feat(char): D-561 — Character/getName + codePointOf over the generated UCD name table

Smell-audited: 2: the first fixed-width index cost +394KB and was
reworked to a delta-varint wire form BEFORE landing (measure-first,
ADR-0172 discipline); final delta mac +279KB (7.35MB) with the README
claim moved to "about 7.8 MB" and the ADR-0172 revision noting Linux at
~94% of ceiling (next size-relevant work measures Linux first).

- scripts/gen_unicode_names.py: UCD 16.0.0 → word-indexed name table
  (40,013 per-cp names; frequency-ordered dict; CPHEX sentinel folds
  the hex-suffixed families' own-hex tokens out of the dictionary;
  pre-compressed raw-deflate blobs — matches the runtime's flate; the
  index is a delta-varint stream decoded once into fixed records).
- src/runtime/char_name.zig: process-lifetime lazy tables + JVM-oracle-
  exact semantics — stored names; pinned control aliases
  (scripts/data/jvm_control_names.txt, captured from live clj 1.12:
  the JVM mixes control-type aliases with abbreviations, 0x01="START OF
  HEADING" but 0x07="BEL"); assigned-but-unnamed = "<BLOCK NAME> <HEX>"
  ("TANGUT 17000" — the JVM does NOT synthesize UCD algorithmic names);
  unassigned nil. codePointOf = trimmed case-insensitive inverse
  accepting exactly getName's output forms (block-hex confirmed by
  forward equality; "HANGUL SYLLABLE GA" rejected like the JVM).
- Character.zig: getName realized (the last feature_not_supported stub
  on the class), codePointOf added; compat_tiers row updated (the
  member gate enforced it). Corpus char_names.txt 10/10; unit tests on
  the oracle values incl. controls + block-hex + unassigned.
- D-561 discharged (active -> discharged).

Author: chaploud

.dev/debt.yaml

@@ -49,24 +49,6 @@ active:
       lastModified/renameTo/createNewFile set, Locale getters, Matcher
       region/replace family).
     last_reviewed: "2026-07-17"
-  - id: "D-561"
-    status: |-
-      open (2026-07-15, Character/* completion cycle). `Character/getName` is
-      the ONE java.lang.Character static not carried (explicit
-      feature_not_supported raise; every other static method + field landed
-      full-Unicode via the generated UCD tables — gen_unicode_case.py).
-      getName needs the UnicodeData col-1 NAME table: raw ~1MB for ~40K
-      non-algorithmic names; the JVM ships it word-indexed at ~90KB. DISCHARGE
-      SHAPE: extend gen_unicode_case.py with a word-dictionary encoding
-      (names → word-index sequences + algorithmic CJK/Hangul/Tangut ranges),
-      decode in a new charset helper; JVM parity incl. nil for unassigned.
-      Weigh the ~100-200KB binary cost against gap area II (Wasm/edge binary
-      size) before landing — if the cost is judged not worth it, re-classify
-      as an AD-NNN accepted divergence instead (the raise is explicit, not
-      silent).
-    category: "host-surface"
-    barrier: "none — implementable any time; the open question is the binary-size trade (gap area II), not a technical blocker"
-    last_reviewed: "2026-07-15"
   - id: "D-560"
     status: |-
       open (2026-07-13, ADR-0150 am1 accepted-cost row). The D-559 fix (the
@@ -2394,6 +2376,12 @@ standing:
     barrier: "open after the D-440 arc completes (R4 + R5 must first rewrite the bulk of 'Phase N' citations to gap-area terms). Premature adoption = ~70 dangling citations with no 1:1 redirect. Trigger: D-440 R4/R5 done + a user nod to re-architect the planning axis (it is depth-4, the project's organizing metaphor)."
     last_reviewed: "2026-06-15"
 discharged:
+  - id: "D-561"
+    status: |-
+      DISCHARGED 2026-07-17 — Character/getName + codePointOf land on the generated UCD 16.0.0 word-indexed name table (scripts/gen_unicode_names.py → unicode_names.zig, pre-compressed raw-deflate blobs + delta-varint index, decoded once per process in char_name.zig). JVM-oracle-exact semantics: stored names, pinned control aliases (scripts/data/jvm_control_names.txt), assigned-but-unnamed = "<BLOCK> <HEX>" (the JVM does NOT synthesize UCD algorithmic names), unassigned nil; codePointOf = trimmed case-insensitive inverse accepting exactly getName output forms. Corpus char_names.txt 10/10. Size: mac +279KB (7.35MB; ADR-0172 revision — README claim 7.8MB; Linux ~94% of ceiling, next size work measures Linux first).
+    category: "host-surface"
+    barrier: "none — implementable any time; the open question is the binary-size trade (gap area II), not a technical blocker"
+    last_reviewed: "2026-07-17"
   - id: "D-563"
     status: |-
       DISCHARGED 2026-07-17 (all sub-items resolved: a/a2/b/c landed, d = AD-043). ORIGINAL: open (2026-07-16, the D-562 inventory's enumerated residuals — each
@@ -6607,7 +6595,6 @@ discharged:
       decompress); varint operands + cross-region constant-pool dedup evaluated in
       the same pass (compaction shrinks with NO decompression copy). est -0.9-1.1MB.
     last_reviewed: "2026-07-16"
-
 conventions: |-
   - Structure: two top-level lists — `active:` and `discharged:`. Active entries carry id / status / category / barrier (+ optional quality_floor / last_reviewed); discharged entries carry id / discharged_at / resolution. Edit as YAML (no Markdown table to align).
   - New debt: grep `.dev/debt.yaml` for class-overlap before adding (per `.claude/rules/debt_dedup.md`).

.dev/decisions/0172_binary_size_budget_and_ledger.md

@@ -406,6 +406,16 @@ dispositions grounded, platform binding declared, peer number measured).
 
 ## Revision history
 
+- **2026-07-17 (D-561 Unicode name table, conscious)**: `Character/getName`
+  + `codePointOf` land the full UCD 16.0.0 name table as pre-compressed
+  word-indexed blobs (40,013 names; 545 KB raw → ~280 KB stored after the
+  delta-varint index rework — the first fixed-width index cost +394 KB and
+  was reworked before landing). macOS 7,073,240 → 7,352,376 B (+279 KB);
+  Linux est. ~8.29 MB (~94% of ceiling — the NEXT size-relevant feature
+  must measure Linux first and likely needs a lever or a ceiling
+  amendment). README claim moved 7.5 → "about 7.8 MB" (covers both
+  platforms within the ±10% gate).
+
 - **2026-07-17 (ADR-0174 feature growth, conscious)**: the host-class
   identity & member-surface campaign grew the binary macOS
   6,974,584 → 7,073,240 B (+98.7 KB — Thread lifecycle worker + stdio

README.md

@@ -33,7 +33,7 @@ still use what other languages have already built.
 
 ## Features
 
-- **Small and quick to start** — one static binary of about 7.5 MB with the
+- **Small and quick to start** — one static binary of about 7.8 MB with the
   Wasm JIT engine included (for scale: babashka's native binary is ~71 MB;
   [measured comparison](docs/works/binary_size.md)), starting in ~5 ms, which
   suits short-lived, start-and-stop workloads (CLI tools, serverless, scripts).

data/compat_tiers.yaml

@@ -195,7 +195,7 @@ tiers:
         # 16.0.0 tables; corpus test/diff/clj_corpus/character.txt locks
         # clj parity). getName is explicit-unsupported (D-561). Instance
         # methods on .char: charValue / compareTo (+ universal Object).
-        methods: [charCount, charValue, codePointAt, codePointBefore, codePointCount, compare, compareTo, digit, forDigit, getDirectionality, getName, getNumericValue, getType, highSurrogate, isAlphabetic, isBmpCodePoint, isDefined, isDigit, isEmoji, isEmojiComponent, isEmojiModifier, isEmojiModifierBase, isEmojiPresentation, isExtendedPictographic, isHighSurrogate, isISOControl, isIdentifierIgnorable, isIdeographic, isJavaIdentifierPart, isJavaIdentifierStart, isJavaLetter, isJavaLetterOrDigit, isLetter, isLetterOrDigit, isLowSurrogate, isLowerCase, isMirrored, isSpace, isSpaceChar, isSupplementaryCodePoint, isSurrogate, isSurrogatePair, isTitleCase, isUnicodeIdentifierPart, isUnicodeIdentifierStart, isUpperCase, isValidCodePoint, isWhitespace, lowSurrogate, offsetByCodePoints, reverseBytes, toChars, toCodePoint, toLowerCase, toTitleCase, toUpperCase, valueOf]
+        methods: [charCount, charValue, codePointAt, codePointBefore, codePointCount, codePointOf, compare, compareTo, digit, forDigit, getDirectionality, getName, getNumericValue, getType, highSurrogate, isAlphabetic, isBmpCodePoint, isDefined, isDigit, isEmoji, isEmojiComponent, isEmojiModifier, isEmojiModifierBase, isEmojiPresentation, isExtendedPictographic, isHighSurrogate, isISOControl, isIdentifierIgnorable, isIdeographic, isJavaIdentifierPart, isJavaIdentifierStart, isJavaLetter, isJavaLetterOrDigit, isLetter, isLetterOrDigit, isLowSurrogate, isLowerCase, isMirrored, isSpace, isSpaceChar, isSupplementaryCodePoint, isSurrogate, isSurrogatePair, isTitleCase, isUnicodeIdentifierPart, isUnicodeIdentifierStart, isUpperCase, isValidCodePoint, isWhitespace, lowSurrogate, offsetByCodePoints, reverseBytes, toChars, toCodePoint, toLowerCase, toTitleCase, toUpperCase, valueOf]
         static_fields: ["BYTES", "COMBINING_SPACING_MARK", "CONNECTOR_PUNCTUATION", "CONTROL", "CURRENCY_SYMBOL", "DASH_PUNCTUATION", "DECIMAL_DIGIT_NUMBER", "DIRECTIONALITY_ARABIC_NUMBER", "DIRECTIONALITY_BOUNDARY_NEUTRAL", "DIRECTIONALITY_COMMON_NUMBER_SEPARATOR", "DIRECTIONALITY_EUROPEAN_NUMBER", "DIRECTIONALITY_EUROPEAN_NUMBER_SEPARATOR", "DIRECTIONALITY_EUROPEAN_NUMBER_TERMINATOR", "DIRECTIONALITY_FIRST_STRONG_ISOLATE", "DIRECTIONALITY_LEFT_TO_RIGHT", "DIRECTIONALITY_LEFT_TO_RIGHT_EMBEDDING", "DIRECTIONALITY_LEFT_TO_RIGHT_ISOLATE", "DIRECTIONALITY_LEFT_TO_RIGHT_OVERRIDE", "DIRECTIONALITY_NONSPACING_MARK", "DIRECTIONALITY_OTHER_NEUTRALS", "DIRECTIONALITY_PARAGRAPH_SEPARATOR", "DIRECTIONALITY_POP_DIRECTIONAL_FORMAT", "DIRECTIONALITY_POP_DIRECTIONAL_ISOLATE", "DIRECTIONALITY_RIGHT_TO_LEFT", "DIRECTIONALITY_RIGHT_TO_LEFT_ARABIC", "DIRECTIONALITY_RIGHT_TO_LEFT_EMBEDDING", "DIRECTIONALITY_RIGHT_TO_LEFT_ISOLATE", "DIRECTIONALITY_RIGHT_TO_LEFT_OVERRIDE", "DIRECTIONALITY_SEGMENT_SEPARATOR", "DIRECTIONALITY_UNDEFINED", "DIRECTIONALITY_WHITESPACE", "ENCLOSING_MARK", "END_PUNCTUATION", "FINAL_QUOTE_PUNCTUATION", "FORMAT", "INITIAL_QUOTE_PUNCTUATION", "LETTER_NUMBER", "LINE_SEPARATOR", "LOWERCASE_LETTER", "MATH_SYMBOL", "MAX_CODE_POINT", "MAX_HIGH_SURROGATE", "MAX_LOW_SURROGATE", "MAX_RADIX", "MAX_SURROGATE", "MAX_VALUE", "MIN_CODE_POINT", "MIN_HIGH_SURROGATE", "MIN_LOW_SURROGATE", "MIN_RADIX", "MIN_SUPPLEMENTARY_CODE_POINT", "MIN_SURROGATE", "MIN_VALUE", "MODIFIER_LETTER", "MODIFIER_SYMBOL", "NON_SPACING_MARK", "OTHER_LETTER", "OTHER_NUMBER", "OTHER_PUNCTUATION", "OTHER_SYMBOL", "PARAGRAPH_SEPARATOR", "PRIVATE_USE", "SIZE", "SPACE_SEPARATOR", "START_PUNCTUATION", "SURROGATE", "TITLECASE_LETTER", "UNASSIGNED", "UPPERCASE_LETTER"]
         opaque_members: [TYPE]
         clojure_peer_vars: []

scripts/data/jvm_control_names.txt

@@ -0,0 +1,70 @@
+# JVM Character.getName oracle for control codepoints (clj 1.12, captured 2026-07-17).
+# Pinned rather than re-deriving NameAliases.txt's preference rule (JVM mixes
+# control-type aliases with abbreviations: 0x07=BEL but 0x01=START OF HEADING).
+0 NULL
+1 START OF HEADING
+2 START OF TEXT
+3 END OF TEXT
+4 END OF TRANSMISSION
+5 ENQUIRY
+6 ACKNOWLEDGE
+7 BEL
+8 BACKSPACE
+9 CHARACTER TABULATION
+10 LINE FEED (LF)
+11 LINE TABULATION
+12 FORM FEED (FF)
+13 CARRIAGE RETURN (CR)
+14 SHIFT OUT
+15 SHIFT IN
+16 DATA LINK ESCAPE
+17 DEVICE CONTROL ONE
+18 DEVICE CONTROL TWO
+19 DEVICE CONTROL THREE
+20 DEVICE CONTROL FOUR
+21 NEGATIVE ACKNOWLEDGE
+22 SYNCHRONOUS IDLE
+23 END OF TRANSMISSION BLOCK
+24 CANCEL
+25 END OF MEDIUM
+26 SUBSTITUTE
+27 ESCAPE
+28 INFORMATION SEPARATOR FOUR
+29 INFORMATION SEPARATOR THREE
+30 INFORMATION SEPARATOR TWO
+31 INFORMATION SEPARATOR ONE
+32 SPACE
+127 DELETE
+128 PADDING CHARACTER
+129 HIGH OCTET PRESET
+130 BREAK PERMITTED HERE
+131 NO BREAK HERE
+132 LATIN 1 SUPPLEMENT 84
+133 NEXT LINE (NEL)
+134 START OF SELECTED AREA
+135 END OF SELECTED AREA
+136 CHARACTER TABULATION SET
+137 CHARACTER TABULATION WITH JUSTIFICATION
+138 LINE TABULATION SET
+139 PARTIAL LINE FORWARD
+140 PARTIAL LINE BACKWARD
+141 REVERSE LINE FEED
+142 SINGLE SHIFT TWO
+143 SINGLE SHIFT THREE
+144 DEVICE CONTROL STRING
+145 PRIVATE USE ONE
+146 PRIVATE USE TWO
+147 SET TRANSMIT STATE
+148 CANCEL CHARACTER
+149 MESSAGE WAITING
+150 START OF GUARDED AREA
+151 END OF GUARDED AREA
+152 START OF STRING
+153 SINGLE GRAPHIC CHARACTER INTRODUCER
+154 SINGLE CHARACTER INTRODUCER
+155 CONTROL SEQUENCE INTRODUCER
+156 STRING TERMINATOR
+157 OPERATING SYSTEM COMMAND
+158 PRIVACY MESSAGE
+159 APPLICATION PROGRAM COMMAND
+160 NO-BREAK SPACE