pdf-codec 3.5.3 → 3.6.0

package/dist/cmap-table.cjs

@@ -3,6 +3,7 @@ const require_sfnt = require("./sfnt.cjs");
 //#region src/cmap-table.ts
 const CMAP_HEADER_SIZE = 4;
 const SUBTABLE_RECORD_SIZE = 8;
+const MAX_UNICODE_CODE_POINT = 1114111;
 const FORMAT_4_HEADER_SIZE = 14;
 function parseFormat4(bytes, subtableOffset) {
 	if (!require_sfnt.hasBytes(bytes, subtableOffset, FORMAT_4_HEADER_SIZE)) return;
@@ -26,7 +27,7 @@ function parseFormat4(bytes, subtableOffset) {
 			idRangeOffset: require_sfnt.u16(bytes, idRangeOffsetPos)
 		});
 	}
-	return (codePoint) => {
+	const lookup = (codePoint) => {
 		if (codePoint > 65535) return;
 		for (const segment of segments) {
 			if (codePoint < segment.startCode || codePoint > segment.endCode) continue;
@@ -37,6 +38,18 @@ function parseFormat4(bytes, subtableOffset) {
 			return glyphId === 0 ? void 0 : glyphId + segment.idDelta & 65535;
 		}
 	};
+	return {
+		lookup,
+		forEachMapping(visit) {
+			for (const segment of segments) {
+				const end = Math.min(segment.endCode, 65534);
+				for (let code = segment.startCode; code <= end; code++) {
+					const glyphId = lookup(code);
+					if (glyphId !== void 0 && glyphId !== 0) visit(code, glyphId);
+				}
+			}
+		}
+	};
 }
 const FORMAT_12_HEADER_SIZE = 16;
 const FORMAT_12_GROUP_SIZE = 12;
@@ -54,8 +67,35 @@ function parseFormat12(bytes, subtableOffset) {
 			startGlyphId: require_sfnt.u32(bytes, recordOffset + 8)
 		});
 	}
-	return (codePoint) => {
-		for (const group of groups) if (codePoint >= group.startCharCode && codePoint <= group.endCharCode) return group.startGlyphId + (codePoint - group.startCharCode);
+	return {
+		lookup(codePoint) {
+			for (const group of groups) if (codePoint >= group.startCharCode && codePoint <= group.endCharCode) return group.startGlyphId + (codePoint - group.startCharCode);
+		},
+		forEachMapping(visit) {
+			for (const group of groups) {
+				const end = Math.min(group.endCharCode, MAX_UNICODE_CODE_POINT);
+				for (let code = group.startCharCode; code <= end; code++) visit(code, group.startGlyphId + (code - group.startCharCode));
+			}
+		}
+	};
+}
+const FORMAT_0_SIZE = 262;
+function parseFormat0(bytes, subtableOffset) {
+	if (!require_sfnt.hasBytes(bytes, subtableOffset, FORMAT_0_SIZE)) return;
+	const glyphIdArrayOffset = subtableOffset + 6;
+	const lookup = (codePoint) => {
+		if (codePoint < 0 || codePoint > 255) return;
+		const glyphId = require_sfnt.u8(bytes, glyphIdArrayOffset + codePoint);
+		return glyphId === 0 ? void 0 : glyphId;
+	};
+	return {
+		lookup,
+		forEachMapping(visit) {
+			for (let code = 0; code <= 255; code++) {
+				const glyphId = lookup(code);
+				if (glyphId !== void 0) visit(code, glyphId);
+			}
+		}
 	};
 }
 const FORMAT_6_HEADER_SIZE = 10;
@@ -65,12 +105,22 @@ function parseFormat6(bytes, subtableOffset) {
 	const entryCount = require_sfnt.u16(bytes, subtableOffset + 8);
 	const glyphIdArrayOffset = subtableOffset + FORMAT_6_HEADER_SIZE;
 	if (!require_sfnt.hasBytes(bytes, glyphIdArrayOffset, entryCount * 2)) return;
-	return (codePoint) => {
+	const lookup = (codePoint) => {
 		const index = codePoint - firstCode;
 		if (index < 0 || index >= entryCount) return;
 		const glyphId = require_sfnt.u16(bytes, glyphIdArrayOffset + index * 2);
 		return glyphId === 0 ? void 0 : glyphId;
 	};
+	return {
+		lookup,
+		forEachMapping(visit) {
+			for (let index = 0; index < entryCount; index++) {
+				const code = firstCode + index;
+				const glyphId = lookup(code);
+				if (glyphId !== void 0) visit(code, glyphId);
+			}
+		}
+	};
 }
 function preferenceRank(subtable) {
 	if (subtable.format === 12) {
@@ -81,28 +131,41 @@ function preferenceRank(subtable) {
 	if (subtable.format === 4) return subtable.platformId === 3 && subtable.encodingId === 1 ? 3 : 4;
 	return 5;
 }
-function buildCmapLookup(font) {
+function parseSubtable(bytes, record) {
+	if (record.format === 12) return parseFormat12(bytes, record.offset);
+	if (record.format === 4) return parseFormat4(bytes, record.offset);
+	if (record.format === 6) return parseFormat6(bytes, record.offset);
+	return record.format === 0 ? parseFormat0(bytes, record.offset) : void 0;
+}
+function readCmapSubtables(font) {
 	const cmapBytes = require_sfnt.sfntTableBytes(font, "cmap");
-	if (cmapBytes === void 0 || !require_sfnt.hasBytes(cmapBytes, 0, CMAP_HEADER_SIZE)) return;
+	if (cmapBytes === void 0 || !require_sfnt.hasBytes(cmapBytes, 0, CMAP_HEADER_SIZE)) return [];
 	const numTables = require_sfnt.u16(cmapBytes, 2);
-	if (!require_sfnt.hasBytes(cmapBytes, CMAP_HEADER_SIZE, numTables * SUBTABLE_RECORD_SIZE)) return;
+	if (!require_sfnt.hasBytes(cmapBytes, CMAP_HEADER_SIZE, numTables * SUBTABLE_RECORD_SIZE)) return [];
 	const subtables = [];
 	for (let i = 0; i < numTables; i++) {
 		const recordOffset = CMAP_HEADER_SIZE + i * SUBTABLE_RECORD_SIZE;
 		const offset = require_sfnt.u32(cmapBytes, recordOffset + 4);
 		if (!require_sfnt.hasBytes(cmapBytes, offset, 2)) continue;
-		subtables.push({
+		const record = {
 			platformId: require_sfnt.u16(cmapBytes, recordOffset),
 			encodingId: require_sfnt.u16(cmapBytes, recordOffset + 2),
 			offset,
 			format: require_sfnt.u16(cmapBytes, offset)
+		};
+		const parsed = parseSubtable(cmapBytes, record);
+		if (parsed !== void 0) subtables.push({
+			platformId: record.platformId,
+			encodingId: record.encodingId,
+			format: record.format,
+			...parsed
 		});
 	}
-	const candidates = subtables.filter((s) => s.format === 4 || s.format === 6 || s.format === 12).sort((a, b) => preferenceRank(a) - preferenceRank(b));
-	for (const candidate of candidates) {
-		const lookup = candidate.format === 12 ? parseFormat12(cmapBytes, candidate.offset) : candidate.format === 4 ? parseFormat4(cmapBytes, candidate.offset) : parseFormat6(cmapBytes, candidate.offset);
-		if (lookup !== void 0) return lookup;
-	}
+	return subtables;
+}
+function buildCmapLookup(font) {
+	return readCmapSubtables(font).filter((s) => s.format === 4 || s.format === 6 || s.format === 12).sort((a, b) => preferenceRank(a) - preferenceRank(b))[0]?.lookup;
 }
 //#endregion
 exports.buildCmapLookup = buildCmapLookup;
+exports.readCmapSubtables = readCmapSubtables;

package/dist/cmap-table.d.cts

@@ -1,6 +1,14 @@
 import { t as SfntFont } from "./sfnt-V08bb5ut.cjs";
 //#region src/cmap-table.d.ts
 type CmapLookup = (codePoint: number) => number | undefined;
+interface CmapSubtable {
+  readonly platformId: number;
+  readonly encodingId: number;
+  readonly format: number;
+  readonly lookup: CmapLookup;
+  forEachMapping(visit: (code: number, glyphId: number) => void): void;
+}
+declare function readCmapSubtables(font: SfntFont): readonly CmapSubtable[];
 declare function buildCmapLookup(font: SfntFont): CmapLookup | undefined;
 //#endregion
-export { CmapLookup, buildCmapLookup };
+export { CmapLookup, CmapSubtable, buildCmapLookup, readCmapSubtables };

package/dist/cmap-table.d.ts

@@ -1,6 +1,14 @@
 import { t as SfntFont } from "./sfnt-V08bb5ut.js";
 //#region src/cmap-table.d.ts
 type CmapLookup = (codePoint: number) => number | undefined;
+interface CmapSubtable {
+  readonly platformId: number;
+  readonly encodingId: number;
+  readonly format: number;
+  readonly lookup: CmapLookup;
+  forEachMapping(visit: (code: number, glyphId: number) => void): void;
+}
+declare function readCmapSubtables(font: SfntFont): readonly CmapSubtable[];
 declare function buildCmapLookup(font: SfntFont): CmapLookup | undefined;
 //#endregion
-export { CmapLookup, buildCmapLookup };
+export { CmapLookup, CmapSubtable, buildCmapLookup, readCmapSubtables };

package/dist/cmap-table.js

@@ -1,7 +1,8 @@
-import { hasBytes, sfntTableBytes, u16, u32 } from "./sfnt.js";
+import { hasBytes, sfntTableBytes, u16, u32, u8 } from "./sfnt.js";
 //#region src/cmap-table.ts
 const CMAP_HEADER_SIZE = 4;
 const SUBTABLE_RECORD_SIZE = 8;
+const MAX_UNICODE_CODE_POINT = 1114111;
 const FORMAT_4_HEADER_SIZE = 14;
 function parseFormat4(bytes, subtableOffset) {
 	if (!hasBytes(bytes, subtableOffset, FORMAT_4_HEADER_SIZE)) return;
@@ -25,7 +26,7 @@ function parseFormat4(bytes, subtableOffset) {
 			idRangeOffset: u16(bytes, idRangeOffsetPos)
 		});
 	}
-	return (codePoint) => {
+	const lookup = (codePoint) => {
 		if (codePoint > 65535) return;
 		for (const segment of segments) {
 			if (codePoint < segment.startCode || codePoint > segment.endCode) continue;
@@ -36,6 +37,18 @@ function parseFormat4(bytes, subtableOffset) {
 			return glyphId === 0 ? void 0 : glyphId + segment.idDelta & 65535;
 		}
 	};
+	return {
+		lookup,
+		forEachMapping(visit) {
+			for (const segment of segments) {
+				const end = Math.min(segment.endCode, 65534);
+				for (let code = segment.startCode; code <= end; code++) {
+					const glyphId = lookup(code);
+					if (glyphId !== void 0 && glyphId !== 0) visit(code, glyphId);
+				}
+			}
+		}
+	};
 }
 const FORMAT_12_HEADER_SIZE = 16;
 const FORMAT_12_GROUP_SIZE = 12;
@@ -53,8 +66,35 @@ function parseFormat12(bytes, subtableOffset) {
 			startGlyphId: u32(bytes, recordOffset + 8)
 		});
 	}
-	return (codePoint) => {
-		for (const group of groups) if (codePoint >= group.startCharCode && codePoint <= group.endCharCode) return group.startGlyphId + (codePoint - group.startCharCode);
+	return {
+		lookup(codePoint) {
+			for (const group of groups) if (codePoint >= group.startCharCode && codePoint <= group.endCharCode) return group.startGlyphId + (codePoint - group.startCharCode);
+		},
+		forEachMapping(visit) {
+			for (const group of groups) {
+				const end = Math.min(group.endCharCode, MAX_UNICODE_CODE_POINT);
+				for (let code = group.startCharCode; code <= end; code++) visit(code, group.startGlyphId + (code - group.startCharCode));
+			}
+		}
+	};
+}
+const FORMAT_0_SIZE = 262;
+function parseFormat0(bytes, subtableOffset) {
+	if (!hasBytes(bytes, subtableOffset, FORMAT_0_SIZE)) return;
+	const glyphIdArrayOffset = subtableOffset + 6;
+	const lookup = (codePoint) => {
+		if (codePoint < 0 || codePoint > 255) return;
+		const glyphId = u8(bytes, glyphIdArrayOffset + codePoint);
+		return glyphId === 0 ? void 0 : glyphId;
+	};
+	return {
+		lookup,
+		forEachMapping(visit) {
+			for (let code = 0; code <= 255; code++) {
+				const glyphId = lookup(code);
+				if (glyphId !== void 0) visit(code, glyphId);
+			}
+		}
 	};
 }
 const FORMAT_6_HEADER_SIZE = 10;
@@ -64,12 +104,22 @@ function parseFormat6(bytes, subtableOffset) {
 	const entryCount = u16(bytes, subtableOffset + 8);
 	const glyphIdArrayOffset = subtableOffset + FORMAT_6_HEADER_SIZE;
 	if (!hasBytes(bytes, glyphIdArrayOffset, entryCount * 2)) return;
-	return (codePoint) => {
+	const lookup = (codePoint) => {
 		const index = codePoint - firstCode;
 		if (index < 0 || index >= entryCount) return;
 		const glyphId = u16(bytes, glyphIdArrayOffset + index * 2);
 		return glyphId === 0 ? void 0 : glyphId;
 	};
+	return {
+		lookup,
+		forEachMapping(visit) {
+			for (let index = 0; index < entryCount; index++) {
+				const code = firstCode + index;
+				const glyphId = lookup(code);
+				if (glyphId !== void 0) visit(code, glyphId);
+			}
+		}
+	};
 }
 function preferenceRank(subtable) {
 	if (subtable.format === 12) {
@@ -80,28 +130,40 @@ function preferenceRank(subtable) {
 	if (subtable.format === 4) return subtable.platformId === 3 && subtable.encodingId === 1 ? 3 : 4;
 	return 5;
 }
-function buildCmapLookup(font) {
+function parseSubtable(bytes, record) {
+	if (record.format === 12) return parseFormat12(bytes, record.offset);
+	if (record.format === 4) return parseFormat4(bytes, record.offset);
+	if (record.format === 6) return parseFormat6(bytes, record.offset);
+	return record.format === 0 ? parseFormat0(bytes, record.offset) : void 0;
+}
+function readCmapSubtables(font) {
 	const cmapBytes = sfntTableBytes(font, "cmap");
-	if (cmapBytes === void 0 || !hasBytes(cmapBytes, 0, CMAP_HEADER_SIZE)) return;
+	if (cmapBytes === void 0 || !hasBytes(cmapBytes, 0, CMAP_HEADER_SIZE)) return [];
 	const numTables = u16(cmapBytes, 2);
-	if (!hasBytes(cmapBytes, CMAP_HEADER_SIZE, numTables * SUBTABLE_RECORD_SIZE)) return;
+	if (!hasBytes(cmapBytes, CMAP_HEADER_SIZE, numTables * SUBTABLE_RECORD_SIZE)) return [];
 	const subtables = [];
 	for (let i = 0; i < numTables; i++) {
 		const recordOffset = CMAP_HEADER_SIZE + i * SUBTABLE_RECORD_SIZE;
 		const offset = u32(cmapBytes, recordOffset + 4);
 		if (!hasBytes(cmapBytes, offset, 2)) continue;
-		subtables.push({
+		const record = {
 			platformId: u16(cmapBytes, recordOffset),
 			encodingId: u16(cmapBytes, recordOffset + 2),
 			offset,
 			format: u16(cmapBytes, offset)
+		};
+		const parsed = parseSubtable(cmapBytes, record);
+		if (parsed !== void 0) subtables.push({
+			platformId: record.platformId,
+			encodingId: record.encodingId,
+			format: record.format,
+			...parsed
 		});
 	}
-	const candidates = subtables.filter((s) => s.format === 4 || s.format === 6 || s.format === 12).sort((a, b) => preferenceRank(a) - preferenceRank(b));
-	for (const candidate of candidates) {
-		const lookup = candidate.format === 12 ? parseFormat12(cmapBytes, candidate.offset) : candidate.format === 4 ? parseFormat4(cmapBytes, candidate.offset) : parseFormat6(cmapBytes, candidate.offset);
-		if (lookup !== void 0) return lookup;
-	}
+	return subtables;
+}
+function buildCmapLookup(font) {
+	return readCmapSubtables(font).filter((s) => s.format === 4 || s.format === 6 || s.format === 12).sort((a, b) => preferenceRank(a) - preferenceRank(b))[0]?.lookup;
 }
 //#endregion
-export { buildCmapLookup };
+export { buildCmapLookup, readCmapSubtables };

package/dist/codec.d.cts

@@ -224,6 +224,7 @@ declare const pdfCodec: z.ZodCodec<z.ZodCustom<Uint8Array<ArrayBuffer>, Uint8Arr
           odf: "odf";
           markdown: "markdown";
           pdf: "pdf";
+          epub: "epub";
         }>;
         xml: z.ZodString;
       }, z.core.$strict>>;
@@ -287,6 +288,7 @@ declare const pdfCodec: z.ZodCodec<z.ZodCustom<Uint8Array<ArrayBuffer>, Uint8Arr
       odf: "odf";
       markdown: "markdown";
       pdf: "pdf";
+      epub: "epub";
     }>;
     xml: z.ZodString;
   }, z.core.$strict>>>;

package/dist/codec.d.ts

@@ -224,6 +224,7 @@ declare const pdfCodec: z.ZodCodec<z.ZodCustom<Uint8Array<ArrayBuffer>, Uint8Arr
           odf: "odf";
           markdown: "markdown";
           pdf: "pdf";
+          epub: "epub";
         }>;
         xml: z.ZodString;
       }, z.core.$strict>>;
@@ -287,6 +288,7 @@ declare const pdfCodec: z.ZodCodec<z.ZodCustom<Uint8Array<ArrayBuffer>, Uint8Arr
       odf: "odf";
       markdown: "markdown";
       pdf: "pdf";
+      epub: "epub";
     }>;
     xml: z.ZodString;
   }, z.core.$strict>>>;

package/dist/content-write.cjs

@@ -1,5 +1,5 @@
 Object.defineProperty(exports, Symbol.toStringTag, { value: "Module" });
-const require_afm_widths = require("./afm-widths-W02fJh-k.cjs");
+const require_afm_widths = require("./afm-widths-BTu0IDp2.cjs");
 const require_objects = require("./objects.cjs");
 const require_bytes_writer = require("./bytes/writer.cjs");
 const require_serialize = require("./serialize.cjs");

package/dist/content-write.js

@@ -1,4 +1,4 @@
-import { u as encodeForShow } from "./afm-widths-B6KdmpRF.js";
+import { h as encodeForShow } from "./afm-widths-De_QMWgC.js";
 import { pdfArray, pdfHexString, pdfNum } from "./objects.js";
 import { ByteWriter } from "./bytes/writer.js";
 import { formatNumber, writeObject } from "./serialize.js";

package/dist/encoding.cjs

@@ -1,9 +1,14 @@
 Object.defineProperty(exports, Symbol.toStringTag, { value: "Module" });
-const require_afm_widths = require("./afm-widths-W02fJh-k.cjs");
+const require_afm_widths = require("./afm-widths-BTu0IDp2.cjs");
+exports.MACROMAN_GLYPH_NAMES = require_afm_widths.MACROMAN_GLYPH_NAMES;
+exports.STANDARD_GLYPH_NAMES = require_afm_widths.STANDARD_GLYPH_NAMES;
 exports.SYMBOL_GLYPH_NAMES = require_afm_widths.SYMBOL_GLYPH_NAMES;
 exports.WINANSI_GLYPH_NAMES = require_afm_widths.WINANSI_GLYPH_NAMES;
 exports.ZAPFDINGBATS_GLYPH_NAMES = require_afm_widths.ZAPFDINGBATS_GLYPH_NAMES;
 exports.glyphNameToUnicode = require_afm_widths.glyphNameToUnicode;
+exports.macRomanGlyphName = require_afm_widths.macRomanGlyphName;
+exports.namedEncodingGlyphName = require_afm_widths.namedEncodingGlyphName;
+exports.standardGlyphName = require_afm_widths.standardGlyphName;
 exports.symbolGlyphName = require_afm_widths.symbolGlyphName;
 exports.winAnsiGlyphName = require_afm_widths.winAnsiGlyphName;
 exports.zapfDingbatsGlyphName = require_afm_widths.zapfDingbatsGlyphName;

package/dist/encoding.d.cts

@@ -4,7 +4,12 @@ declare function winAnsiGlyphName(code: number): string | undefined;
 declare const SYMBOL_GLYPH_NAMES: readonly string[];
 declare function symbolGlyphName(code: number): string | undefined;
 declare const ZAPFDINGBATS_GLYPH_NAMES: readonly string[];
+declare const STANDARD_GLYPH_NAMES: readonly string[];
+declare function standardGlyphName(code: number): string | undefined;
+declare const MACROMAN_GLYPH_NAMES: readonly string[];
+declare function macRomanGlyphName(code: number): string | undefined;
+declare function namedEncodingGlyphName(encodingName: string): ((code: number) => string | undefined) | undefined;
 declare function zapfDingbatsGlyphName(code: number): string | undefined;
 declare function glyphNameToUnicode(name: string): number | undefined;
 //#endregion
-export { SYMBOL_GLYPH_NAMES, WINANSI_GLYPH_NAMES, ZAPFDINGBATS_GLYPH_NAMES, glyphNameToUnicode, symbolGlyphName, winAnsiGlyphName, zapfDingbatsGlyphName };
+export { MACROMAN_GLYPH_NAMES, STANDARD_GLYPH_NAMES, SYMBOL_GLYPH_NAMES, WINANSI_GLYPH_NAMES, ZAPFDINGBATS_GLYPH_NAMES, glyphNameToUnicode, macRomanGlyphName, namedEncodingGlyphName, standardGlyphName, symbolGlyphName, winAnsiGlyphName, zapfDingbatsGlyphName };

package/dist/encoding.d.ts

@@ -4,7 +4,12 @@ declare function winAnsiGlyphName(code: number): string | undefined;
 declare const SYMBOL_GLYPH_NAMES: readonly string[];
 declare function symbolGlyphName(code: number): string | undefined;
 declare const ZAPFDINGBATS_GLYPH_NAMES: readonly string[];
+declare const STANDARD_GLYPH_NAMES: readonly string[];
+declare function standardGlyphName(code: number): string | undefined;
+declare const MACROMAN_GLYPH_NAMES: readonly string[];
+declare function macRomanGlyphName(code: number): string | undefined;
+declare function namedEncodingGlyphName(encodingName: string): ((code: number) => string | undefined) | undefined;
 declare function zapfDingbatsGlyphName(code: number): string | undefined;
 declare function glyphNameToUnicode(name: string): number | undefined;
 //#endregion
-export { SYMBOL_GLYPH_NAMES, WINANSI_GLYPH_NAMES, ZAPFDINGBATS_GLYPH_NAMES, glyphNameToUnicode, symbolGlyphName, winAnsiGlyphName, zapfDingbatsGlyphName };
+export { MACROMAN_GLYPH_NAMES, STANDARD_GLYPH_NAMES, SYMBOL_GLYPH_NAMES, WINANSI_GLYPH_NAMES, ZAPFDINGBATS_GLYPH_NAMES, glyphNameToUnicode, macRomanGlyphName, namedEncodingGlyphName, standardGlyphName, symbolGlyphName, winAnsiGlyphName, zapfDingbatsGlyphName };

package/dist/encoding.js

@@ -1,2 +1,2 @@
-import { a as ZAPFDINGBATS_GLYPH_NAMES, c as winAnsiGlyphName, i as WINANSI_GLYPH_NAMES, l as zapfDingbatsGlyphName, o as glyphNameToUnicode, r as SYMBOL_GLYPH_NAMES, s as symbolGlyphName } from "./afm-widths-B6KdmpRF.js";
-export { SYMBOL_GLYPH_NAMES, WINANSI_GLYPH_NAMES, ZAPFDINGBATS_GLYPH_NAMES, glyphNameToUnicode, symbolGlyphName, winAnsiGlyphName, zapfDingbatsGlyphName };
+import { a as SYMBOL_GLYPH_NAMES, c as glyphNameToUnicode, d as standardGlyphName, f as symbolGlyphName, i as STANDARD_GLYPH_NAMES, l as macRomanGlyphName, m as zapfDingbatsGlyphName, o as WINANSI_GLYPH_NAMES, p as winAnsiGlyphName, r as MACROMAN_GLYPH_NAMES, s as ZAPFDINGBATS_GLYPH_NAMES, u as namedEncodingGlyphName } from "./afm-widths-De_QMWgC.js";
+export { MACROMAN_GLYPH_NAMES, STANDARD_GLYPH_NAMES, SYMBOL_GLYPH_NAMES, WINANSI_GLYPH_NAMES, ZAPFDINGBATS_GLYPH_NAMES, glyphNameToUnicode, macRomanGlyphName, namedEncodingGlyphName, standardGlyphName, symbolGlyphName, winAnsiGlyphName, zapfDingbatsGlyphName };

package/dist/font-read.cjs

@@ -1,7 +1,8 @@
 Object.defineProperty(exports, Symbol.toStringTag, { value: "Module" });
-const require_afm_widths = require("./afm-widths-W02fJh-k.cjs");
+const require_afm_widths = require("./afm-widths-BTu0IDp2.cjs");
 const require_objects = require("./objects.cjs");
 const require_filters = require("./filters.cjs");
+const require_builtin_encoding = require("./builtin-encoding.cjs");
 const require_cmap = require("./cmap.cjs");
 const require_fonts = require("./fonts.cjs");
 const require_font_style = require("./font-style.cjs");
@@ -23,11 +24,26 @@ function builtinSymbolGlyphNameLookup(baseFamily) {
 	if (baseFamily === "Symbol") return require_afm_widths.symbolGlyphName;
 	if (baseFamily === "ZapfDingbats") return require_afm_widths.zapfDingbatsGlyphName;
 }
-function glyphNameToUnicodeWithUniFallback(name) {
-	const direct = require_afm_widths.glyphNameToUnicode(name);
-	if (direct !== void 0) return direct;
-	const uniMatch = /^uni([0-9A-Fa-f]{4,6})$/.exec(name);
-	return uniMatch?.[1] !== void 0 ? Number.parseInt(uniMatch[1], 16) : void 0;
+function readFontProgram(descriptor, context) {
+	if (descriptor === void 0) return;
+	for (const key of [
+		"FontFile2",
+		"FontFile3",
+		"FontFile"
+	]) {
+		const stream = context.resolver.resolve(require_objects.dictGet(descriptor, key));
+		if (stream?.kind !== "stream") continue;
+		const decoded = require_filters.decodeStream(stream.raw, stream.dict, context.sink);
+		const encoding = require_builtin_encoding.readFontProgramEncoding(decoded.bytes);
+		if (encoding !== void 0) return encoding;
+	}
+}
+function lazyFontProgram(descriptor, context) {
+	let read;
+	return () => {
+		read ??= { encoding: readFontProgram(descriptor, context) };
+		return read.encoding;
+	};
 }
 function readDifferencesMap(differences) {
 	const map = /* @__PURE__ */ new Map();
@@ -71,13 +87,36 @@ function buildSimpleFont(fontDict, context) {
 	const encodingObj = require_objects.dictGet(fontDict, "Encoding");
 	const encodingDict = context.resolver.resolveDict(encodingObj);
 	const baseEncodingName = require_objects.asName(encodingDict !== void 0 ? require_objects.dictGet(encodingDict, "BaseEncoding") : encodingObj);
-	if (baseEncodingName !== void 0 && baseEncodingName !== "WinAnsiEncoding") context.sink({
+	const namedBaseGlyphName = baseEncodingName !== void 0 ? require_afm_widths.namedEncodingGlyphName(baseEncodingName) : void 0;
+	if (baseEncodingName !== void 0 && namedBaseGlyphName === void 0) context.sink({
 		code: "char/encoding-approximated",
 		severity: "info",
 		message: `base encoding "${baseEncodingName}" has no dedicated table here and is approximated as WinAnsiEncoding`
 	});
 	const differencesMap = readDifferencesMap(encodingDict !== void 0 ? require_objects.asArray(require_objects.dictGet(encodingDict, "Differences")) : void 0);
-	const fallbackGlyphName = builtinGlyphName ?? (symbolic ? () => void 0 : require_afm_widths.winAnsiGlyphName);
+	const programEncoding = lazyFontProgram(descriptor, context);
+	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);
+	};
+	const fromProgram = (code) => programEncoding()?.codeToUnicode(code);
+	const namedBase = namedBaseGlyphName === void 0 ? void 0 : byGlyphName(namedBaseGlyphName);
+	const resolvers = [byGlyphName((code) => differencesMap.get(code)), ...symbolEncoded ? [
+		fromProgram,
+		builtinGlyphName === void 0 ? void 0 : byGlyphName(builtinGlyphName),
+		namedBase
+	] : [
+		namedBase,
+		fromProgram,
+		byGlyphName(require_afm_widths.winAnsiGlyphName)
+	]].filter((resolver) => resolver !== void 0);
+	const unicodeForCode = (code) => {
+		for (const resolve of resolvers) {
+			const unicode = resolve(code);
+			if (unicode !== void 0) return unicode;
+		}
+	};
 	const decodeToUnicode = (codes) => {
 		let out = "";
 		let unmapped = 0;
@@ -87,10 +126,9 @@ function buildSimpleFont(fontDict, context) {
 				out += viaToUnicode;
 				continue;
 			}
-			const name = differencesMap.get(code) ?? fallbackGlyphName(code);
-			const unicode = name !== void 0 ? glyphNameToUnicodeWithUniFallback(name) : void 0;
+			const unicode = unicodeForCode(code);
 			if (unicode !== void 0) {
-				out += String.fromCharCode(unicode);
+				out += String.fromCodePoint(unicode);
 				continue;
 			}
 			unmapped++;
@@ -149,6 +187,8 @@ function buildCompositeFont(fontDict, context) {
 	const widthMap = readCidWidths(descendantDict !== void 0 ? require_objects.asArray(require_objects.dictGet(descendantDict, "W")) : void 0);
 	const widthOf = (cid) => widthMap.get(cid) ?? dw;
 	const toUnicode = readToUnicodeCMap(fontDict, context);
+	const cidToGidMap = descendantDict !== void 0 ? require_objects.dictGet(descendantDict, "CIDToGIDMap") : void 0;
+	const programEncoding = require_objects.isName(require_objects.dictGet(fontDict, "Encoding"), "Identity-H") && (cidToGidMap === void 0 || require_objects.isName(cidToGidMap, "Identity")) ? lazyFontProgram(descriptor, context) : void 0;
 	const decodeToUnicode = (codes) => {
 		let out = "";
 		let glyphCount = 0;
@@ -161,6 +201,11 @@ function buildCompositeFont(fontDict, context) {
 				out += mapped;
 				continue;
 			}
+			const fromProgram = programEncoding?.()?.glyphIdToUnicode(cid);
+			if (fromProgram !== void 0) {
+				out += String.fromCodePoint(fromProgram);
+				continue;
+			}
 			unmapped++;
 			out += REPLACEMENT_CHARACTER;
 		}

package/dist/font-read.js

@@ -1,6 +1,7 @@
-import { c as winAnsiGlyphName, l as zapfDingbatsGlyphName, n as widthOfCode, o as glyphNameToUnicode, s as symbolGlyphName } from "./afm-widths-B6KdmpRF.js";
-import { asArray, asName, asNumber, dictGet } from "./objects.js";
+import { c as glyphNameToUnicode, f as symbolGlyphName, m as zapfDingbatsGlyphName, n as widthOfCode, p as winAnsiGlyphName, u as namedEncodingGlyphName } from "./afm-widths-De_QMWgC.js";
+import { asArray, asName, asNumber, dictGet, isName } from "./objects.js";
 import { decodeStream } from "./filters.js";
+import { readFontProgramEncoding } from "./builtin-encoding.js";
 import { parseToUnicodeCMap } from "./cmap.js";
 import { resolveStandardFont } from "./fonts.js";
 import { styleFromBaseFontName } from "./font-style.js";
@@ -22,11 +23,26 @@ function builtinSymbolGlyphNameLookup(baseFamily) {
 	if (baseFamily === "Symbol") return symbolGlyphName;
 	if (baseFamily === "ZapfDingbats") return zapfDingbatsGlyphName;
 }
-function glyphNameToUnicodeWithUniFallback(name) {
-	const direct = glyphNameToUnicode(name);
-	if (direct !== void 0) return direct;
-	const uniMatch = /^uni([0-9A-Fa-f]{4,6})$/.exec(name);
-	return uniMatch?.[1] !== void 0 ? Number.parseInt(uniMatch[1], 16) : void 0;
+function readFontProgram(descriptor, context) {
+	if (descriptor === void 0) return;
+	for (const key of [
+		"FontFile2",
+		"FontFile3",
+		"FontFile"
+	]) {
+		const stream = context.resolver.resolve(dictGet(descriptor, key));
+		if (stream?.kind !== "stream") continue;
+		const decoded = decodeStream(stream.raw, stream.dict, context.sink);
+		const encoding = readFontProgramEncoding(decoded.bytes);
+		if (encoding !== void 0) return encoding;
+	}
+}
+function lazyFontProgram(descriptor, context) {
+	let read;
+	return () => {
+		read ??= { encoding: readFontProgram(descriptor, context) };
+		return read.encoding;
+	};
 }
 function readDifferencesMap(differences) {
 	const map = /* @__PURE__ */ new Map();
@@ -70,13 +86,36 @@ function buildSimpleFont(fontDict, context) {
 	const encodingObj = dictGet(fontDict, "Encoding");
 	const encodingDict = context.resolver.resolveDict(encodingObj);
 	const baseEncodingName = asName(encodingDict !== void 0 ? dictGet(encodingDict, "BaseEncoding") : encodingObj);
-	if (baseEncodingName !== void 0 && baseEncodingName !== "WinAnsiEncoding") context.sink({
+	const namedBaseGlyphName = baseEncodingName !== void 0 ? namedEncodingGlyphName(baseEncodingName) : void 0;
+	if (baseEncodingName !== void 0 && namedBaseGlyphName === void 0) context.sink({
 		code: "char/encoding-approximated",
 		severity: "info",
 		message: `base encoding "${baseEncodingName}" has no dedicated table here and is approximated as WinAnsiEncoding`
 	});
 	const differencesMap = readDifferencesMap(encodingDict !== void 0 ? asArray(dictGet(encodingDict, "Differences")) : void 0);
-	const fallbackGlyphName = builtinGlyphName ?? (symbolic ? () => void 0 : winAnsiGlyphName);
+	const programEncoding = lazyFontProgram(descriptor, context);
+	const symbolEncoded = symbolic || builtinGlyphName !== void 0;
+	const byGlyphName = (source) => (code) => {
+		const name = source(code);
+		return name === void 0 ? void 0 : 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 ? [
+		fromProgram,
+		builtinGlyphName === void 0 ? void 0 : byGlyphName(builtinGlyphName),
+		namedBase
+	] : [
+		namedBase,
+		fromProgram,
+		byGlyphName(winAnsiGlyphName)
+	]].filter((resolver) => resolver !== void 0);
+	const unicodeForCode = (code) => {
+		for (const resolve of resolvers) {
+			const unicode = resolve(code);
+			if (unicode !== void 0) return unicode;
+		}
+	};
 	const decodeToUnicode = (codes) => {
 		let out = "";
 		let unmapped = 0;
@@ -86,10 +125,9 @@ function buildSimpleFont(fontDict, context) {
 				out += viaToUnicode;
 				continue;
 			}
-			const name = differencesMap.get(code) ?? fallbackGlyphName(code);
-			const unicode = name !== void 0 ? glyphNameToUnicodeWithUniFallback(name) : void 0;
+			const unicode = unicodeForCode(code);
 			if (unicode !== void 0) {
-				out += String.fromCharCode(unicode);
+				out += String.fromCodePoint(unicode);
 				continue;
 			}
 			unmapped++;
@@ -148,6 +186,8 @@ function buildCompositeFont(fontDict, context) {
 	const widthMap = readCidWidths(descendantDict !== void 0 ? asArray(dictGet(descendantDict, "W")) : void 0);
 	const widthOf = (cid) => widthMap.get(cid) ?? dw;
 	const toUnicode = readToUnicodeCMap(fontDict, context);
+	const cidToGidMap = descendantDict !== void 0 ? dictGet(descendantDict, "CIDToGIDMap") : void 0;
+	const programEncoding = isName(dictGet(fontDict, "Encoding"), "Identity-H") && (cidToGidMap === void 0 || isName(cidToGidMap, "Identity")) ? lazyFontProgram(descriptor, context) : void 0;
 	const decodeToUnicode = (codes) => {
 		let out = "";
 		let glyphCount = 0;
@@ -160,6 +200,11 @@ function buildCompositeFont(fontDict, context) {
 				out += mapped;
 				continue;
 			}
+			const fromProgram = programEncoding?.()?.glyphIdToUnicode(cid);
+			if (fromProgram !== void 0) {
+				out += String.fromCodePoint(fromProgram);
+				continue;
+			}
 			unmapped++;
 			out += REPLACEMENT_CHARACTER;
 		}