Ensure that worker-thread image caching doesn't break optional content (issue 14824)

Currently we only insert optionalContent-data into the operatorList the first time that an image is parsed, which will (in hindsight) obviously cause problems for cached images.
Hence we also need to insert the optionalContent-data in the various worker-thread image caches, such that it can be accessed in the fast-paths that are used to skip re-parsing of images.

In order to reduce the amount of repeated code, this patch also adds a new `OperatorList`-method that takes care of inserting the necessary data in the operatorList.

src/core/evaluator.js

@@ -592,12 +592,8 @@ class PartialEvaluator {
         resources
       );
     }
-    if (optionalContent !== undefined) {
-      operatorList.addOp(OPS.beginMarkedContentProps, ["OC", optionalContent]);
-    }
 
     const imageMask = dict.get("IM", "ImageMask") || false;
-    const interpolate = dict.get("I", "Interpolate");
     let imgData, args;
     if (imageMask) {
       // This depends on a tmpCanvas being filled with the
@@ -605,6 +601,7 @@ class PartialEvaluator {
       // data can't be done here. Instead of creating a
       // complete PDFImage, only read the information needed
       // for later.
+      const interpolate = dict.get("I", "Interpolate");
       const bitStrideLength = (w + 7) >> 3;
       const imgArray = image.getBytes(
         bitStrideLength * h,
@@ -625,17 +622,19 @@ class PartialEvaluator {
         imgData.cached = !!cacheKey;
         args = [imgData];
 
-        operatorList.addOp(OPS.paintImageMaskXObject, args);
+        operatorList.addImageOps(
+          OPS.paintImageMaskXObject,
+          args,
+          optionalContent
+        );
+
         if (cacheKey) {
           localImageCache.set(cacheKey, imageRef, {
             fn: OPS.paintImageMaskXObject,
             args,
+            optionalContent,
           });
         }
-
-        if (optionalContent !== undefined) {
-          operatorList.addOp(OPS.endMarkedContent, []);
-        }
         return;
       }
 
@@ -651,17 +650,19 @@ class PartialEvaluator {
       if (imgData.isSingleOpaquePixel) {
         // Handles special case of mainly LaTeX documents which use image
         // masks to draw lines with the current fill style.
-        operatorList.addOp(OPS.paintSolidColorImageMask, []);
+        operatorList.addImageOps(
+          OPS.paintSolidColorImageMask,
+          [],
+          optionalContent
+        );
+
         if (cacheKey) {
           localImageCache.set(cacheKey, imageRef, {
             fn: OPS.paintSolidColorImageMask,
             args: [],
+            optionalContent,
           });
         }
-
-        if (optionalContent !== undefined) {
-          operatorList.addOp(OPS.endMarkedContent, []);
-        }
         return;
       }
 
@@ -678,18 +679,19 @@ class PartialEvaluator {
           count: 1,
         },
       ];
+      operatorList.addImageOps(
+        OPS.paintImageMaskXObject,
+        args,
+        optionalContent
+      );
 
-      operatorList.addOp(OPS.paintImageMaskXObject, args);
       if (cacheKey) {
         localImageCache.set(cacheKey, imageRef, {
           fn: OPS.paintImageMaskXObject,
           args,
+          optionalContent,
         });
       }
-
-      if (optionalContent !== undefined) {
-        operatorList.addOp(OPS.endMarkedContent, []);
-      }
       return;
     }
 
@@ -710,11 +712,11 @@ class PartialEvaluator {
       // We force the use of RGBA_32BPP images here, because we can't handle
       // any other kind.
       imgData = imageObj.createImageData(/* forceRGBA = */ true);
-      operatorList.addOp(OPS.paintInlineImageXObject, [imgData]);
-
-      if (optionalContent !== undefined) {
-        operatorList.addOp(OPS.endMarkedContent, []);
-      }
+      operatorList.addImageOps(
+        OPS.paintInlineImageXObject,
+        [imgData],
+        optionalContent
+      );
       return;
     }
 
@@ -762,11 +764,13 @@ class PartialEvaluator {
         return this._sendImgData(objId, /* imgData = */ null, cacheGlobally);
       });
 
-    operatorList.addOp(OPS.paintImageXObject, args);
+    operatorList.addImageOps(OPS.paintImageXObject, args, optionalContent);
+
     if (cacheKey) {
       localImageCache.set(cacheKey, imageRef, {
         fn: OPS.paintImageXObject,
         args,
+        optionalContent,
       });
 
       if (imageRef) {
@@ -778,15 +782,12 @@ class PartialEvaluator {
             objId,
             fn: OPS.paintImageXObject,
             args,
+            optionalContent,
             byteSize: 0, // Temporary entry, note `addByteSize` above.
           });
         }
       }
     }
-
-    if (optionalContent !== undefined) {
-      operatorList.addOp(OPS.endMarkedContent, []);
-    }
   }
 
   handleSMask(
@@ -1700,7 +1701,12 @@ class PartialEvaluator {
             if (isValidName) {
               const localImage = localImageCache.getByName(name);
               if (localImage) {
-                operatorList.addOp(localImage.fn, localImage.args);
+                operatorList.addImageOps(
+                  localImage.fn,
+                  localImage.args,
+                  localImage.optionalContent
+                );
+
                 if (
                   localImage.fn === OPS.paintImageMaskXObject &&
                   localImage.args[0] &&
@@ -1723,7 +1729,12 @@ class PartialEvaluator {
                 if (xobj instanceof Ref) {
                   const localImage = localImageCache.getByRef(xobj);
                   if (localImage) {
-                    operatorList.addOp(localImage.fn, localImage.args);
+                    operatorList.addImageOps(
+                      localImage.fn,
+                      localImage.args,
+                      localImage.optionalContent
+                    );
+
                     if (
                       localImage.fn === OPS.paintImageMaskXObject &&
                       localImage.args[0] &&
@@ -1741,7 +1752,11 @@ class PartialEvaluator {
                   );
                   if (globalImage) {
                     operatorList.addDependency(globalImage.objId);
-                    operatorList.addOp(globalImage.fn, globalImage.args);
+                    operatorList.addImageOps(
+                      globalImage.fn,
+                      globalImage.args,
+                      globalImage.optionalContent
+                    );
 
                     resolveXObject();
                     return;
@@ -1846,7 +1861,12 @@ class PartialEvaluator {
             if (cacheKey) {
               const localImage = localImageCache.getByName(cacheKey);
               if (localImage) {
-                operatorList.addOp(localImage.fn, localImage.args);
+                operatorList.addImageOps(
+                  localImage.fn,
+                  localImage.args,
+                  localImage.optionalContent
+                );
+
                 if (
                   localImage.fn === OPS.paintImageMaskXObject &&
                   localImage.args[0] &&

src/core/operator_list.js

@@ -622,6 +622,18 @@ class OperatorList {
     }
   }
 
+  addImageOps(fn, args, optionalContent) {
+    if (optionalContent !== undefined) {
+      this.addOp(OPS.beginMarkedContentProps, ["OC", optionalContent]);
+    }
+
+    this.addOp(fn, args);
+
+    if (optionalContent !== undefined) {
+      this.addOp(OPS.endMarkedContent, []);
+    }
+  }
+
   addDependency(dependency) {
     if (this.dependencies.has(dependency)) {
       return;

test/pdfs/issue14824.pdf.link

@@ -0,0 +1 @@
+https://github.com/mozilla/pdf.js/files/8540275/PDF.pdf

test/test_manifest.json

@@ -2965,6 +2965,18 @@
        "annotations": true,
        "about": "LinkAnnotation with a relative link, and a /Catalog Base-URI."
     },
+    {  "id": "issue14824",
+       "file": "pdfs/issue14824.pdf",
+       "md5": "7b8d061ab0a342e3606a3b3ba1925d5b",
+       "rounds": 1,
+       "link": true,
+       "lastPage": 4,
+       "type": "eq",
+       "optionalContent": {
+         "7R": false
+       },
+       "about": "Need to test *at least* three pages, since the `GlobalImageCache` is involved."
+    },
     {  "id": "issue1127-text",
        "file": "pdfs/issue1127.pdf",
        "md5": "4fb2be5ffefeafda4ba977de2a1bb4d8",