Update ICU to 78 - Part 4: MS patches, test expectations, and patch records

.gitattributes

@@ -87,3 +87,12 @@ configure text
 # Use JSONC for syntax highlighting on GitHub.com
 *.json linguist-language=jsonc
 
+# Codepoint trie golden tests compare bytes; keep generated TOML line endings stable.
+icu/icu4c/source/test/testdata/codepointtrie/*.toml text eol=lf
+
+# Shell script fragments assembled into icu-config (an executable shell script).
+# They must have LF line endings so WSL/Linux can execute the generated file.
+icu/icu4c/source/config/icu-config-top    text eol=lf
+icu/icu4c/source/config/icu-config-bottom text eol=lf
+icu/icu4c/source/config/mh-*              text eol=lf
+

icu-patches/dev_report.md

@@ -0,0 +1,74 @@
+# ICU Upgrade Dev Report
+
+Generated: 2026-05-19 09:01:27
+
+## 1. Version Summary
+
+| Field | Value |
+|-------|-------|
+| ICU Version | 78.3.0.0 |
+| Upstream Hash | `21d1eb0f30...` |
+
+## 2. Source Changes
+
+Diff from `icu72-pre-swap` to HEAD (icu/ directory only):
+
+```
+ 6919 files changed, 693820 insertions(+), 545058 deletions(-)
+```
+
+### Top file types changed
+
+| Extension | Count |
+|-----------|-------|
+| .txt | 5117 |
+| .cpp | 799 |
+| .h | 552 |
+| .c | 105 |
+| .vcxproj | 77 |
+| .json | 46 |
+| .filters | 30 |
+| .xml | 29 |
+| .java | 28 |
+| .in | 25 |
+
+## 3. Patch Audit Results
+
+*No audit report found in `icu-patches`. Final patch records are summarized from `icu-patches/patches` instead.*
+
+- Latest patch-record commit: `70ae0ef8eab Refresh patch record for Saudi Riyal symbol override`
+- Patch files present: 20
+- Refreshed ICU 78 patch records: `002`, `017`, `018`, `020`, `022`, `023`
+- Validation used for refreshed records: `git apply --cached --check --reverse` against the final ICU index
+
+## 4. Build Results
+
+| Config | Status | Log |
+|--------|--------|-----|
+| Debug-ARM64 | ✅ PASS | build-Debug-ARM64.log |
+| Debug-Win32 | ✅ PASS | build-Debug-Win32.log |
+| Debug-x64 | ✅ PASS | build-Debug-x64.log |
+| Release-ARM64 | ✅ PASS | build-Release-ARM64.log |
+| Release-Win32 | ✅ PASS | build-Release-Win32.log |
+| Release-x64 | ✅ PASS | build-Release-x64.log |
+
+## 5. Test Results
+
+| Suite | Status | Log |
+|-------|--------|-----|
+| cintltst-Release-x64 | ✅ PASS | test-cintltst-Release-x64.log |
+| intlRelease-x64 | ✅ PASS | test-intltest-Release-x64.log |
+| ioRelease-x64 | ✅ PASS | test-iotest-Release-x64.log |
+
+Post-report targeted validation for patch `023`: `genrb` parsed `icu/icu4c/source/data/curr/ar_SA.txt`, and `currtest.c` passed C syntax checking with the available compiler.
+
+## 6. Action Items
+
+- [x] Patch conflicts resolved and ICU 78 patch records refreshed
+- [x] Saudi Riyal symbol override for `ar_SA` added and recorded as patch `023`
+- [x] CLDR data rebuild consumed by ICU data generation
+- [x] Timezone data check/regeneration completed
+- [x] Full Stage 8 build/test pass completed
+- [x] Refreshed dev report generated
+- [ ] Create PR/PR stack (owner-handled)
+

icu-patches/patches/017-MSFT-Patch_ICU_test_changes_for_extra_CLDR-MS_locales.patch

@@ -1,131 +1,58 @@
 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Jeff Genovy <29107334+jefgen@users.noreply.github.com>
-Date: Wed, 7 Apr 2021 20:47:08 -0700
-Subject: MSFT-PATCH: Modify ICU tests to pass with extra locales from CLDR-MS.
+From: Arvind Oruganti <aoruganti@microsoft.com>
+Date: Tue, 19 May 2026 06:45:00 +0530
+Subject: MSFT-PATCH: Refresh ICU tests for extra CLDR-MS locales on ICU 78
 
-This changes modifies the ICU tests to pass with extra locales from
-CLDR-MS.
+Refreshes the CLDR-MS extra-locale test expectations for ICU 78.
+
+The ICU 72-era patch no longer matched the refactored ICU 78 tests, so the applicable 029 locale expectations were re-derived from the final passing Stage 7/8 tree.
 
 diff --git a/icu/icu4c/source/test/cintltst/cldrtest.c b/icu/icu4c/source/test/cintltst/cldrtest.c
-index 179539d5178b5cfd8b5350043c539e89e8740185..bfb35c35a449f60c61c9a8eb6282c307e41fd161 100644
+index e014a1e1d59a48fb6ef78923fb7de4160f69779b..8cf8e5dc1ba8f3059e4ce6b0d1bb6369cda51ca7 100644
 --- a/icu/icu4c/source/test/cintltst/cldrtest.c
 +++ b/icu/icu4c/source/test/cintltst/cldrtest.c
-@@ -970,7 +970,14 @@ static void VerifyTranslation(void) {
-             uprv_strncmp(currLoc, "oc", 2) == 0 || uprv_strncmp(currLoc, "oc_FR", 5) == 0     ||
-             uprv_strncmp(currLoc, "syr", 3) == 0 || uprv_strncmp(currLoc, "syr_SY", 6) == 0   ||
-             uprv_strncmp(currLoc, "tig", 3) == 0 || uprv_strncmp(currLoc, "tig_ER", 6) == 0   ||
--            uprv_strncmp(currLoc, "wal", 3) == 0 || uprv_strncmp(currLoc, "wal_ET", 6) == 0
-+            uprv_strncmp(currLoc, "wal", 3) == 0 || uprv_strncmp(currLoc, "wal_ET", 6) == 0   ||
-+            uprv_strncmp(currLoc, "bin", 3) == 0 || uprv_strncmp(currLoc, "bin_NG", 6) == 0   ||
-+            uprv_strncmp(currLoc, "la", 2) == 0 || uprv_strncmp(currLoc, "la_VA", 6) == 0    ||
-+            uprv_strncmp(currLoc, "jv_Java", 7) == 0 || uprv_strncmp(currLoc, "jv_Java_ID", 10) == 0   ||
-+            uprv_strncmp(currLoc, "ks_Deva", 7) == 0 || uprv_strncmp(currLoc, "ks_Deva_IN", 10) == 0   ||
-+            uprv_strncmp(currLoc, "pap", 3) == 0 || uprv_strncmp(currLoc, "pap_029", 7) == 0           ||
-+            uprv_strncmp(currLoc, "tzm_Arab", 8) == 0 || uprv_strncmp(currLoc, "tzm_Arab_MA", 11) == 0 ||
-+            uprv_strncmp(currLoc, "tzm_Tfng", 8) == 0 || uprv_strncmp(currLoc, "tzm_Tfng_MA", 11) == 0
-         ) {
-             log_knownIssue("0", "MSFT Change: skipping test for %s which has issues due to CLDR Seed data.", currLoc);
-         }
-@@ -1110,7 +1117,8 @@ static void VerifyTranslation(void) {
+@@ -1122,7 +1122,7 @@ static void VerifyTranslation(void) {
                 if (U_FAILURE(errorCode)) {
                     log_err("ulocdata_getMeasurementSystem failed for locale %s with error: %s \n", currLoc, u_errorName(errorCode));
                 } else {
 -                   if ( strstr(fullLoc, "_US")!=NULL || strstr(fullLoc, "_LR")!=NULL ) {
-+                   /* MSFT Change: CLDR-MS adds 029 region with US measurement */
 +                   if ( strstr(fullLoc, "_US")!=NULL || strstr(fullLoc, "_LR")!=NULL || strstr(fullLoc, "_029")!=NULL ) {
                         if(measurementSystem != UMS_US){
                              log_err("ulocdata_getMeasurementSystem did not return expected data for locale %s \n", currLoc);
                         }
-@@ -1128,10 +1136,11 @@ static void VerifyTranslation(void) {
-                if (U_FAILURE(errorCode)) {
-                    log_err("ulocdata_getPaperSize failed for locale %s with error: %s \n", currLoc, u_errorName(errorCode));
-                } else {
-+                   /* MSFT Change: CLDR-MS adds 029 region with US paper size */
+@@ -1143,7 +1143,7 @@ static void VerifyTranslation(void) {
                     if ( strstr(fullLoc, "_US")!=NULL || strstr(fullLoc, "_BZ")!=NULL || strstr(fullLoc, "_CA")!=NULL || strstr(fullLoc, "_CL")!=NULL ||
                          strstr(fullLoc, "_CO")!=NULL || strstr(fullLoc, "_CR")!=NULL || strstr(fullLoc, "_GT")!=NULL || strstr(fullLoc, "_MX")!=NULL ||
                          strstr(fullLoc, "_NI")!=NULL || strstr(fullLoc, "_PA")!=NULL || strstr(fullLoc, "_PH")!=NULL || strstr(fullLoc, "_PR")!=NULL ||
 -                        strstr(fullLoc, "_SV")!=NULL || strstr(fullLoc, "_VE")!=NULL ) {
-+                        strstr(fullLoc, "_SV")!=NULL || strstr(fullLoc, "_VE")!=NULL || strstr(fullLoc, "_029")!=NULL ) {
++                         strstr(fullLoc, "_SV")!=NULL || strstr(fullLoc, "_VE")!=NULL || strstr(fullLoc, "_029")!=NULL ) {
                         if (height != 279 || width != 216) {
                              log_err("ulocdata_getPaperSize did not return expected data for locale %s \n", currLoc);
                         }
-@@ -1251,8 +1260,14 @@ static void TestExemplarSet(void){
-
-             if (existsInScript == FALSE){
-                 /* MSFT Change */
--                if (uprv_strncmp(locale, "oc", 2) == 0 || uprv_strncmp(locale, "oc_FR", 5) == 0) {
--                    log_knownIssue("0", "MSFT Change: oc and oc_FR have ExemplarSet issues.");
-+                if (uprv_strncmp(locale, "oc", 2) == 0 || uprv_strncmp(locale, "oc_FR", 5) == 0 ||
-+                    uprv_strncmp(locale, "jv_Java", 7) == 0 || uprv_strncmp(locale, "jv_Java_ID", 10) == 0 ||
-+                    uprv_strncmp(locale, "la", 2) == 0 || uprv_strncmp(locale, "la_VA", 6) == 0 ||
-+                    uprv_strncmp(locale, "pap", 3) == 0 || uprv_strncmp(locale, "pap_029", 7) == 0 ||
-+                    uprv_strncmp(locale, "tzm_Arab", 8) == 0 || uprv_strncmp(locale, "tzm_Arab_MA", 11) == 0 ||
-+                    uprv_strncmp(locale, "tzm_Tfng", 8) == 0 || uprv_strncmp(locale, "tzm_Tfng_MA", 11) == 0
-+                ) {
-+                    log_knownIssue("0", "MSFT Change: skipping test for %s which has issue due to CLDR seed data.", locale);
-                     continue;
-                 } else {
-                     log_err("ExemplarSet containment failed for locale : %s\n", locale);
 diff --git a/icu/icu4c/source/test/cintltst/cnmdptst.c b/icu/icu4c/source/test/cintltst/cnmdptst.c
-index 99a9ce23be7ef93923493b2b8312a254706c3f2d..a319b99e3faf6f6fc749e70852a298ce42268696 100644
+index 98504e0220ffb26a5f02877e1bbf7343e07e5a64..b7ce012361befc52b1b23e068017ffb71bdc4b4a 100644
 --- a/icu/icu4c/source/test/cintltst/cnmdptst.c
 +++ b/icu/icu4c/source/test/cintltst/cnmdptst.c
-@@ -811,7 +811,8 @@ static void TestGetKeywordValuesForLocale(void) {
+@@ -813,7 +813,7 @@ static void TestGetKeywordValuesForLocale(void) {
              { "und",                "USD", "USN", NULL },
   /*           { "und_ZZ",             "USD", NULL, NULL },  -- temporarily remove as this locale now has 15 entries */
              { "en_US",              "USD", "USN", NULL },
 -            { "en_029",             "USD", "USN", NULL },
-+        /* MSFT Change: CLDR-MS adds en_029 locale with XCD currency */
 +            { "en_029",             "XCD", NULL, NULL },
              { "en_TH",              "THB", NULL, NULL },
              { "de",                 "EUR", NULL, NULL },
              { "de_DE",              "EUR", NULL, NULL },
-@@ -825,11 +826,13 @@ static void TestGetKeywordValuesForLocale(void) {
+@@ -827,11 +827,11 @@ static void TestGetKeywordValuesForLocale(void) {
              { "en_US@currency=CAD;rg=THZZZZ", "THB", NULL, NULL },
      };
      const int32_t EXPECTED_SIZE[PREFERRED_SIZE] = {
 -            2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 1, 1, 1, 2, 1
-+        /* MSFT Change: CLDR-MS adds en_029 locale with XCD currency */
 +            2, 2, 2, 1, 1, 1, 1, 1, 2, 2, 1, 1, 1, 2, 1
      };
      /* ucurr_forLocale results for same locales; "" if no result expected */
      const char *FORLOCALE[PREFERRED_SIZE] = {
 -            "",    "",    "USD", "",
-+        /* MSFT Change: CLDR-MS adds en_029 locale with XCD currency */
 +            "",    "",    "USD", "XCD",
              "THB", "",    "EUR", "",
              "ILS", "CAD", "ZZZ", "DEM",
              "THB", "USD", "CAD"
-diff --git a/icu/icu4c/source/test/intltest/dtptngts.cpp b/icu/icu4c/source/test/intltest/dtptngts.cpp
-index dc86355f93fa68f509a0023dc95fc17a548170c1..c83e27bb68c212c3e9569c71ed5b417a96b7f9de 100644
---- a/icu/icu4c/source/test/intltest/dtptngts.cpp
-+++ b/icu/icu4c/source/test/intltest/dtptngts.cpp
-@@ -1408,7 +1408,8 @@ void IntlTestDateTimePatternGeneratorAPI::testJjMapping() {
-                         uprv_strncmp(localeID, "ssy_ER", 6) == 0 || uprv_strncmp(localeID, "st_LS", 5) == 0   ||
-                         uprv_strncmp(localeID, "syr", 3) == 0 || uprv_strncmp(localeID, "syr_SY", 6) == 0   ||
-                         uprv_strncmp(localeID, "tig", 3) == 0 || uprv_strncmp(localeID, "tig_ER", 6) == 0   ||
--                        uprv_strncmp(localeID, "wal", 3) == 0 || uprv_strncmp(localeID, "wal_ET", 6) == 0
-+                        uprv_strncmp(localeID, "wal", 3) == 0 || uprv_strncmp(localeID, "wal_ET", 6) == 0  ||
-+                        uprv_strncmp(localeID, "ks_Deva", 7) == 0 || uprv_strncmp(localeID, "ks_Deva_IN", 10) == 0
-                      ) {
-                          logKnownIssue("0", "MSFT Change: skipping test for %s which has issues due to CLDR Seed data.", localeID);
-                      }
-diff --git a/icu/icu4c/source/test/intltest/numfmtst.cpp b/icu/icu4c/source/test/intltest/numfmtst.cpp
-index bc072bd8e3775554b8a8f96a81baf850b5e8feb7..a63c329c8eabf287512b9c152ac91d862d333264 100644
---- a/icu/icu4c/source/test/intltest/numfmtst.cpp
-+++ b/icu/icu4c/source/test/intltest/numfmtst.cpp
-@@ -8094,10 +8094,12 @@ void NumberFormatTest::TestAccountingCurrency() {
-         (Formattable)(double)-1000.5, UnicodeString("(\\uFFE51,000)").unescape(), FALSE, status);
-     expect(NumberFormat::createInstance("de_DE", style, status),
-         (Formattable)(double)-23456.7, UnicodeString("-23.456,70\\u00A0\\u20AC").unescape(), TRUE, status);
-+    /* MSFT Change: We add a CLDR-MS locale en_ID */
-     expect(NumberFormat::createInstance("en_ID", style, status),
--        (Formattable)(double)0, UnicodeString("IDR\\u00A00.00").unescape(), TRUE, status);
-+        (Formattable)(double)0, UnicodeString("IDR\\u00A00,00").unescape(), TRUE, status);
-     expect(NumberFormat::createInstance("en_ID", style, status),
--        (Formattable)(double)-0.2, UnicodeString("(IDR\\u00A00.20)").unescape(), TRUE, status);
-+        (Formattable)(double)-0.2, UnicodeString("(IDR\\u00A00,20)").unescape(), TRUE, status);
-+    /* MSFT Change: End */
-     expect(NumberFormat::createInstance("sh_ME", style, status),
-         (Formattable)(double)0, UnicodeString("0,00\\u00A0\\u20AC").unescape(), TRUE, status);
-     expect(NumberFormat::createInstance("sh_ME", style, status),

icu-patches/patches/018-MSFT-Patch_ICU_toolutil_increase_string_store_for_extra_locales.patch

@@ -1,19 +1,22 @@
 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Jeff Genovy <29107334+jefgen@users.noreply.github.com>
-Date: Wed, 7 Apr 2021 20:47:31 -0700
-Subject: MSFT-PATCH: Increase the STRING_STORE_SIZE due to extra locales from CLDR-MS
+From: Arvind Oruganti <aoruganti@microsoft.com>
+Date: Tue, 19 May 2026 06:45:00 +0530
+Subject: MSFT-PATCH: Increase ICU string store for CLDR 48 extra locales
 
+Increases STRING_STORE_SIZE for the larger CLDR 48 / CLDR-MS locale set.
+
+The original 120000 value was carried from the older patch, but this ICU 78 upgrade settled on 200000 before the CLDR data build.
 
 diff --git a/icu/icu4c/source/tools/toolutil/package.h b/icu/icu4c/source/tools/toolutil/package.h
-index ea60c13a74a57b94057882e39b6aa05314c4ca1f..6c411ae5fd04c8463a089307efc056f4d2f622f9 100644
+index ea60c13a74a57b94057882e39b6aa05314c4ca1f..3a1d3cd23d9ee75ad2881d52c0f34413dbd554f1 100644
 --- a/icu/icu4c/source/tools/toolutil/package.h
 +++ b/icu/icu4c/source/tools/toolutil/package.h
 @@ -27,7 +27,7 @@
 
  // .dat package file representation ---------------------------------------- ***
 
 -#define STRING_STORE_SIZE 100000
-+#define STRING_STORE_SIZE 120000
++#define STRING_STORE_SIZE 200000
  #define MAX_PKG_NAME_LENGTH 64
 
  typedef void CheckDependency(void *context, const char *itemName, const char *targetName);