@dev-pi2pie/word-counter 0.1.2 → 0.1.3-canary.1

package/dist/esm/bin.mjs

@@ -146,7 +146,7 @@ function createDebugChannel(options) {
 
 //#endregion
 //#region src/cli/path/filter.ts
-const DEFAULT_INCLUDE_EXTENSIONS = new Set([
+const DEFAULT_INCLUDE_EXTENSIONS = Object.freeze([
 	".md",
 	".markdown",
 	".mdx",
@@ -223,13 +223,13 @@ function shouldIncludeFromDirectoryRegex(relativePath, filter) {
 
 //#endregion
 //#region src/cli/total-of.ts
-const TOTAL_OF_PARTS = [
+const TOTAL_OF_PARTS = Object.freeze([
 	"words",
 	"emoji",
 	"symbols",
 	"punctuation",
 	"whitespace"
-];
+]);
 const TOTAL_OF_PART_ALIASES = {
 	word: "words",
 	words: "words",
@@ -342,8 +342,11 @@ const PATH_MODE_CHOICES = ["auto", "manual"];
 function collectPathValue(value, previous = []) {
 	return [...previous, value];
 }
+function collectLatinHintValue(value, previous = []) {
+	return [...previous, value];
+}
 function configureProgramOptions(program, parseMode) {
-	program.addOption(new Option("-m, --mode <mode>", "breakdown mode").choices(MODE_CHOICES).argParser(parseMode).default("chunk")).addOption(new Option("-f, --format <format>", "output format").choices(FORMAT_CHOICES).default("standard")).addOption(new Option("--section <section>", "document section mode").choices(SECTION_CHOICES).default("all")).addOption(new Option("--path-mode <mode>", "path resolution mode: auto (default) expands directories; manual treats --path values as literal files").choices(PATH_MODE_CHOICES).default("auto")).option("--latin-language <language>", "hint a language tag for Latin script text").option("--latin-tag <tag>", "hint a BCP 47 tag for Latin script text").option("--latin-locale <locale>", "legacy alias of --latin-language").option("--han-language <language>", "hint a language tag for Han script text").option("--han-tag <tag>", "hint a BCP 47 tag for Han script text").option("--non-words", "collect emoji, symbols, and punctuation (excludes whitespace)").option("--include-whitespace", "include whitespace counts (implies with --non-words; same as --misc)").option("--misc", "collect non-words plus whitespace (alias for --include-whitespace)").option("--total-of <parts>", "override total composition (comma-separated): words,emoji,symbols,punctuation,whitespace", parseTotalOfOption).option("--pretty", "pretty print JSON output", false).option("--debug", "enable debug diagnostics on stderr").option("--verbose", "emit verbose per-file debug diagnostics (requires --debug)").option("--debug-report [path]", "write debug diagnostics to a report file").option("--debug-report-tee", "mirror debug diagnostics to both report file and stderr").option("--debug-tee", "alias of --debug-report-tee").option("--merged", "show merged aggregate output (default)").option("--per-file", "show per-file output plus merged summary").option("--no-progress", "disable batch progress indicator").option("--keep-progress", "keep final batch progress line visible in standard mode").option("--no-recursive", "disable recursive directory traversal").option("--quiet-skips", "hide skip diagnostics (applies when --debug is enabled)").option("--include-ext <exts>", "comma-separated extensions to include during directory scanning", collectExtensionOption, []).option("--exclude-ext <exts>", "comma-separated extensions to exclude during directory scanning", collectExtensionOption, []).option("--regex <pattern>", "regex filter for directory-scanned paths (applies to --path directories only)").option("-p, --path <path>", "read input from file or directory (directories expand in auto mode by default)", collectPathValue, []).argument("[text...]", "text to count").showHelpAfterError();
+	program.addOption(new Option("-m, --mode <mode>", "breakdown mode").choices(MODE_CHOICES).argParser(parseMode).default("chunk")).addOption(new Option("-f, --format <format>", "output format").choices(FORMAT_CHOICES).default("standard")).addOption(new Option("--section <section>", "document section mode").choices(SECTION_CHOICES).default("all")).addOption(new Option("--path-mode <mode>", "path resolution mode: auto (default) expands directories; manual treats --path values as literal files").choices(PATH_MODE_CHOICES).default("auto")).option("--latin-language <language>", "hint a language tag for Latin script text").option("--latin-tag <tag>", "hint a BCP 47 tag for Latin script text").option("--latin-locale <locale>", "legacy alias of --latin-language").option("--latin-hint <tag>=<pattern>", "add a custom Latin hint rule (repeatable)", collectLatinHintValue, []).option("--latin-hints-file <path>", "load custom Latin hint rules from a JSON file").option("--no-default-latin-hints", "disable built-in Latin hint rules").option("--han-language <language>", "hint a language tag for Han script text").option("--han-tag <tag>", "hint a BCP 47 tag for Han script text").option("--non-words", "collect emoji, symbols, and punctuation (excludes whitespace)").option("--include-whitespace", "include whitespace counts (implies with --non-words; same as --misc)").option("--misc", "collect non-words plus whitespace (alias for --include-whitespace)").option("--total-of <parts>", "override total composition (comma-separated): words,emoji,symbols,punctuation,whitespace", parseTotalOfOption).option("--pretty", "pretty print JSON output", false).option("--debug", "enable debug diagnostics on stderr").option("--verbose", "emit verbose per-file debug diagnostics (requires --debug)").option("--debug-report [path]", "write debug diagnostics to a report file").option("--debug-report-tee", "mirror debug diagnostics to both report file and stderr").option("--debug-tee", "alias of --debug-report-tee").option("--merged", "show merged aggregate output (default)").option("--per-file", "show per-file output plus merged summary").option("--no-progress", "disable batch progress indicator").option("--keep-progress", "keep final batch progress line visible in standard mode").option("--no-recursive", "disable recursive directory traversal").option("--quiet-skips", "hide skip diagnostics (applies when --debug is enabled)").option("--include-ext <exts>", "comma-separated extensions to include during directory scanning", collectExtensionOption, []).option("--exclude-ext <exts>", "comma-separated extensions to exclude during directory scanning", collectExtensionOption, []).option("--regex <pattern>", "regex filter for directory-scanned paths (applies to --path directories only)").option("-p, --path <path>", "read input from file or directory (directories expand in auto mode by default)", collectPathValue, []).argument("[text...]", "text to count").showHelpAfterError();
 }
 
 //#endregion
@@ -418,7 +421,7 @@ var require_picocolors = /* @__PURE__ */ __commonJSMin(((exports, module) => {
 //#endregion
 //#region src/cli/program/version-embedded.ts
 var import_picocolors = /* @__PURE__ */ __toESM(require_picocolors(), 1);
-const EMBEDDED_PACKAGE_VERSION = "0.1.2";
+const EMBEDDED_PACKAGE_VERSION = "0.1.3-canary.1";
 
 //#endregion
 //#region src/cli/program/version.ts
@@ -1605,10 +1608,53 @@ function resolveMode(input, fallback = "chunk") {
 	return normalizeMode(input) ?? fallback;
 }
 
+//#endregion
+//#region src/wc/latin-hints.ts
+const DEFAULT_LATIN_HINT_RULES_SOURCE = [
+	{
+		tag: "de",
+		pattern: "[äöüÄÖÜß]"
+	},
+	{
+		tag: "es",
+		pattern: "[ñÑ¿¡]"
+	},
+	{
+		tag: "pt",
+		pattern: "[ãõÃÕ]"
+	},
+	{
+		tag: "fr",
+		pattern: "[œŒæÆ]"
+	},
+	{
+		tag: "pl",
+		pattern: "[ąćęłńśźżĄĆĘŁŃŚŹŻ]"
+	},
+	{
+		tag: "tr",
+		pattern: "[ıİğĞşŞ]"
+	},
+	{
+		tag: "ro",
+		pattern: "[ăĂâÂîÎșȘțȚ]"
+	},
+	{
+		tag: "hu",
+		pattern: "[őŐűŰ]"
+	},
+	{
+		tag: "is",
+		pattern: "[ðÐþÞ]"
+	}
+];
+const DEFAULT_LATIN_HINT_RULES = Object.freeze(DEFAULT_LATIN_HINT_RULES_SOURCE.map((rule) => Object.freeze({ ...rule })));
+
 //#endregion
 //#region src/wc/locale-detect.ts
 const DEFAULT_LOCALE = "und-Latn";
-const DEFAULT_HAN_TAG = "zh-Hani";
+const DEFAULT_HAN_TAG = "und-Hani";
+const MAX_LATIN_HINT_PATTERN_LENGTH = 256;
 const regex = {
 	hiragana: /\p{Script=Hiragana}/u,
 	katakana: /\p{Script=Katakana}/u,
@@ -1620,31 +1666,10 @@ const regex = {
 	devanagari: /\p{Script=Devanagari}/u,
 	thai: /\p{Script=Thai}/u
 };
-const latinLocaleHints = [
-	{
-		locale: "de",
-		regex: /[äöüÄÖÜß]/
-	},
-	{
-		locale: "es",
-		regex: /[ñÑ¿¡]/
-	},
-	{
-		locale: "pt",
-		regex: /[ãõÃÕ]/
-	},
-	{
-		locale: "fr",
-		regex: /[œŒæÆ]/
-	}
-];
-const latinLocales = new Set([DEFAULT_LOCALE, ...latinLocaleHints.map((hint) => hint.locale)]);
-function isLatinLocale(locale) {
-	return latinLocales.has(locale);
-}
-function detectLatinLocale(char) {
-	for (const hint of latinLocaleHints) if (hint.regex.test(char)) return hint.locale;
-	return DEFAULT_LOCALE;
+const defaultLatinLocales = new Set([DEFAULT_LOCALE, ...DEFAULT_LATIN_HINT_RULES.map((hint) => hint.tag)]);
+function isLatinLocale(locale, context) {
+	if (context) return context.latinLocales.has(locale);
+	return defaultLatinLocales.has(locale);
 }
 function resolveLatinHint(options) {
 	const latinTagHint = options.latinTagHint?.trim();
@@ -1660,7 +1685,82 @@ function resolveHanHint(options) {
 	const hanLanguageHint = options.hanLanguageHint?.trim();
 	if (hanLanguageHint) return hanLanguageHint;
 }
-function detectLocaleForChar(char, previousLocale, options = {}) {
+function compileLatinHintPattern(pattern, label) {
+	const source = typeof pattern === "string" ? pattern : pattern.source;
+	const hasUnicodeMode = typeof pattern !== "string" && (pattern.flags.includes("u") || pattern.flags.includes("v"));
+	const flags = typeof pattern === "string" ? "u" : hasUnicodeMode ? pattern.flags : `${pattern.flags}u`;
+	if (source.length === 0) throw new Error(`${label}: pattern must not be empty.`);
+	if (source.length > MAX_LATIN_HINT_PATTERN_LENGTH) throw new Error(`${label}: pattern must be at most ${MAX_LATIN_HINT_PATTERN_LENGTH} characters.`);
+	try {
+		return new RegExp(source, flags);
+	} catch (error) {
+		const message = error instanceof Error ? error.message : String(error);
+		throw new Error(`${label}: invalid Unicode regex pattern (${message}).`);
+	}
+}
+function normalizeLatinHintPriority(priority, label) {
+	if (priority === void 0) return 0;
+	if (typeof priority !== "number" || !Number.isFinite(priority)) throw new Error(`${label}: priority must be a finite number when provided.`);
+	return priority;
+}
+function compileLatinHintRule(rule, order, label) {
+	const tag = typeof rule.tag === "string" ? rule.tag.trim() : "";
+	if (!tag) throw new Error(`${label}: tag must be a non-empty string.`);
+	return {
+		tag,
+		pattern: compileLatinHintPattern(rule.pattern, label),
+		priority: normalizeLatinHintPriority(rule.priority, label),
+		order
+	};
+}
+function resolveLatinHintRules$1(options) {
+	const useDefaultLatinHints = options.useDefaultLatinHints !== false;
+	const customRules = options.latinHintRules ?? [];
+	const combinedRules = [];
+	for (let index = 0; index < customRules.length; index += 1) {
+		const rule = customRules[index];
+		if (!rule) continue;
+		combinedRules.push({
+			rule,
+			label: `Invalid custom Latin hint rule at index ${index}`
+		});
+	}
+	if (useDefaultLatinHints) for (let index = 0; index < DEFAULT_LATIN_HINT_RULES.length; index += 1) {
+		const rule = DEFAULT_LATIN_HINT_RULES[index];
+		if (!rule) continue;
+		combinedRules.push({
+			rule,
+			label: `Invalid default Latin hint rule at index ${index}`
+		});
+	}
+	const resolvedRules = combinedRules.map((entry, index) => compileLatinHintRule(entry.rule, index, entry.label));
+	resolvedRules.sort((left, right) => {
+		if (left.priority !== right.priority) return right.priority - left.priority;
+		return left.order - right.order;
+	});
+	return resolvedRules;
+}
+function resolveLocaleDetectContext(options = {}) {
+	const latinHint = resolveLatinHint(options);
+	const latinHintRules = resolveLatinHintRules$1(options);
+	const latinLocales = new Set([DEFAULT_LOCALE]);
+	for (const rule of latinHintRules) latinLocales.add(rule.tag);
+	if (latinHint) latinLocales.add(latinHint);
+	return {
+		latinHint,
+		hanHint: resolveHanHint(options),
+		latinHintRules,
+		latinLocales
+	};
+}
+function detectLatinLocale(char, context) {
+	for (const hint of context.latinHintRules) {
+		hint.pattern.lastIndex = 0;
+		if (hint.pattern.test(char)) return hint.tag;
+	}
+	return DEFAULT_LOCALE;
+}
+function detectLocaleForChar(char, previousLocale, options = {}, context = resolveLocaleDetectContext(options)) {
 	if (regex.hiragana.test(char) || regex.katakana.test(char)) return "ja";
 	if (regex.hangul.test(char)) return "ko";
 	if (regex.arabic.test(char)) return "ar";
@@ -1669,14 +1769,13 @@ function detectLocaleForChar(char, previousLocale, options = {}) {
 	if (regex.thai.test(char)) return "th";
 	if (regex.han.test(char)) {
 		if (previousLocale && previousLocale.startsWith("ja")) return previousLocale;
-		return resolveHanHint(options) ?? DEFAULT_HAN_TAG;
+		return context.hanHint ?? DEFAULT_HAN_TAG;
 	}
 	if (regex.latin.test(char)) {
-		const hintedLocale = detectLatinLocale(char);
+		const hintedLocale = detectLatinLocale(char, context);
 		if (hintedLocale !== DEFAULT_LOCALE) return hintedLocale;
-		if (previousLocale && isLatinLocale(previousLocale) && previousLocale !== DEFAULT_LOCALE) return previousLocale;
-		const latinHint = resolveLatinHint(options);
-		if (latinHint) return latinHint;
+		if (previousLocale && isLatinLocale(previousLocale, context) && previousLocale !== DEFAULT_LOCALE) return previousLocale;
+		if (context.latinHint) return context.latinHint;
 		return DEFAULT_LOCALE;
 	}
 	return null;
@@ -1685,12 +1784,13 @@ function detectLocaleForChar(char, previousLocale, options = {}) {
 //#endregion
 //#region src/wc/segment.ts
 function segmentTextByLocale(text, options = {}) {
+	const context = resolveLocaleDetectContext(options);
 	const chunks = [];
 	let currentLocale = DEFAULT_LOCALE;
 	let buffer = "";
 	let bufferHasScript = false;
 	for (const char of text) {
-		const detected = detectLocaleForChar(char, currentLocale, options);
+		const detected = detectLocaleForChar(char, currentLocale, options, context);
 		const targetLocale = detected ?? currentLocale;
 		if (buffer === "") {
 			currentLocale = targetLocale;
@@ -1705,7 +1805,7 @@ function segmentTextByLocale(text, options = {}) {
 			continue;
 		}
 		if (targetLocale !== currentLocale && detected !== null) {
-			if (currentLocale === DEFAULT_LOCALE && isLatinLocale(targetLocale)) {
+			if (currentLocale === DEFAULT_LOCALE && isLatinLocale(targetLocale, context)) {
 				currentLocale = targetLocale;
 				buffer += char;
 				bufferHasScript = true;
@@ -1758,6 +1858,8 @@ function wordCounter(text, options = {}) {
 		latinLanguageHint: options.latinLanguageHint,
 		latinTagHint: options.latinTagHint,
 		latinLocaleHint: options.latinLocaleHint,
+		latinHintRules: options.latinHintRules,
+		useDefaultLatinHints: options.useDefaultLatinHints,
 		hanLanguageHint: options.hanLanguageHint,
 		hanTagHint: options.hanTagHint
 	});
@@ -2477,6 +2579,57 @@ function resolveDebugReportPathOption(rawValue) {
 	if (rawValue === void 0 || rawValue === false) return;
 	if (typeof rawValue === "string") return rawValue;
 }
+function parseInlineLatinHintRule(value) {
+	const separatorIndex = value.indexOf("=");
+	if (separatorIndex <= 0) throw new Error("`--latin-hint` must use `<tag>=<pattern>` format.");
+	const tag = value.slice(0, separatorIndex).trim();
+	const pattern = value.slice(separatorIndex + 1);
+	if (!tag) throw new Error("`--latin-hint` tag must be non-empty.");
+	if (!pattern) throw new Error("`--latin-hint` pattern must be non-empty.");
+	return {
+		tag,
+		pattern
+	};
+}
+function parseLatinHintsFileRule(value, index, sourcePath) {
+	if (typeof value !== "object" || value === null) throw new Error(`Invalid Latin hint rule at ${sourcePath}#${index}: rule must be an object.`);
+	const tag = "tag" in value ? value.tag : void 0;
+	const pattern = "pattern" in value ? value.pattern : void 0;
+	const priority = "priority" in value ? value.priority : void 0;
+	if (typeof tag !== "string" || tag.trim().length === 0) throw new Error(`Invalid Latin hint rule at ${sourcePath}#${index}: tag must be a non-empty string.`);
+	if (typeof pattern !== "string") throw new Error(`Invalid Latin hint rule at ${sourcePath}#${index}: pattern must be a string.`);
+	if (priority !== void 0 && (typeof priority !== "number" || !Number.isFinite(priority))) throw new Error(`Invalid Latin hint rule at ${sourcePath}#${index}: priority must be a finite number.`);
+	return {
+		tag,
+		pattern,
+		...priority !== void 0 ? { priority } : {}
+	};
+}
+function parseLatinHintsFile(path) {
+	let raw;
+	try {
+		raw = readFileSync(path, "utf8");
+	} catch (error) {
+		const message = error instanceof Error ? error.message : String(error);
+		throw new Error(`Failed to read Latin hint file (${path}): ${message}`);
+	}
+	let parsed;
+	try {
+		parsed = JSON.parse(raw);
+	} catch (error) {
+		const message = error instanceof Error ? error.message : String(error);
+		throw new Error(`Invalid JSON in Latin hint file (${path}): ${message}`);
+	}
+	if (!Array.isArray(parsed)) throw new Error(`Latin hint file (${path}) must contain a JSON array.`);
+	return parsed.map((rule, index) => parseLatinHintsFileRule(rule, index, path));
+}
+function resolveLatinHintRules(options) {
+	const inlineRules = (options.latinHint ?? []).map((value) => parseInlineLatinHintRule(value));
+	const fileRules = typeof options.latinHintsFile === "string" && options.latinHintsFile.length > 0 ? parseLatinHintsFile(options.latinHintsFile) : [];
+	const mergedRules = [...inlineRules, ...fileRules];
+	if (mergedRules.length === 0) return;
+	return mergedRules;
+}
 function resolveCountRunOptions(options) {
 	const useSection = options.section !== "all";
 	const totalOfParts = options.totalOf;
@@ -2495,6 +2648,8 @@ function resolveCountRunOptions(options) {
 			latinLanguageHint: options.latinLanguage,
 			latinTagHint: options.latinTag,
 			latinLocaleHint: options.latinLocale,
+			latinHintRules: resolveLatinHintRules(options),
+			useDefaultLatinHints: options.defaultLatinHints !== false,
 			hanLanguageHint: options.hanLanguage,
 			hanTagHint: options.hanTag,
 			nonWords: enableNonWords,