feat(rules): 建立共享排版契约并收束 WebKit 与 Unicode 基线

2026-09-04 21:44:11 +08:00
parent 8ee9895247
commit 993cf576c6
42 changed files with 8511 additions and 1422 deletions
@@ -0,0 +1,61 @@
+# 聚珍共享規則
+
+本目錄是聚珍跨平台排版語義的機讀權威來源。Web／CSS 實作與未來其他 adapter
+都應以 [`juzhen-rules.json`](./juzhen-rules.json) 為準；程式內常數不得另立一套未經
+驗證的規則。`golden-vectors.json` 則固定跨 adapter 必須得到的語義結果，向量只描述
+字符分類、間距、禁斷與段末保字等結果，不指定 DOM 結構或 CSS 技巧。
+
+## 檔案
+
+- `juzhen-rules.json`：版本化的共享規則資料。
+- `juzhen-rules.schema.json`：JSON Schema 2020-12，約束共享規則格式。
+- `golden-vectors.json`：跨平台黃金向量。
+- `golden-vectors.schema.json`：JSON Schema 2020-12，約束向量格式。
+
+套件公開子路徑分別為 `cjk-autospace/rules`、`cjk-autospace/rules/schema`、
+`cjk-autospace/rules/vectors` 與 `cjk-autospace/rules/vectors-schema`。
+
+## 版本
+
+`schemaVersion` 是整數格式版本；讀取器遇到不認識的值必須拒絕，不得猜測。
+`rulesetVersion` 採語義化版本：
+
+- patch：澄清文字或新增不改變既有結果的向量；
+- minor：向後相容地擴充字符或新增可選語義；
+- major：既有輸入的分類、間距、禁斷、預設政策或 adapter 義務改變。
+
+Unicode profile 另行記錄。更新 Unicode 範圍時，即使 JSON 結構不變，也必須同步提升
+`rulesetVersion`、補向量並驗證所有 adapter。
+
+## 變更協議
+
+任何規則變更須在同一變更集中完成以下事項：
+
+1. 更新 canonical JSON 及必要的 schema；
+2. 新增或修訂能暴露行為差異的黃金向量，不以刪除反例來換取通過；
+3. 更新 Web adapter 的對照測試；未來其他 adapter 亦須執行同一批語義向量；
+4. 說明版本提升級別、相容性與 Unicode 依據；
+5. 通過 Ajv 2020 strict schema 驗證、內部常數漂移檢查與全部 adapter 測試後方可發佈。
+
+字符覆寫固定為追加語義：不得經 override 移除或重分類內置字符。若未來需要破壞此
+規則，屬 major 變更，須先作產品政策決策，不能由單一 adapter 私自實作。
+
+## Adapter 邊界
+
+字符集合一律以大寫十六進位 Unicode code point 的閉區間／單點表示；陣列須按起點
+遞增、彼此不重疊。adapter 可序列化成自身所需的 regex、查表或其他等價結構，但該
+序列化不是 canonical 格式。
+
+共享層只定義平台中立結果：Unicode 字符類、base＋至多一枚 VS 的邏輯 atom、locale
+分類與政策、標點半形語義、中西間距與作者空格處置、禁斷關係、段末保字數、功能級別
+及語義依賴。
+
+adapter 的節點表示、ownership／revert 帳務、渲染原語、生成式斷行提示、重跑生命
+週期、layout 量測與平台 workaround 均不屬共享規則。例如 Web 目前可使用 `jz-*` DOM、
+CSS margin／OpenType `halt` 與 MutationObserver；未來 TeX 類 adapter 也可能選擇 node、
+glue、penalty 或 macro。這些都只是**非規範示例**，adapter 可採任何能通過黃金向量的
+機制，不得把示例反寫為 canonical 要求。
+
+Web 目前保留的 `hanging`／`biaodian`／`emphasis`／`ruby` 相容 no-op 亦屬
+**非規範 adapter 相容性資訊**，不列入 canonical 規則，也不代表其他 adapter 必須公開
+同名選項或未來功能介面已定案。
@@ -0,0 +1,373 @@
+{
+  "$schema": "./golden-vectors.schema.json",
+  "schemaVersion": 1,
+  "rulesetVersion": "1.0.0",
+  "vectors": [
+    {
+      "id": "locale-ja",
+      "category": "language-classification",
+      "input": { "lang": "ja" },
+      "expected": { "langClass": "ja" }
+    },
+    {
+      "id": "locale-ja-jp",
+      "category": "language-classification",
+      "input": { "lang": "ja-JP" },
+      "expected": { "langClass": "ja" }
+    },
+    {
+      "id": "locale-zh-bare",
+      "category": "language-classification",
+      "input": { "lang": "zh" },
+      "expected": { "langClass": "zh-Hant" }
+    },
+    {
+      "id": "locale-zh-hant",
+      "category": "language-classification",
+      "input": { "lang": "zh-Hant" },
+      "expected": { "langClass": "zh-Hant" }
+    },
+    {
+      "id": "locale-zh-hans",
+      "category": "language-classification",
+      "input": { "lang": "zh-Hans" },
+      "expected": { "langClass": "zh-Hans" }
+    },
+    {
+      "id": "locale-zh-tw",
+      "category": "language-classification",
+      "input": { "lang": "zh-TW" },
+      "expected": { "langClass": "zh-Hant" }
+    },
+    {
+      "id": "locale-zh-hk",
+      "category": "language-classification",
+      "input": { "lang": "zh-HK" },
+      "expected": { "langClass": "zh-Hant" }
+    },
+    {
+      "id": "locale-zh-cn",
+      "category": "language-classification",
+      "input": { "lang": "zh-CN" },
+      "expected": { "langClass": "zh-Hans" }
+    },
+    {
+      "id": "locale-zhx-boundary",
+      "category": "language-classification",
+      "input": { "lang": "zhx" },
+      "expected": { "langClass": "other" }
+    },
+    {
+      "id": "locale-en",
+      "category": "language-classification",
+      "input": { "lang": "en" },
+      "expected": { "langClass": "other" }
+    },
+    {
+      "id": "punct-stop",
+      "category": "punctuation-classification",
+      "input": { "character": "．" },
+      "expected": { "punctuationClass": "stop" }
+    },
+    {
+      "id": "punct-pause",
+      "category": "punctuation-classification",
+      "input": { "character": "，" },
+      "expected": { "punctuationClass": "pause" }
+    },
+    {
+      "id": "punct-open",
+      "category": "punctuation-classification",
+      "input": { "character": "「" },
+      "expected": { "punctuationClass": "open" }
+    },
+    {
+      "id": "punct-close",
+      "category": "punctuation-classification",
+      "input": { "character": "」" },
+      "expected": { "punctuationClass": "close" }
+    },
+    {
+      "id": "punct-middle",
+      "category": "punctuation-classification",
+      "input": { "character": "·" },
+      "expected": { "punctuationClass": "middle" }
+    },
+    {
+      "id": "punct-liga",
+      "category": "punctuation-classification",
+      "input": { "character": "…" },
+      "expected": { "punctuationClass": "liga" }
+    },
+    {
+      "id": "punct-base-with-vs",
+      "category": "punctuation-classification",
+      "input": { "character": "。︀" },
+      "expected": { "punctuationClass": "stop" }
+    },
+    {
+      "id": "punct-other",
+      "category": "punctuation-classification",
+      "input": { "character": "A" },
+      "expected": { "punctuationClass": null }
+    },
+    {
+      "id": "style-quanjiao-pause",
+      "category": "punctuation-style",
+      "input": { "style": "quanjiao", "punctuationClass": "pause" },
+      "expected": { "halfWidth": false }
+    },
+    {
+      "id": "style-kaiming-pause",
+      "category": "punctuation-style",
+      "input": { "style": "kaiming", "punctuationClass": "pause" },
+      "expected": { "halfWidth": true }
+    },
+    {
+      "id": "style-kaiming-open",
+      "category": "punctuation-style",
+      "input": { "style": "kaiming", "punctuationClass": "open" },
+      "expected": { "halfWidth": true }
+    },
+    {
+      "id": "style-kaiming-stop",
+      "category": "punctuation-style",
+      "input": { "style": "kaiming", "punctuationClass": "stop" },
+      "expected": { "halfWidth": false }
+    },
+    {
+      "id": "style-banjiao-stop",
+      "category": "punctuation-style",
+      "input": { "style": "banjiao", "punctuationClass": "stop" },
+      "expected": { "halfWidth": true }
+    },
+    {
+      "id": "style-banjiao-liga",
+      "category": "punctuation-style",
+      "input": { "style": "banjiao", "punctuationClass": "liga" },
+      "expected": { "halfWidth": false }
+    },
+    {
+      "id": "adjacency-pause-open",
+      "category": "punctuation-adjacency",
+      "input": { "left": "pause", "right": "open" },
+      "expected": { "squeezeLeft": true, "squeezeRight": true }
+    },
+    {
+      "id": "adjacency-stop-open",
+      "category": "punctuation-adjacency",
+      "input": { "left": "stop", "right": "open" },
+      "expected": { "squeezeLeft": false, "squeezeRight": true }
+    },
+    {
+      "id": "adjacency-stop-close",
+      "category": "punctuation-adjacency",
+      "input": { "left": "stop", "right": "close" },
+      "expected": { "squeezeLeft": true, "squeezeRight": false }
+    },
+    {
+      "id": "adjacency-close-open",
+      "category": "punctuation-adjacency",
+      "input": { "left": "close", "right": "open" },
+      "expected": { "squeezeLeft": true, "squeezeRight": true }
+    },
+    {
+      "id": "adjacency-middle-liga",
+      "category": "punctuation-adjacency",
+      "input": { "left": "middle", "right": "liga" },
+      "expected": { "squeezeLeft": false, "squeezeRight": false }
+    },
+    {
+      "id": "spacing-cjk-latin",
+      "category": "spacing",
+      "input": { "text": "中A" },
+      "expected": { "managedGaps": 1, "normalizedText": "中A", "gapEm": 0.25 }
+    },
+    {
+      "id": "spacing-latin-cjk",
+      "category": "spacing",
+      "input": { "text": "A中" },
+      "expected": { "managedGaps": 1, "normalizedText": "A中", "gapEm": 0.25 }
+    },
+    {
+      "id": "spacing-cjk-at",
+      "category": "spacing",
+      "input": { "text": "中@" },
+      "expected": { "managedGaps": 1, "normalizedText": "中@", "gapEm": 0.25 }
+    },
+    {
+      "id": "spacing-at-cjk-asymmetric",
+      "category": "spacing",
+      "input": { "text": "@中" },
+      "expected": { "managedGaps": 0, "normalizedText": "@中" }
+    },
+    {
+      "id": "spacing-author-space-cjk-latin",
+      "category": "spacing",
+      "input": { "text": "中 A" },
+      "expected": {
+        "managedGaps": 1,
+        "normalizedText": "中A",
+        "authorSpaceDisposition": "remove-and-use-managed-gap"
+      }
+    },
+    {
+      "id": "spacing-author-space-latin-cjk",
+      "category": "spacing",
+      "input": { "text": "A 中" },
+      "expected": {
+        "managedGaps": 1,
+        "normalizedText": "A中",
+        "authorSpaceDisposition": "remove-and-use-managed-gap"
+      }
+    },
+    {
+      "id": "spacing-author-space-cjk-cjk",
+      "category": "spacing",
+      "input": { "text": "中 文" },
+      "expected": {
+        "managedGaps": 0,
+        "normalizedText": "中文",
+        "authorSpaceDisposition": "remove-without-gap"
+      }
+    },
+    {
+      "id": "spacing-cjk-punctuation-latin",
+      "category": "spacing",
+      "input": { "text": "中。A" },
+      "expected": { "managedGaps": 0, "normalizedText": "中。A" }
+    },
+    {
+      "id": "spacing-astral-ivs-latin",
+      "category": "spacing",
+      "input": { "text": "𠀀󠄀A" },
+      "expected": { "managedGaps": 1, "normalizedText": "𠀀󠄀A", "gapEm": 0.25 }
+    },
+    {
+      "id": "atom-bmp-vs",
+      "category": "ivs-atom",
+      "input": { "text": "中︀A" },
+      "expected": { "atoms": [ "中︀", "A" ] }
+    },
+    {
+      "id": "atom-astral-ivs",
+      "category": "ivs-atom",
+      "input": { "text": "𠀀󠄀A" },
+      "expected": { "atoms": [ "𠀀󠄀", "A" ] }
+    },
+    {
+      "id": "atom-at-most-one-vs",
+      "category": "ivs-atom",
+      "input": { "text": "中︀︁" },
+      "expected": { "atoms": [ "中︀", "︁" ] }
+    },
+    {
+      "id": "atom-isolated-vs",
+      "category": "ivs-atom",
+      "input": { "text": "︀A" },
+      "expected": { "atoms": [ "︀", "A" ] }
+    },
+    {
+      "id": "jinze-before-stop",
+      "category": "jinze",
+      "input": { "left": "A", "right": "。" },
+      "expected": { "prohibitedBreak": true, "reason": "before-stop" }
+    },
+    {
+      "id": "jinze-before-close",
+      "category": "jinze",
+      "input": { "left": "中", "right": "」" },
+      "expected": { "prohibitedBreak": true, "reason": "before-close" }
+    },
+    {
+      "id": "jinze-after-open",
+      "category": "jinze",
+      "input": { "left": "「", "right": "A" },
+      "expected": { "prohibitedBreak": true, "reason": "after-open" }
+    },
+    {
+      "id": "jinze-ordinary-boundary",
+      "category": "jinze",
+      "input": { "left": "中", "right": "A" },
+      "expected": { "prohibitedBreak": false }
+    },
+    {
+      "id": "jinze-supplementary-ivs-before-stop",
+      "category": "jinze",
+      "input": { "left": "𠀀\uDB40\uDD00", "right": "。" },
+      "expected": {
+        "prohibitedBreak": true,
+        "reason": "before-stop",
+        "atoms": [ "𠀀\uDB40\uDD00", "。" ]
+      }
+    },
+    {
+      "id": "jinze-second-vs-before-stop",
+      "category": "jinze",
+      "input": { "left": "𠀀\uDB40\uDD00\uDB40\uDD01", "right": "。" },
+      "expected": {
+        "prohibitedBreak": true,
+        "reason": "before-stop",
+        "atoms": [ "𠀀\uDB40\uDD00", "\uDB40\uDD01", "。" ]
+      }
+    },
+    {
+      "id": "jinze-open-before-second-isolated-vs",
+      "category": "jinze",
+      "input": { "left": "「\uFE00", "right": "\uDB40\uDD00" },
+      "expected": {
+        "prohibitedBreak": true,
+        "reason": "after-open",
+        "atoms": [ "「\uFE00", "\uDB40\uDD00" ]
+      }
+    },
+    {
+      "id": "orphan-default-two",
+      "category": "orphan",
+      "input": { "text": "天地玄黃", "minimum": 2 },
+      "expected": { "meaningfulCjkCount": 4, "protectedSuffix": "玄黃" }
+    },
+    {
+      "id": "orphan-trailing-punctuation",
+      "category": "orphan",
+      "input": { "text": "天地。", "minimum": 2 },
+      "expected": { "meaningfulCjkCount": 2, "protectedSuffix": "天地。" }
+    },
+    {
+      "id": "orphan-insufficient-cjk",
+      "category": "orphan",
+      "input": { "text": "天。", "minimum": 2 },
+      "expected": { "meaningfulCjkCount": 1, "protectedSuffix": null }
+    },
+    {
+      "id": "orphan-supplementary-ivs",
+      "category": "orphan",
+      "input": { "text": "甲𠀀\uDB40\uDD00", "minimum": 2 },
+      "expected": {
+        "meaningfulCjkCount": 2,
+        "protectedSuffix": "甲𠀀\uDB40\uDD00",
+        "atoms": [ "甲", "𠀀\uDB40\uDD00" ]
+      }
+    },
+    {
+      "id": "orphan-second-vs",
+      "category": "orphan",
+      "input": { "text": "甲𠀀\uDB40\uDD00\uDB40\uDD01", "minimum": 2 },
+      "expected": {
+        "meaningfulCjkCount": 2,
+        "protectedSuffix": "甲𠀀\uDB40\uDD00\uDB40\uDD01",
+        "atoms": [ "甲", "𠀀\uDB40\uDD00", "\uDB40\uDD01" ]
+      }
+    },
+    {
+      "id": "orphan-isolated-vs-prefix",
+      "category": "orphan",
+      "input": { "text": "\uDB40\uDD00甲𠀀\uDB40\uDD01", "minimum": 2 },
+      "expected": {
+        "meaningfulCjkCount": 2,
+        "protectedSuffix": "甲𠀀\uDB40\uDD01",
+        "atoms": [ "\uDB40\uDD00", "甲", "𠀀\uDB40\uDD01" ]
+      }
+    }
+  ]
+}
@@ -0,0 +1,267 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://juzhen.dev/schemas/golden-vectors.schema.json",
+  "title": "Juzhen shared rules golden vectors",
+  "type": "object",
+  "additionalProperties": false,
+  "required": [ "schemaVersion", "rulesetVersion", "vectors" ],
+  "properties": {
+    "$schema": { "type": "string" },
+    "schemaVersion": { "const": 1 },
+    "rulesetVersion": {
+      "type": "string",
+      "pattern": "^(0|[1-9][0-9]*)\\.(0|[1-9][0-9]*)\\.(0|[1-9][0-9]*)$"
+    },
+    "vectors": {
+      "type": "array",
+      "minItems": 20,
+      "items": {
+        "oneOf": [
+          { "$ref": "#/$defs/languageVector" },
+          { "$ref": "#/$defs/punctuationClassVector" },
+          { "$ref": "#/$defs/punctuationStyleVector" },
+          { "$ref": "#/$defs/punctuationAdjacencyVector" },
+          { "$ref": "#/$defs/spacingVector" },
+          { "$ref": "#/$defs/ivsAtomVector" },
+          { "$ref": "#/$defs/jinzeVector" },
+          { "$ref": "#/$defs/orphanVector" }
+        ]
+      }
+    }
+  },
+  "$defs": {
+    "id": {
+      "type": "string",
+      "pattern": "^[a-z0-9]+(?:-[a-z0-9]+)*$"
+    },
+    "punctuationClass": {
+      "enum": [ "stop", "pause", "open", "close", "middle", "liga" ]
+    },
+    "style": {
+      "enum": [ "quanjiao", "kaiming", "banjiao" ]
+    },
+    "atoms": {
+      "type": "array",
+      "minItems": 1,
+      "items": { "type": "string", "minLength": 1 }
+    },
+    "languageVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "language-classification" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "lang" ],
+          "properties": {
+            "lang": { "type": "string" }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "langClass" ],
+          "properties": {
+            "langClass": { "enum": [ "zh-Hans", "zh-Hant", "ja", "other" ] }
+          }
+        }
+      }
+    },
+    "punctuationClassVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "punctuation-classification" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "character" ],
+          "properties": {
+            "character": { "type": "string", "minLength": 1 }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "punctuationClass" ],
+          "properties": {
+            "punctuationClass": {
+              "oneOf": [ { "$ref": "#/$defs/punctuationClass" }, { "type": "null" } ]
+            }
+          }
+        }
+      }
+    },
+    "punctuationStyleVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "punctuation-style" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "style", "punctuationClass" ],
+          "properties": {
+            "style": { "$ref": "#/$defs/style" },
+            "punctuationClass": { "$ref": "#/$defs/punctuationClass" }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "halfWidth" ],
+          "properties": {
+            "halfWidth": { "type": "boolean" }
+          }
+        }
+      }
+    },
+    "punctuationAdjacencyVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "punctuation-adjacency" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "left", "right" ],
+          "properties": {
+            "left": { "$ref": "#/$defs/punctuationClass" },
+            "right": { "$ref": "#/$defs/punctuationClass" }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "squeezeLeft", "squeezeRight" ],
+          "properties": {
+            "squeezeLeft": { "type": "boolean" },
+            "squeezeRight": { "type": "boolean" }
+          }
+        }
+      }
+    },
+    "spacingVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "spacing" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "text" ],
+          "properties": {
+            "text": { "type": "string" }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "managedGaps", "normalizedText" ],
+          "properties": {
+            "managedGaps": { "type": "integer", "minimum": 0 },
+            "normalizedText": { "type": "string" },
+            "gapEm": { "const": 0.25 },
+            "authorSpaceDisposition": {
+              "enum": [ "remove-and-use-managed-gap", "remove-without-gap" ]
+            }
+          }
+        }
+      }
+    },
+    "ivsAtomVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "ivs-atom" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "text" ],
+          "properties": {
+            "text": { "type": "string", "minLength": 1 }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "atoms" ],
+          "properties": {
+            "atoms": { "$ref": "#/$defs/atoms" }
+          }
+        }
+      }
+    },
+    "jinzeVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "jinze" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "left", "right" ],
+          "properties": {
+            "left": { "type": "string", "minLength": 1 },
+            "right": { "type": "string", "minLength": 1 }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "prohibitedBreak" ],
+          "properties": {
+            "prohibitedBreak": { "type": "boolean" },
+            "reason": { "enum": [ "before-stop", "before-close", "after-open" ] },
+            "atoms": { "$ref": "#/$defs/atoms" }
+          }
+        }
+      }
+    },
+    "orphanVector": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "id", "category", "input", "expected" ],
+      "properties": {
+        "id": { "$ref": "#/$defs/id" },
+        "category": { "const": "orphan" },
+        "input": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "text", "minimum" ],
+          "properties": {
+            "text": { "type": "string", "minLength": 1 },
+            "minimum": { "type": "integer", "minimum": 1 }
+          }
+        },
+        "expected": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "meaningfulCjkCount", "protectedSuffix" ],
+          "properties": {
+            "meaningfulCjkCount": { "type": "integer", "minimum": 0 },
+            "protectedSuffix": {
+              "oneOf": [ { "type": "string" }, { "type": "null" } ]
+            },
+            "atoms": { "$ref": "#/$defs/atoms" }
+          }
+        }
+      }
+    }
+  }
+}
@@ -0,0 +1,255 @@
+{
+  "$schema": "./juzhen-rules.schema.json",
+  "schemaVersion": 1,
+  "rulesetVersion": "1.0.0",
+  "unicodeProfile": {
+    "version": "17.0.0",
+    "coverage": "CJK Unified Ideographs through Extension J, compatibility ideographs, kana, bopomofo, radicals, and enclosed CJK characters"
+  },
+  "characterClasses": {
+    "representation": "unicode-code-point-ranges",
+    "cjk": [
+      { "start": "2E80", "end": "2EFF" },
+      { "start": "2F00", "end": "2FDF" },
+      { "start": "3040", "end": "309F" },
+      { "start": "30A0", "end": "30FA" },
+      { "start": "30FC", "end": "30FF" },
+      { "start": "3100", "end": "312F" },
+      { "start": "3200", "end": "32FF" },
+      { "start": "3400", "end": "4DBF" },
+      { "start": "4E00", "end": "9FFF" },
+      { "start": "F900", "end": "FAFF" },
+      { "start": "20000", "end": "2A6DF" },
+      { "start": "2A700", "end": "2B73F" },
+      { "start": "2B740", "end": "2B81F" },
+      { "start": "2B820", "end": "2CEAF" },
+      { "start": "2CEB0", "end": "2EBEF" },
+      { "start": "2EBF0", "end": "2EE5F" },
+      { "start": "2F800", "end": "2FA1F" },
+      { "start": "30000", "end": "3134F" },
+      { "start": "31350", "end": "323AF" },
+      { "start": "323B0", "end": "3347F" }
+    ],
+    "ansAfterCjk": [
+      { "start": "0024", "end": "0026" },
+      { "start": "002A", "end": "002B" },
+      { "codePoint": "002D" },
+      { "start": "0030", "end": "0039" },
+      { "codePoint": "003D" },
+      { "codePoint": "0040" },
+      { "start": "0041", "end": "005A" },
+      { "codePoint": "005C" },
+      { "codePoint": "005E" },
+      { "start": "0061", "end": "007A" },
+      { "start": "00A1", "end": "00B6" },
+      { "start": "00B8", "end": "00FF" },
+      { "start": "0100", "end": "024F" },
+      { "start": "0370", "end": "03FF" },
+      { "start": "0400", "end": "04FF" },
+      { "start": "20A0", "end": "20BF" },
+      { "start": "2150", "end": "21FF" },
+      { "start": "2200", "end": "22FF" },
+      { "start": "2700", "end": "27BF" }
+    ],
+    "ansBeforeCjk": [
+      { "start": "0024", "end": "0026" },
+      { "start": "002A", "end": "002B" },
+      { "codePoint": "002D" },
+      { "start": "0030", "end": "0039" },
+      { "codePoint": "003D" },
+      { "start": "0041", "end": "005A" },
+      { "codePoint": "005C" },
+      { "codePoint": "005E" },
+      { "start": "0061", "end": "007A" },
+      { "start": "00A1", "end": "00B6" },
+      { "start": "00B8", "end": "00FF" },
+      { "start": "0100", "end": "024F" },
+      { "start": "0370", "end": "03FF" },
+      { "start": "0400", "end": "04FF" },
+      { "start": "20A0", "end": "20BF" },
+      { "start": "2150", "end": "21FF" },
+      { "start": "2200", "end": "22FF" },
+      { "start": "2700", "end": "27BF" }
+    ],
+    "variationSelectors": [
+      { "start": "FE00", "end": "FE0F" },
+      { "start": "E0100", "end": "E01EF" }
+    ],
+    "biaodian": {
+      "classificationOrder": [ "stop", "pause", "open", "close", "middle", "liga" ],
+      "classes": {
+        "stop": [
+          { "codePoint": "3002" },
+          { "codePoint": "FF01" },
+          { "codePoint": "FF0E" },
+          { "codePoint": "FF1F" }
+        ],
+        "pause": [
+          { "codePoint": "3001" },
+          { "codePoint": "FF0C" },
+          { "codePoint": "FF1A" },
+          { "codePoint": "FF1B" }
+        ],
+        "open": [
+          { "codePoint": "2018" },
+          { "codePoint": "201C" },
+          { "codePoint": "3008" },
+          { "codePoint": "300A" },
+          { "codePoint": "300C" },
+          { "codePoint": "300E" },
+          { "codePoint": "3010" },
+          { "codePoint": "3014" },
+          { "codePoint": "3016" },
+          { "codePoint": "FF08" },
+          { "codePoint": "FF3B" },
+          { "codePoint": "FF5B" }
+        ],
+        "close": [
+          { "codePoint": "2019" },
+          { "codePoint": "201D" },
+          { "codePoint": "3009" },
+          { "codePoint": "300B" },
+          { "codePoint": "300D" },
+          { "codePoint": "300F" },
+          { "codePoint": "3011" },
+          { "codePoint": "3015" },
+          { "codePoint": "3017" },
+          { "codePoint": "FF09" },
+          { "codePoint": "FF3D" },
+          { "codePoint": "FF5D" }
+        ],
+        "middle": [
+          { "codePoint": "00B7" },
+          { "codePoint": "2027" },
+          { "codePoint": "30FB" }
+        ],
+        "liga": [
+          { "codePoint": "2014" },
+          { "codePoint": "2026" }
+        ]
+      }
+    },
+    "overrideSemantics": {
+      "mode": "append",
+      "westernTargets": [ "ansAfterCjk", "ansBeforeCjk" ],
+      "removalAllowed": false,
+      "reclassificationAllowed": false
+    }
+  },
+  "atoms": {
+    "unit": "base-code-point-plus-optional-variation-selector",
+    "maxVariationSelectors": 1,
+    "variationSelectorMustFollowBase": true,
+    "isolatedVariationSelector": "independent-atom"
+  },
+  "locales": {
+    "classes": [ "zh-Hans", "zh-Hant", "ja", "other" ],
+    "primaryLanguageBoundary": "exact-or-hyphen-subtag",
+    "classificationOrder": [
+      "ja-primary-language",
+      "zh-explicit-script",
+      "zh-traditional-region",
+      "zh-bare-default",
+      "zh-remaining",
+      "other"
+    ],
+    "zh": {
+      "explicitTraditionalSubtag": "hant",
+      "explicitSimplifiedSubtag": "hans",
+      "traditionalRegions": [ "tw", "hk", "mo" ],
+      "bare": "zh-Hant",
+      "remaining": "zh-Hans"
+    },
+    "ja": "ja",
+    "defaultPolicies": {
+      "zh-Hans": "kaiming",
+      "zh-Hant": "quanjiao",
+      "ja": "quanjiao",
+      "other": "quanjiao"
+    },
+    "resolutionPriority": [
+      "element-attribute-or-style-policy-nearest-ancestor",
+      "locale-class-policy",
+      "global-style",
+      "default-policy"
+    ]
+  },
+  "punctuationStyles": {
+    "baselineHalfWidthClasses": {
+      "quanjiao": [],
+      "kaiming": [ "pause", "open", "close" ],
+      "banjiao": [ "stop", "pause", "open", "close" ]
+    },
+    "adjacency": {
+      "rule": "squeeze-each-facing-blank-side-independently",
+      "sentenceBoundaryException": {
+        "left": "stop",
+        "right": "open",
+        "preserve": "left-right-blank"
+      }
+    }
+  },
+  "spacing": {
+    "gapEm": 0.25,
+    "directions": {
+      "cjkToAns": "ansAfterCjk",
+      "ansToCjk": "ansBeforeCjk",
+      "asymmetry": "at-sign-only-after-cjk"
+    },
+    "authorSpaceDisposition": {
+      "cjkAns": "remove-and-use-managed-gap",
+      "ansCjk": "remove-and-use-managed-gap",
+      "cjkCjk": "remove-without-gap",
+      "revertRestoresRemovedSpace": false
+    }
+  },
+  "lineBreaking": {
+    "prohibitedBreaks": {
+      "before": [ "close", "pause", "stop", "middle", "liga" ],
+      "after": [ "open" ]
+    },
+    "orphan": {
+      "defaultMeaningfulCjkChars": 2,
+      "counts": "cjk-atoms-not-classified-as-biaodian",
+      "excludes": [ "biaodian", "non-cjk" ]
+    }
+  },
+  "applicabilityLevels": {
+    "text": [ "spacing", "punctuation-classification-and-width" ],
+    "paragraph": [ "prohibited-breaks", "orphan-minimum", "long-word-hyphenation" ]
+  },
+  "semanticProcessing": {
+    "dependencies": [
+      { "before": "atomize", "after": "classify" },
+      { "before": "classify", "after": "derive-spacing" },
+      { "before": "classify", "after": "derive-punctuation-width" },
+      { "before": "classify", "after": "derive-break-constraints" },
+      { "before": "classify", "after": "derive-orphan-constraint" }
+    ],
+    "conditionalLineEdgeOutcomes": [
+      "punctuation-half-width-at-line-edge",
+      "remove-managed-gap-at-line-end"
+    ]
+  },
+  "adapterBoundary": {
+    "platformNeutral": [
+      "unicode-character-classes",
+      "logical-atom-boundaries",
+      "locale-classification-and-policy",
+      "punctuation-width-semantics",
+      "spacing-decisions-and-distance",
+      "prohibited-break-relations",
+      "orphan-count",
+      "applicability-levels-and-semantic-dependencies"
+    ],
+    "excludedAdapterMechanisms": [
+      "generated-node-or-element-representation",
+      "ownership-and-reversion-bookkeeping",
+      "rendering-primitive-selection",
+      "generated-line-break-hints",
+      "mutation-observation-or-rerun-lifecycle",
+      "layout-measurement",
+      "platform-specific-workarounds"
+    ]
+  }
+}
@@ -0,0 +1,389 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://juzhen.dev/schemas/juzhen-rules.schema.json",
+  "title": "Juzhen shared typography rules",
+  "type": "object",
+  "additionalProperties": false,
+  "required": [
+    "schemaVersion",
+    "rulesetVersion",
+    "unicodeProfile",
+    "characterClasses",
+    "atoms",
+    "locales",
+    "punctuationStyles",
+    "spacing",
+    "lineBreaking",
+    "applicabilityLevels",
+    "semanticProcessing",
+    "adapterBoundary"
+  ],
+  "properties": {
+    "$schema": { "type": "string" },
+    "schemaVersion": { "const": 1 },
+    "rulesetVersion": {
+      "type": "string",
+      "pattern": "^(0|[1-9][0-9]*)\\.(0|[1-9][0-9]*)\\.(0|[1-9][0-9]*)$"
+    },
+    "unicodeProfile": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "version", "coverage" ],
+      "properties": {
+        "version": { "const": "17.0.0" },
+        "coverage": { "type": "string", "minLength": 1 }
+      }
+    },
+    "characterClasses": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [
+        "representation",
+        "cjk",
+        "ansAfterCjk",
+        "ansBeforeCjk",
+        "variationSelectors",
+        "biaodian",
+        "overrideSemantics"
+      ],
+      "properties": {
+        "representation": { "const": "unicode-code-point-ranges" },
+        "cjk": { "$ref": "#/$defs/codePointSet" },
+        "ansAfterCjk": { "$ref": "#/$defs/codePointSet" },
+        "ansBeforeCjk": { "$ref": "#/$defs/codePointSet" },
+        "variationSelectors": { "$ref": "#/$defs/codePointSet" },
+        "biaodian": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "classificationOrder", "classes" ],
+          "properties": {
+            "classificationOrder": {
+              "type": "array",
+              "prefixItems": [
+                { "const": "stop" },
+                { "const": "pause" },
+                { "const": "open" },
+                { "const": "close" },
+                { "const": "middle" },
+                { "const": "liga" }
+              ],
+              "items": false,
+              "minItems": 6,
+              "maxItems": 6
+            },
+            "classes": {
+              "type": "object",
+              "additionalProperties": false,
+              "required": [ "stop", "pause", "open", "close", "middle", "liga" ],
+              "properties": {
+                "stop": { "$ref": "#/$defs/codePointSet" },
+                "pause": { "$ref": "#/$defs/codePointSet" },
+                "open": { "$ref": "#/$defs/codePointSet" },
+                "close": { "$ref": "#/$defs/codePointSet" },
+                "middle": { "$ref": "#/$defs/codePointSet" },
+                "liga": { "$ref": "#/$defs/codePointSet" }
+              }
+            }
+          }
+        },
+        "overrideSemantics": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "mode", "westernTargets", "removalAllowed", "reclassificationAllowed" ],
+          "properties": {
+            "mode": { "const": "append" },
+            "westernTargets": {
+              "type": "array",
+              "prefixItems": [ { "const": "ansAfterCjk" }, { "const": "ansBeforeCjk" } ],
+              "items": false,
+              "minItems": 2,
+              "maxItems": 2
+            },
+            "removalAllowed": { "const": false },
+            "reclassificationAllowed": { "const": false }
+          }
+        }
+      }
+    },
+    "atoms": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [
+        "unit",
+        "maxVariationSelectors",
+        "variationSelectorMustFollowBase",
+        "isolatedVariationSelector"
+      ],
+      "properties": {
+        "unit": { "const": "base-code-point-plus-optional-variation-selector" },
+        "maxVariationSelectors": { "const": 1 },
+        "variationSelectorMustFollowBase": { "const": true },
+        "isolatedVariationSelector": { "const": "independent-atom" }
+      }
+    },
+    "locales": { "$ref": "#/$defs/locales" },
+    "punctuationStyles": { "$ref": "#/$defs/punctuationStyles" },
+    "spacing": { "$ref": "#/$defs/spacing" },
+    "lineBreaking": { "$ref": "#/$defs/lineBreaking" },
+    "applicabilityLevels": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "text", "paragraph" ],
+      "properties": {
+        "text": { "$ref": "#/$defs/stringArray" },
+        "paragraph": { "$ref": "#/$defs/stringArray" }
+      }
+    },
+    "semanticProcessing": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "dependencies", "conditionalLineEdgeOutcomes" ],
+      "properties": {
+        "dependencies": {
+          "type": "array",
+          "minItems": 1,
+          "items": {
+            "type": "object",
+            "additionalProperties": false,
+            "required": [ "before", "after" ],
+            "properties": {
+              "before": { "$ref": "#/$defs/semanticStage" },
+              "after": { "$ref": "#/$defs/semanticStage" }
+            }
+          }
+        },
+        "conditionalLineEdgeOutcomes": {
+          "type": "array",
+          "prefixItems": [
+            { "const": "punctuation-half-width-at-line-edge" },
+            { "const": "remove-managed-gap-at-line-end" }
+          ],
+          "items": false,
+          "minItems": 2,
+          "maxItems": 2
+        }
+      }
+    },
+    "adapterBoundary": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "platformNeutral", "excludedAdapterMechanisms" ],
+      "properties": {
+        "platformNeutral": { "$ref": "#/$defs/stringArray" },
+        "excludedAdapterMechanisms": { "$ref": "#/$defs/stringArray" }
+      }
+    }
+  },
+  "$defs": {
+    "hexCodePoint": {
+      "type": "string",
+      "pattern": "^(?:[0-9A-F]{4}|[0-9A-F]{5}|10[0-9A-F]{4})$"
+    },
+    "codePointEntry": {
+      "oneOf": [
+        {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "codePoint" ],
+          "properties": {
+            "codePoint": { "$ref": "#/$defs/hexCodePoint" }
+          }
+        },
+        {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "start", "end" ],
+          "properties": {
+            "start": { "$ref": "#/$defs/hexCodePoint" },
+            "end": { "$ref": "#/$defs/hexCodePoint" }
+          }
+        }
+      ]
+    },
+    "codePointSet": {
+      "type": "array",
+      "minItems": 1,
+      "uniqueItems": true,
+      "items": { "$ref": "#/$defs/codePointEntry" }
+    },
+    "stringArray": {
+      "type": "array",
+      "items": { "type": "string", "minLength": 1 },
+      "uniqueItems": true
+    },
+    "style": {
+      "enum": [ "quanjiao", "kaiming", "banjiao" ]
+    },
+    "semanticStage": {
+      "enum": [
+        "atomize",
+        "classify",
+        "derive-spacing",
+        "derive-punctuation-width",
+        "derive-break-constraints",
+        "derive-orphan-constraint"
+      ]
+    },
+    "locales": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [
+        "classes",
+        "primaryLanguageBoundary",
+        "classificationOrder",
+        "zh",
+        "ja",
+        "defaultPolicies",
+        "resolutionPriority"
+      ],
+      "properties": {
+        "classes": {
+          "type": "array",
+          "prefixItems": [
+            { "const": "zh-Hans" },
+            { "const": "zh-Hant" },
+            { "const": "ja" },
+            { "const": "other" }
+          ],
+          "items": false,
+          "minItems": 4,
+          "maxItems": 4
+        },
+        "primaryLanguageBoundary": { "const": "exact-or-hyphen-subtag" },
+        "classificationOrder": { "$ref": "#/$defs/stringArray" },
+        "zh": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "explicitTraditionalSubtag", "explicitSimplifiedSubtag", "traditionalRegions", "bare", "remaining" ],
+          "properties": {
+            "explicitTraditionalSubtag": { "const": "hant" },
+            "explicitSimplifiedSubtag": { "const": "hans" },
+            "traditionalRegions": {
+              "type": "array",
+              "prefixItems": [ { "const": "tw" }, { "const": "hk" }, { "const": "mo" } ],
+              "items": false,
+              "minItems": 3,
+              "maxItems": 3
+            },
+            "bare": { "const": "zh-Hant" },
+            "remaining": { "const": "zh-Hans" }
+          }
+        },
+        "ja": { "const": "ja" },
+        "defaultPolicies": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "zh-Hans", "zh-Hant", "ja", "other" ],
+          "properties": {
+            "zh-Hans": { "$ref": "#/$defs/style" },
+            "zh-Hant": { "$ref": "#/$defs/style" },
+            "ja": { "$ref": "#/$defs/style" },
+            "other": { "$ref": "#/$defs/style" }
+          }
+        },
+        "resolutionPriority": {
+          "type": "array",
+          "prefixItems": [
+            { "const": "element-attribute-or-style-policy-nearest-ancestor" },
+            { "const": "locale-class-policy" },
+            { "const": "global-style" },
+            { "const": "default-policy" }
+          ],
+          "items": false,
+          "minItems": 4,
+          "maxItems": 4
+        }
+      }
+    },
+    "punctuationStyles": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "baselineHalfWidthClasses", "adjacency" ],
+      "properties": {
+        "baselineHalfWidthClasses": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "quanjiao", "kaiming", "banjiao" ],
+          "properties": {
+            "quanjiao": { "$ref": "#/$defs/stringArray" },
+            "kaiming": { "$ref": "#/$defs/stringArray" },
+            "banjiao": { "$ref": "#/$defs/stringArray" }
+          }
+        },
+        "adjacency": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "rule", "sentenceBoundaryException" ],
+          "properties": {
+            "rule": { "const": "squeeze-each-facing-blank-side-independently" },
+            "sentenceBoundaryException": {
+              "type": "object",
+              "additionalProperties": false,
+              "required": [ "left", "right", "preserve" ],
+              "properties": {
+                "left": { "const": "stop" },
+                "right": { "const": "open" },
+                "preserve": { "const": "left-right-blank" }
+              }
+            }
+          }
+        }
+      }
+    },
+    "spacing": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "gapEm", "directions", "authorSpaceDisposition" ],
+      "properties": {
+        "gapEm": { "const": 0.25 },
+        "directions": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "cjkToAns", "ansToCjk", "asymmetry" ],
+          "properties": {
+            "cjkToAns": { "const": "ansAfterCjk" },
+            "ansToCjk": { "const": "ansBeforeCjk" },
+            "asymmetry": { "const": "at-sign-only-after-cjk" }
+          }
+        },
+        "authorSpaceDisposition": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "cjkAns", "ansCjk", "cjkCjk", "revertRestoresRemovedSpace" ],
+          "properties": {
+            "cjkAns": { "const": "remove-and-use-managed-gap" },
+            "ansCjk": { "const": "remove-and-use-managed-gap" },
+            "cjkCjk": { "const": "remove-without-gap" },
+            "revertRestoresRemovedSpace": { "const": false }
+          }
+        }
+      }
+    },
+    "lineBreaking": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [ "prohibitedBreaks", "orphan" ],
+      "properties": {
+        "prohibitedBreaks": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "before", "after" ],
+          "properties": {
+            "before": { "$ref": "#/$defs/stringArray" },
+            "after": { "$ref": "#/$defs/stringArray" }
+          }
+        },
+        "orphan": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [ "defaultMeaningfulCjkChars", "counts", "excludes" ],
+          "properties": {
+            "defaultMeaningfulCjkChars": { "const": 2 },
+            "counts": { "const": "cjk-atoms-not-classified-as-biaodian" },
+            "excludes": { "$ref": "#/$defs/stringArray" }
+          }
+        }
+      }
+    }
+  }
+}