pdf-codec 5.7.2 → 5.7.4

package/dist/encoding.js

@@ -1,2 +1,271 @@
-import { a as glyphNameToUnicode, i as WINANSI_GLYPH_NAMES, o as namedEncodingGlyphName, s as winAnsiGlyphName } from "./afm-widths-BNI8qa2S.js";
-export { WINANSI_GLYPH_NAMES, glyphNameToUnicode, namedEncodingGlyphName, winAnsiGlyphName };
+import { macRomanGlyphName, standardGlyphName } from "./encoding-standard-tables.js";
+//#region src/encoding.ts
+const WINANSI_GLYPH_NAMES = [
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"",
+	"space",
+	"exclam",
+	"quotedbl",
+	"numbersign",
+	"dollar",
+	"percent",
+	"ampersand",
+	"quotesingle",
+	"parenleft",
+	"parenright",
+	"asterisk",
+	"plus",
+	"comma",
+	"hyphen",
+	"period",
+	"slash",
+	"zero",
+	"one",
+	"two",
+	"three",
+	"four",
+	"five",
+	"six",
+	"seven",
+	"eight",
+	"nine",
+	"colon",
+	"semicolon",
+	"less",
+	"equal",
+	"greater",
+	"question",
+	"at",
+	"A",
+	"B",
+	"C",
+	"D",
+	"E",
+	"F",
+	"G",
+	"H",
+	"I",
+	"J",
+	"K",
+	"L",
+	"M",
+	"N",
+	"O",
+	"P",
+	"Q",
+	"R",
+	"S",
+	"T",
+	"U",
+	"V",
+	"W",
+	"X",
+	"Y",
+	"Z",
+	"bracketleft",
+	"backslash",
+	"bracketright",
+	"asciicircum",
+	"underscore",
+	"grave",
+	"a",
+	"b",
+	"c",
+	"d",
+	"e",
+	"f",
+	"g",
+	"h",
+	"i",
+	"j",
+	"k",
+	"l",
+	"m",
+	"n",
+	"o",
+	"p",
+	"q",
+	"r",
+	"s",
+	"t",
+	"u",
+	"v",
+	"w",
+	"x",
+	"y",
+	"z",
+	"braceleft",
+	"bar",
+	"braceright",
+	"asciitilde",
+	"bullet",
+	"Euro",
+	"bullet",
+	"quotesinglbase",
+	"florin",
+	"quotedblbase",
+	"ellipsis",
+	"dagger",
+	"daggerdbl",
+	"circumflex",
+	"perthousand",
+	"Scaron",
+	"guilsinglleft",
+	"OE",
+	"bullet",
+	"Zcaron",
+	"bullet",
+	"bullet",
+	"quoteleft",
+	"quoteright",
+	"quotedblleft",
+	"quotedblright",
+	"bullet",
+	"endash",
+	"emdash",
+	"tilde",
+	"trademark",
+	"scaron",
+	"guilsinglright",
+	"oe",
+	"bullet",
+	"zcaron",
+	"Ydieresis",
+	"space",
+	"exclamdown",
+	"cent",
+	"sterling",
+	"currency",
+	"yen",
+	"brokenbar",
+	"section",
+	"dieresis",
+	"copyright",
+	"ordfeminine",
+	"guillemotleft",
+	"logicalnot",
+	"hyphen",
+	"registered",
+	"macron",
+	"degree",
+	"plusminus",
+	"twosuperior",
+	"threesuperior",
+	"acute",
+	"mu",
+	"paragraph",
+	"periodcentered",
+	"cedilla",
+	"onesuperior",
+	"ordmasculine",
+	"guillemotright",
+	"onequarter",
+	"onehalf",
+	"threequarters",
+	"questiondown",
+	"Agrave",
+	"Aacute",
+	"Acircumflex",
+	"Atilde",
+	"Adieresis",
+	"Aring",
+	"AE",
+	"Ccedilla",
+	"Egrave",
+	"Eacute",
+	"Ecircumflex",
+	"Edieresis",
+	"Igrave",
+	"Iacute",
+	"Icircumflex",
+	"Idieresis",
+	"Eth",
+	"Ntilde",
+	"Ograve",
+	"Oacute",
+	"Ocircumflex",
+	"Otilde",
+	"Odieresis",
+	"multiply",
+	"Oslash",
+	"Ugrave",
+	"Uacute",
+	"Ucircumflex",
+	"Udieresis",
+	"Yacute",
+	"Thorn",
+	"germandbls",
+	"agrave",
+	"aacute",
+	"acircumflex",
+	"atilde",
+	"adieresis",
+	"aring",
+	"ae",
+	"ccedilla",
+	"egrave",
+	"eacute",
+	"ecircumflex",
+	"edieresis",
+	"igrave",
+	"iacute",
+	"icircumflex",
+	"idieresis",
+	"eth",
+	"ntilde",
+	"ograve",
+	"oacute",
+	"ocircumflex",
+	"otilde",
+	"odieresis",
+	"divide",
+	"oslash",
+	"ugrave",
+	"uacute",
+	"ucircumflex",
+	"udieresis",
+	"yacute",
+	"thorn",
+	"ydieresis"
+];
+function winAnsiGlyphName(code) {
+	const name = WINANSI_GLYPH_NAMES[code];
+	return name === void 0 || name === "" ? void 0 : name;
+}
+function namedEncodingGlyphName(encodingName) {
+	if (encodingName === "WinAnsiEncoding") return winAnsiGlyphName;
+	if (encodingName === "MacRomanEncoding") return macRomanGlyphName;
+	if (encodingName === "StandardEncoding") return standardGlyphName;
+}
+//#endregion
+export { WINANSI_GLYPH_NAMES, namedEncodingGlyphName, winAnsiGlyphName };

package/dist/font-read.cjs

@@ -1,12 +1,14 @@
 Object.defineProperty(exports, Symbol.toStringTag, { value: "Module" });
-const require_afm_widths = require("./afm-widths-sjaRvmQX.cjs");
+const require_encoding = require("./encoding.cjs");
+const require_afm_widths = require("./afm-widths.cjs");
 const require_objects = require("./objects.cjs");
 const require_filters = require("./filters.cjs");
+const require_font_style = require("./font-style.cjs");
+const require_glyph_names = require("./glyph-names.cjs");
 const require_builtin_encoding = require("./builtin-encoding.cjs");
 const require_cmap = require("./cmap.cjs");
 const require_encoding_symbol_tables = require("./encoding-symbol-tables.cjs");
 const require_fonts = require("./fonts.cjs");
-const require_font_style = require("./font-style.cjs");
 //#region src/font-read.ts
 const BITS_PER_BYTE = 8;
 const SYMBOLIC_FLAG_BIT = 4;
@@ -26,7 +28,7 @@ function builtinSymbolGlyphNameLookup(baseFamily) {
 	if (baseFamily === "Symbol") return require_encoding_symbol_tables.symbolGlyphName;
 	if (baseFamily === "ZapfDingbats") return require_encoding_symbol_tables.zapfDingbatsGlyphName;
 }
-function readFontProgram(descriptor, context) {
+function readFontProgram(descriptor, baseFont, context) {
 	if (descriptor === void 0) return;
 	for (const key of [
 		"FontFile2",
@@ -36,14 +38,14 @@ function readFontProgram(descriptor, context) {
 		const stream = context.resolver.resolve(require_objects.dictGet(descriptor, key));
 		if (stream?.kind !== "stream") continue;
 		const decoded = require_filters.decodeStream(stream.raw, stream.dict, context.sink);
-		const encoding = require_builtin_encoding.readFontProgramEncoding(decoded.bytes);
+		const encoding = require_builtin_encoding.readFontProgramEncoding(decoded.bytes, baseFont);
 		if (encoding !== void 0) return encoding;
 	}
 }
-function lazyFontProgram(descriptor, context) {
+function lazyFontProgram(descriptor, baseFont, context) {
 	let read;
 	return () => {
-		read ??= { encoding: readFontProgram(descriptor, context) };
+		read ??= { encoding: readFontProgram(descriptor, baseFont, context) };
 		return read.encoding;
 	};
 }
@@ -113,31 +115,40 @@ function buildSimpleFont(fontDict, context) {
 	const encodingObj = require_objects.dictGet(fontDict, "Encoding");
 	const encodingDict = context.resolver.resolveDict(encodingObj);
 	const baseEncodingName = require_objects.asName(encodingDict !== void 0 ? require_objects.dictGet(encodingDict, "BaseEncoding") : encodingObj);
-	const namedBaseGlyphName = baseEncodingName !== void 0 ? require_afm_widths.namedEncodingGlyphName(baseEncodingName) : void 0;
+	const namedBaseGlyphName = baseEncodingName !== void 0 ? require_encoding.namedEncodingGlyphName(baseEncodingName) : void 0;
 	if (baseEncodingName !== void 0 && namedBaseGlyphName === void 0) context.sink({
 		code: "char/encoding-approximated",
 		severity: "info",
 		message: `base encoding "${baseEncodingName}" has no dedicated table here and is approximated as WinAnsiEncoding`
 	});
 	const differencesMap = readDifferencesMap(encodingDict !== void 0 ? resolvedArray(require_objects.dictGet(encodingDict, "Differences"), context) : void 0);
-	const programEncoding = lazyFontProgram(descriptor, context);
+	const programEncoding = lazyFontProgram(descriptor, baseFont, context);
+	const unicodeForDifferencesName = (name) => {
+		const named = require_glyph_names.glyphNameToUnicode(name);
+		if (named !== void 0) return named;
+		const program = programEncoding();
+		const glyphId = program?.glyphIdOfName(name);
+		return glyphId === void 0 ? void 0 : program?.glyphIdToUnicode(glyphId);
+	};
 	const symbolEncoded = symbolic || builtinGlyphName !== void 0;
 	const byGlyphName = (source) => (code) => {
 		const name = source(code);
-		return name === void 0 ? void 0 : require_afm_widths.glyphNameToUnicode(name);
+		return name === void 0 ? void 0 : require_glyph_names.glyphNameToUnicode(name);
 	};
 	const fromProgram = (code) => programEncoding()?.codeToUnicode(code);
 	const namedBase = namedBaseGlyphName === void 0 ? void 0 : byGlyphName(namedBaseGlyphName);
-	const resolvers = [byGlyphName((code) => differencesMap.get(code)), ...symbolEncoded ? [
+	const resolvers = (symbolEncoded ? [
 		fromProgram,
 		builtinGlyphName === void 0 ? void 0 : byGlyphName(builtinGlyphName),
 		namedBase
 	] : [
 		namedBase,
 		fromProgram,
-		byGlyphName(require_afm_widths.winAnsiGlyphName)
-	]].filter((resolver) => resolver !== void 0);
+		byGlyphName(require_encoding.winAnsiGlyphName)
+	]).filter((resolver) => resolver !== void 0);
 	const unicodeForCode = (code) => {
+		const differencesName = differencesMap.get(code);
+		if (differencesName !== void 0) return unicodeForDifferencesName(differencesName);
 		for (const resolve of resolvers) {
 			const unicode = resolve(code);
 			if (unicode !== void 0) return unicode;
@@ -154,7 +165,7 @@ function buildSimpleFont(fontDict, context) {
 			}
 			const unicode = unicodeForCode(code);
 			if (unicode !== void 0) {
-				out += String.fromCodePoint(unicode);
+				out += unicode;
 				continue;
 			}
 			unmapped++;
@@ -265,6 +276,29 @@ function readsVertically(fontDict, context) {
 	if (encoding?.kind === "stream") return require_objects.asNumber(require_objects.dictGet(encoding.dict, "WMode")) === VERTICAL_WRITING_MODE;
 	return false;
 }
+const CID_TO_GID_ENTRY_SIZE = 2;
+/**
+* How a composite font's codes reach the embedded program's glyphs, so the program itself can say what a glyph is when the font dictionary's /ToUnicode CMap is absent or incomplete: through the glyph's own name, its program's Unicode mapping read backwards, or the glyph order of the core face the program is.
+*
+* With /Identity-H or /Identity-V a code is its CID. A /CIDToGIDMap that is absent or /Identity then makes the CID the glyph index, and a /CIDToGIDMap stream gives each CID its glyph index explicitly (ISO 32000-1 9.7.4.2). Any other /Encoding is a CMap from codes to CIDs this reader does not apply, and then the program is not consulted at all rather than being read against the wrong glyph.
+* @param fontDict - the Type0 font dictionary
+* @param descendantDict - its descendant CIDFont dictionary, if it has a readable one
+* @param context - the font read whose resolver follows references and whose sink receives stream-decoding diagnostics
+* @returns a lookup from CID to glyph index, answering undefined for a CID the map does not reach, or undefined when codes do not reach glyphs in a way this reader applies
+*/
+function cidToGlyphId(fontDict, descendantDict, context) {
+	const encodingName = require_objects.dictGet(fontDict, "Encoding");
+	if (!require_objects.isName(encodingName, "Identity-H") && !require_objects.isName(encodingName, "Identity-V")) return;
+	const map = context.resolver.resolve(descendantDict !== void 0 ? require_objects.dictGet(descendantDict, "CIDToGIDMap") : void 0);
+	if (map === void 0 || require_objects.isName(map, "Identity")) return (cid) => cid;
+	if (map.kind !== "stream") return;
+	const bytes = require_filters.decodeStream(map.raw, map.dict, context.sink).bytes;
+	return (cid) => {
+		const at = cid * CID_TO_GID_ENTRY_SIZE;
+		if (at + 1 >= bytes.length) return;
+		return (bytes[at] ?? 0) << BITS_PER_BYTE | (bytes[at + 1] ?? 0);
+	};
+}
 function buildCompositeFont(fontDict, context) {
 	const baseFont = require_objects.asName(require_objects.dictGet(fontDict, "BaseFont")) ?? "Helvetica";
 	const descendants = resolvedArray(require_objects.dictGet(fontDict, "DescendantFonts"), context);
@@ -287,9 +321,8 @@ function buildCompositeFont(fontDict, context) {
 		};
 	};
 	const toUnicode = readToUnicodeCMap(fontDict, context);
-	const cidToGidMap = descendantDict !== void 0 ? require_objects.dictGet(descendantDict, "CIDToGIDMap") : void 0;
-	const encodingName = require_objects.dictGet(fontDict, "Encoding");
-	const programEncoding = (require_objects.isName(encodingName, "Identity-H") || require_objects.isName(encodingName, "Identity-V")) && (cidToGidMap === void 0 || require_objects.isName(cidToGidMap, "Identity")) ? lazyFontProgram(descriptor, context) : void 0;
+	const glyphIdOf = cidToGlyphId(fontDict, descendantDict, context);
+	const programEncoding = glyphIdOf !== void 0 ? lazyFontProgram(descriptor, baseFont, context) : void 0;
 	const decodeToUnicode = (codes) => {
 		let out = "";
 		let glyphCount = 0;
@@ -302,9 +335,10 @@ function buildCompositeFont(fontDict, context) {
 				out += mapped;
 				continue;
 			}
-			const fromProgram = programEncoding?.()?.glyphIdToUnicode(cid);
+			const glyphId = glyphIdOf?.(cid);
+			const fromProgram = glyphId === void 0 ? void 0 : programEncoding?.()?.glyphIdToUnicode(glyphId);
 			if (fromProgram !== void 0) {
-				out += String.fromCodePoint(fromProgram);
+				out += fromProgram;
 				continue;
 			}
 			unmapped++;

package/dist/font-read.js

@@ -1,11 +1,13 @@
-import { a as glyphNameToUnicode, n as afmWidthOfCode, o as namedEncodingGlyphName, s as winAnsiGlyphName } from "./afm-widths-BNI8qa2S.js";
+import { namedEncodingGlyphName, winAnsiGlyphName } from "./encoding.js";
+import { afmWidthOfCode } from "./afm-widths.js";
 import { asArray, asName, asNumber, dictGet, isName } from "./objects.js";
 import { decodeStream } from "./filters.js";
+import { styleFromBaseFontName } from "./font-style.js";
+import { glyphNameToUnicode } from "./glyph-names.js";
 import { readFontProgramEncoding } from "./builtin-encoding.js";
 import { parseToUnicodeCMap } from "./cmap.js";
 import { symbolGlyphName, zapfDingbatsGlyphName } from "./encoding-symbol-tables.js";
 import { resolveStandardFont } from "./fonts.js";
-import { styleFromBaseFontName } from "./font-style.js";
 //#region src/font-read.ts
 const BITS_PER_BYTE = 8;
 const SYMBOLIC_FLAG_BIT = 4;
@@ -25,7 +27,7 @@ function builtinSymbolGlyphNameLookup(baseFamily) {
 	if (baseFamily === "Symbol") return symbolGlyphName;
 	if (baseFamily === "ZapfDingbats") return zapfDingbatsGlyphName;
 }
-function readFontProgram(descriptor, context) {
+function readFontProgram(descriptor, baseFont, context) {
 	if (descriptor === void 0) return;
 	for (const key of [
 		"FontFile2",
@@ -35,14 +37,14 @@ function readFontProgram(descriptor, context) {
 		const stream = context.resolver.resolve(dictGet(descriptor, key));
 		if (stream?.kind !== "stream") continue;
 		const decoded = decodeStream(stream.raw, stream.dict, context.sink);
-		const encoding = readFontProgramEncoding(decoded.bytes);
+		const encoding = readFontProgramEncoding(decoded.bytes, baseFont);
 		if (encoding !== void 0) return encoding;
 	}
 }
-function lazyFontProgram(descriptor, context) {
+function lazyFontProgram(descriptor, baseFont, context) {
 	let read;
 	return () => {
-		read ??= { encoding: readFontProgram(descriptor, context) };
+		read ??= { encoding: readFontProgram(descriptor, baseFont, context) };
 		return read.encoding;
 	};
 }
@@ -119,7 +121,14 @@ function buildSimpleFont(fontDict, context) {
 		message: `base encoding "${baseEncodingName}" has no dedicated table here and is approximated as WinAnsiEncoding`
 	});
 	const differencesMap = readDifferencesMap(encodingDict !== void 0 ? resolvedArray(dictGet(encodingDict, "Differences"), context) : void 0);
-	const programEncoding = lazyFontProgram(descriptor, context);
+	const programEncoding = lazyFontProgram(descriptor, baseFont, context);
+	const unicodeForDifferencesName = (name) => {
+		const named = glyphNameToUnicode(name);
+		if (named !== void 0) return named;
+		const program = programEncoding();
+		const glyphId = program?.glyphIdOfName(name);
+		return glyphId === void 0 ? void 0 : program?.glyphIdToUnicode(glyphId);
+	};
 	const symbolEncoded = symbolic || builtinGlyphName !== void 0;
 	const byGlyphName = (source) => (code) => {
 		const name = source(code);
@@ -127,7 +136,7 @@ function buildSimpleFont(fontDict, context) {
 	};
 	const fromProgram = (code) => programEncoding()?.codeToUnicode(code);
 	const namedBase = namedBaseGlyphName === void 0 ? void 0 : byGlyphName(namedBaseGlyphName);
-	const resolvers = [byGlyphName((code) => differencesMap.get(code)), ...symbolEncoded ? [
+	const resolvers = (symbolEncoded ? [
 		fromProgram,
 		builtinGlyphName === void 0 ? void 0 : byGlyphName(builtinGlyphName),
 		namedBase
@@ -135,8 +144,10 @@ function buildSimpleFont(fontDict, context) {
 		namedBase,
 		fromProgram,
 		byGlyphName(winAnsiGlyphName)
-	]].filter((resolver) => resolver !== void 0);
+	]).filter((resolver) => resolver !== void 0);
 	const unicodeForCode = (code) => {
+		const differencesName = differencesMap.get(code);
+		if (differencesName !== void 0) return unicodeForDifferencesName(differencesName);
 		for (const resolve of resolvers) {
 			const unicode = resolve(code);
 			if (unicode !== void 0) return unicode;
@@ -153,7 +164,7 @@ function buildSimpleFont(fontDict, context) {
 			}
 			const unicode = unicodeForCode(code);
 			if (unicode !== void 0) {
-				out += String.fromCodePoint(unicode);
+				out += unicode;
 				continue;
 			}
 			unmapped++;
@@ -264,6 +275,29 @@ function readsVertically(fontDict, context) {
 	if (encoding?.kind === "stream") return asNumber(dictGet(encoding.dict, "WMode")) === VERTICAL_WRITING_MODE;
 	return false;
 }
+const CID_TO_GID_ENTRY_SIZE = 2;
+/**
+* How a composite font's codes reach the embedded program's glyphs, so the program itself can say what a glyph is when the font dictionary's /ToUnicode CMap is absent or incomplete: through the glyph's own name, its program's Unicode mapping read backwards, or the glyph order of the core face the program is.
+*
+* With /Identity-H or /Identity-V a code is its CID. A /CIDToGIDMap that is absent or /Identity then makes the CID the glyph index, and a /CIDToGIDMap stream gives each CID its glyph index explicitly (ISO 32000-1 9.7.4.2). Any other /Encoding is a CMap from codes to CIDs this reader does not apply, and then the program is not consulted at all rather than being read against the wrong glyph.
+* @param fontDict - the Type0 font dictionary
+* @param descendantDict - its descendant CIDFont dictionary, if it has a readable one
+* @param context - the font read whose resolver follows references and whose sink receives stream-decoding diagnostics
+* @returns a lookup from CID to glyph index, answering undefined for a CID the map does not reach, or undefined when codes do not reach glyphs in a way this reader applies
+*/
+function cidToGlyphId(fontDict, descendantDict, context) {
+	const encodingName = dictGet(fontDict, "Encoding");
+	if (!isName(encodingName, "Identity-H") && !isName(encodingName, "Identity-V")) return;
+	const map = context.resolver.resolve(descendantDict !== void 0 ? dictGet(descendantDict, "CIDToGIDMap") : void 0);
+	if (map === void 0 || isName(map, "Identity")) return (cid) => cid;
+	if (map.kind !== "stream") return;
+	const bytes = decodeStream(map.raw, map.dict, context.sink).bytes;
+	return (cid) => {
+		const at = cid * CID_TO_GID_ENTRY_SIZE;
+		if (at + 1 >= bytes.length) return;
+		return (bytes[at] ?? 0) << BITS_PER_BYTE | (bytes[at + 1] ?? 0);
+	};
+}
 function buildCompositeFont(fontDict, context) {
 	const baseFont = asName(dictGet(fontDict, "BaseFont")) ?? "Helvetica";
 	const descendants = resolvedArray(dictGet(fontDict, "DescendantFonts"), context);
@@ -286,9 +320,8 @@ function buildCompositeFont(fontDict, context) {
 		};
 	};
 	const toUnicode = readToUnicodeCMap(fontDict, context);
-	const cidToGidMap = descendantDict !== void 0 ? dictGet(descendantDict, "CIDToGIDMap") : void 0;
-	const encodingName = dictGet(fontDict, "Encoding");
-	const programEncoding = (isName(encodingName, "Identity-H") || isName(encodingName, "Identity-V")) && (cidToGidMap === void 0 || isName(cidToGidMap, "Identity")) ? lazyFontProgram(descriptor, context) : void 0;
+	const glyphIdOf = cidToGlyphId(fontDict, descendantDict, context);
+	const programEncoding = glyphIdOf !== void 0 ? lazyFontProgram(descriptor, baseFont, context) : void 0;
 	const decodeToUnicode = (codes) => {
 		let out = "";
 		let glyphCount = 0;
@@ -301,9 +334,10 @@ function buildCompositeFont(fontDict, context) {
 				out += mapped;
 				continue;
 			}
-			const fromProgram = programEncoding?.()?.glyphIdToUnicode(cid);
+			const glyphId = glyphIdOf?.(cid);
+			const fromProgram = glyphId === void 0 ? void 0 : programEncoding?.()?.glyphIdToUnicode(glyphId);
 			if (fromProgram !== void 0) {
-				out += String.fromCodePoint(fromProgram);
+				out += fromProgram;
 				continue;
 			}
 			unmapped++;

package/dist/font-style.cjs

@@ -16,8 +16,18 @@ function stripStyleSuffix(name) {
 	}
 	return name;
 }
+/**
+* A font name with any subset tag removed.
+*
+* Exactly six uppercase letters followed by '+' (ISO 32000-1 9.6.4) marks a subsetted font's unique tag, meaningless to a reader and never part of the real font name. The same tag prefixes the PostScript name inside the embedded program a producer subsetted.
+* @param fontName - a /BaseFont value or an embedded program's own PostScript name
+* @returns the name without its leading subset tag, or the name unchanged when it carries none
+*/
+function withoutSubsetTag(fontName) {
+	return fontName.replace(/^[A-Z]{6}\+/, "");
+}
 function styleFromBaseFontName(baseFont, flags) {
-	const withoutSubset = baseFont.replace(/^[A-Z]{6}\+/, "");
+	const withoutSubset = withoutSubsetTag(baseFont);
 	const lower = withoutSubset.toLowerCase();
 	const nameBold = lower.includes("bold");
 	const nameItalic = /italic|oblique/.test(lower);
@@ -31,3 +41,4 @@ function styleFromBaseFontName(baseFont, flags) {
 }
 //#endregion
 exports.styleFromBaseFontName = styleFromBaseFontName;
+exports.withoutSubsetTag = withoutSubsetTag;

package/dist/font-style.d.cts

@@ -9,6 +9,14 @@ interface FontNameStyle {
   readonly bold: boolean;
   readonly italic: boolean;
 }
+/**
+ * A font name with any subset tag removed.
+ *
+ * Exactly six uppercase letters followed by '+' (ISO 32000-1 9.6.4) marks a subsetted font's unique tag, meaningless to a reader and never part of the real font name. The same tag prefixes the PostScript name inside the embedded program a producer subsetted.
+ * @param fontName - a /BaseFont value or an embedded program's own PostScript name
+ * @returns the name without its leading subset tag, or the name unchanged when it carries none
+ */
+declare function withoutSubsetTag(fontName: string): string;
 declare function styleFromBaseFontName(baseFont: string, flags?: FontStyleFlags): FontNameStyle;
 //#endregion
-export { FontNameStyle, FontStyleFlags, styleFromBaseFontName };
+export { FontNameStyle, FontStyleFlags, styleFromBaseFontName, withoutSubsetTag };

package/dist/font-style.d.ts

@@ -9,6 +9,14 @@ interface FontNameStyle {
   readonly bold: boolean;
   readonly italic: boolean;
 }
+/**
+ * A font name with any subset tag removed.
+ *
+ * Exactly six uppercase letters followed by '+' (ISO 32000-1 9.6.4) marks a subsetted font's unique tag, meaningless to a reader and never part of the real font name. The same tag prefixes the PostScript name inside the embedded program a producer subsetted.
+ * @param fontName - a /BaseFont value or an embedded program's own PostScript name
+ * @returns the name without its leading subset tag, or the name unchanged when it carries none
+ */
+declare function withoutSubsetTag(fontName: string): string;
 declare function styleFromBaseFontName(baseFont: string, flags?: FontStyleFlags): FontNameStyle;
 //#endregion
-export { FontNameStyle, FontStyleFlags, styleFromBaseFontName };
+export { FontNameStyle, FontStyleFlags, styleFromBaseFontName, withoutSubsetTag };

package/dist/font-style.js

@@ -15,8 +15,18 @@ function stripStyleSuffix(name) {
 	}
 	return name;
 }
+/**
+* A font name with any subset tag removed.
+*
+* Exactly six uppercase letters followed by '+' (ISO 32000-1 9.6.4) marks a subsetted font's unique tag, meaningless to a reader and never part of the real font name. The same tag prefixes the PostScript name inside the embedded program a producer subsetted.
+* @param fontName - a /BaseFont value or an embedded program's own PostScript name
+* @returns the name without its leading subset tag, or the name unchanged when it carries none
+*/
+function withoutSubsetTag(fontName) {
+	return fontName.replace(/^[A-Z]{6}\+/, "");
+}
 function styleFromBaseFontName(baseFont, flags) {
-	const withoutSubset = baseFont.replace(/^[A-Z]{6}\+/, "");
+	const withoutSubset = withoutSubsetTag(baseFont);
 	const lower = withoutSubset.toLowerCase();
 	const nameBold = lower.includes("bold");
 	const nameItalic = /italic|oblique/.test(lower);
@@ -29,4 +39,4 @@ function styleFromBaseFontName(baseFont, flags) {
 	};
 }
 //#endregion
-export { styleFromBaseFontName };
+export { styleFromBaseFontName, withoutSubsetTag };