Add toothless G. Fix #442, #443

utility/export-data/supported-languages.js

@@ -0,0 +1,71 @@
+const cldr = require("cldr");
+const fs = require("fs-extra");
+const path = require("path");
+
+module.exports = async function(charMapPath) {
+	const charMap = await fs.readJson(charMapPath);
+
+	const supportLocaleSet = new Set();
+	const codePointSet = new Set();
+	for (const ch of charMap) for (const unicode of ch[1]) codePointSet.add(unicode);
+
+	for (const locale of cldr.localeIds) {
+		const exemplar = cldr.extractCharacters(locale).exemplar;
+		if (!exemplar) continue;
+		const basicChars = [...(exemplar.default || [])];
+		const fullChars = [
+			...basicChars,
+			...(exemplar.auxiliary || []),
+			...(exemplar.index || []),
+			...(exemplar.numbers || []),
+			...(exemplar.punctuation || [])
+		].join("");
+
+		let fullSupport = true;
+		let basicSupport = true;
+		for (const ch of basicChars) {
+			if (!codePointSet.has(ch.codePointAt(0))) basicSupport = false;
+		}
+		for (const ch of fullChars) {
+			if (!codePointSet.has(ch.codePointAt(0))) fullSupport = false;
+		}
+
+		if (basicSupport) {
+			supportLocaleSet.add(locale);
+		}
+	}
+	for (const loc of supportLocaleSet) {
+		const seg = loc.split("_");
+		if (seg.length < 2) continue;
+		for (let m = 1; m < seg.length; m++) {
+			const upperLoc = seg.slice(0, m).join("_");
+			if (upperLoc && supportLocaleSet.has(upperLoc)) {
+				supportLocaleSet.delete(loc);
+			}
+		}
+	}
+	const supportLangSet = new Set();
+	for (const loc of supportLocaleSet) {
+		const seg = loc.split("_");
+		let displayName = null;
+		for (let m = 1; m <= seg.length; m++) {
+			const upperLoc = seg.slice(0, m).join("_");
+			const subDisplayName = cldr.extractLanguageDisplayNames("en")[upperLoc];
+			if (subDisplayName)
+				displayName = subDisplayName + (upperLoc === loc ? "" : "\u00A0(" + loc + ")");
+		}
+		if (displayName) supportLangSet.add(displayName);
+	}
+
+	const unicodeCoverage = new Map();
+	for (const [gn, codes, cl] of charMap) for (const u of codes) unicodeCoverage.set(u, cl);
+
+	return {
+		stats: {
+			glyphCount: charMap.length,
+			codePointCount: unicodeCoverage.size
+		},
+		unicodeCoverage: Array.from(unicodeCoverage).sort((a, b) => a[0] - b[0]),
+		languages: Array.from(supportLangSet).sort()
+	};
+};