pdf-codec 5.5.1 → 5.6.1

package/dist/read.d.ts

@@ -18,7 +18,7 @@ interface ReadPdfOptions {
   /**
    * Which document-level surfaces to read. Each is read unless its entry is `false`, so omitting the setting, or an entry, reads it as before. Independent of `pages`: it applies to a full read and a page-selected read alike.
    *
-   * A surface turned off is never parsed (its part of the catalog is not walked, so a malformed outline, form, layer list or structure tree reports no diagnostic) and its field is absent from the result whatever the file holds; the result lists it in `skippedSurfaces`, which is how a caller tells "not read" from "none". What each one covers, and what page items lose without it:
+   * A surface turned off is never parsed (its part of the catalog is not walked, so a malformed outline, form, layer list or structure tree reports no diagnostic, and no image a page draws is decoded or diagnosed) and its field is absent from the result whatever the file holds (`images` is left empty instead); the result lists it in `skippedSurfaces`, which is how a caller tells "not read" from "none". What each one covers, and what page items lose without it:
    *
    * - `outline`: the `/Outlines` bookmark tree (`outline`). Page items do not depend on it, except through minted names: a destination stated as a direct array is named with the first unused `destN` in the order the reader meets it, and the outline is read before any page, so an outline entry's direct destination takes a lower number than any link's. Without the outline, the names minted for page links' direct destinations are renumbered, and `destinations` loses the entries only the outline minted; each link still resolves to the same target.
    * - `destinations`: the named destinations of the catalog `/Dests` dictionary and the `/Names /Dests` name tree, and the registry links and outline entries name their targets through (`destinations`). Without it nothing can name an internal target, so a link whose target is internal (a `/Dest`, or a `/GoTo` action, named or direct) is left out of its page's items, and a read outline entry carries no `destination`; a link to a URI is kept as usual.
@@ -26,6 +26,7 @@ interface ReadPdfOptions {
    * - `layers`: optional content, the `/OCProperties` groups and their default visibility (`layers`). Without it no item carries a `layer`, since an `/OC` membership resolves to no name; the items themselves are all still read, so content the default view hides is indistinguishable from visible content.
    * - `structure`: the `/StructTreeRoot` element tree and its `/ParentTree` ownership map (`structure`). Without it no item carries a `structure` owner, and the parent-tree diagnostics are not reported.
    * - `residue`: the package-level residue rows (`source`), including the per-page `page-boxes` row. Page items do not depend on it; `metadata` still reads the XMP packet it falls back to.
+   * - `images`: the image registry (`images`), which holds every image XObject and inline image a read page draws, decoded to PNG or passed through as JPEG. Without it no image's stream is decoded, `images` is an empty record (every `LayoutDocument` carries one), and each page's `image` items are left out, since they would name assets that were never read. Decoding an image costs memory in proportion to its pixels, so a read that wants only a page's text leaves this out.
    *
    * Apart from those stated losses and the renumbering a skipped outline causes, every page item is exactly what a read with the surface on produces.
    */

package/dist/read.js

@@ -16,7 +16,7 @@ import { NOOP_DIAGNOSTIC_SINK, PdfPageSelectionError, PdfParseError } from "./di
 import { openPdfDocument } from "./document.js";
 import { throwIfAborted } from "./util/abort.js";
 import { createFontResolver } from "./font-read.js";
-import { t as interpretContentStream } from "./interpret---0lZPlv.js";
+import { t as interpretContentStream } from "./interpret-BeMKZGBm.js";
 import { createDestinationRegistry, readOutline } from "./navigation.js";
 //#region src/read.ts
 function skippedSurfacesOf(surfaces) {
@@ -58,11 +58,14 @@ function readPdf(bytes, options) {
 		sink
 	});
 	const images = {};
-	const imageIdCache = /* @__PURE__ */ new Map();
 	const pageDicts = doc.pages();
 	const selection = options?.pages === void 0 ? void 0 : resolvePageSelection(options.pages, pageDicts.length);
 	const skippedSurfaces = skippedSurfacesOf(options?.surfaces);
 	const reads = (surface) => !skippedSurfaces.includes(surface);
+	const imageRegistry = reads("images") ? {
+		images,
+		imageIdCache: /* @__PURE__ */ new Map()
+	} : void 0;
 	const destinationRegistry = reads("destinations") ? createDestinationRegistry(doc.catalog, doc, (obj) => doc.pageIndex(obj), sink) : UNREAD_DESTINATION_REGISTRY;
 	const outline = reads("outline") ? readOutline(doc.catalog, destinationRegistry, doc, sink) : void 0;
 	const attachments = selection === void 0 ? readAttachments(doc.catalog, pageDicts, doc, sink) : [];
@@ -74,7 +77,7 @@ function readPdf(bytes, options) {
 	const pages = pageDicts.flatMap((pageDict, index) => {
 		if (selection !== void 0 && !selection.has(index)) return [];
 		throwIfAborted(signal);
-		const read = readPage(index, pageDict, doc, fontResolver, images, imageIdCache, destinationRegistry, optionalContent?.layerNameOf ?? noLayerName, structure?.ownerOf ?? noStructureOwner, sink);
+		const read = readPage(index, pageDict, doc, fontResolver, imageRegistry, destinationRegistry, optionalContent?.layerNameOf ?? noLayerName, structure?.ownerOf ?? noStructureOwner, sink);
 		if (read.boxRow !== void 0) pageBoxRows.push(read.boxRow);
 		return [selection === void 0 ? read.page : {
 			...read.page,
@@ -308,7 +311,7 @@ function pageBoxResidueEntry(pageIndex, page, mediaBox, cropBox) {
 	for (const [key, value] of declaredBoxes) if (value !== void 0) entries[key] = value;
 	return pdfDict(entries);
 }
-function readPage(pageIndex, page, resolver, fontResolver, images, imageIdCache, destinationRegistry, layerNameOf, structureOwnerOf, sink) {
+function readPage(pageIndex, page, resolver, fontResolver, imageRegistry, destinationRegistry, layerNameOf, structureOwnerOf, sink) {
 	const resources = resolver.resolveDict(dictGet(page, "Resources"));
 	const mediaBox = readMediaBox(page);
 	let cropBox = readDeclaredPageBox(page, "CropBox") ?? mediaBox;
@@ -337,7 +340,7 @@ function readPage(pageIndex, page, resolver, fontResolver, images, imageIdCache,
 			layerNameOf
 		});
 		for (const item of extracted) {
-			const converted = convertExtractedItem(item, pageMatrix, fontResolver, images, imageIdCache, resolver, sink);
+			const converted = convertExtractedItem(item, pageMatrix, fontResolver, imageRegistry, resolver, sink);
 			if (converted === void 0 || !itemIntersectsVisibleRegion(converted, widthPt, heightPt)) continue;
 			const owner = item.mcid === void 0 ? void 0 : structureOwnerOf(pageIndex, item.mcid);
 			items.push(owner === void 0 ? converted : withStructure(converted, owner));

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pdf-codec",
-  "version": "5.5.1",
+  "version": "5.6.1",
   "description": "Hand-written, dependency-minimal PDF codec: parses arbitrary real-world PDFs and generates new ones, built on its own codec-owned LayoutDocument item model and Zod 4 codecs.",
   "type": "module",
   "repository": {