ICU-22707 Age 6:2, run generate.sh

.bazeliskrc

@@ -6,4 +6,4 @@
 # for running Bazel commands while ensuring, through configuration, that only a
 # specific version of Bazel is executed.
 
-USE_BAZEL_VERSION=6.0.0
+USE_BAZEL_VERSION=7.1.1

icu4c/source/common/uchar.cpp

@@ -524,8 +524,8 @@ U_CAPI void U_EXPORT2
 u_charAge(UChar32 c, UVersionInfo versionArray) {
  if(versionArray!=nullptr) {
  uint32_t version=u_getUnicodeProperties(c, 0)>>UPROPS_AGE_SHIFT;
- versionArray[0]=(uint8_t)(version>>4);
- versionArray[1]=(uint8_t)(version&0xf);
+ versionArray[0]=(uint8_t)(version>>2);
+ versionArray[1]=(uint8_t)(version&3);
  versionArray[2]=versionArray[3]=0;
  }
 }

icu4c/source/common/uprops.h

@@ -120,7 +120,8 @@ enum {
 /*
  * Properties in vector word 0
  * Bits
- * 31..24 DerivedAge version major/minor one nibble each
+ * 31..26 Age major version (0..63)
+ * 25..24 Age minor version (0..3)
  * 23..22 3..1: Bits 21..20 & 7..0 = Script_Extensions index
  * 3: Script value from Script_Extensions
  * 2: Script=Inherited
@@ -132,7 +133,6 @@ enum {
  * 7.. 0 UScriptCode, or index to Script_Extensions
  */
 
-/* derived age: one nibble each for major and minor version numbers */
 #define UPROPS_AGE_MASK 0xff000000
 #define UPROPS_AGE_SHIFT 24
 
@@ -164,6 +164,9 @@ enum {
 
 namespace {
 
+inline constexpr uint8_t UPROPS_AGE_MAJOR_MAX = 63;
+inline constexpr uint8_t UPROPS_AGE_MINOR_MAX = 3;
+
 inline uint32_t uprops_mergeScriptCodeOrIndex(uint32_t scriptX) {
  return
  ((scriptX & UPROPS_SCRIPT_HIGH_MASK) >> UPROPS_SCRIPT_HIGH_SHIFT) |
@@ -236,6 +239,8 @@ enum {
 
 #ifdef __cplusplus
 
+namespace {
+
 // https://www.unicode.org/reports/tr39/#Identifier_Status_and_Type
 // The Identifier_Type maps each code point to a *set* of one or more values.
 // Some can be combined with others, some can only occur alone.
@@ -296,6 +301,8 @@ inline constexpr uint8_t uprops_idTypeToEncoded[] = {
  UPROPS_ID_TYPE_RECOMMENDED
 };
 
+} // namespace
+
 #endif // __cplusplus
 
 #define UPROPS_LB_MASK 0x03f00000

icu4c/source/i18n/collationfcd.cpp

@@ -148,10 +148,10 @@ const uint8_t CollationFCD::lcccIndex[2048]={
 
 const uint32_t CollationFCD::lcccBits[73]={
 0,0xffffffff,0xffff7fff,0xffff,0xf8,0xfffe0000,0xbfffffff,0xb6,0x7ff0000,0xfffff800,0x10000,0x9fc00000,0x3d9f,0x20000,0xffff0000,0x7ff,
-0x200ff800,0xfbc00000,0x3eef,0xe000000,0xff000000,0xfffffc00,0xfffffffb,0x10000000,0x1e2000,0x2000,0x40000000,0x602000,0x18000000,0x400,0x7000000,0xf00,
+0x200ff800,0xfbc00000,0x3eef,0xe000000,0xff800000,0xfffffc00,0xfffffffb,0x10000000,0x1e2000,0x2000,0x40000000,0x602000,0x18000000,0x400,0x7000000,0xf00,
 0x3000000,0x2a00000,0x3c3e0000,0xdf,0x40,0x6800000,0xe0000000,0x300000,0x100000,0x20040000,0x200,0x1800000,0x9fe00001,0xbfff0000,0x7fff,0x10,
 0xff800,0xc00,0xc0040,0x800000,0xfff70000,0x31021fd,0x1fff0000,0x1ffe2,0x38000,0x80000000,0xfc00,0x6000000,0x3ff08000,0xc0000000,0x30000,0x1000,
-0x3ffff,0x3800,0x80000,1,0xc19d0000,2,0x400000,0xc0000fd,0x7108000
+0x3ffff,0x3800,0x80000,1,0xc19d0000,2,0x400000,0xd0000fd,0x7108000
 };
 
 const uint8_t CollationFCD::tcccIndex[2048]={
@@ -288,12 +288,12 @@ const uint8_t CollationFCD::tcccIndex[2048]={
 const uint32_t CollationFCD::tcccBits[123]={
 0,0xffffffff,0x3e7effbf,0xbe7effbf,0xfffcffff,0x7ef1ff3f,0xfff3f1f8,0x7fffff3f,0x18003,0xdfffe000,0xff31ffcf,0xcfffffff,0xfffc0,0xffff7fff,0xffff,0x1d760,
 0x1fc00,0x187c00,0x200708b,0x2000000,0x708b0000,0xc00000,0xf8,0xfccf0006,0x33ffcfc,0xfffe0000,0xbfffffff,0xb6,0x7ff0000,0x7c,0xfffff800,0x10000,
-0x9fc80005,0x3d9f,0x20000,0xffff0000,0x7ff,0x200ff800,0xfbc00000,0x3eef,0xe000000,0xff000000,0xfffffc00,0xfffffffb,0x10120200,0xff1e2000,0x10000000,0xb0002000,
+0x9fc80005,0x3d9f,0x20000,0xffff0000,0x7ff,0x200ff800,0xfbc00000,0x3eef,0xe000000,0xff800000,0xfffffc00,0xfffffffb,0x10120200,0xff1e2000,0x10000000,0xb0002000,
 0x40000000,0x10480000,0x4e002000,0x2000,0x30002000,0x602100,0x18000000,0x24000400,0x7000000,0xf00,0x3000000,0x2a00000,0x3d7e0000,0xdf,0x40,0x6800000,
 0xe0000000,0x300000,0x100000,0x20040000,0x200,0x1800000,0x9fe00001,0xbfff0000,0x7fff,0x10,0xff800,0xc00,0xc0040,0x800000,0xfff70000,0x31021fd,
 0xbffffff,0x3ffffff,0x3f3fffff,0xaaff3f3f,0x3fffffff,0x1fdfffff,0xefcfffde,0x1fdc7fff,0x1fff0000,0x1ffe2,0x800,0xc000000,0x4000,0xe000,0x1210,0x50,
 0x292,0x333e005,0x333,0xf000,0x3c0f,0x38000,0x80000000,0xfc00,0x55555000,0x36db02a5,0x46100000,0x47900000,0x3ff08000,0xc0000000,0x30000,0x1000,
-0x3ffff,0x3800,0x80000,1,0xc19d0000,2,0x400000,0xc0000fd,0x7108000,0x5f7ffc00,0x7fdb
+0x3ffff,0x3800,0x80000,1,0xc19d0000,2,0x400000,0xd0000ff,0x7108000,0x5f7ffc00,0x7fdb
 };
 
 U_NAMESPACE_END

icu4c/source/tools/toolutil/swapimpl.cpp

@@ -182,7 +182,7 @@ uprops_swap(const UDataSwapper *ds,
  pInfo->dataFormat[1]==0x50 &&
  pInfo->dataFormat[2]==0x72 &&
  pInfo->dataFormat[3]==0x6f &&
- (3<=pInfo->formatVersion[0] && pInfo->formatVersion[0]<=8) &&
+ (3<=pInfo->formatVersion[0] && pInfo->formatVersion[0]<=9) &&
  (pInfo->formatVersion[0]>=7 ||
  (pInfo->formatVersion[2]==UTRIE_SHIFT &&
  pInfo->formatVersion[3]==UTRIE_INDEX_SHIFT))

icu4j/main/core/src/main/java/com/ibm/icu/impl/UCharacterProperty.java

@@ -294,10 +294,8 @@ public int getAdditional(int codepoint, int column) {
  */
  public VersionInfo getAge(int codepoint)
  {
- int version = getAdditional(codepoint, 0) >> AGE_SHIFT_;
- return VersionInfo.getInstance(
- (version >> FIRST_NIBBLE_SHIFT_) & LAST_NIBBLE_MASK_,
- version & LAST_NIBBLE_MASK_, 0, 0);
+ int version = getAdditional(codepoint, 0) >>> AGE_SHIFT_;
+ return VersionInfo.getInstance(version >> 2, version & 3, 0, 0);
  }
 
  private static final int GC_CN_MASK = getMask(UCharacter.UNASSIGNED);
@@ -1546,14 +1544,6 @@ public static final int mergeScriptCodeOrIndex(int scriptX) {
  */
  private static final int DECOMPOSITION_TYPE_MASK_ = 0x0000001f;
 
- /**
- * First nibble shift
- */
- private static final int FIRST_NIBBLE_SHIFT_ = 0x4;
- /**
- * Second nibble mask
- */
- private static final int LAST_NIBBLE_MASK_ = 0xF;
  /**
  * Age value shift
  */
@@ -1633,7 +1623,7 @@ private UCharacterProperty() throws IOException
  private static final class IsAcceptable implements ICUBinary.Authenticate {
  @Override
  public boolean isDataVersionAcceptable(byte version[]) {
- return version[0] == 8;
+ return version[0] == 9;
  }
  }
  private static final int DATA_FORMAT = 0x5550726F; // "UPro"

tools/unicode/c/genprops/corepropsbuilder.cpp

@@ -47,7 +47,7 @@ the udata API for loading ICU data. Especially, a UDataInfo structure
 precedes the actual data. It contains platform properties values and the
 file format version.
 
-The following is a description of format version 8.0 .
+The following is a description of format version 9.0 .
 
 Data contents:
 
@@ -295,6 +295,10 @@ The 6 bits in vector word 2 that stored emoji properties are unused again.
 
 ICU 75 uses the vector word 2 bits 31..26 for encoded Identifier_Type bit sets.
 
+--- Changes in format version 9.0 (ICU 76) ---
+
+Age major:minor version bit fields changed from 4:4 to 6:2 so that age=16.0 fits.
+
 ----------------------------------------------------------------------------- */
 
 U_NAMESPACE_USE
@@ -312,8 +316,8 @@ UDataInfo dataInfo={
  0,
 
  { 0x55, 0x50, 0x72, 0x6f }, /* dataFormat="UPro" */
- { 8, 0, 0, 0 }, /* formatVersion */
- { 15, 1, 0, 0 } /* dataVersion */
+ { 9, 0, 0, 0 }, /* formatVersion */
+ { 16, 0, 0, 0 } /* dataVersion */
 };
 
 inline uint32_t splitScriptCodeOrIndex(uint32_t v) {
@@ -712,9 +716,10 @@ CorePropsBuilder::setProps(const UniProps &props, const UnicodeSet &newValues,
 
  UChar32 start=props.start;
  UChar32 end=props.end;
+ UChar32 pvecEnd=end;
  if(start==0 && end==0x10ffff) {
  // Also set bits for initialValue and errorValue.
- end=UPVEC_MAX_CP;
+ pvecEnd=UPVEC_MAX_CP;
  }
 
  if(newValues.containsSome(0, UCHAR_BINARY_LIMIT-1)) {
@@ -724,7 +729,7 @@ CorePropsBuilder::setProps(const UniProps &props, const UnicodeSet &newValues,
  if(newValues.contains(p2b.prop)) {
  uint32_t mask=U_MASK(p2b.vecShift);
  uint32_t value= props.binProps[p2b.prop] ? mask : 0;
- upvec_setValue(pv, start, end, p2b.vecWord, value, mask, &errorCode);
+ upvec_setValue(pv, start, pvecEnd, p2b.vecWord, value, mask, &errorCode);
  }
  }
  }
@@ -738,20 +743,21 @@ CorePropsBuilder::setProps(const UniProps &props, const UnicodeSet &newValues,
  uint32_t mask=p2e.vecMask;
  uint32_t value=(uint32_t)(props.getIntProp(p2e.prop)<<p2e.vecShift);
  U_ASSERT((value&mask)==value);
- upvec_setValue(pv, start, end, p2e.vecWord, value, mask, &errorCode);
+ upvec_setValue(pv, start, pvecEnd, p2e.vecWord, value, mask, &errorCode);
  }
  }
  }
  if(newValues.contains(UCHAR_AGE)) {
- if(props.age[0]>15 || props.age[1]>15 || props.age[2]!=0 || props.age[3]!=0) {
+ if(props.age[0]>UPROPS_AGE_MAJOR_MAX || props.age[1]>UPROPS_AGE_MINOR_MAX ||
+ props.age[2]!=0 || props.age[3]!=0) {
  char buffer[U_MAX_VERSION_STRING_LENGTH];
  u_versionToString(props.age, buffer);
  fprintf(stderr, "genprops error: age %s cannot be encoded\n", buffer);
  errorCode=U_ILLEGAL_ARGUMENT_ERROR;
  return;
  }
- uint32_t version=(props.age[0]<<4)|props.age[1];
- upvec_setValue(pv, start, end,
+ uint32_t version=(props.age[0]<<2)|props.age[1];
+ upvec_setValue(pv, start, pvecEnd,
  0, version<<UPROPS_AGE_SHIFT, UPROPS_AGE_MASK,
  &errorCode);
  }
@@ -773,7 +779,7 @@ CorePropsBuilder::setProps(const UniProps &props, const UnicodeSet &newValues,
  // Use UPROPS_SCRIPT_X_MASK:
  // When writing a Script code, remove Script_Extensions bits as well.
  // If needed, they will get written again.
- upvec_setValue(pv, start, end, 0, value, UPROPS_SCRIPT_X_MASK, &errorCode);
+ upvec_setValue(pv, start, pvecEnd, 0, value, UPROPS_SCRIPT_X_MASK, &errorCode);
  }
  // Write a new (Script, Script_Extensions) value if there are Script_Extensions
  // and either Script or Script_Extensions are new on the current line.
@@ -820,12 +826,12 @@ CorePropsBuilder::setProps(const UniProps &props, const UnicodeSet &newValues,
  return;
  }
  scriptX|=splitScriptCodeOrIndex(index);
- upvec_setValue(pv, start, end, 0, scriptX, UPROPS_SCRIPT_X_MASK, &errorCode);
+ upvec_setValue(pv, start, pvecEnd, 0, scriptX, UPROPS_SCRIPT_X_MASK, &errorCode);
  }
  if(newValues.contains(UCHAR_IDENTIFIER_TYPE)) {
  uint32_t encodedType=encodeIdentifierType(props.idType, start==0xA9CF && start==end, errorCode);
  upvec_setValue(
- pv, start, end, 2,
+ pv, start, pvecEnd, 2,
  encodedType << UPROPS_2_ID_TYPE_SHIFT, UPROPS_2_ID_TYPE_MASK,
  &errorCode);
  }