Add support for basic structure tree for accessibility.

When a PDF is "marked" we now generate a separate DOM that represents
the structure tree from the PDF.  This DOM is inserted into the <canvas>
element and allows screen readers to walk the tree and have more
information about headings, images, links, etc. To link the structure
tree DOM (which is empty) to the text layer aria-owns is used. This
required modifying the text layer creation so that marked items are
now tracked.

src/core/document.js

@@ -58,6 +58,7 @@ import { calculateMD5 } from "./crypto.js";
 import { Linearization } from "./parser.js";
 import { OperatorList } from "./operator_list.js";
 import { PartialEvaluator } from "./evaluator.js";
+import { StructTreePage } from "./struct_tree.js";
 import { XFAFactory } from "./xfa/factory.js";
 
 const DEFAULT_USER_UNIT = 1.0;
@@ -104,6 +105,10 @@ class Page {
       static createObjId() {
         return `p${pageIndex}_${++idCounters.obj}`;
       }
+
+      static getPageObjId() {
+        return `page${ref.toString()}`;
+      }
     };
   }
 
@@ -406,6 +411,7 @@ class Page {
     handler,
     task,
     normalizeWhitespace,
+    includeMarkedContent,
     sink,
     combineTextItems,
   }) {
@@ -437,12 +443,23 @@ class Page {
         task,
         resources: this.resources,
         normalizeWhitespace,
+        includeMarkedContent,
         combineTextItems,
         sink,
+        pageObjId: this.ref.toString(),
       });
     });
   }
 
+  async getStructTree() {
+    const structTreeRoot = await this.pdfManager.ensureCatalog(
+      "structTreeRoot"
+    );
+    const tree = new StructTreePage(structTreeRoot, this.pageDict);
+    tree.parse();
+    return tree;
+  }
+
   getAnnotationsData(intent) {
     return this._parsedAnnotations.then(function (annotations) {
       const annotationsData = [];
@@ -604,6 +621,10 @@ class PDFDocument {
       static createObjId() {
         unreachable("Abstract method `createObjId` called.");
       }
+
+      static getPageObjId() {
+        unreachable("Abstract method `getPageObjId` called.");
+      }
     };
   }
 

src/core/evaluator.js

@@ -1913,7 +1913,10 @@ class PartialEvaluator {
               return;
             }
             // Other marked content types aren't supported yet.
-            args = [args[0].name];
+            args = [
+              args[0].name,
+              args[1] && isDict(args[1]) ? args[1].get("MCID") : null,
+            ];
 
             break;
           case OPS.beginMarkedContent:
@@ -1973,8 +1976,10 @@ class PartialEvaluator {
     stateManager = null,
     normalizeWhitespace = false,
     combineTextItems = false,
+    includeMarkedContent = false,
     sink,
     seenStyles = new Set(),
+    pageObjId,
   }) {
     // Ensure that `resources`/`stateManager` is correctly initialized,
     // even if the provided parameter is e.g. `null`.
@@ -2573,8 +2578,10 @@ class PartialEvaluator {
                     stateManager: xObjStateManager,
                     normalizeWhitespace,
                     combineTextItems,
+                    includeMarkedContent,
                     sink: sinkWrapper,
                     seenStyles,
+                    pageObjId,
                   })
                   .then(function () {
                     if (!sinkWrapper.enqueueInvoked) {
@@ -2650,6 +2657,38 @@ class PartialEvaluator {
               })
             );
             return;
+          case OPS.beginMarkedContent:
+            if (includeMarkedContent) {
+              textContent.items.push({
+                type: "beginMarkedContent",
+                tag: isName(args[0]) ? args[0].name : null,
+              });
+            }
+            break;
+          case OPS.beginMarkedContentProps:
+            if (includeMarkedContent) {
+              flushTextContentItem();
+              let mcid = null;
+              if (isDict(args[1])) {
+                mcid = args[1].get("MCID");
+              }
+              textContent.items.push({
+                type: "beginMarkedContentProps",
+                id: Number.isInteger(mcid)
+                  ? `${self.idFactory.getPageObjId()}_mcid${mcid}`
+                  : null,
+                tag: isName(args[0]) ? args[0].name : null,
+              });
+            }
+            break;
+          case OPS.endMarkedContent:
+            if (includeMarkedContent) {
+              flushTextContentItem();
+              textContent.items.push({
+                type: "endMarkedContent",
+              });
+            }
+            break;
         } // switch
         if (textContent.items.length >= sink.desiredSize) {
           // Wait for ready, if we reach highWaterMark.

src/core/obj.js

@@ -60,6 +60,7 @@ import { CipherTransformFactory } from "./crypto.js";
 import { ColorSpace } from "./colorspace.js";
 import { GlobalImageCache } from "./image_utils.js";
 import { MetadataParser } from "./metadata_parser.js";
+import { StructTreeRoot } from "./struct_tree.js";
 
 function fetchDestination(dest) {
   return isDict(dest) ? dest.get("D") : dest;
@@ -200,6 +201,32 @@ class Catalog {
     return markInfo;
   }
 
+  get structTreeRoot() {
+    let structTree = null;
+    try {
+      structTree = this._readStructTreeRoot();
+    } catch (ex) {
+      if (ex instanceof MissingDataException) {
+        throw ex;
+      }
+      warn("Unable read to structTreeRoot info.");
+    }
+    return shadow(this, "structTreeRoot", structTree);
+  }
+
+  /**
+   * @private
+   */
+  _readStructTreeRoot() {
+    const obj = this._catDict.get("StructTreeRoot");
+    if (!isDict(obj)) {
+      return null;
+    }
+    const root = new StructTreeRoot(obj);
+    root.init();
+    return root;
+  }
+
   get toplevelPagesDict() {
     const pagesObj = this._catDict.get("Pages");
     if (!isDict(pagesObj)) {
@@ -2626,4 +2653,4 @@ const ObjectLoader = (function () {
   return ObjectLoader;
 })();
 
-export { Catalog, FileSpec, ObjectLoader, XRef };
+export { Catalog, FileSpec, NumberTree, ObjectLoader, XRef };