6.5.3 clean

util/cmake/pro2cmake.py

@@ -4666,7 +4666,7 @@ def create_top_level_cmake_conf():
     conf_file_name = ".cmake.conf"
     try:
         with open(conf_file_name, "x") as file:
-            file.write('set(QT_REPO_MODULE_VERSION "6.5.1")\n')
+            file.write('set(QT_REPO_MODULE_VERSION "6.5.3")\n')
     except FileExistsError:
         pass
 

util/locale_database/cldr.py

@@ -356,7 +356,6 @@ class CldrAccess (object):
     def __checkEnum(given, proper, scraps,
                     remap = { 'å': 'a', 'ã': 'a', 'ç': 'c', 'é': 'e', 'í': 'i', 'ü': 'u'},
                     prefix = { 'St.': 'Saint', 'U.S.': 'United States' },
-                    suffixes = ( 'Han', ),
                     skip = '\u02bc'):
         # Each is a { code: full name } mapping
         for code, name in given.items():
@@ -376,8 +375,6 @@ class CldrAccess (object):
                 try: f, t = ok.index('('), ok.index(')')
                 except ValueError: break
                 ok = ok[:f].rstrip() + ' ' + ok[t:].lstrip()
-            if any(name == ok + ' ' + s for s in suffixes):
-                continue
             if ''.join(ch for ch in name.lower() if not ch.isspace()) in ''.join(
                 remap.get(ch, ch) for ch in ok.lower() if ch.isalpha() and ch not in skip):
                 continue
@@ -448,7 +445,7 @@ enumdata.py (keeping the old name as an alias):
             wid, code = attrs['other'], attrs['territory']
             data = dict(windowsId = wid,
                         territoryCode = code,
-                        ianaList = attrs['type'])
+                        ianaList = ' '.join(attrs['type'].split()))
 
             try:
                 key = lookup[wid]
@@ -699,7 +696,13 @@ enumdata.py (keeping the old name as an alias):
             except (KeyError, ValueError, TypeError):
                 pass
             else:
-                if key not in seen or 'alt' not in elt.attributes:
+                # Prefer stand-alone forms of names when present, ignore other
+                # alt="..." entries. For example, Traditional and Simplified
+                # Han omit "Han" in the plain form, but include it for
+                # stand-alone. As the stand-alone version appears later, it
+                # over-writes the plain one.
+                if (key not in seen or 'alt' not in elt.attributes
+                    or elt.attributes['alt'].nodeValue == 'stand-alone'):
                     yield key, value
                     seen.add(key)
 
@@ -708,7 +711,8 @@ enumdata.py (keeping the old name as an alias):
     def __parentLocale(self, cache = {}):
         # see http://www.unicode.org/reports/tr35/#Parent_Locales
         if not cache:
-            for tag, attrs in self.__supplementalData.find('parentLocales'):
+            for tag, attrs in self.__supplementalData.find('parentLocales',
+                                                           ('component',)):
                 parent = attrs.get('parent', '')
                 for child in attrs['locales'].split():
                     cache[child] = parent

util/locale_database/enumdata.py

@@ -361,13 +361,21 @@ language_map = {
     329: ("Nheengatu",                   "yrl"),
     # added in CLDR v42
     330: ("Haryanvi",                    "bgc"),
-    331: ("Moksha",                      "mdf"),
-    332: ("Northern Frisian",            "frr"),
-    333: ("Obolo",                       "ann"),
-    334: ("Pijin",                       "pis"),
-    335: ("Rajasthani",                  "raj"),
-    336: ("Toki Pona",                   "tok"),
+    331: ("Northern Frisian",            "frr"),
+    332: ("Rajasthani",                  "raj"),
+    333: ("Moksha",                      "mdf"),
+    334: ("Toki Pona",                   "tok"),
+    335: ("Pijin",                       "pis"),
+    336: ("Obolo",                       "ann"),
+    # added in CLDR v43
+    337: ("Baluchi",                     "bal"),
+    338: ("Ligurian",                    "lij"),
+    339: ("Rohingya",                    "rhg"),
+    340: ("Torwali",                     "trw"),
 }
+# Don't add languages just because they exist; check CLDR does provide
+# substantial data for locales using it; and check, once added, they
+# don't show up in cldr2qlocalexmo.py's unused listing.
 
 language_aliases = {
     # Renamings prior to Qt 6.0 (CLDR v37):
@@ -837,6 +845,8 @@ script_map = {
     139: ("Vai",                    "Vaii"),
     140: ("Varang Kshiti",          "Wara"),
     141: ("Yi",                     "Yiii"),
+    # Added at CLDR v43
+    142: ("Hanifi",                 "Rohg"), # Used for Rohingya
 }
 
 script_aliases = {

util/locale_database/ldml.py

@@ -166,10 +166,23 @@ class XmlScanner (object):
         return elts
 
 class Supplement (XmlScanner):
-    def find(self, xpath):
+    def find(self, xpath, exclude=()):
+        """Finds nodes by matching a specified xpath.
+
+        If exclude is passed, it should be a sequence of attribute names (its
+        default is empty). Any matches to the given xpath that also have any
+        attribute in this sequence will be excluded.
+
+        For each childless node matching the xpath, or child of a node matching
+        the xpath, this yields a twople (name, attrs) where name is the
+        nodeName and attrs is a dict mapping the node's attribute's names to
+        their values. For attribute values that are not simple strings, the
+        nodeValue of the attribute node is used."""
         elts = self.findNodes(xpath)
-        for elt in _iterateEach(e.dom.childNodes if e.dom.childNodes else (e.dom,)
-                                for e in elts):
+        for elt in _iterateEach(e.dom.childNodes or (e.dom,)
+                                for e in elts
+                                if not any(a in e.dom.attributes
+                                           for a in exclude)):
             if elt.attributes:
                 yield (elt.nodeName,
                        dict((k, v if isinstance(v, str) else v.nodeValue)
@@ -257,7 +270,13 @@ class LocaleScanner (object):
         stem = f'numbers/symbols[numberSystem={system}]/'
         decimal = self.find(f'{stem}decimal')
         group = self.find(f'{stem}group')
-        assert decimal != group, (self.name, system, decimal)
+        if decimal == group:
+            # mn_Mong_MN @v43 :-(
+            clean = Node.draftScore('approved')
+            decimal = self.find(f'{stem}decimal', draft=clean)
+            group = self.find(f'{stem}group', draft=clean)
+            assert decimal != group, (self.name, system, decimal)
+
         yield 'decimal', decimal
         yield 'group', group
         yield 'percent', self.find(f'{stem}percentSign')

util/locale_database/qlocalexml.py

@@ -324,9 +324,9 @@ class QLocaleXmlWriter (object):
         self.__languages = set(p[1] for p in language_map.values()
                                if not p[1].isspace())
         self.__scripts = set(p[1] for p in script_map.values()
-                             if p[1] != 'ZZ')
+                             if p[1] != 'Zzzz')
         self.__territories = set(p[1] for p in territory_map.values()
-                                 if p[1] != 'Zzzz')
+                                 if p[1] != 'ZZ')
 
     def likelySubTags(self, entries):
         self.__openTag('likelySubtags')
@@ -360,7 +360,7 @@ class QLocaleXmlWriter (object):
             self.__write('</localeDatabase>')
         self.__rawOutput = self.__complain
 
-        if self.__languages or self.__scripts or self.territories:
+        if self.__languages or self.__scripts or self.__territories:
             grumble('Some enum members are unused, corresponding to these tags:\n')
             import textwrap
             def kvetch(kind, seq, g = grumble, w = textwrap.wrap):