    CharString*& actual;
 };
 
-inline LocaleBased::LocaleBased(char* validAlias, char* actualAlias) :
+inline LocaleBased::LocaleBased(CharString*& validAlias, CharString*& actualAlias) :
     valid(validAlias), actual(actualAlias) {
 }
 
-inline LocaleBased::LocaleBased(const char* validAlias,
-                                const char* actualAlias) :
-    // ugh: cast away const
-    valid(const_cast<char*>(validAlias)), actual(const_cast<char*>(actualAlias)) {
+inline void LocaleBased::setValidLocaleID(const CharString* id, UErrorCode& status) {
+    setLocaleID(id, valid, status);
+}
+inline void LocaleBased::setActualLocaleID(const CharString* id, UErrorCode& status) {
+    setLocaleID(id, actual, status);
+}
+inline void LocaleBased::setValidLocaleID(const char* id, UErrorCode& status) {
+    setLocaleID(id, valid, status);
+}
+inline void LocaleBased::setActualLocaleID(const char* id, UErrorCode& status) {
+    setLocaleID(id, actual, status);
 }
 
 U_NAMESPACE_END

thirdparty/icu4c/common/locdispnames.cpp

@@ -19,6 +19,8 @@
 *   that then do not depend on resource bundle code and display name data.
 */
 
+#include <string_view>
+
 #include "unicode/utypes.h"
 #include "unicode/brkiter.h"
 #include "unicode/locid.h"
@@ -359,7 +361,7 @@ _getStringOrCopyKey(const char *path, const char *locale,
     return u_terminateUChars(dest, destCapacity, length, &errorCode);
 }
 
-using UDisplayNameGetter = icu::CharString(const char*, UErrorCode&);
+using UDisplayNameGetter = icu::CharString(std::string_view, UErrorCode&);
 
 int32_t
 _getDisplayNameForComponent(const char *locale,
@@ -377,6 +379,10 @@ _getDisplayNameForComponent(const char *locale,
         return 0;
     }
 
+    if (locale == nullptr) {
+        locale = uloc_getDefault();
+    }
+
     localStatus = U_ZERO_ERROR;
     icu::CharString localeBuffer = (*getter)(locale, localStatus);
     if (U_FAILURE(localStatus)) {

thirdparty/icu4c/common/locid.cpp

@@ -1828,8 +1828,13 @@ ulocimp_isCanonicalizedLocaleForTest(const char* localeName)
 
 U_NAMESPACE_BEGIN
 
-/*This function initializes a Locale from a C locale ID*/
 Locale& Locale::init(const char* localeID, UBool canonicalize)
+{
+    return localeID == nullptr ? *this = getDefault() : init(StringPiece{localeID}, canonicalize);
+}
+
+/*This function initializes a Locale from a C locale ID*/
+Locale& Locale::init(StringPiece localeID, UBool canonicalize)
 {
     fIsBogus = false;
     /* Free our current storage */
@@ -1854,19 +1859,28 @@ Locale& Locale::init(const char* localeID, UBool canonicalize)
         int32_t length;
         UErrorCode err;
 
-        if(localeID == nullptr) {
-            // not an error, just set the default locale
-            return *this = getDefault();
-        }
-
         /* preset all fields to empty */
         language[0] = script[0] = country[0] = 0;
 
+        const auto parse = [canonicalize](std::string_view localeID,
+                                          char* name,
+                                          int32_t nameCapacity,
+                                          UErrorCode& status) {
+            return ByteSinkUtil::viaByteSinkToTerminatedChars(
+                name, nameCapacity,
+                [&](ByteSink& sink, UErrorCode& status) {
+                    if (canonicalize) {
+                        ulocimp_canonicalize(localeID, sink, status);
+                    } else {
+                        ulocimp_getName(localeID, sink, status);
+                    }
+                },
+                status);
+        };
+
         // "canonicalize" the locale ID to ICU/Java format
         err = U_ZERO_ERROR;
-        length = canonicalize ?
-            uloc_canonicalize(localeID, fullName, sizeof(fullNameBuffer), &err) :
-            uloc_getName(localeID, fullName, sizeof(fullNameBuffer), &err);
+        length = parse(localeID, fullName, sizeof fullNameBuffer, err);
 
         if (err == U_BUFFER_OVERFLOW_ERROR || length >= static_cast<int32_t>(sizeof(fullNameBuffer))) {
             U_ASSERT(baseName == nullptr);
@@ -1877,9 +1891,7 @@ Locale& Locale::init(const char* localeID, UBool canonicalize)
             }
             fullName = newFullName;
             err = U_ZERO_ERROR;
-            length = canonicalize ?
-                uloc_canonicalize(localeID, fullName, length+1, &err) :
-                uloc_getName(localeID, fullName, length+1, &err);
+            length = parse(localeID, fullName, length + 1, err);
         }
         if(U_FAILURE(err) || err == U_STRING_NOT_TERMINATED_WARNING) {
             /* should never occur */
@@ -2200,6 +2212,13 @@ Locale::createFromName (const char *name)
     }
 }
 
+Locale U_EXPORT2
+Locale::createFromName(StringPiece name) {
+    Locale loc("");
+    loc.init(name, false);
+    return loc;
+}
+
 Locale U_EXPORT2
 Locale::createCanonical(const char* name) {
     Locale loc("");

thirdparty/icu4c/common/loclikely.cpp

@@ -300,6 +300,9 @@ ulocimp_addLikelySubtags(const char* localeID,
                          icu::ByteSink& sink,
                          UErrorCode& status) {
     if (U_FAILURE(status)) { return; }
+    if (localeID == nullptr) {
+        localeID = uloc_getDefault();
+    }
     icu::CharString localeBuffer = ulocimp_canonicalize(localeID, status);
     _uloc_addLikelySubtags(localeBuffer.data(), sink, status);
 }
@@ -334,6 +337,9 @@ ulocimp_minimizeSubtags(const char* localeID,
                         bool favorScript,
                         UErrorCode& status) {
     if (U_FAILURE(status)) { return; }
+    if (localeID == nullptr) {
+        localeID = uloc_getDefault();
+    }
     icu::CharString localeBuffer = ulocimp_canonicalize(localeID, status);
     _uloc_minimizeSubtags(localeBuffer.data(), sink, favorScript, status);
 }
@@ -349,7 +355,9 @@ uloc_isRightToLeft(const char *locale) {
     UErrorCode errorCode = U_ZERO_ERROR;
     icu::CharString lang;
     icu::CharString script;
-    ulocimp_getSubtags(locale, &lang, &script, nullptr, nullptr, nullptr, errorCode);
+    ulocimp_getSubtags(
+        locale == nullptr ? uloc_getDefault() : locale,
+        &lang, &script, nullptr, nullptr, nullptr, errorCode);
     if (U_FAILURE(errorCode) || script.isEmpty()) {
         // Fastpath: We know the likely scripts and their writing direction
         // for some common languages.
@@ -369,7 +377,7 @@ uloc_isRightToLeft(const char *locale) {
         if (U_FAILURE(errorCode)) {
             return false;
         }
-        ulocimp_getSubtags(likely.data(), nullptr, &script, nullptr, nullptr, nullptr, errorCode);
+        ulocimp_getSubtags(likely.toStringPiece(), nullptr, &script, nullptr, nullptr, nullptr, errorCode);
         if (U_FAILURE(errorCode) || script.isEmpty()) {
             return false;
         }
@@ -430,7 +438,7 @@ ulocimp_getRegionForSupplementalData(const char *localeID, bool inferRegion,
     icu::CharString rgBuf = GetRegionFromKey(localeID, "rg", status);
     if (U_SUCCESS(status) && rgBuf.isEmpty()) {
         // No valid rg keyword value, try for unicode_region_subtag
-        rgBuf = ulocimp_getRegion(localeID, status);
+        rgBuf = ulocimp_getRegion(localeID == nullptr ? uloc_getDefault() : localeID, status);
         if (U_SUCCESS(status) && rgBuf.isEmpty() && inferRegion) {
             // Second check for sd keyword value
             rgBuf = GetRegionFromKey(localeID, "sd", status);
@@ -439,7 +447,7 @@ ulocimp_getRegionForSupplementalData(const char *localeID, bool inferRegion,
                 UErrorCode rgStatus = U_ZERO_ERROR;
                 icu::CharString locBuf = ulocimp_addLikelySubtags(localeID, rgStatus);
                 if (U_SUCCESS(rgStatus)) {
-                    rgBuf = ulocimp_getRegion(locBuf.data(), status);
+                    rgBuf = ulocimp_getRegion(locBuf.toStringPiece(), status);
                 }
             }
         }

thirdparty/icu4c/common/loclikelysubtags.cpp

@@ -527,7 +527,7 @@ LSR LikelySubtags::makeMaximizedLsrFrom(const Locale &locale,
         return {};
     }
     const char *name = locale.getName();
-    if (uprv_isAtSign(name[0]) && name[1] == 'x' && name[2] == '=') {  // name.startsWith("@x=")
+    if (!returnInputIfUnmatch && uprv_isAtSign(name[0]) && name[1] == 'x' && name[2] == '=') {  // name.startsWith("@x=")
         // Private use language tag x-subtag-subtag... which CLDR changes to
         // und-x-subtag-subtag...
         return LSR(name, "", "", LSR::EXPLICIT_LSR);

thirdparty/icu4c/common/locresdata.cpp

@@ -161,6 +161,9 @@ _uloc_getOrientationHelper(const char* localeId,
 
     if (U_FAILURE(status)) { return result; }
 
+    if (localeId == nullptr) {
+        localeId = uloc_getDefault();
+    }
     icu::CharString localeBuffer = ulocimp_canonicalize(localeId, status);
 
     if (U_FAILURE(status)) { return result; }

thirdparty/icu4c/common/punycode.cpp

@@ -193,7 +193,7 @@ u_strToPunycode(const char16_t *src, int32_t srcLength,
         return 0;
     }
 
-    if(src==nullptr || srcLength<-1 || (dest==nullptr && destCapacity!=0)) {
+    if(src==nullptr || srcLength<-1 || destCapacity<0 || (dest==nullptr && destCapacity!=0)) {
         *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
         return 0;
     }

thirdparty/icu4c/common/putil.cpp

@@ -76,7 +76,7 @@
 #include <float.h>
 
 #ifndef U_COMMON_IMPLEMENTATION
-#error U_COMMON_IMPLEMENTATION not set - must be set for all ICU source files in common/ - see https://unicode-org.github.io/icu/userguide/howtouseicu
+#error U_COMMON_IMPLEMENTATION not set - must be set for all ICU source files in common/ - see https://unicode-org.github.io/icu/userguide/icu/howtouseicu.html
 #endif
 
 

thirdparty/icu4c/common/rbbinode.cpp

@@ -47,7 +47,10 @@ static int  gLastSerial = 0;
 //    Constructor.   Just set the fields to reasonable default values.
 //
 //-------------------------------------------------------------------------
-RBBINode::RBBINode(NodeType t) : UMemory() {
+RBBINode::RBBINode(NodeType t, UErrorCode& status) : UMemory() {
+    if (U_FAILURE(status)) {
+        return;
+    }
 #ifdef RBBI_DEBUG
     fSerialNum    = ++gLastSerial;
 #endif
@@ -65,10 +68,13 @@ RBBINode::RBBINode(NodeType t) : UMemory() {
     fVal          = 0;
     fPrecedence   = precZero;
 
-    UErrorCode     status = U_ZERO_ERROR;
-    fFirstPosSet  = new UVector(status);  // TODO - get a real status from somewhere
+    fFirstPosSet  = new UVector(status);
     fLastPosSet   = new UVector(status);
     fFollowPos    = new UVector(status);
+    if (U_SUCCESS(status) &&
+        (fFirstPosSet == nullptr || fLastPosSet == nullptr || fFollowPos == nullptr)) {
+        status =  U_MEMORY_ALLOCATION_ERROR;
+    }
     if      (t==opCat)    {fPrecedence = precOpCat;}
     else if (t==opOr)     {fPrecedence = precOpOr;}
     else if (t==opStart)  {fPrecedence = precStart;}
@@ -77,7 +83,10 @@ RBBINode::RBBINode(NodeType t) : UMemory() {
 }
 
 
-RBBINode::RBBINode(const RBBINode &other) : UMemory(other) {
+RBBINode::RBBINode(const RBBINode &other, UErrorCode& status) : UMemory(other) {
+    if (U_FAILURE(status)) {
+        return;
+    }
 #ifdef RBBI_DEBUG
     fSerialNum   = ++gLastSerial;
 #endif
@@ -94,10 +103,13 @@ RBBINode::RBBINode(const RBBINode &other) : UMemory(other) {
     fVal         = other.fVal;
     fRuleRoot    = false;
     fChainIn     = other.fChainIn;
-    UErrorCode     status = U_ZERO_ERROR;
     fFirstPosSet = new UVector(status);   // TODO - get a real status from somewhere
     fLastPosSet  = new UVector(status);
     fFollowPos   = new UVector(status);
+    if (U_SUCCESS(status) &&
+        (fFirstPosSet == nullptr || fLastPosSet == nullptr || fFollowPos == nullptr)) {
+        status =  U_MEMORY_ALLOCATION_ERROR;
+    }
 }
 
 
@@ -193,27 +205,54 @@ void RBBINode::NRDeleteNode(RBBINode *node) {
 //                  references in preparation for generating the DFA tables.
 //
 //-------------------------------------------------------------------------
-RBBINode *RBBINode::cloneTree() {
+constexpr int kRecursiveDepthLimit = 3500;
+RBBINode *RBBINode::cloneTree(UErrorCode &status, int depth) {
+    if (U_FAILURE(status)) {
+        return nullptr;
+    }
+    // If the depth of the stack is too deep, we return U_INPUT_TOO_LONG_ERROR
+    // to avoid stack overflow crash.
+    if (depth > kRecursiveDepthLimit) {
+        status = U_INPUT_TOO_LONG_ERROR;
+        return nullptr;
+    }
     RBBINode    *n;
 
     if (fType == RBBINode::varRef) {
         // If the current node is a variable reference, skip over it
         //   and clone the definition of the variable instead.
-        n = fLeftChild->cloneTree();
+        n = fLeftChild->cloneTree(status, depth+1);
+        if (U_FAILURE(status)) {
+            return nullptr;
+        }
     } else if (fType == RBBINode::uset) {
         n = this;
     } else {
-        n = new RBBINode(*this);
+        n = new RBBINode(*this, status);
+        if (U_FAILURE(status)) {
+            delete n;
+            return nullptr;
+        }
         // Check for null pointer.
-        if (n != nullptr) {
-            if (fLeftChild != nullptr) {
-                n->fLeftChild          = fLeftChild->cloneTree();
-                n->fLeftChild->fParent = n;
+        if (n == nullptr) {
+            status =  U_MEMORY_ALLOCATION_ERROR;
+            return nullptr;
+        }
+        if (fLeftChild != nullptr) {
+            n->fLeftChild          = fLeftChild->cloneTree(status, depth+1);
+            if (U_FAILURE(status)) {
+                delete n;
+                return nullptr;
             }
-            if (fRightChild != nullptr) {
-                n->fRightChild          = fRightChild->cloneTree();
-                n->fRightChild->fParent = n;
+            n->fLeftChild->fParent = n;
+        }
+        if (fRightChild != nullptr) {
+            n->fRightChild          = fRightChild->cloneTree(status, depth+1);
+            if (U_FAILURE(status)) {
+                delete n;
+                return nullptr;
             }
+            n->fRightChild->fParent = n;
         }
     }
     return n;
@@ -239,7 +278,6 @@ RBBINode *RBBINode::cloneTree() {
 //                      nested references are handled by cloneTree(), not here.
 //
 //-------------------------------------------------------------------------
-constexpr int kRecursiveDepthLimit = 3500;
 RBBINode *RBBINode::flattenVariables(UErrorCode& status, int depth) {
     if (U_FAILURE(status)) {
         return this;
@@ -251,21 +289,34 @@ RBBINode *RBBINode::flattenVariables(UErrorCode& status, int depth) {
         return this;
     }
     if (fType == varRef) {
-        RBBINode *retNode  = fLeftChild->cloneTree();
-        if (retNode != nullptr) {
-            retNode->fRuleRoot = this->fRuleRoot;
-            retNode->fChainIn  = this->fChainIn;
+        RBBINode *retNode  = fLeftChild->cloneTree(status, depth+1);
+        if (U_FAILURE(status)) {
+            return this;
         }
+        retNode->fRuleRoot = this->fRuleRoot;
+        retNode->fChainIn  = this->fChainIn;
         delete this;   // TODO: undefined behavior. Fix.
         return retNode;
     }
 
     if (fLeftChild != nullptr) {
         fLeftChild = fLeftChild->flattenVariables(status, depth+1);
+        if (fLeftChild == nullptr) {
+            status = U_MEMORY_ALLOCATION_ERROR;
+        }
+        if (U_FAILURE(status)) {
+            return this;
+        }
         fLeftChild->fParent  = this;
     }
     if (fRightChild != nullptr) {
         fRightChild = fRightChild->flattenVariables(status, depth+1);
+        if (fRightChild == nullptr) {
+            status = U_MEMORY_ALLOCATION_ERROR;
+        }
+        if (U_FAILURE(status)) {
+            return this;
+        }
         fRightChild->fParent = this;
     }
     return this;
@@ -280,7 +331,16 @@ RBBINode *RBBINode::flattenVariables(UErrorCode& status, int depth) {
 //                 the left child of the uset node.
 //
 //-------------------------------------------------------------------------
-void RBBINode::flattenSets() {
+void RBBINode::flattenSets(UErrorCode &status, int depth) {
+    if (U_FAILURE(status)) {
+        return;
+    }
+    // If the depth of the stack is too deep, we return U_INPUT_TOO_LONG_ERROR
+    // to avoid stack overflow crash.
+    if (depth > kRecursiveDepthLimit) {
+        status = U_INPUT_TOO_LONG_ERROR;
+        return;
+    }
     U_ASSERT(fType != setRef);
 
     if (fLeftChild != nullptr) {
@@ -288,11 +348,15 @@ void RBBINode::flattenSets() {
             RBBINode *setRefNode = fLeftChild;
             RBBINode *usetNode   = setRefNode->fLeftChild;
             RBBINode *replTree   = usetNode->fLeftChild;
-            fLeftChild           = replTree->cloneTree();
+            fLeftChild           = replTree->cloneTree(status, depth+1);
+            if (U_FAILURE(status)) {
+                delete setRefNode;
+                return;
+            }
             fLeftChild->fParent  = this;
             delete setRefNode;
         } else {
-            fLeftChild->flattenSets();
+            fLeftChild->flattenSets(status, depth+1);
         }
     }
 
@@ -301,11 +365,15 @@ void RBBINode::flattenSets() {
             RBBINode *setRefNode = fRightChild;
             RBBINode *usetNode   = setRefNode->fLeftChild;
             RBBINode *replTree   = usetNode->fLeftChild;
-            fRightChild           = replTree->cloneTree();
+            fRightChild           = replTree->cloneTree(status, depth+1);
+            if (U_FAILURE(status)) {
+                delete setRefNode;
+                return;
+            }
             fRightChild->fParent  = this;
             delete setRefNode;
         } else {
-            fRightChild->flattenSets();
+            fRightChild->flattenSets(status, depth+1);
         }
     }
 }

thirdparty/icu4c/common/rbbinode.h

@@ -91,14 +91,14 @@ class RBBINode : public UMemory {
         UVector       *fFollowPos;
 
 
-        RBBINode(NodeType t);
-        RBBINode(const RBBINode &other);
+        RBBINode(NodeType t, UErrorCode& status);
+        RBBINode(const RBBINode &other, UErrorCode& status);
         ~RBBINode();
         static void  NRDeleteNode(RBBINode *node);
         
-        RBBINode    *cloneTree();
+        RBBINode    *cloneTree(UErrorCode &status, int depth=0);
         RBBINode    *flattenVariables(UErrorCode &status, int depth=0);
-        void         flattenSets();
+        void         flattenSets(UErrorCode &status, int depth=0);
         void         findNodes(UVector *dest, RBBINode::NodeType kind, UErrorCode &status);
 
 #ifdef RBBI_DEBUG

thirdparty/icu4c/common/rbbiscan.cpp

@@ -767,15 +767,24 @@ void RBBIRuleScanner::findSetFor(const UnicodeString &s, RBBINode *node, Unicode
             c = s.char32At(0);
             setToAdopt = new UnicodeSet(c, c);
         }
+        if (setToAdopt == nullptr) {
+            error(U_MEMORY_ALLOCATION_ERROR);
+            return;
+        }
     }
 
     //
     // Make a new uset node to refer to this UnicodeSet
     // This new uset node becomes the child of the caller's setReference node.
     //
-    RBBINode *usetNode    = new RBBINode(RBBINode::uset);
+    UErrorCode localStatus = U_ZERO_ERROR;
+    RBBINode *usetNode    = new RBBINode(RBBINode::uset, localStatus);
     if (usetNode == nullptr) {
-        error(U_MEMORY_ALLOCATION_ERROR);
+        localStatus = U_MEMORY_ALLOCATION_ERROR;
+    }
+    if (U_FAILURE(localStatus)) {
+        delete usetNode;
+        error(localStatus);
         delete setToAdopt;
         return;
     }
@@ -1191,7 +1200,7 @@ RBBINode  *RBBIRuleScanner::pushNewNode(RBBINode::NodeType  t) {
         return nullptr;
     }
     fNodeStackPtr++;
-    fNodeStack[fNodeStackPtr] = new RBBINode(t);
+    fNodeStack[fNodeStackPtr] = new RBBINode(t, *fRB->fStatus);
     if (fNodeStack[fNodeStackPtr] == nullptr) {
         *fRB->fStatus = U_MEMORY_ALLOCATION_ERROR;
     }

thirdparty/icu4c/common/rbbisetb.cpp

@@ -375,7 +375,11 @@ void  RBBISetBuilder::addValToSets(UVector *sets, uint32_t val) {
 }
 
 void  RBBISetBuilder::addValToSet(RBBINode *usetNode, uint32_t val) {
-    RBBINode *leafNode = new RBBINode(RBBINode::leafChar);
+    RBBINode *leafNode = new RBBINode(RBBINode::leafChar, *fStatus);
+    if (U_FAILURE(*fStatus)) {
+        delete leafNode;
+        return;
+    }
     if (leafNode == nullptr) {
         *fStatus = U_MEMORY_ALLOCATION_ERROR;
         return;
@@ -388,9 +392,13 @@ void  RBBISetBuilder::addValToSet(RBBINode *usetNode, uint32_t val) {
         // There are already input symbols present for this set.
         // Set up an OR node, with the previous stuff as the left child
         //   and the new value as the right child.
-        RBBINode *orNode = new RBBINode(RBBINode::opOr);
+        RBBINode *orNode = new RBBINode(RBBINode::opOr, *fStatus);
         if (orNode == nullptr) {
             *fStatus = U_MEMORY_ALLOCATION_ERROR;
+        }
+        if (U_FAILURE(*fStatus)) {
+            delete orNode;
+            delete leafNode;
             return;
         }
         orNode->fLeftChild  = usetNode->fLeftChild;

thirdparty/icu4c/common/rbbitblb.cpp

@@ -99,13 +99,22 @@ void  RBBITableBuilder::buildForwardTable() {
     //   {bof} fake character.
     // 
     if (fRB->fSetBuilder->sawBOF()) {
-        RBBINode *bofTop    = new RBBINode(RBBINode::opCat);
-        RBBINode *bofLeaf   = new RBBINode(RBBINode::leafChar);
-        // Delete and exit if memory allocation failed.
-        if (bofTop == nullptr || bofLeaf == nullptr) {
+        RBBINode *bofTop    = new RBBINode(RBBINode::opCat, *fStatus);
+        if (bofTop == nullptr) {
             *fStatus = U_MEMORY_ALLOCATION_ERROR;
+        }
+        if (U_FAILURE(*fStatus)) {
             delete bofTop;
+            return;
+        }
+        RBBINode *bofLeaf   = new RBBINode(RBBINode::leafChar, *fStatus);
+        // Delete and exit if memory allocation failed.
+        if (bofLeaf == nullptr) {
+            *fStatus = U_MEMORY_ALLOCATION_ERROR;
+        }
+        if (U_FAILURE(*fStatus)) {
             delete bofLeaf;
+            delete bofTop;
             return;
         }
         bofTop->fLeftChild  = bofLeaf;
@@ -120,18 +129,23 @@ void  RBBITableBuilder::buildForwardTable() {
     //   Appears as a cat-node, left child being the original tree,
     //   right child being the end marker.
     //
-    RBBINode *cn = new RBBINode(RBBINode::opCat);
+    RBBINode *cn = new RBBINode(RBBINode::opCat, *fStatus);
     // Exit if memory allocation failed.
     if (cn == nullptr) {
         *fStatus = U_MEMORY_ALLOCATION_ERROR;
+    }
+    if (U_FAILURE(*fStatus)) {
+        delete cn;
         return;
     }
     cn->fLeftChild = fTree;
     fTree->fParent = cn;
-    RBBINode *endMarkerNode = cn->fRightChild = new RBBINode(RBBINode::endMark);
+    RBBINode *endMarkerNode = cn->fRightChild = new RBBINode(RBBINode::endMark, *fStatus);
     // Delete and exit if memory allocation failed.
     if (cn->fRightChild == nullptr) {
         *fStatus = U_MEMORY_ALLOCATION_ERROR;
+    }
+    if (U_FAILURE(*fStatus)) {
         delete cn;
         return;
     }
@@ -142,7 +156,7 @@ void  RBBITableBuilder::buildForwardTable() {
     //  Replace all references to UnicodeSets with the tree for the equivalent
     //      expression.
     //
-    fTree->flattenSets();
+    fTree->flattenSets(*fStatus, 0);
 #ifdef RBBI_DEBUG
     if (fRB->fDebugEnv && uprv_strstr(fRB->fDebugEnv, "stree")) {
         RBBIDebugPuts("\nParse tree after flattening Unicode Set references.");

thirdparty/icu4c/common/resbund.cpp

@@ -388,7 +388,7 @@ const Locale &ResourceBundle::getLocale() const {
     return ncThis->fLocale != nullptr ? *ncThis->fLocale : Locale::getDefault();
 }
 
-const Locale ResourceBundle::getLocale(ULocDataLocaleType type, UErrorCode &status) const
+Locale ResourceBundle::getLocale(ULocDataLocaleType type, UErrorCode &status) const
 {
   return ures_getLocaleByType(fResource, type, &status);
 }

thirdparty/icu4c/common/ucnvmbcs.cpp

@@ -3146,11 +3146,8 @@ ucnv_MBCSGetNextUChar(UConverterToUnicodeArgs *pArgs,
     if(c<0) {
         if(U_SUCCESS(*pErrorCode) && source==sourceLimit && lastSource<source) {
             /* incomplete character byte sequence */
-            uint8_t *bytes=cnv->toUBytes;
             cnv->toULength = static_cast<int8_t>(source - lastSource);
-            do {
-                *bytes++=*lastSource++;
-            } while(lastSource<source);
+            uprv_memcpy(cnv->toUBytes, lastSource, cnv->toULength);
             *pErrorCode=U_TRUNCATED_CHAR_FOUND;
         } else if(U_FAILURE(*pErrorCode)) {
             /* callback(illegal) */

thirdparty/icu4c/common/ucurr.cpp

@@ -372,12 +372,8 @@ struct CReg : public icu::UMemory {
     CReg(const char16_t* _iso, const char* _id)
         : next(nullptr)
     {
-        int32_t len = static_cast<int32_t>(uprv_strlen(_id));
-        if (len > static_cast<int32_t>(sizeof(id) - 1)) {
-            len = (sizeof(id)-1);
-        }
-        uprv_strncpy(id, _id, len);
-        id[len] = 0;
+        uprv_strncpy(id, _id, sizeof(id)-1);
+        id[sizeof(id)-1] = 0;
         u_memcpy(iso, _iso, ISO_CURRENCY_CODE_LENGTH);
         iso[ISO_CURRENCY_CODE_LENGTH] = 0;
     }
@@ -682,6 +678,9 @@ ucurr_getName(const char16_t* currency,
     // this function.
     UErrorCode ec2 = U_ZERO_ERROR;
 
+    if (locale == nullptr) {
+        locale = uloc_getDefault();
+    }
     CharString loc = ulocimp_getName(locale, ec2);
     if (U_FAILURE(ec2)) {
         *ec = U_ILLEGAL_ARGUMENT_ERROR;
@@ -780,6 +779,9 @@ ucurr_getPluralName(const char16_t* currency,
     // this function.
     UErrorCode ec2 = U_ZERO_ERROR;
 
+    if (locale == nullptr) {
+        locale = uloc_getDefault();
+    }
     CharString loc = ulocimp_getName(locale, ec2);
     if (U_FAILURE(ec2)) {
         *ec = U_ILLEGAL_ARGUMENT_ERROR;
@@ -973,6 +975,9 @@ collectCurrencyNames(const char* locale,
     // Look up the Currencies resource for the given locale.
     UErrorCode ec2 = U_ZERO_ERROR;
 
+    if (locale == nullptr) {
+        locale = uloc_getDefault();
+    }
     CharString loc = ulocimp_getName(locale, ec2);
     if (U_FAILURE(ec2)) {
         ec = U_ILLEGAL_ARGUMENT_ERROR;

thirdparty/icu4c/common/uloc_tag.cpp

@@ -1043,7 +1043,7 @@ _initializeULanguageTag(ULanguageTag* langtag) {
 }
 
 void
-_appendLanguageToLanguageTag(const char* localeID, icu::ByteSink& sink, bool strict, UErrorCode& status) {
+_appendLanguageToLanguageTag(std::string_view localeID, icu::ByteSink& sink, bool strict, UErrorCode& status) {
     UErrorCode tmpStatus = U_ZERO_ERROR;
 
     if (U_FAILURE(status)) {
@@ -1088,7 +1088,7 @@ _appendLanguageToLanguageTag(const char* localeID, icu::ByteSink& sink, bool str
 }
 
 void
-_appendScriptToLanguageTag(const char* localeID, icu::ByteSink& sink, bool strict, UErrorCode& status) {
+_appendScriptToLanguageTag(std::string_view localeID, icu::ByteSink& sink, bool strict, UErrorCode& status) {
     UErrorCode tmpStatus = U_ZERO_ERROR;
 
     if (U_FAILURE(status)) {
@@ -1118,7 +1118,7 @@ _appendScriptToLanguageTag(const char* localeID, icu::ByteSink& sink, bool stric
 }
 
 void
-_appendRegionToLanguageTag(const char* localeID, icu::ByteSink& sink, bool strict, UErrorCode& status) {
+_appendRegionToLanguageTag(std::string_view localeID, icu::ByteSink& sink, bool strict, UErrorCode& status) {
     UErrorCode tmpStatus = U_ZERO_ERROR;
 
     if (U_FAILURE(status)) {
@@ -1169,7 +1169,7 @@ void _sortVariants(VariantListEntry* first) {
 }
 
 void
-_appendVariantsToLanguageTag(const char* localeID, icu::ByteSink& sink, bool strict, bool& hadPosix, UErrorCode& status) {
+_appendVariantsToLanguageTag(std::string_view localeID, icu::ByteSink& sink, bool strict, bool& hadPosix, UErrorCode& status) {
     if (U_FAILURE(status)) { return; }
 
     UErrorCode tmpStatus = U_ZERO_ERROR;
@@ -1872,7 +1872,7 @@ _appendKeywords(ULanguageTag* langtag, icu::ByteSink& sink, UErrorCode& status)
 }
 
 void
-_appendPrivateuseToLanguageTag(const char* localeID, icu::ByteSink& sink, bool strict, bool /*hadPosix*/, UErrorCode& status) {
+_appendPrivateuseToLanguageTag(std::string_view localeID, icu::ByteSink& sink, bool strict, bool /*hadPosix*/, UErrorCode& status) {
     if (U_FAILURE(status)) { return; }
 
     UErrorCode tmpStatus = U_ZERO_ERROR;
@@ -2596,6 +2596,9 @@ ulocimp_toLanguageTag(const char* localeID,
     bool hadPosix = false;
     const char* pKeywordStart;
 
+    if (localeID == nullptr) {
+        localeID = uloc_getDefault();
+    }
     /* Note: uloc_canonicalize returns "en_US_POSIX" for input locale ID "".  See #6835 */
     icu::CharString canonical = ulocimp_canonicalize(localeID, tmpStatus);
     if (U_FAILURE(tmpStatus)) {
@@ -2604,7 +2607,7 @@ ulocimp_toLanguageTag(const char* localeID,
     }
 
     /* For handling special case - private use only tag */
-    pKeywordStart = locale_getKeywordsStart(canonical.data());
+    pKeywordStart = locale_getKeywordsStart(canonical.toStringPiece());
     if (pKeywordStart == canonical.data()) {
         int kwdCnt = 0;
         bool done = false;
@@ -2642,12 +2645,12 @@ ulocimp_toLanguageTag(const char* localeID,
         }
     }
 
-    _appendLanguageToLanguageTag(canonical.data(), sink, strict, status);
-    _appendScriptToLanguageTag(canonical.data(), sink, strict, status);
-    _appendRegionToLanguageTag(canonical.data(), sink, strict, status);
-    _appendVariantsToLanguageTag(canonical.data(), sink, strict, hadPosix, status);
+    _appendLanguageToLanguageTag(canonical.toStringPiece(), sink, strict, status);
+    _appendScriptToLanguageTag(canonical.toStringPiece(), sink, strict, status);
+    _appendRegionToLanguageTag(canonical.toStringPiece(), sink, strict, status);
+    _appendVariantsToLanguageTag(canonical.toStringPiece(), sink, strict, hadPosix, status);
     _appendKeywordsToLanguageTag(canonical.data(), sink, strict, hadPosix, status);
-    _appendPrivateuseToLanguageTag(canonical.data(), sink, strict, hadPosix, status);
+    _appendPrivateuseToLanguageTag(canonical.toStringPiece(), sink, strict, hadPosix, status);
 }
 
 

thirdparty/icu4c/common/ulocale.cpp

@@ -10,7 +10,6 @@
 #include "unicode/locid.h"
 
 #include "bytesinkutil.h"
-#include "charstr.h"
 #include "cmemory.h"
 
 U_NAMESPACE_USE
@@ -24,9 +23,7 @@ ulocale_openForLocaleID(const char* localeID, int32_t length, UErrorCode* err) {
     if (length < 0) {
         return EXTERNAL(icu::Locale::createFromName(localeID).clone());
     }
-    CharString str(localeID, length, *err);  // Make a NUL terminated copy.
-    if (U_FAILURE(*err)) { return nullptr; }
-    return EXTERNAL(icu::Locale::createFromName(str.data()).clone());
+    return EXTERNAL(icu::Locale::createFromName(StringPiece{localeID, length}).clone());
 }
 
 ULocale*

thirdparty/icu4c/common/ulocimp.h

@@ -68,42 +68,42 @@ U_EXPORT std::optional<std::string_view>
 ulocimp_toLegacyTypeWithFallback(std::string_view keyword, std::string_view value);
 
 U_EXPORT icu::CharString
-ulocimp_getKeywords(const char* localeID,
+ulocimp_getKeywords(std::string_view localeID,
                     char prev,
                     bool valuesToo,
                     UErrorCode& status);
 
 U_EXPORT void
-ulocimp_getKeywords(const char* localeID,
+ulocimp_getKeywords(std::string_view localeID,
                     char prev,
                     icu::ByteSink& sink,
                     bool valuesToo,
                     UErrorCode& status);
 
 U_EXPORT icu::CharString
-ulocimp_getName(const char* localeID,
+ulocimp_getName(std::string_view localeID,
                 UErrorCode& err);
 
 U_EXPORT void
-ulocimp_getName(const char* localeID,
+ulocimp_getName(std::string_view localeID,
                 icu::ByteSink& sink,
                 UErrorCode& err);
 
 U_EXPORT icu::CharString
-ulocimp_getBaseName(const char* localeID,
+ulocimp_getBaseName(std::string_view localeID,
                     UErrorCode& err);
 
 U_EXPORT void
-ulocimp_getBaseName(const char* localeID,
+ulocimp_getBaseName(std::string_view localeID,
                     icu::ByteSink& sink,
                     UErrorCode& err);
 
 U_EXPORT icu::CharString
-ulocimp_canonicalize(const char* localeID,
+ulocimp_canonicalize(std::string_view localeID,
                      UErrorCode& err);
 
 U_EXPORT void
-ulocimp_canonicalize(const char* localeID,
+ulocimp_canonicalize(std::string_view localeID,
                      icu::ByteSink& sink,
                      UErrorCode& err);
 
@@ -119,16 +119,16 @@ ulocimp_getKeywordValue(const char* localeID,
                         UErrorCode& status);
 
 U_EXPORT icu::CharString
-ulocimp_getLanguage(const char* localeID, UErrorCode& status);
+ulocimp_getLanguage(std::string_view localeID, UErrorCode& status);
 
 U_EXPORT icu::CharString
-ulocimp_getScript(const char* localeID, UErrorCode& status);
+ulocimp_getScript(std::string_view localeID, UErrorCode& status);
 
 U_EXPORT icu::CharString
-ulocimp_getRegion(const char* localeID, UErrorCode& status);
+ulocimp_getRegion(std::string_view localeID, UErrorCode& status);
 
 U_EXPORT icu::CharString
-ulocimp_getVariant(const char* localeID, UErrorCode& status);
+ulocimp_getVariant(std::string_view localeID, UErrorCode& status);
 
 U_EXPORT void
 ulocimp_setKeywordValue(std::string_view keywordName,
@@ -145,7 +145,7 @@ ulocimp_setKeywordValue(std::string_view keywords,
 
 U_EXPORT void
 ulocimp_getSubtags(
-        const char* localeID,
+        std::string_view localeID,
         icu::CharString* language,
         icu::CharString* script,
         icu::CharString* region,
@@ -155,7 +155,7 @@ ulocimp_getSubtags(
 
 U_EXPORT void
 ulocimp_getSubtags(
-        const char* localeID,
+        std::string_view localeID,
         icu::ByteSink* language,
         icu::ByteSink* script,
         icu::ByteSink* region,
@@ -165,7 +165,7 @@ ulocimp_getSubtags(
 
 inline void
 ulocimp_getSubtags(
-        const char* localeID,
+        std::string_view localeID,
         std::nullptr_t,
         std::nullptr_t,
         std::nullptr_t,
@@ -364,7 +364,7 @@ ulocimp_minimizeSubtags(const char* localeID,
                         UErrorCode& err);
 
 U_CAPI const char * U_EXPORT2
-locale_getKeywordsStart(const char *localeID);
+locale_getKeywordsStart(std::string_view localeID);
 
 bool
 ultag_isExtensionSubtags(const char* s, int32_t len);

thirdparty/icu4c/common/umapfile.cpp

@@ -237,8 +237,13 @@ typedef HANDLE MemoryMap;
         pData->map = (char *)data + length;
         pData->pHeader=(const DataHeader *)data;
         pData->mapAddr = data;
-#if U_PLATFORM == U_PF_IPHONE
+#if U_PLATFORM == U_PF_IPHONE || U_PLATFORM == U_PF_ANDROID
+    // Apparently supported from Android 23 and higher:
+    //   https://github.com/ggml-org/llama.cpp/pull/3631
+    // Checking for the flag itself is safer than checking for __ANDROID_API__.
+#   ifdef POSIX_MADV_RANDOM
         posix_madvise(data, length, POSIX_MADV_RANDOM);
+#   endif
 #endif
         return true;
     }

thirdparty/icu4c/common/unicode/brkiter.h

@@ -58,6 +58,8 @@ U_NAMESPACE_END
 
 U_NAMESPACE_BEGIN
 
+class CharString;
+
 /**
  * The BreakIterator class implements methods for finding the location
  * of boundaries in text. BreakIterator is an abstract base class.
@@ -646,9 +648,9 @@ protected:
 private:
 
     /** @internal (private) */
-    char actualLocale[ULOC_FULLNAME_CAPACITY];
-    char validLocale[ULOC_FULLNAME_CAPACITY];
-    char requestLocale[ULOC_FULLNAME_CAPACITY];
+    CharString* actualLocale = nullptr;
+    CharString* validLocale = nullptr;
+    CharString* requestLocale = nullptr;
 };
 
 #ifndef U_HIDE_DEPRECATED_API

thirdparty/icu4c/common/unicode/char16ptr.h

@@ -9,10 +9,13 @@
 
 #include "unicode/utypes.h"
 
-#if U_SHOW_CPLUSPLUS_API
+#if U_SHOW_CPLUSPLUS_API || U_SHOW_CPLUSPLUS_HEADER_API
 
 #include <cstddef>
 #include <string_view>
+#include <type_traits>
+
+#endif
 
 /**
  * \file
@@ -21,8 +24,6 @@
  *        Also conversion functions from char16_t * to UChar * and OldUChar *.
  */
 
-U_NAMESPACE_BEGIN
-
 /**
  * \def U_ALIASING_BARRIER
  * Barrier for pointer anti-aliasing optimizations even across function boundaries.
@@ -36,6 +37,11 @@ U_NAMESPACE_BEGIN
 #   define U_ALIASING_BARRIER(ptr)
 #endif
 
+// ICU DLL-exported
+#if U_SHOW_CPLUSPLUS_API
+
+U_NAMESPACE_BEGIN
+
 /**
  * char16_t * wrapper with implicit conversion from distinct but bit-compatible pointer types.
  * @stable ICU 59
@@ -251,6 +257,60 @@ const char16_t *ConstChar16Ptr::get() const { return u_.cp; }
 #endif
 /// \endcond
 
+U_NAMESPACE_END
+
+#endif  // U_SHOW_CPLUSPLUS_API
+
+// Usable in header-only definitions
+#if U_SHOW_CPLUSPLUS_API || U_SHOW_CPLUSPLUS_HEADER_API
+
+namespace U_ICU_NAMESPACE_OR_INTERNAL {
+
+#ifndef U_FORCE_HIDE_INTERNAL_API
+/** @internal */
+template<typename T, typename = std::enable_if_t<std::is_same_v<T, UChar>>>
+inline const char16_t *uprv_char16PtrFromUChar(const T *p) {
+    if constexpr (std::is_same_v<UChar, char16_t>) {
+        return p;
+    } else {
+#if U_SHOW_CPLUSPLUS_API
+        return ConstChar16Ptr(p).get();
+#else
+#ifdef U_ALIASING_BARRIER
+        U_ALIASING_BARRIER(p);
+#endif
+        return reinterpret_cast<const char16_t *>(p);
+#endif
+    }
+}
+#if !U_CHAR16_IS_TYPEDEF && (!defined(_LIBCPP_VERSION) || _LIBCPP_VERSION < 180000)
+/** @internal */
+inline const char16_t *uprv_char16PtrFromUint16(const uint16_t *p) {
+#if U_SHOW_CPLUSPLUS_API
+    return ConstChar16Ptr(p).get();
+#else
+#ifdef U_ALIASING_BARRIER
+    U_ALIASING_BARRIER(p);
+#endif
+    return reinterpret_cast<const char16_t *>(p);
+#endif
+}
+#endif
+#if U_SIZEOF_WCHAR_T==2
+/** @internal */
+inline const char16_t *uprv_char16PtrFromWchar(const wchar_t *p) {
+#if U_SHOW_CPLUSPLUS_API
+    return ConstChar16Ptr(p).get();
+#else
+#ifdef U_ALIASING_BARRIER
+    U_ALIASING_BARRIER(p);
+#endif
+    return reinterpret_cast<const char16_t *>(p);
+#endif
+}
+#endif
+#endif
+
 /**
  * Converts from const char16_t * to const UChar *.
  * Includes an aliasing barrier if available.
@@ -307,6 +367,15 @@ inline OldUChar *toOldUCharPtr(char16_t *p) {
     return reinterpret_cast<OldUChar *>(p);
 }
 
+}  // U_ICU_NAMESPACE_OR_INTERNAL
+
+#endif  // U_SHOW_CPLUSPLUS_API || U_SHOW_CPLUSPLUS_HEADER_API
+
+// ICU DLL-exported
+#if U_SHOW_CPLUSPLUS_API
+
+U_NAMESPACE_BEGIN
+
 #ifndef U_FORCE_HIDE_INTERNAL_API
 /**
  * Is T convertible to a std::u16string_view or some other 16-bit string view?
@@ -379,6 +448,6 @@ inline std::u16string_view toU16StringViewNullable(const T& text) {
 
 U_NAMESPACE_END
 
-#endif /* U_SHOW_CPLUSPLUS_API */
+#endif  // U_SHOW_CPLUSPLUS_API
 
 #endif  // __CHAR16PTR_H__

thirdparty/icu4c/common/unicode/locid.h

@@ -449,6 +449,11 @@ public:
      */
     static Locale U_EXPORT2 createFromName(const char *name);
 
+#ifndef U_HIDE_INTERNAL_API
+    /** @internal */
+    static Locale U_EXPORT2 createFromName(StringPiece name);
+#endif  /* U_HIDE_INTERNAL_API */
+
     /**
      * Creates a locale from the given string after canonicalizing
      * the string according to CLDR by calling uloc_canonicalize().
@@ -1133,7 +1138,9 @@ private:
      * @param cLocaleID The new locale name.
      * @param canonicalize whether to call uloc_canonicalize on cLocaleID
      */
-    Locale& init(const char* cLocaleID, UBool canonicalize);
+    Locale& init(const char* localeID, UBool canonicalize);
+    /** @internal */
+    Locale& init(StringPiece localeID, UBool canonicalize);
 
     /*
      * Internal constructor to allow construction of a locale object with

thirdparty/icu4c/common/unicode/resbund.h

@@ -450,7 +450,7 @@ public:
      * @return a Locale object
      * @stable ICU 2.8
      */
-    const Locale
+    Locale
       getLocale(ULocDataLocaleType type, UErrorCode &status) const;
 #ifndef U_HIDE_INTERNAL_API
     /**

thirdparty/icu4c/common/unicode/uchar.h

@@ -675,14 +675,14 @@ typedef enum UProperty {
      * @stable ICU 63
      */
     UCHAR_VERTICAL_ORIENTATION=0x1018,
-#ifndef U_HIDE_DRAFT_API
     /**
      * Enumerated property Identifier_Status.
      * Used for UTS #39 General Security Profile for Identifiers
      * (https://www.unicode.org/reports/tr39/#General_Security_Profile).
-     * @draft ICU 75
+     * @stable ICU 75
      */
     UCHAR_IDENTIFIER_STATUS=0x1019,
+#ifndef U_HIDE_DRAFT_API
     /**
      * Enumerated property Indic_Conjunct_Break.
      * Used in the grapheme cluster break algorithm in UAX #29.
@@ -796,7 +796,6 @@ typedef enum UProperty {
     UCHAR_SCRIPT_EXTENSIONS=0x7000,
     /** First constant for Unicode properties with unusual value types. @stable ICU 4.6 */
     UCHAR_OTHER_PROPERTY_START=UCHAR_SCRIPT_EXTENSIONS,
-#ifndef U_HIDE_DRAFT_API
     /**
      * Miscellaneous property Identifier_Type.
      * Used for UTS #39 General Security Profile for Identifiers
@@ -808,10 +807,9 @@ typedef enum UProperty {
      *
      * @see u_hasIDType
      * @see u_getIDTypes
-     * @draft ICU 75
+     * @stable ICU 75
      */
     UCHAR_IDENTIFIER_TYPE=0x7001,
-#endif  // U_HIDE_DRAFT_API
 #ifndef U_HIDE_DEPRECATED_API
     /**
      * One more than the last constant for Unicode properties with unusual value types.
@@ -2791,13 +2789,12 @@ typedef enum UVerticalOrientation {
     U_VO_UPRIGHT,
 } UVerticalOrientation;
 
-#ifndef U_HIDE_DRAFT_API
 /**
  * Identifier Status constants.
  * See https://www.unicode.org/reports/tr39/#Identifier_Status_and_Type.
  *
  * @see UCHAR_IDENTIFIER_STATUS
- * @draft ICU 75
+ * @stable ICU 75
  */
 typedef enum UIdentifierStatus {
     /*
@@ -2806,9 +2803,9 @@ typedef enum UIdentifierStatus {
      *     U_ID_STATUS_<Unicode Identifier_Status value name>
      */
 
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_STATUS_RESTRICTED,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_STATUS_ALLOWED,
 } UIdentifierStatus;
 
@@ -2817,7 +2814,7 @@ typedef enum UIdentifierStatus {
  * See https://www.unicode.org/reports/tr39/#Identifier_Status_and_Type.
  *
  * @see UCHAR_IDENTIFIER_TYPE
- * @draft ICU 75
+ * @stable ICU 75
  */
 typedef enum UIdentifierType {
     /*
@@ -2826,32 +2823,31 @@ typedef enum UIdentifierType {
      *     U_ID_TYPE_<Unicode Identifier_Type value name>
      */
 
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_NOT_CHARACTER,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_DEPRECATED,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_DEFAULT_IGNORABLE,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_NOT_NFKC,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_NOT_XID,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_EXCLUSION,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_OBSOLETE,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_TECHNICAL,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_UNCOMMON_USE,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_LIMITED_USE,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_INCLUSION,
-    /** @draft ICU 75 */
+    /** @stable ICU 75 */
     U_ID_TYPE_RECOMMENDED,
 } UIdentifierType;
-#endif  // U_HIDE_DRAFT_API
 
 /**
  * Check a binary Unicode property for a code point.
@@ -4057,7 +4053,6 @@ u_isIDStart(UChar32 c);
 U_CAPI UBool U_EXPORT2
 u_isIDPart(UChar32 c);
 
-#ifndef U_HIDE_DRAFT_API
 /**
  * Does the set of Identifier_Type values code point c contain the given type?
  *
@@ -4069,7 +4064,7 @@ u_isIDPart(UChar32 c);
  * @param c code point
  * @param type Identifier_Type to check
  * @return true if type is in Identifier_Type(c)
- * @draft ICU 75
+ * @stable ICU 75
  */
 U_CAPI bool U_EXPORT2
 u_hasIDType(UChar32 c, UIdentifierType type);
@@ -4104,11 +4099,10 @@ u_hasIDType(UChar32 c, UIdentifierType type);
  *                   function chaining. (See User Guide for details.)
  * @return number of values in c's Identifier_Type,
  *         written to types unless U_BUFFER_OVERFLOW_ERROR indicates insufficient capacity
- * @draft ICU 75
+ * @stable ICU 75
  */
 U_CAPI int32_t U_EXPORT2
 u_getIDTypes(UChar32 c, UIdentifierType *types, int32_t capacity, UErrorCode *pErrorCode);
-#endif  // U_HIDE_DRAFT_API
 
 /**
  * Determines if the specified character should be regarded

thirdparty/icu4c/common/unicode/uniset.h

@@ -1173,10 +1173,12 @@ public:
     inline U_HEADER_NESTED_NAMESPACE::USetStrings strings() const {
         return U_HEADER_NESTED_NAMESPACE::USetStrings(toUSet());
     }
+#endif  // U_HIDE_DRAFT_API
 
+#ifndef U_HIDE_DRAFT_API
     /**
      * Returns a C++ iterator for iterating over all of the elements of this set.
-     * Convenient all-in one iteration, but creates a UnicodeString for each
+     * Convenient all-in one iteration, but creates a std::u16string for each
      * code point or string.
      * (Similar to how Java UnicodeSet *is an* Iterable<String>.)
      *
@@ -1185,13 +1187,14 @@ public:
      * \code
      * UnicodeSet set(u"[abcçカ🚴{}{abc}{de}]", errorCode);
      * for (auto el : set) {
+     *     UnicodeString us(el);
      *     std::string u8;
-     *     printf("set.string length %ld \"%s\"\n", (long)el.length(), el.toUTF8String(u8).c_str());
+     *     printf("set.element length %ld \"%s\"\n", (long)us.length(), us.toUTF8String(u8).c_str());
      * }
      * \endcode
      *
      * @return an all-elements iterator.
-     * @draft ICU 76
+     * @draft ICU 77
      * @see end
      * @see codePoints
      * @see ranges
@@ -1203,7 +1206,7 @@ public:
 
     /**
      * @return an exclusive-end sentinel for iterating over all of the elements of this set.
-     * @draft ICU 76
+     * @draft ICU 77
      * @see begin
      * @see codePoints
      * @see ranges

thirdparty/icu4c/common/unicode/uset.h

@@ -32,12 +32,13 @@
 #include "unicode/utypes.h"
 #include "unicode/uchar.h"
 
-#if U_SHOW_CPLUSPLUS_API
+#if U_SHOW_CPLUSPLUS_API || U_SHOW_CPLUSPLUS_HEADER_API
+#include <string>
 #include <string_view>
 #include "unicode/char16ptr.h"
 #include "unicode/localpointer.h"
-#include "unicode/unistr.h"
-#endif   // U_SHOW_CPLUSPLUS_API
+#include "unicode/utf16.h"
+#endif
 
 #ifndef USET_DEFINED
 
@@ -1392,8 +1393,8 @@ public:
 private:
     friend class USetCodePoints;
 
-    USetCodePointIterator(const USet *uset, int32_t rangeIndex, int32_t rangeCount)
-            : uset(uset), rangeIndex(rangeIndex), rangeCount(rangeCount),
+    USetCodePointIterator(const USet *pUset, int32_t nRangeIndex, int32_t nRangeCount)
+            : uset(pUset), rangeIndex(nRangeIndex), rangeCount(nRangeCount),
                 c(U_SENTINEL), end(U_SENTINEL) {
         // Fetch the first range.
         operator++();
@@ -1429,7 +1430,7 @@ public:
      * Constructs a C++ "range" object over the code points of the USet.
      * @draft ICU 76
      */
-    USetCodePoints(const USet *uset) : uset(uset), rangeCount(uset_getRangeCount(uset)) {}
+    USetCodePoints(const USet *pUset) : uset(pUset), rangeCount(uset_getRangeCount(pUset)) {}
 
     /** @draft ICU 76 */
     USetCodePoints(const USetCodePoints &other) = default;
@@ -1460,7 +1461,7 @@ struct CodePointRange {
     /** @draft ICU 76 */
     struct iterator {
         /** @draft ICU 76 */
-        iterator(UChar32 c) : c(c) {}
+        iterator(UChar32 aC) : c(aC) {}
 
         /** @draft ICU 76 */
         bool operator==(const iterator &other) const { return c == other.c; }
@@ -1573,8 +1574,8 @@ public:
 private:
     friend class USetRanges;
 
-    USetRangeIterator(const USet *uset, int32_t rangeIndex, int32_t rangeCount)
-            : uset(uset), rangeIndex(rangeIndex), rangeCount(rangeCount) {}
+    USetRangeIterator(const USet *pUset, int32_t nRangeIndex, int32_t nRangeCount)
+            : uset(pUset), rangeIndex(nRangeIndex), rangeCount(nRangeCount) {}
 
     const USet *uset;
     int32_t rangeIndex;
@@ -1610,7 +1611,7 @@ public:
      * Constructs a C++ "range" object over the code point ranges of the USet.
      * @draft ICU 76
      */
-    USetRanges(const USet *uset) : uset(uset), rangeCount(uset_getRangeCount(uset)) {}
+    USetRanges(const USet *pUset) : uset(pUset), rangeCount(uset_getRangeCount(pUset)) {}
 
     /** @draft ICU 76 */
     USetRanges(const USetRanges &other) = default;
@@ -1657,7 +1658,7 @@ public:
             int32_t length;
             const UChar *uchars = uset_getString(uset, index, &length);
             // assert uchars != nullptr;
-            return {ConstChar16Ptr(uchars), static_cast<uint32_t>(length)};
+            return {uprv_char16PtrFromUChar(uchars), static_cast<size_t>(length)};
         }
         return {};
     }
@@ -1684,8 +1685,8 @@ public:
 private:
     friend class USetStrings;
 
-    USetStringIterator(const USet *uset, int32_t index, int32_t count)
-            : uset(uset), index(index), count(count) {}
+    USetStringIterator(const USet *pUset, int32_t nIndex, int32_t nCount)
+            : uset(pUset), index(nIndex), count(nCount) {}
 
     const USet *uset;
     int32_t index;
@@ -1699,9 +1700,11 @@ private:
  * using U_HEADER_NESTED_NAMESPACE::USetStrings;
  * LocalUSetPointer uset(uset_openPattern(u"[abcçカ🚴{}{abc}{de}]", -1, &errorCode));
  * for (auto s : USetStrings(uset.getAlias())) {
- *     UnicodeString us(s);
- *     std::string u8;
- *     printf("uset.string length %ld \"%s\"\n", (long)s.length(), us.toUTF8String(u8).c_str());
+ *     int32_t len32 = s.length();
+ *     char utf8[200];
+ *     u_strToUTF8WithSub(utf8, int32_t{sizeof(utf8) - 1}, nullptr,
+ *                        s.data(), len32, 0xFFFD, nullptr, errorCode);
+ *     printf("uset.string length %ld \"%s\"\n", long{len32}, utf8);
  * }
  * \endcode
  *
@@ -1718,7 +1721,7 @@ public:
      * Constructs a C++ "range" object over the strings of the USet.
      * @draft ICU 76
      */
-    USetStrings(const USet *uset) : uset(uset), count(uset_getStringCount(uset)) {}
+    USetStrings(const USet *pUset) : uset(pUset), count(uset_getStringCount(pUset)) {}
 
     /** @draft ICU 76 */
     USetStrings(const USetStrings &other) = default;
@@ -1737,17 +1740,19 @@ private:
     const USet *uset;
     int32_t count;
 };
+#endif  // U_HIDE_DRAFT_API
 
+#ifndef U_HIDE_DRAFT_API
 /**
  * Iterator returned by USetElements.
- * @draft ICU 76
+ * @draft ICU 77
  */
 class USetElementIterator {
 public:
-    /** @draft ICU 76 */
+    /** @draft ICU 77 */
     USetElementIterator(const USetElementIterator &other) = default;
 
-    /** @draft ICU 76 */
+    /** @draft ICU 77 */
     bool operator==(const USetElementIterator &other) const {
         // No need to compare rangeCount & end given private constructor
         // and assuming we don't compare iterators across the set being modified.
@@ -1756,26 +1761,28 @@ public:
         return uset == other.uset && c == other.c && index == other.index;
     }
 
-    /** @draft ICU 76 */
+    /** @draft ICU 77 */
     bool operator!=(const USetElementIterator &other) const { return !operator==(other); }
 
-    /** @draft ICU 76 */
-    UnicodeString operator*() const {
+    /** @draft ICU 77 */
+    std::u16string operator*() const {
         if (c >= 0) {
-            return UnicodeString(c);
+            return c <= 0xffff ?
+                std::u16string({static_cast<char16_t>(c)}) :
+                std::u16string({U16_LEAD(c), U16_TRAIL(c)});
         } else if (index < totalCount) {
             int32_t length;
             const UChar *uchars = uset_getString(uset, index - rangeCount, &length);
             // assert uchars != nullptr;
-            return UnicodeString(uchars, length);
+            return {uprv_char16PtrFromUChar(uchars), static_cast<size_t>(length)};
         } else {
-            return UnicodeString();
+            return {};
         }
     }
 
     /**
      * Pre-increment.
-     * @draft ICU 76
+     * @draft ICU 77
      */
     USetElementIterator &operator++() {
         if (c < end) {
@@ -1800,7 +1807,7 @@ public:
 
     /**
      * Post-increment.
-     * @draft ICU 76
+     * @draft ICU 77
      */
     USetElementIterator operator++(int) {
         USetElementIterator result(*this);
@@ -1811,8 +1818,8 @@ public:
 private:
     friend class USetElements;
 
-    USetElementIterator(const USet *uset, int32_t index, int32_t rangeCount, int32_t totalCount)
-            : uset(uset), index(index), rangeCount(rangeCount), totalCount(totalCount),
+    USetElementIterator(const USet *pUset, int32_t nIndex, int32_t nRangeCount, int32_t nTotalCount)
+            : uset(pUset), index(nIndex), rangeCount(nRangeCount), totalCount(nTotalCount),
                 c(U_SENTINEL), end(U_SENTINEL) {
         if (index < rangeCount) {
             // Fetch the first range.
@@ -1840,7 +1847,7 @@ private:
 
 /**
  * A C++ "range" for iterating over all of the elements of a USet.
- * Convenient all-in one iteration, but creates a UnicodeString for each
+ * Convenient all-in one iteration, but creates a std::u16string for each
  * code point or string.
  *
  * Code points are returned first, then empty and multi-character strings.
@@ -1849,15 +1856,18 @@ private:
  * using U_HEADER_NESTED_NAMESPACE::USetElements;
  * LocalUSetPointer uset(uset_openPattern(u"[abcçカ🚴{}{abc}{de}]", -1, &errorCode));
  * for (auto el : USetElements(uset.getAlias())) {
- *     std::string u8;
- *     printf("uset.string length %ld \"%s\"\n", (long)el.length(), el.toUTF8String(u8).c_str());
+ *     int32_t len32 = el.length();
+ *     char utf8[200];
+ *     u_strToUTF8WithSub(utf8, int32_t{sizeof(utf8) - 1}, nullptr,
+ *                        el.data(), len32, 0xFFFD, nullptr, errorCode);
+ *     printf("uset.element length %ld \"%s\"\n", long{len32}, utf8);
  * }
  * \endcode
  *
  * C++ UnicodeSet has member functions for iteration, including begin() and end().
  *
  * @return an all-elements iterator.
- * @draft ICU 76
+ * @draft ICU 77
  * @see USetCodePoints
  * @see USetRanges
  * @see USetStrings
@@ -1866,21 +1876,21 @@ class USetElements {
 public:
     /**
      * Constructs a C++ "range" object over all of the elements of the USet.
-     * @draft ICU 76
+     * @draft ICU 77
      */
-    USetElements(const USet *uset)
-        : uset(uset), rangeCount(uset_getRangeCount(uset)),
-            stringCount(uset_getStringCount(uset)) {}
+    USetElements(const USet *pUset)
+        : uset(pUset), rangeCount(uset_getRangeCount(pUset)),
+            stringCount(uset_getStringCount(pUset)) {}
 
-    /** @draft ICU 76 */
+    /** @draft ICU 77 */
     USetElements(const USetElements &other) = default;
 
-    /** @draft ICU 76 */
+    /** @draft ICU 77 */
     USetElementIterator begin() const {
         return USetElementIterator(uset, 0, rangeCount, rangeCount + stringCount);
     }
 
-    /** @draft ICU 76 */
+    /** @draft ICU 77 */
     USetElementIterator end() const {
         return USetElementIterator(uset, rangeCount + stringCount, rangeCount, rangeCount + stringCount);
     }

thirdparty/icu4c/common/unicode/utf8.h

@@ -124,7 +124,7 @@
  * @internal
  */
 U_CAPI UChar32 U_EXPORT2
-utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict);
+utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, int8_t strict);
 
 /**
  * Function for handling "append code point" with error-checking.
@@ -148,7 +148,7 @@ utf8_appendCharSafeBody(uint8_t *s, int32_t i, int32_t length, UChar32 c, UBool
  * @internal
  */
 U_CAPI UChar32 U_EXPORT2
-utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict);
+utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, int8_t strict);
 
 /**
  * Function for handling "skip backward one code point" with error-checking.

thirdparty/icu4c/common/unicode/utypes.h

@@ -598,12 +598,13 @@ typedef enum UErrorCode {
     U_MF_DUPLICATE_DECLARATION_ERROR, /**< The same variable is declared in more than one .local or .input declaration. @internal ICU 75 technology preview @deprecated This API is for technology preview only. */
     U_MF_OPERAND_MISMATCH_ERROR,     /**< An operand provided to a function does not have the required form for that function @internal ICU 75 technology preview @deprecated This API is for technology preview only. */
     U_MF_DUPLICATE_VARIANT_ERROR, /**< A message includes a variant with the same key list as another variant. @internal ICU 76 technology preview @deprecated This API is for technology preview only. */
+    U_MF_BAD_OPTION,             /**< An option value provided to a function does not have the required form for that option. @internal ICU 77 technology preview @deprecated This API is for technology preview only. */
 #ifndef U_HIDE_DEPRECATED_API
     /**
      * One more than the highest normal formatting API error code.
      * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420.
      */
-    U_FMT_PARSE_ERROR_LIMIT = 0x10120,
+    U_FMT_PARSE_ERROR_LIMIT = 0x10121,
 #endif  // U_HIDE_DEPRECATED_API
 
     /*

thirdparty/icu4c/common/unicode/uvernum.h

@@ -53,7 +53,7 @@
  *  This value will change in the subsequent releases of ICU
  *  @stable ICU 2.4
  */
-#define U_ICU_VERSION_MAJOR_NUM 76
+#define U_ICU_VERSION_MAJOR_NUM 77
 
 /** The current ICU minor version as an integer.
  *  This value will change in the subsequent releases of ICU
@@ -79,7 +79,7 @@
  *  This value will change in the subsequent releases of ICU
  *  @stable ICU 2.6
  */
-#define U_ICU_VERSION_SUFFIX _76
+#define U_ICU_VERSION_SUFFIX _77
 
 /**
  * \def U_DEF2_ICU_ENTRY_POINT_RENAME
@@ -132,7 +132,7 @@
  *  This value will change in the subsequent releases of ICU
  *  @stable ICU 2.4
  */
-#define U_ICU_VERSION "76.1"
+#define U_ICU_VERSION "77.1"
 
 /**
  * The current ICU library major version number as a string, for library name suffixes.
@@ -145,13 +145,13 @@
  *
  * @stable ICU 2.6
  */
-#define U_ICU_VERSION_SHORT "76"
+#define U_ICU_VERSION_SHORT "77"
 
 #ifndef U_HIDE_INTERNAL_API
 /** Data version in ICU4C.
  * @internal ICU 4.4 Internal Use Only
  **/
-#define U_ICU_DATA_VERSION "76.1"
+#define U_ICU_DATA_VERSION "77.1"
 #endif  /* U_HIDE_INTERNAL_API */
 
 /*===========================================================================

thirdparty/icu4c/common/unicode/uversion.h

@@ -125,7 +125,7 @@ typedef uint8_t UVersionInfo[U_MAX_VERSION_LENGTH];
         U_NAMESPACE_USE
 #   endif
 
-#ifndef U_HIDE_DRAFT_API
+#ifndef U_FORCE_HIDE_DRAFT_API
 /**
  * \def U_HEADER_NESTED_NAMESPACE
  * Nested namespace used inside U_ICU_NAMESPACE for header-only APIs.
@@ -150,22 +150,37 @@ typedef uint8_t UVersionInfo[U_MAX_VERSION_LENGTH];
  * @draft ICU 76
  */
 
+/**
+ * \def U_ICU_NAMESPACE_OR_INTERNAL
+ * Namespace used for header-only APIs that used to be regular C++ APIs.
+ * Different when used inside ICU to prevent public use of internal instantiations.
+ * Similar to U_HEADER_ONLY_NAMESPACE, but the public definition is the same as U_ICU_NAMESPACE.
+ * "U_ICU_NAMESPACE" or "U_ICU_NAMESPACE::internal".
+ *
+ * @draft ICU 77
+ */
+
 // The first test is the same as for defining U_EXPORT for Windows.
 #if defined(_MSC_VER) || (UPRV_HAS_DECLSPEC_ATTRIBUTE(__dllexport__) && \
                           UPRV_HAS_DECLSPEC_ATTRIBUTE(__dllimport__))
 #   define U_HEADER_NESTED_NAMESPACE header
+#   define U_ICU_NAMESPACE_OR_INTERNAL U_ICU_NAMESPACE
 #elif defined(U_COMBINED_IMPLEMENTATION) || defined(U_COMMON_IMPLEMENTATION) || \
         defined(U_I18N_IMPLEMENTATION) || defined(U_IO_IMPLEMENTATION) || \
         defined(U_LAYOUTEX_IMPLEMENTATION) || defined(U_TOOLUTIL_IMPLEMENTATION)
 #   define U_HEADER_NESTED_NAMESPACE internal
+#   define U_ICU_NAMESPACE_OR_INTERNAL U_ICU_NAMESPACE::internal
+    namespace U_ICU_NAMESPACE_OR_INTERNAL {}
+    using namespace U_ICU_NAMESPACE_OR_INTERNAL;
 #else
 #   define U_HEADER_NESTED_NAMESPACE header
+#   define U_ICU_NAMESPACE_OR_INTERNAL U_ICU_NAMESPACE
 #endif
 
 #define U_HEADER_ONLY_NAMESPACE U_ICU_NAMESPACE::U_HEADER_NESTED_NAMESPACE
 
 namespace U_HEADER_ONLY_NAMESPACE {}
-#endif  // U_HIDE_DRAFT_API
+#endif  // U_FORCE_HIDE_DRAFT_API
 
 #endif /* __cplusplus */
 

thirdparty/icu4c/common/unistr.cpp

@@ -1945,6 +1945,13 @@ UnicodeString::cloneArrayIfNeeded(int32_t newCapacity,
       growCapacity = newCapacity;
     } else if(newCapacity <= US_STACKBUF_SIZE && growCapacity > US_STACKBUF_SIZE) {
       growCapacity = US_STACKBUF_SIZE;
+    } else if(newCapacity > growCapacity) {
+      setToBogus();
+      return false;  // bad inputs
+    }
+    if(growCapacity > kMaxCapacity) {
+      setToBogus();
+      return false;
     }
 
     // save old values

thirdparty/icu4c/common/uresbund.cpp

@@ -2716,6 +2716,9 @@ ures_openWithType(UResourceBundle *r, const char* path, const char* localeID,
 
     UResourceDataEntry *entry;
     if(openType != URES_OPEN_DIRECT) {
+        if (localeID == nullptr) {
+            localeID = uloc_getDefault();
+        }
         /* first "canonicalize" the locale ID */
         CharString canonLocaleID = ulocimp_getBaseName(localeID, *status);
         if(U_FAILURE(*status)) {
@@ -3080,6 +3083,9 @@ ures_getFunctionalEquivalent(char *result, int32_t resultCapacity,
             kwVal.clear();
         }
     }
+    if (locid == nullptr) {
+        locid = uloc_getDefault();
+    }
     CharString base = ulocimp_getBaseName(locid, subStatus);
 #if defined(URES_TREE_DEBUG)
     fprintf(stderr, "getFunctionalEquivalent: \"%s\" [%s=%s] in %s - %s\n", 
@@ -3244,7 +3250,7 @@ ures_getFunctionalEquivalent(char *result, int32_t resultCapacity,
             const char *validLoc = ures_getLocaleByType(res, ULOC_VALID_LOCALE, &subStatus);
             if (U_SUCCESS(subStatus) && validLoc != nullptr && validLoc[0] != 0 && uprv_strcmp(validLoc, "root") != 0) {
                 CharString validLang = ulocimp_getLanguage(validLoc, subStatus);
-                CharString parentLang = ulocimp_getLanguage(parent.data(), subStatus);
+                CharString parentLang = ulocimp_getLanguage(parent.toStringPiece(), subStatus);
                 if (U_SUCCESS(subStatus) && validLang != parentLang) {
                     // validLoc is not root and has a different language than parent, use it instead
                     found.clear().append(validLoc, subStatus);

thirdparty/icu4c/common/uscript.cpp

@@ -59,6 +59,9 @@ getCodesFromLocale(const char *locale,
     if (U_FAILURE(*err)) { return 0; }
     icu::CharString lang;
     icu::CharString script;
+    if (locale == nullptr) {
+        locale = uloc_getDefault();
+    }
     ulocimp_getSubtags(locale, &lang, &script, nullptr, nullptr, nullptr, *err);
     if (U_FAILURE(*err)) { return 0; }
     // Multi-script languages, equivalent to the LocaleScript data

thirdparty/icu4c/common/ushape.cpp

@@ -28,6 +28,7 @@
 #include "ubidi_props.h"
 #include "uassert.h"
 
+#include <limits>
 /*
  * This implementation is designed for 16-bit Unicode strings.
  * The main assumption is that the Arabic characters and their
@@ -747,6 +748,10 @@ handleGeneratedSpaces(char16_t *dest, int32_t sourceLength,
         }
     }
 
+    if (static_cast<size_t>(sourceLength) + 1 > std::numeric_limits<size_t>::max() / U_SIZEOF_UCHAR) {
+        *pErrorCode = U_INDEX_OUTOFBOUNDS_ERROR;
+        return 0;
+    }
     tempbuffer = static_cast<char16_t*>(uprv_malloc((sourceLength + 1) * U_SIZEOF_UCHAR));
     /* Test for nullptr */
     if(tempbuffer == nullptr) {

thirdparty/icu4c/common/usprep.cpp

@@ -126,7 +126,7 @@ compareEntries(const UHashTok p1, const UHashTok p2) {
     name2.pointer = b2->name;
     path1.pointer = b1->path;
     path2.pointer = b2->path;
-    return uhash_compareChars(name1, name2) & uhash_compareChars(path1, path2);
+    return uhash_compareChars(name1, name2) && uhash_compareChars(path1, path2);
 }
 
 static void 

thirdparty/icu4c/common/utf_impl.cpp

@@ -124,11 +124,9 @@ errorValue(int32_t count, int8_t strict) {
  * >0  Obsolete "strict" behavior of UTF8_NEXT_CHAR_SAFE(..., true):
  *     Same as the obsolete "safe" behavior, but non-characters are also treated
  *     like illegal sequences.
- *
- * Note that a UBool is the same as an int8_t.
  */
 U_CAPI UChar32 U_EXPORT2
-utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict) {
+utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, int8_t strict) {
     // *pi is one after byte c.
     int32_t i=*pi;
     // length can be negative for NUL-terminated strings: Read and validate one byte at a time.
@@ -233,7 +231,7 @@ utf8_appendCharSafeBody(uint8_t *s, int32_t i, int32_t length, UChar32 c, UBool
 }
 
 U_CAPI UChar32 U_EXPORT2
-utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict) {
+utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, int8_t strict) {
     // *pi is the index of byte c.
     int32_t i=*pi;
     if(U8_IS_TRAIL(c) && i>start) {

thirdparty/icu4c/common/utypes.cpp

@@ -140,7 +140,8 @@ _uFmtErrorName[U_FMT_PARSE_ERROR_LIMIT - U_FMT_PARSE_ERROR_START] = {
     "U_MF_MISSING_SELECTOR_ANNOTATION_ERROR",
     "U_MF_DUPLICATE_DECLARATION_ERROR",
     "U_MF_OPERAND_MISMATCH_ERROR",
-    "U_MF_DUPLICATE_VARIANT_ERROR"
+    "U_MF_DUPLICATE_VARIANT_ERROR",
+    "U_MF_BAD_OPTION"
 };
 
 static const char * const

thirdparty/icu4c/i18n/scriptset.cpp

@@ -285,19 +285,19 @@ uhash_equalsScriptSet(const UElement key1, const UElement key2) {
     return (*s1 == *s2);
 }
 
-U_CAPI int8_t U_EXPORT2
+U_CAPI int32_t U_EXPORT2
 uhash_compareScriptSet(UElement key0, UElement key1) {
     icu::ScriptSet *s0 = static_cast<icu::ScriptSet *>(key0.pointer);
     icu::ScriptSet *s1 = static_cast<icu::ScriptSet *>(key1.pointer);
     int32_t diff = s0->countMembers() - s1->countMembers();
-    if (diff != 0) return static_cast<UBool>(diff);
+    if (diff != 0) return diff;
     int32_t i0 = s0->nextSetBit(0);
     int32_t i1 = s1->nextSetBit(0);
     while ((diff = i0-i1) == 0 && i0 > 0) {
         i0 = s0->nextSetBit(i0+1);
         i1 = s1->nextSetBit(i1+1);
     }
-    return (int8_t)diff;
+    return diff;
 }
 
 U_CAPI int32_t U_EXPORT2

thirdparty/icu4c/i18n/scriptset.h

@@ -74,7 +74,7 @@ class U_I18N_API ScriptSet: public UMemory {
 
 U_NAMESPACE_END
 
-U_CAPI UBool U_EXPORT2
+U_CAPI int32_t U_EXPORT2
 uhash_compareScriptSet(const UElement key1, const UElement key2);
 
 U_CAPI int32_t U_EXPORT2

thirdparty/icu4c/i18n/ucln_in.h

@@ -39,6 +39,7 @@ typedef enum ECleanupI18NType {
     UCLN_I18N_HEBREW_CALENDAR,
     UCLN_I18N_ASTRO_CALENDAR,
     UCLN_I18N_DANGI_CALENDAR,
+    UCLN_I18N_PERSIAN_CALENDAR,
     UCLN_I18N_CALENDAR,
     UCLN_I18N_TIMEZONEFORMAT,
     UCLN_I18N_TZDBTIMEZONENAMES,
@@ -62,6 +63,7 @@ typedef enum ECleanupI18NType {
     UCLN_I18N_REGION,
     UCLN_I18N_LIST_FORMATTER,
     UCLN_I18N_NUMSYS,
+    UCLN_I18N_MF2_UNISETS,
     UCLN_I18N_COUNT /* This must be last */
 } ECleanupI18NType;