From ff4dae05b0f8349b6277d45e2d282ad893a4b3af Mon Sep 17 00:00:00 2001
From: Jonas Jenwald <jonas.jenwald@gmail.com>
Date: Sun, 11 Apr 2021 12:00:14 +0200
Subject: [PATCH 1/2] Ensure that `getStructTree` won't break with
`disableAutoFetch = true` set (PR 13171 follow-up)
Open http://localhost:8888/web/viewer.html?file=/test/pdfs/pdf.pdf#disableStream=true&disableAutoFetch=true and observe the following message in the console (repeated for each page of the document):
```
Uncaught (in promise)
Object { message: "Missing data [19787293, 19787294)", name: "UnknownErrorException", details: "MissingDataException: Missing data [19787293, 19787294)", stack: "BaseExceptionClosure@http://localhost:8888/src/shared/util.js:458:29\n@http://localhost:8888/src/shared/util.js:462:3\n" }
```
---
src/core/document.js | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/src/core/document.js b/src/core/document.js
index 11830f0b0..27cbede28 100644
--- a/src/core/document.js
+++ b/src/core/document.js
@@ -454,6 +454,13 @@ class Page {
const structTreeRoot = await this.pdfManager.ensureCatalog(
"structTreeRoot"
);
+ return this.pdfManager.ensure(this, "_parseStructTree", [structTreeRoot]);
+ }
+
+ /**
+ * @private
+ */
+ _parseStructTree(structTreeRoot) {
const tree = new StructTreePage(structTreeRoot, this.pageDict);
tree.parse();
return tree;
From 5adee0cdd14017ca79a56ef00c4ec07659f17084 Mon Sep 17 00:00:00 2001
From: Jonas Jenwald <jonas.jenwald@gmail.com>
Date: Sun, 11 Apr 2021 12:04:29 +0200
Subject: [PATCH 2/2] [api-minor] Let `PDFPageProxy.getStructTree` return
`null`, rather than an empty structTree, for documents without any
accessibility data (PR 13171 follow-up)
This is first of all consistent with existing API-methods, where we return `null` when the data in question doesn't exist. Secondly, it should also be (slightly) more efficient since there's less dummy-data that we need to transfer between threads.
Finally, this prevents us from adding an empty/unnecessary span to *every* single page even in documents without any structure tree data.
---
src/core/struct_tree.js | 4 +++
src/display/api.js | 3 +-
test/unit/api_spec.js | 63 +++++++++++++++++++++++++++++++++++++++++
web/pdf_page_view.js | 3 ++
4 files changed, 72 insertions(+), 1 deletion(-)
diff --git a/src/core/struct_tree.js b/src/core/struct_tree.js
index 41587d45c..a07d99b96 100644
--- a/src/core/struct_tree.js
+++ b/src/core/struct_tree.js
@@ -328,6 +328,10 @@ class StructTreePage {
}
nodeToSerializable(child, root);
}
+
+ if (root.children.length === 0) {
+ return null;
+ }
return root;
}
}
diff --git a/src/display/api.js b/src/display/api.js
index 9b5f70536..b05fd4641 100644
--- a/src/display/api.js
+++ b/src/display/api.js
@@ -1522,7 +1522,8 @@ class PDFPageProxy {
/**
* @returns {Promise<StructTreeNode>} A promise that is resolved with a
- * {@link StructTreeNode} object that represents the page's structure tree.
+ * {@link StructTreeNode} object that represents the page's structure tree,
+ * or `null` when no structure tree is present for the current page.
*/
getStructTree() {
return (this._structTreePromise ||= this._transport.getStructTree(
diff --git a/test/unit/api_spec.js b/test/unit/api_spec.js
index cc0548335..af9a494bf 100644
--- a/test/unit/api_spec.js
+++ b/test/unit/api_spec.js
@@ -1702,6 +1702,69 @@ describe("api", function () {
.catch(done.fail);
});
+ it("gets empty structure tree", async function () {
+ const tree = await page.getStructTree();
+
+ expect(tree).toEqual(null);
+ });
+ it("gets simple structure tree", async function () {
+ const loadingTask = getDocument(
+ buildGetDocumentParams("structure_simple.pdf")
+ );
+ const pdfDoc = await loadingTask.promise;
+ const pdfPage = await pdfDoc.getPage(1);
+ const tree = await pdfPage.getStructTree();
+
+ expect(tree).toEqual({
+ role: "Root",
+ children: [
+ {
+ role: "Document",
+ children: [
+ {
+ role: "H1",
+ children: [
+ {
+ role: "NonStruct",
+ children: [{ type: "content", id: "page2R_mcid0" }],
+ },
+ ],
+ },
+ {
+ role: "P",
+ children: [
+ {
+ role: "NonStruct",
+ children: [{ type: "content", id: "page2R_mcid1" }],
+ },
+ ],
+ },
+ {
+ role: "H2",
+ children: [
+ {
+ role: "NonStruct",
+ children: [{ type: "content", id: "page2R_mcid2" }],
+ },
+ ],
+ },
+ {
+ role: "P",
+ children: [
+ {
+ role: "NonStruct",
+ children: [{ type: "content", id: "page2R_mcid3" }],
+ },
+ ],
+ },
+ ],
+ },
+ ],
+ });
+
+ await loadingTask.destroy();
+ });
+
it("gets operator list", function (done) {
const promise = page.getOperatorList();
promise
diff --git a/web/pdf_page_view.js b/web/pdf_page_view.js
index f83b0568d..8d6771129 100644
--- a/web/pdf_page_view.js
+++ b/web/pdf_page_view.js
@@ -619,6 +619,9 @@ class PDFPageView {
this.eventBus._off("textlayerrendered", this._onTextLayerRendered);
this._onTextLayerRendered = null;
this.pdfPage.getStructTree().then(tree => {
+ if (!tree) {
+ return;
+ }
const treeDom = this.structTreeLayer.render(tree);
treeDom.classList.add("structTree");
this.canvas.appendChild(treeDom);