Always flush the current item with MarkedContent stuff when getting t…

…ext (mozilla#15094)
calixteman · Jun 25, 2022 · f161929 · f161929
1 parent 23fcdab
commit f161929
Show file tree

Hide file tree

Showing 2 changed files with 24 additions and 3 deletions.
diff --git a/src/core/evaluator.js b/src/core/evaluator.js
@@ -3290,6 +3290,7 @@ class PartialEvaluator {
             );
             return;
           case OPS.beginMarkedContent:
+            flushTextContentItem();
             if (includeMarkedContent) {
               textContent.items.push({
                 type: "beginMarkedContent",
@@ -3298,8 +3299,8 @@ class PartialEvaluator {
             }
             break;
           case OPS.beginMarkedContentProps:
+            flushTextContentItem();
             if (includeMarkedContent) {
-              flushTextContentItem();
               let mcid = null;
               if (args[1] instanceof Dict) {
                 mcid = args[1].get("MCID");
@@ -3314,8 +3315,8 @@ class PartialEvaluator {
             }
             break;
           case OPS.endMarkedContent:
+            flushTextContentItem();
             if (includeMarkedContent) {
-              flushTextContentItem();
               textContent.items.push({
                 type: "endMarkedContent",
               });

diff --git a/test/unit/api_spec.js b/test/unit/api_spec.js
@@ -78,7 +78,9 @@ describe("api", function () {
   }
 
   function mergeText(items) {
-    return items.map(chunk => chunk.str + (chunk.hasEOL ? "\n" : "")).join("");
+    return items
+      .map(chunk => (chunk.str ?? "") + (chunk.hasEOL ? "\n" : ""))
+      .join("");
   }
 
   describe("getDocument", function () {
@@ -2275,6 +2277,24 @@ Caron Broadcasting, Inc., an Ohio corporation (“Lessee”).`)
       await loadingTask.destroy();
     });
 
+    it("gets text content with or without includeMarkedContent, and compare (issue 15094)", async function () {
+      const loadingTask = getDocument(buildGetDocumentParams("pdf.pdf"));
+      const pdfDoc = await loadingTask.promise;
+      const pdfPage = await pdfDoc.getPage(568);
+      let { items } = await pdfPage.getTextContent({
+        includeMarkedContent: false,
+      });
+      const textWithoutMC = mergeText(items);
+      ({ items } = await pdfPage.getTextContent({
+        includeMarkedContent: true,
+      }));
+      const textWithMC = mergeText(items);
+
+      expect(textWithoutMC).toEqual(textWithMC);
+
+      await loadingTask.destroy();
+    });
+
     it("gets empty structure tree", async function () {
       const tree = await page.getStructTree();