@sushichan044/kg-core 0.10.1 → 0.10.2

package/README.md

@@ -150,6 +150,24 @@ tate-chu-yoko unit. Each render unit carries its selected `presentation` as a
 string. A `CombinedGlyphUnit` renders the two digits together; the composer owns
 that decision, so renderers do not compare source group ranges.
 
+ASCII sentences are recognized before individual words are classified. A continuous
+stretch of ASCII letters, digits, punctuation, and spaces containing a letter and
+multiple alphanumeric groups separated by spaces or punctuation renders sideways.
+Japanese characters, fullwidth spaces, and newlines end that context. Thus `I love you`,
+`This is a Web novel`, `I'm happy!`, and `STR 10` keep their initials, abbreviations,
+numbers, and punctuation sideways. Standalone `I`, `BGM`, and `Web` retain their
+upright presentation, and standalone `12` still forms a tate-chu-yoko unit.
+Standalone dotted abbreviations whose parts match the upright abbreviation rule,
+such as `U.S.` and `Ph.D.`, keep their letters upright and their periods in the
+existing mixed presentation. In a sentence such as `I live in the U.S.`, the
+abbreviation and its periods render sideways with the surrounding words.
+
+Sentences can wrap between words. Words, including attached punctuation,
+apostrophes, and hyphens, stay together; ASCII spaces retain
+their source positions and supply line-break opportunities. Ruby and inline styling
+do not interrupt sentence recognition. Non-ASCII Western text keeps the existing
+presentation rules.
+
 An upright Western character sets solid against the kana or kanji beside it rather than
 taking the quarter em 3.2.6 puts around a sideways run, following JLReq 3.2.4: it classes
 with kanji, the same way 3.2.5 classes a tate-chu-yoko run. `QR` and `URL` set upright by

package/dist/describe-cause-SABI_1ex.mjs

@@ -0,0 +1,9 @@
+//#region src/internal/describe-cause.ts
+/**
+* Renders an unknown throwable for display, without assuming it is an Error.
+*/
+function describeCause(cause) {
+	return cause instanceof Error ? cause.message : String(cause);
+}
+//#endregion
+export { describeCause as t };

package/dist/index.mjs

@@ -1,6 +1,6 @@
-import { c as graphemeSegmenter, d as ManuscriptDiagnostic, l as ValidationIssue, m as ManuscriptRange, p as NamespacedId, u as ManuscriptResult, v as readonlyObject } from "./question-or-exclamation-spacing-Cp3eOGlA.mjs";
-import { a as PaperSize, d as ManuscriptOffsets, f as ManuscriptAppearanceSettings, g as ParsedManuscript, h as FontPresetId, i as ManuscriptGeometry, l as NovelCompositionSettings, m as FontSizePt, n as novelComposer, p as PaperSizeId, u as NovelFlowSettings } from "./novel-composer-D_orH6ET.mjs";
-import { n as assertNever, t as describeCause } from "./describe-cause-BV8mZ3Um.mjs";
+import { c as assertNever, d as ManuscriptResult, f as ManuscriptDiagnostic, h as ManuscriptRange, l as graphemeSegmenter, m as NamespacedId, u as ValidationIssue, y as readonlyObject } from "./question-or-exclamation-spacing-C5cFLeKr.mjs";
+import { a as PaperSize, d as ManuscriptOffsets, f as ManuscriptAppearanceSettings, g as ParsedManuscript, h as FontPresetId, i as ManuscriptGeometry, l as NovelCompositionSettings, m as FontSizePt, n as novelComposer, p as PaperSizeId, u as NovelFlowSettings } from "./novel-composer-CHNWyvn8.mjs";
+import { t as describeCause } from "./describe-cause-SABI_1ex.mjs";
 import * as v from "valibot";
 const Rejection = { schema: readonlyObject({ reason: v.string() }) };
 //#endregion

package/dist/lint.mjs

@@ -1,5 +1,5 @@
-import { _ as readonlyArray, a as CLOSING_BRACKETS, c as graphemeSegmenter, d as ManuscriptDiagnostic, f as DiagnosticSeverity, l as ValidationIssue, m as ManuscriptRange, n as leadingSpaces, o as OPENING_BRACKETS, p as NamespacedId, s as characterClass, t as questionOrExclamationSpacings, u as ManuscriptResult, v as readonlyObject } from "./question-or-exclamation-spacing-Cp3eOGlA.mjs";
-import { n as assertNever, t as describeCause } from "./describe-cause-BV8mZ3Um.mjs";
+import { a as CLOSING_BRACKETS, c as assertNever, d as ManuscriptResult, f as ManuscriptDiagnostic, h as ManuscriptRange, l as graphemeSegmenter, m as NamespacedId, n as leadingSpaces, o as OPENING_BRACKETS, p as DiagnosticSeverity, s as characterClass, t as questionOrExclamationSpacings, u as ValidationIssue, v as readonlyArray, y as readonlyObject } from "./question-or-exclamation-spacing-C5cFLeKr.mjs";
+import { t as describeCause } from "./describe-cause-SABI_1ex.mjs";
 import * as v from "valibot";
 //#region src/proofreading/invalid-rule-options.ts
 const InvalidRuleOptions = { describe: (error) => `ルール ${error.ruleId} のオプション "${error.option}" が不正です: ${error.issues.map(({ message }) => message).join(", ")}` };

package/dist/{novel-composer-D_orH6ET.mjs → novel-composer-CHNWyvn8.mjs}

@@ -1,4 +1,4 @@
-import { _ as readonlyArray, a as CLOSING_BRACKETS, c as graphemeSegmenter, g as nonNegativeInteger, h as TextRange, i as TATE_CHU_YOKO_MARKS, m as ManuscriptRange, o as OPENING_BRACKETS, p as NamespacedId, r as DIVIDING_PUNCTUATION, s as characterClass, t as questionOrExclamationSpacings, u as ManuscriptResult, v as readonlyObject } from "./question-or-exclamation-spacing-Cp3eOGlA.mjs";
+import { _ as nonNegativeInteger, a as CLOSING_BRACKETS, c as assertNever, d as ManuscriptResult, g as TextRange, h as ManuscriptRange, i as TATE_CHU_YOKO_MARKS, l as graphemeSegmenter, m as NamespacedId, o as OPENING_BRACKETS, r as DIVIDING_PUNCTUATION, s as characterClass, t as questionOrExclamationSpacings, v as readonlyArray, y as readonlyObject } from "./question-or-exclamation-spacing-C5cFLeKr.mjs";
 import * as v from "valibot";
 import { eastAsianWidth } from "get-east-asian-width";
 const BoldAnnotation = { schema: readonlyObject({
@@ -365,6 +365,9 @@ const ASCII_TWO_DIGITS = /^[0-9]{2}$/u;
 const UPRIGHT_LATIN_ABBREVIATION = /^(?:[A-Z]+|[A-Z][a-z]{1,2})$/u;
 const TATE_CHU_YOKO_MARK = new RegExp(`^[${characterClass(TATE_CHU_YOKO_MARKS)}]$`, "u");
 const FULLWIDTH_ALPHANUMERIC = /^[Ａ-Ｚａ-ｚ０-９]$/u;
+const ASCII_TEXT = /^[ -~]$/u;
+const ASCII_WORD = /^[!-~]$/u;
+const WESTERN_CONTEXT = /[A-Za-z0-9][^A-Za-z0-9]+[A-Za-z0-9]/u;
 function stretchEnd(graphemes, start, pattern) {
 	let end = start + 1;
 	while (pattern.test(graphemes[end]?.value ?? "")) end += 1;
@@ -377,6 +380,10 @@ function presentationOf(text) {
 	}
 	return FULLWIDTH_ALPHANUMERIC.test(text) ? "upright" : "mixed";
 }
+/**
+* Recognize sentence-wide presentation while keeping words as separate line-break units. Standalone
+* dotted abbreviations retain the existing per-token presentation.
+*/
 function recognize(graphemes) {
 	const runs = [];
 	const push = (members, presentation) => {
@@ -388,9 +395,25 @@ function recognize(graphemes) {
 		});
 	};
 	let cursor = 0;
+	let checkedThrough = 0;
 	while (cursor < graphemes.length) {
 		const first = graphemes[cursor];
 		if (first === void 0) break;
+		if (cursor >= checkedThrough && ASCII_TEXT.test(first.value)) {
+			const contextEnd = stretchEnd(graphemes, cursor, ASCII_TEXT);
+			checkedThrough = contextEnd;
+			const text = graphemes.slice(cursor, contextEnd).map(({ value }) => value).join("");
+			const abbreviationParts = text.trim().replace(/\.$/u, "").split(".");
+			const standaloneDottedAbbreviation = abbreviationParts.length > 1 && abbreviationParts.every((part) => UPRIGHT_LATIN_ABBREVIATION.test(part));
+			if (/[A-Za-z]/u.test(text) && WESTERN_CONTEXT.test(text) && !standaloneDottedAbbreviation) {
+				while (cursor < contextEnd) {
+					const end = ASCII_WORD.test(graphemes[cursor]?.value ?? "") ? stretchEnd(graphemes, cursor, ASCII_WORD) : cursor + 1;
+					push(graphemes.slice(cursor, end), "sideways");
+					cursor = end;
+				}
+				continue;
+			}
+		}
 		if (TATE_CHU_YOKO_MARK.test(first.value)) {
 			const end = stretchEnd(graphemes, cursor, TATE_CHU_YOKO_MARK);
 			const members = graphemes.slice(cursor, end);
@@ -959,29 +982,35 @@ const WESTERN = /^[\p{Script=Latin}\p{Script=Greek}\p{Script=Cyrillic}]$/u;
 * the same as a tate-chu-yoko run's 3.2.5.
 */
 function classify({ value, presentation }) {
-	if (presentation === "tate-chu-yoko") return COMBINED_DIVIDING_MARKS.has(value) ? "cl-04" : "cl-30";
-	if (presentation === "upright") return "cl-19";
-	if (OPENING.has(value)) return "cl-01";
-	if (CLOSING.has(value)) return "cl-02";
-	if (HYPHENS.has(value)) return "cl-03";
-	if (DIVIDING_MARKS.has(value)) return "cl-04";
-	if (MIDDLE_DOTS.has(value)) return "cl-05";
-	if (value === "。" || value === "．") return "cl-06";
-	if (value === "、" || value === "，") return "cl-07";
-	if (INSEPARABLE.has(value)) return "cl-08";
-	if (ITERATION_MARKS.has(value)) return "cl-09";
-	if (value === "ー") return "cl-10";
-	if (SMALL_KANA.has(value)) return "cl-11";
-	if (PREFIXED_ABBREVIATIONS.has(value)) return "cl-12";
-	if (POSTFIXED_ABBREVIATIONS.has(value)) return "cl-13";
-	if (value === "　") return "cl-14";
-	if (HIRAGANA.test(value)) return "cl-15";
-	if (KATAKANA.test(value)) return "cl-16";
-	if (GROUPED_NUMERAL.test(value)) return "cl-24";
-	if (UNIT_SYMBOLS.has(value)) return "cl-25";
-	if (value === " ") return "cl-26";
-	if (WESTERN.test(value)) return "cl-27";
-	return "cl-19";
+	switch (presentation) {
+		case "tate-chu-yoko": return COMBINED_DIVIDING_MARKS.has(value) ? "cl-04" : "cl-30";
+		case "upright": return "cl-19";
+		case "sideways":
+		case "mixed":
+			if (presentation === "sideways" && /^[!-~]$/u.test(value) && !/^[A-Za-z0-9]$/u.test(value)) return "cl-27";
+			if (OPENING.has(value)) return "cl-01";
+			if (CLOSING.has(value)) return "cl-02";
+			if (HYPHENS.has(value)) return "cl-03";
+			if (DIVIDING_MARKS.has(value)) return "cl-04";
+			if (MIDDLE_DOTS.has(value)) return "cl-05";
+			if (value === "。" || value === "．") return "cl-06";
+			if (value === "、" || value === "，") return "cl-07";
+			if (INSEPARABLE.has(value)) return "cl-08";
+			if (ITERATION_MARKS.has(value)) return "cl-09";
+			if (value === "ー") return "cl-10";
+			if (SMALL_KANA.has(value)) return "cl-11";
+			if (PREFIXED_ABBREVIATIONS.has(value)) return "cl-12";
+			if (POSTFIXED_ABBREVIATIONS.has(value)) return "cl-13";
+			if (value === "　") return "cl-14";
+			if (HIRAGANA.test(value)) return "cl-15";
+			if (KATAKANA.test(value)) return "cl-16";
+			if (GROUPED_NUMERAL.test(value)) return "cl-24";
+			if (UNIT_SYMBOLS.has(value)) return "cl-25";
+			if (value === " ") return "cl-26";
+			if (WESTERN.test(value)) return "cl-27";
+			return "cl-19";
+		default: return assertNever(presentation);
+	}
 }
 /**
 * 無印 in 表1 with no colour in 表6: the pair is set solid and line adjustment leaves it alone.

package/dist/plugin.mjs

@@ -1,5 +1,5 @@
-import { l as ValidationIssue, p as NamespacedId, u as ManuscriptResult, v as readonlyObject } from "./question-or-exclamation-spacing-Cp3eOGlA.mjs";
-import { c as MeasurementTextRange, g as ParsedManuscript, h as FontPresetId, m as FontSizePt, o as RunMeasurement, r as logicalRunMeasurer, s as MeasuredCluster, t as createNovelComposer } from "./novel-composer-D_orH6ET.mjs";
+import { d as ManuscriptResult, m as NamespacedId, u as ValidationIssue, y as readonlyObject } from "./question-or-exclamation-spacing-C5cFLeKr.mjs";
+import { c as MeasurementTextRange, g as ParsedManuscript, h as FontPresetId, m as FontSizePt, o as RunMeasurement, r as logicalRunMeasurer, s as MeasuredCluster, t as createNovelComposer } from "./novel-composer-CHNWyvn8.mjs";
 import * as v from "valibot";
 const PresentationKind = { schema: v.picklist([
 	"mixed",

package/dist/{question-or-exclamation-spacing-Cp3eOGlA.mjs → question-or-exclamation-spacing-C5cFLeKr.mjs}

@@ -171,6 +171,15 @@ const ValidationIssue = {
 */
 const graphemeSegmenter = new Intl.Segmenter("ja", { granularity: "grapheme" });
 //#endregion
+//#region src/internal/assert-never.ts
+/**
+* Marks a branch the compiler proved unreachable. Reaching it at runtime means a variant was added
+* to a union without updating the switch that produced this call.
+*/
+function assertNever(value) {
+	throw new TypeError(`unhandled variant: ${JSON.stringify(value)}`);
+}
+//#endregion
 //#region src/internal/japanese-brackets.ts
 /**
 * Brackets and quotes a Japanese paragraph may open with. Halfwidth forms are included because a
@@ -259,4 +268,4 @@ function questionOrExclamationSpacings(line) {
 	return [...line.matchAll(MARKS)].map((marks) => analyzeMatch(line, marks)).filter((spacing) => spacing !== void 0);
 }
 //#endregion
-export { readonlyArray as _, CLOSING_BRACKETS as a, graphemeSegmenter as c, ManuscriptDiagnostic as d, DiagnosticSeverity as f, nonNegativeInteger as g, TextRange as h, TATE_CHU_YOKO_MARKS as i, ValidationIssue as l, ManuscriptRange as m, leadingSpaces as n, OPENING_BRACKETS as o, NamespacedId as p, DIVIDING_PUNCTUATION as r, characterClass as s, questionOrExclamationSpacings as t, ManuscriptResult as u, readonlyObject as v };
+export { nonNegativeInteger as _, CLOSING_BRACKETS as a, assertNever as c, ManuscriptResult as d, ManuscriptDiagnostic as f, TextRange as g, ManuscriptRange as h, TATE_CHU_YOKO_MARKS as i, graphemeSegmenter as l, NamespacedId as m, leadingSpaces as n, OPENING_BRACKETS as o, DiagnosticSeverity as p, DIVIDING_PUNCTUATION as r, characterClass as s, questionOrExclamationSpacings as t, ValidationIssue as u, readonlyArray as v, readonlyObject as y };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sushichan044/kg-core",
-  "version": "0.10.1",
+  "version": "0.10.2",
   "description": "Framework-independent parsing, composition, and proofreading for Japanese manuscripts",
   "keywords": [
     "japanese",

package/dist/describe-cause-BV8mZ3Um.mjs

@@ -1,18 +0,0 @@
-//#region src/internal/assert-never.ts
-/**
-* Marks a branch the compiler proved unreachable. Reaching it at runtime means a variant was added
-* to a union without updating the switch that produced this call.
-*/
-function assertNever(value) {
-	throw new TypeError(`unhandled variant: ${JSON.stringify(value)}`);
-}
-//#endregion
-//#region src/internal/describe-cause.ts
-/**
-* Renders an unknown throwable for display, without assuming it is an Error.
-*/
-function describeCause(cause) {
-	return cause instanceof Error ? cause.message : String(cause);
-}
-//#endregion
-export { assertNever as n, describeCause as t };