Parse block headers (#1317)

eggrobin · web-flow · commit b0362cfbc4ab · 2026-03-24T15:44:39.000+01:00
* Parse block headers

* nonuniform unassigned

* more eggsceptions

* NO_SPLIT
diff --git a/unicodetools/src/main/java/org/unicode/props/IndexUnicodeProperties.java b/unicodetools/src/main/java/org/unicode/props/IndexUnicodeProperties.java
@@ -922,11 +922,11 @@ protected List _getValueAliases(String valueAlias, List result) {
             }
             return result;
         }
-        //        @Override
-        //        public boolean hasUniformUnassigned() {
-        //            //throw new UnsupportedOperationException();
-        //            return false;
-        //        }
+
+        @Override
+        public boolean hasUniformUnassigned() {
+            return false;
+        }
     }
 
     {
diff --git a/unicodetools/src/main/java/org/unicode/props/PropertyParsingInfo.java b/unicodetools/src/main/java/org/unicode/props/PropertyParsingInfo.java
@@ -892,6 +892,23 @@ private static void parseNamesListFile(
             IndexUnicodeProperties nextProperties,
             Set<PropertyParsingInfo> propInfoSet) {
         final var namesListChar = Pattern.compile("[0-9A-F]{4,6}");
+        final var blockHeaderPropInfo =
+                property2PropertyInfo.get(UcdProperty.Names_List_Block_Header);
+        final var nextBlockHeader =
+                nextProperties == null
+                        ? null
+                        : nextProperties.getProperty(UcdProperty.Names_List_Block_Header);
+        final UnicodeMap<String> blockHeaderData =
+                indexUnicodeProperties.property2UnicodeMap.get(UcdProperty.Names_List_Block_Header);
+        final var blockHeaderNoticePropInfo =
+                property2PropertyInfo.get(UcdProperty.Names_List_Block_Header_Notice);
+        final var nextBlockHeaderNotice =
+                nextProperties == null
+                        ? null
+                        : nextProperties.getProperty(UcdProperty.Names_List_Block_Header_Notice);
+        final UnicodeMap<String> blockHeaderNoticeData =
+                indexUnicodeProperties.property2UnicodeMap.get(
+                        UcdProperty.Names_List_Block_Header_Notice);
         final var subheaderPropInfo = property2PropertyInfo.get(UcdProperty.Names_List_Subheader);
         final var nextSubheader =
                 nextProperties == null
@@ -933,10 +950,14 @@ private static void parseNamesListFile(
 
         aliasPropInfo.multivaluedSplit = NO_SPLIT;
         commentPropInfo.multivaluedSplit = NO_SPLIT;
+        blockHeaderNoticePropInfo.multivaluedSplit = NO_SPLIT;
 
+        String blockHeader = null;
+        String blockHeaderNotice = null;
         String subheader = null;
         String subheaderNotice = null;
         IntRange codePoint = null;
+        IntRange blockRange = null;
         for (String line : lines) {
             String[] parts = line.split("\t+");
             if (parts.length == 2 && namesListChar.matcher(parts[0]).matches()) {
@@ -1000,16 +1021,31 @@ private static void parseNamesListFile(
                 codePoint = null;
             }
             if (parts.length == 4 && parts[0].equals("@@")) {
-                // New block header, clear the current subheader.
+                blockRange = new IntRange();
+                blockRange.set(parts[1] + ".." + parts[3]);
+                blockHeaderPropInfo.put(
+                        blockHeaderData,
+                        blockRange,
+                        parts[2],
+                        nextBlockHeader,
+                        indexUnicodeProperties.getUcdVersion());
+                blockHeaderNotice = null;
                 subheader = null;
                 subheaderNotice = null;
                 codePoint = null;
             }
-            if (parts.length == 2
-                    && parts[0].equals("@+")
-                    && codePoint == null
-                    && subheader != null) {
-                subheaderNotice = parts[1];
+            if (parts.length == 2 && parts[0].equals("@+") && codePoint == null) {
+                if (subheader != null) {
+                    subheaderNotice = parts[1];
+                } else if (blockRange != null) {
+                    blockHeaderNoticePropInfo.put(
+                            blockHeaderNoticeData,
+                            blockRange,
+                            parts[1],
+                            IndexUnicodeProperties.MULTIVALUED_JOINER,
+                            nextBlockHeaderNotice,
+                            indexUnicodeProperties.getUcdVersion());
+                }
             }
         }
     }
diff --git a/unicodetools/src/main/java/org/unicode/props/UcdProperty.java b/unicodetools/src/main/java/org/unicode/props/UcdProperty.java
@@ -184,6 +184,16 @@ public enum UcdProperty {
             null,
             ValueCardinality.Unordered,
             "Names_List_Alias"),
+    Names_List_Block_Header(
+            PropertyType.Miscellaneous,
+            DerivedPropertyStatus.UCDNonProperty,
+            "Names_List_Block_Header"),
+    Names_List_Block_Header_Notice(
+            PropertyType.Miscellaneous,
+            DerivedPropertyStatus.UCDNonProperty,
+            null,
+            ValueCardinality.Unordered,
+            "Names_List_Block_Header_Notice"),
     Names_List_Comment(
             PropertyType.Miscellaneous,
             DerivedPropertyStatus.UCDNonProperty,
diff --git a/unicodetools/src/main/java/org/unicode/props/UcdPropertyValues.java b/unicodetools/src/main/java/org/unicode/props/UcdPropertyValues.java
@@ -2020,6 +2020,8 @@ public static Math_Class_Ex_Values forName(String name) {
     // Named_Sequences
     // Named_Sequences_Prov
     // Names_List_Alias
+    // Names_List_Block_Header
+    // Names_List_Block_Header_Notice
     // Names_List_Comment
     // Names_List_Cross_Ref
     // Names_List_Subheader
diff --git a/unicodetools/src/main/resources/org/unicode/props/ExtraPropertyAliases.txt b/unicodetools/src/main/resources/org/unicode/props/ExtraPropertyAliases.txt
@@ -242,10 +242,12 @@ kEH_Func                 ; kEH_Func ; Provisional
 kEH_FVal                 ; kEH_FVal ; Provisional
 kEH_UniK                 ; kEH_UniK ; Provisional
 
-Names_List_Subheader        ; Names_List_Subheader ; subhead ; UCDNonProperty
-Names_List_Subheader_Notice ; Names_List_Subheader_Notice    ; UCDNonProperty
-Names_List_Alias            ; Names_List_Alias               ; UCDNonProperty
-Names_List_Comment          ; Names_List_Comment             ; UCDNonProperty
+Names_List_Block_Header        ; Names_List_Block_Header        ; UCDNonProperty
+Names_List_Block_Header_Notice ; Names_List_Block_Header_Notice ; UCDNonProperty
+Names_List_Subheader           ; Names_List_Subheader ; subhead ; UCDNonProperty
+Names_List_Subheader_Notice    ; Names_List_Subheader_Notice    ; UCDNonProperty
+Names_List_Alias               ; Names_List_Alias               ; UCDNonProperty
+Names_List_Comment             ; Names_List_Comment             ; UCDNonProperty
 
 Math_Entity_Name          ; Math_Entity_Name          ; NonUCDNonProperty
 Math_Entity_Set           ; Math_Entity_Set           ; NonUCDNonProperty
diff --git a/unicodetools/src/main/resources/org/unicode/props/IndexPropertyRegex.txt b/unicodetools/src/main/resources/org/unicode/props/IndexPropertyRegex.txt
@@ -83,6 +83,8 @@ Emoji_SB ;                    SINGLE_VALUED ;               [0-9A-F]{4}
 emoji_variation_sequence ;    SINGLE_VALUED ;               text style|emoji style
 Named_Sequences ;             SINGLE_VALUED ;               $name
 Named_Sequences_Prov ;        SINGLE_VALUED ;               $name
+Names_List_Block_Header  ;    SINGLE_VALUED ;               .*
+Names_List_Block_Header_Notice ; MULTI_VALUED ;             .*
 Names_List_Alias  ;           MULTI_VALUED ;                .*
 Names_List_Comment ;          MULTI_VALUED ;                .*
 Names_List_Cross_Ref ;        MULTI_VALUED ;                .*
diff --git a/unicodetools/src/main/resources/org/unicode/props/IndexUnicodeProperties.txt b/unicodetools/src/main/resources/org/unicode/props/IndexUnicodeProperties.txt
@@ -602,6 +602,8 @@ Unikemet ; kEH_NoRotate
 Unikemet ; kEH_UniK
 Unikemet ; kEH_AltSeq
 
+NamesList ; Names_List_Block_Header
+NamesList ; Names_List_Block_Header_Notice
 NamesList ; Names_List_Alias
 NamesList ; Names_List_Cross_Ref
 NamesList ; Names_List_Comment
diff --git a/unicodetools/src/main/resources/org/unicode/text/UCD/UnicodeInvariantTest.txt b/unicodetools/src/main/resources/org/unicode/text/UCD/UnicodeInvariantTest.txt
@@ -1548,6 +1548,46 @@ In \P{U6.0:Math_Class_Ex=None}, U6.0:Math_Class = U6.0:Math_Class_Ex
 
 OnPairsOf $code_points, EqualityOf Block ⇔ EqualityOf Pretty_Block
 
+# Names_List_Block_Header is different:
+# - for blocks without a names list, the block header gives the assigned range,
+#   not the range of the block;
+# - for planes 15 and 16, the block header with the block name only covers the
+#   last eight columns of the plane, for the chart that shows the noncharacters;
+# - for planes 1–14, there is a block header Unassigned covering the last eight
+#   columns of the plane, again for the noncharacters.
+Let $uncharted_block_ends := [
+    [
+        \p{Block=Hangul Syllables}
+        \p{Block=Tangut Supplement}
+        \p{Block=Jurchen}
+        \p{Block=/^CJK.Unified.Ideographs/}
+    ] & \p{gc=Cn}
+]
+Let $noncharacter_charts := [
+    \x{1FF80}-\x{1FFFF}
+    \x{2FF80}-\x{2FFFF}
+    \x{3FF80}-\x{3FFFF}
+    \x{4FF80}-\x{4FFFF}
+    \x{5FF80}-\x{5FFFF}
+    \x{6FF80}-\x{6FFFF}
+    \x{7FF80}-\x{7FFFF}
+    \x{8FF80}-\x{8FFFF}
+    \x{9FF80}-\x{9FFFF}
+    \x{AFF80}-\x{AFFFF}
+    \x{BFF80}-\x{BFFFF}
+    \x{CFF80}-\x{CFFFF}
+    \x{DFF80}-\x{DFFFF}
+    \x{EFF80}-\x{EFFFF}
+]
+Let $uncharted_pua_planes := [
+    \x{F0000}-\x{FFF7F}
+    \x{100000}-\x{10FF7F}
+]
+OnPairsOf [$code_points - $noncharacter_charts - $uncharted_block_ends - $uncharted_pua_planes],
+    EqualityOf Block ⇔ EqualityOf Names_List_Block_Header
+[$uncharted_block_ends $uncharted_pua_planes] ⊂ \p{Names_List_Block_Header=@none@}
+[$noncharacter_charts] = \p{Names_List_Block_Header=Unassigned}
+
 # Basic Propertywise tests.
 Ignoring Name:
 
diff --git a/unicodetools/src/test/java/org/unicode/propstest/TestInvariants.java b/unicodetools/src/test/java/org/unicode/propstest/TestInvariants.java
@@ -164,13 +164,24 @@ public void TestUniformUnassigned() {
                         UcdProperty.Names_List_Cross_Ref,
                         UcdProperty.Names_List_Comment,
                         UcdProperty.Names_List_Subheader,
-                        UcdProperty.Names_List_Subheader_Notice));
+                        UcdProperty.Names_List_Subheader_Notice,
+                        UcdProperty.Names_List_Block_Header,
+                        UcdProperty.Names_List_Block_Header_Notice));
         exceptions.putAll(
                 General_Category_Values.Private_Use,
-                Arrays.asList(UcdProperty.Age, UcdProperty.Block, UcdProperty.Pretty_Block));
+                Arrays.asList(
+                        UcdProperty.Age,
+                        UcdProperty.Block,
+                        UcdProperty.Pretty_Block,
+                        UcdProperty.Names_List_Block_Header,
+                        UcdProperty.Names_List_Block_Header_Notice));
         exceptions.putAll(
                 General_Category_Values.Surrogate,
-                Arrays.asList(UcdProperty.Block, UcdProperty.Pretty_Block));
+                Arrays.asList(
+                        UcdProperty.Block,
+                        UcdProperty.Pretty_Block,
+                        UcdProperty.Names_List_Block_Header,
+                        UcdProperty.Names_List_Block_Header_Notice));
 
         List<UcdProperty> ordered = new ArrayList<>();
         // ordered.add(UcdProperty.Bidi_Class);

Original file line number	Diff line number	Diff line change
`@@ -922,11 +922,11 @@ protected List _getValueAliases(String valueAlias, List result) {`
`922`	`922`	`}`
`923`	`923`	`return result;`
`924`	`924`	`}`
`925`		`- // @Override`
`926`		`- // public boolean hasUniformUnassigned() {`
`927`		`- // //throw new UnsupportedOperationException();`
`928`		`- // return false;`
`929`		`- // }`
	`925`	`+`
	`926`	`+ @Override`
	`927`	`+ public boolean hasUniformUnassigned() {`
	`928`	`+ return false;`
	`929`	`+ }`
`930`	`930`	`}`
`931`	`931`
`932`	`932`	`{`