@markuplint/parser-utils 5.0.0-rc.1 → 5.0.0-rc.4

package/CHANGELOG.md

@@ -3,6 +3,34 @@
 All notable changes to this project will be documented in this file.
 See [Conventional Commits](https://conventionalcommits.org) for commit guidelines.
 
+# [5.0.0-rc.4](https://github.com/markuplint/markuplint/compare/v5.0.0-rc.3...v5.0.0-rc.4) (2026-04-19)
+
+**Note:** Version bump only for package @markuplint/parser-utils
+
+# [5.0.0-rc.3](https://github.com/markuplint/markuplint/compare/v5.0.0-rc.2...v5.0.0-rc.3) (2026-04-19)
+
+**Note:** Version bump only for package @markuplint/parser-utils
+
+# [5.0.0-rc.2](https://github.com/markuplint/markuplint/compare/v5.0.0-rc.1...v5.0.0-rc.2) (2026-04-15)
+
+- fix(parser-utils)!: align unquoted attribute value tokenizer with HTML spec ([5168c04](https://github.com/markuplint/markuplint/commit/5168c041a1d37a7080f12e3c7569b5bbacd5a2f1))
+- build!: remove ESLint and replace with oxlint ([1e0a337](https://github.com/markuplint/markuplint/commit/1e0a337707f76b903b16beeeb8c4d4fc0d8fc9e4))
+- feat(parser-utils)!: remove deprecated getLine and getCol functions ([51cd7a2](https://github.com/markuplint/markuplint/commit/51cd7a28b7fcdd4a9d52d57f35c55b97b3b196fe))
+
+### BREAKING CHANGES
+
+- The default `endOfUnquotedValueChars` no longer includes `/`.
+  Unquoted attribute values preserve `/` as part of the value, matching the
+  WHATWG HTML "attribute value (unquoted) state". Consumers that relied on `/`
+  as a terminator must pass `endOfUnquotedValueChars: ['\t', '\n', '\f', '\r',
+' ', '/', '>']` explicitly to `visitAttr()` / `attrTokenizer()`.
+
+Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
+
+- ESLint is no longer used. Use oxlint instead.
+- getLine() and getCol() have been removed.
+  Use getPosition() instead.
+
 # [5.0.0-rc.1](https://github.com/markuplint/markuplint/compare/v5.0.0-rc.0...v5.0.0-rc.1) (2026-03-27)
 
 **Note:** Version bump only for package @markuplint/parser-utils

package/SKILL.md

@@ -101,6 +101,7 @@ Customize attribute parsing behavior for a specific parser. Follow recipe #4 in
    - `quoteSet` -- custom quote delimiters (e.g., `{` `}` for JSX)
    - `startState` -- initial AttrState (usually `BeforeName`)
    - `noQuoteValueType` -- value type for unquoted values
+   - `endOfUnquotedValueChars` -- characters that terminate an unquoted value (default: whitespace and `>`, matching the WHATWG HTML spec)
 
 ### Step 2: Post-process
 

package/docs/maintenance.ja.md

@@ -120,6 +120,27 @@ visitAttr(token: Token) {
 }
 ```
 
+#### 引用符なし属性値の終端文字のカスタマイズ
+
+`endOfUnquotedValueChars` のデフォルトは WHATWG HTML の "attribute value (unquoted) state" に準拠し、空白類と `>` のみが値を終端する。`/` は値の一部として扱われる。
+
+ホスト言語が異なる規則を要求する場合はカスタムリストを渡す:
+
+```ts
+// Pug: 文字による終端を無効化 — Pug レクサが属性境界を既に区切っている
+super.visitAttr(token, {
+  quoteSet: [],
+  endOfUnquotedValueChars: [],
+});
+```
+
+```ts
+// `/` を終端文字として扱う（例: `a=x/>` を属性 `a="x"` + 自己閉じ `/>` として読む）
+super.visitAttr(token, {
+  endOfUnquotedValueChars: ['\t', '\n', '\f', '\r', ' ', '/', '>'],
+});
+```
+
 ### 5. IDL属性マッピングの追加
 
 IDL属性マップは `src/idl-attributes.ts` で定義されています。新しいマッピングを追加するには:

package/docs/maintenance.md

@@ -120,6 +120,27 @@ visitAttr(token: Token) {
 }
 ```
 
+#### Customizing Unquoted Value Terminators
+
+The default `endOfUnquotedValueChars` follows the WHATWG HTML "attribute value (unquoted) state": only whitespace and `>` terminate the value. `/` is part of the value.
+
+Pass a custom list when the host language requires different rules:
+
+```ts
+// Pug: never terminate by character — the Pug lexer already delimits attrs
+super.visitAttr(token, {
+  quoteSet: [],
+  endOfUnquotedValueChars: [],
+});
+```
+
+```ts
+// Treat `/` as a terminator (e.g., to read `a=x/>` as attribute `a="x"` + self-closing `/>`)
+super.visitAttr(token, {
+  endOfUnquotedValueChars: ['\t', '\n', '\f', '\r', ' ', '/', '>'],
+});
+```
+
 ### 5. Adding an IDL Attribute Mapping
 
 The IDL attribute map is defined in `src/idl-attributes.ts`. To add a new mapping:

package/docs/parser-class.ja.md

@@ -282,6 +282,17 @@ visitAttr(
 
 また、`visitSpreadAttr()` を通じてスプレッド属性の検出も試みます。
 
+#### オプション
+
+| オプション                | デフォルト                                                                             | 説明                                                                                                                                                                                                                                                                                |
+| ------------------------- | -------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `quoteSet`                | `[{ start: '"', end: '"', type: 'string' }, { start: "'", end: "'", type: 'string' }]` | 属性値として認識される引用符のセット。JSX / Vue / Svelte / Astro / MDX では `{` `}` を `type: 'script'` として追加する。                                                                                                                                                            |
+| `noQuoteValueType`        | `'string'`                                                                             | 引用符で囲まれていない値に割り当てられる値の型。                                                                                                                                                                                                                                    |
+| `endOfUnquotedValueChars` | `['\t', '\n', '\f', '\r', ' ', '>']`                                                   | 引用符なし属性値を終端する文字。WHATWG HTML の "attribute value (unquoted) state" に準拠し、空白類と `>` のみ。`/` は終端ではなく値に含まれる（自己閉じの `/` は前段の before/after attribute name state で処理される）。終端判定を完全に無効化するには `[]` を渡す（Pug で利用）。 |
+| `startState`              | `AttrState.BeforeName`                                                                 | 属性トークナイザの初期状態。JSX の `{...spread}` のような式のみの属性を扱う場合は `AttrState.BeforeValue` を指定する。                                                                                                                                                              |
+
+> 参照: [WHATWG HTML — attribute value (unquoted) state](<https://html.spec.whatwg.org/multipage/parsing.html#attribute-value-(unquoted)-state>)
+
 ### visitSpreadAttr()
 
 ```ts

package/docs/parser-class.md

@@ -282,6 +282,17 @@ If the raw string contains multiple attributes, only the first is parsed and the
 
 Also attempts to detect spread attributes via `visitSpreadAttr()`.
 
+#### Options
+
+| Option                    | Default                                                                                | Description                                                                                                                                                                                                                                                                                                                                 |
+| ------------------------- | -------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `quoteSet`                | `[{ start: '"', end: '"', type: 'string' }, { start: "'", end: "'", type: 'string' }]` | Quote delimiters recognized for the attribute value. Override to add expression delimiters such as `{` `}` (JSX / Vue / Svelte / Astro / MDX) with `type: 'script'`.                                                                                                                                                                        |
+| `noQuoteValueType`        | `'string'`                                                                             | Value type assigned when the value is unquoted.                                                                                                                                                                                                                                                                                             |
+| `endOfUnquotedValueChars` | `['\t', '\n', '\f', '\r', ' ', '>']`                                                   | Characters that terminate an unquoted attribute value. Matches the WHATWG HTML "attribute value (unquoted) state" — whitespace and `>` only. `/` is NOT a terminator and is included in the value (self-closing `/` is handled earlier, in the before/after attribute name state). Pass `[]` to disable termination entirely (used by Pug). |
+| `startState`              | `AttrState.BeforeName`                                                                 | Initial state of the attribute tokenizer. Use `AttrState.BeforeValue` to start directly at the value (e.g., for bare expression attributes like JSX `{...spread}`).                                                                                                                                                                         |
+
+> See: [WHATWG HTML — attribute value (unquoted) state](<https://html.spec.whatwg.org/multipage/parsing.html#attribute-value-(unquoted)-state>)
+
 ### visitSpreadAttr()
 
 ```ts

package/lib/attr-tokenizer.d.ts

@@ -4,6 +4,7 @@ import { AttrState } from './enums.js';
  * @see https://html.spec.whatwg.org/multipage/parsing.html#tag-name-state
  * @see https://html.spec.whatwg.org/multipage/parsing.html#before-attribute-name-state
  * @see https://html.spec.whatwg.org/multipage/parsing.html#attribute-name-state
+ * @see https://html.spec.whatwg.org/multipage/parsing.html#attribute-value-(unquoted)-state
  */
 export declare function attrTokenizer(raw: string, quoteSet?: readonly QuoteSet[], startState?: AttrState, noQuoteValueType?: ValueType, endOfUnquotedValueChars?: ReadonlyArray<string>): {
     spacesBeforeAttrName: string;

package/lib/attr-tokenizer.js

@@ -11,8 +11,13 @@ const EQUAL = '=';
  * @see https://html.spec.whatwg.org/multipage/parsing.html#tag-name-state
  * @see https://html.spec.whatwg.org/multipage/parsing.html#before-attribute-name-state
  * @see https://html.spec.whatwg.org/multipage/parsing.html#attribute-name-state
+ * @see https://html.spec.whatwg.org/multipage/parsing.html#attribute-value-(unquoted)-state
  */
-export function attrTokenizer(raw, quoteSet = defaultQuoteSet, startState = AttrState.BeforeName, noQuoteValueType = 'string', endOfUnquotedValueChars = [...defaultSpaces, '/', '>']) {
+export function attrTokenizer(raw, quoteSet = defaultQuoteSet, startState = AttrState.BeforeName, noQuoteValueType = 'string', 
+// Default follows the spec referenced above: only whitespace and `>` terminate
+// the value. `/` is handled by the outer tag tokenizer (before/after attribute
+// name state), not here.
+endOfUnquotedValueChars = [...defaultSpaces, '>']) {
     let state = startState;
     let spacesBeforeAttrName = '';
     let attrName = '';

package/lib/const.js

@@ -94,6 +94,7 @@ export const svgElementList = [
     'tref',
     'vkern',
 ];
+// eslint-disable-next-line no-control-regex -- WHATWG HTML spec requires matching NULL character
 export const reTagName = /^[a-z][^\0\t\n\f />]*/i;
 export const reSplitterTag = /<[^>]+>/g;
 /**

package/lib/get-location.d.ts

@@ -1,12 +1,4 @@
 export { getPosition } from '@markuplint/shared';
-/**
- * @deprecated Use {@link getPosition} instead. Will be removed in v5.0.0.
- */
-export declare function getLine(rawCodeFragment: string, startOffset: number): number;
-/**
- * @deprecated Use {@link getPosition} instead. Will be removed in v5.0.0.
- */
-export declare function getCol(rawCodeFragment: string, startOffset: number): number;
 export declare function getEndLine(rawCodeFragment: string, startLine: number): number;
 export declare function getEndCol(rawCodeFragment: string, startCol: number): number;
 /**

package/lib/get-location.js

@@ -1,18 +1,5 @@
 export { getPosition } from '@markuplint/shared';
 const LINE_BREAK = '\n';
-/**
- * @deprecated Use {@link getPosition} instead. Will be removed in v5.0.0.
- */
-export function getLine(rawCodeFragment, startOffset) {
-    return rawCodeFragment.slice(0, startOffset).split(LINE_BREAK).length;
-}
-/**
- * @deprecated Use {@link getPosition} instead. Will be removed in v5.0.0.
- */
-export function getCol(rawCodeFragment, startOffset) {
-    const lines = rawCodeFragment.slice(0, startOffset).split(LINE_BREAK);
-    return (lines.at(-1) ?? '').length + 1;
-}
 export function getEndLine(rawCodeFragment, startLine) {
     return rawCodeFragment.split(LINE_BREAK).length - 1 + startLine;
 }

package/package.json

@@ -1,8 +1,12 @@
 {
 	"name": "@markuplint/parser-utils",
-	"version": "5.0.0-rc.1",
+	"version": "5.0.0-rc.4",
 	"description": "Utility module for markuplint parser plugin",
-	"repository": "git@github.com:markuplint/markuplint.git",
+	"repository": {
+		"type": "git",
+		"url": "https://github.com/markuplint/markuplint.git",
+		"directory": "packages/@markuplint/parser-utils"
+	},
 	"author": "Yusuke Hirao <yusukehirao@me.com>",
 	"license": "MIT",
 	"engines": {
@@ -31,16 +35,16 @@
 		"clean": "tsc --build --clean tsconfig.build.json"
 	},
 	"dependencies": {
-		"@markuplint/ml-ast": "5.0.0-rc.1",
-		"@markuplint/ml-spec": "5.0.0-rc.1",
-		"@markuplint/shared": "5.0.0-rc.1",
-		"@markuplint/types": "5.0.0-rc.1",
+		"@markuplint/ml-ast": "5.0.0-rc.4",
+		"@markuplint/ml-spec": "5.0.0-rc.4",
+		"@markuplint/shared": "5.0.0-rc.4",
+		"@markuplint/types": "5.0.0-rc.4",
 		"debug": "4.4.3",
 		"espree": "11.2.0",
-		"type-fest": "5.5.0"
+		"type-fest": "5.6.0"
 	},
 	"devDependencies": {
-		"@typescript-eslint/typescript-estree": "8.57.2"
+		"@typescript-eslint/typescript-estree": "8.58.2"
 	},
-	"gitHead": "0d6b4324d9a7d6b9e1ba57d4a57e45d36975cba9"
+	"gitHead": "97a6339bbae23f556de5d307b3ce2ef7cfd9402d"
 }