Support parsing encrypted documents in XRef.indexObjects (issue 15893)

*Please note:* The reduced test-case is *not* a perfect reproduction of the original PDF document, since this one fails to open in e.g. Adobe Reader, but I do believe that it captures the most important points here.

For corrupt *and* encrypted PDF documents, it's possible that only some trailer dictionaries actually contain an /Encrypt-entry. Previously we'd could easily miss that, since we generally pick the first not obviously corrupt trailer dictionary, and the solution implemented here is to simply pre-parse all trailer dictionaries to see if there's any /Encrypt-entries.

src/core/xref.js

@@ -581,16 +581,11 @@ class XRef {
       this.startXRefQueue.push(xrefStm);
       this.readXRef(/* recoveryMode */ true);
     }
-    // finding main trailer
-    let trailerDict, trailerError;
-    for (const trailer of [...trailers, "generationFallback", ...trailers]) {
-      if (trailer === "generationFallback") {
-        if (!trailerError) {
-          break; // No need to fallback if there were no validation errors.
-        }
-        this._generationFallback = true;
-        continue;
-      }
+
+    const trailerDicts = [];
+    // Pre-parsing the trailers to check if the document is possibly encrypted.
+    let isEncrypted = false;
+    for (const trailer of trailers) {
       stream.pos = trailer;
       const parser = new Parser({
         lexer: new Lexer(stream),
@@ -607,6 +602,23 @@ class XRef {
       if (!(dict instanceof Dict)) {
         continue;
       }
+      trailerDicts.push(dict);
+
+      if (dict.has("Encrypt")) {
+        isEncrypted = true;
+      }
+    }
+
+    // finding main trailer
+    let trailerDict, trailerError;
+    for (const dict of [...trailerDicts, "genFallback", ...trailerDicts]) {
+      if (dict === "genFallback") {
+        if (!trailerError) {
+          break; // No need to fallback if there were no validation errors.
+        }
+        this._generationFallback = true;
+        continue;
+      }
       // Do some basic validation of the trailer/root dictionary candidate.
       let validPagesDict = false;
       try {
@@ -628,7 +640,11 @@ class XRef {
         continue;
       }
       // taking the first one with 'ID'
-      if (validPagesDict && dict.has("ID")) {
+      if (
+        validPagesDict &&
+        (!isEncrypted || dict.has("Encrypt")) &&
+        dict.has("ID")
+      ) {
         return dict;
       }
       // The current dictionary is a candidate, but continue searching.

test/pdfs/.gitignore

@@ -413,6 +413,7 @@
 !issue2128r.pdf
 !bug1703683_page2_reduced.pdf
 !issue5540.pdf
+!issue15893_reduced.pdf
 !issue5549.pdf
 !visibility_expressions.pdf
 !issue5475.pdf

test/test_manifest.json

@@ -386,6 +386,13 @@
       "lastPage": 4,
       "type": "eq"
     },
+    {  "id": "issue15893_reduced",
+       "file": "pdfs/issue15893_reduced.pdf",
+       "md5": "cf889b927f9f53d164622a99378bf39d",
+       "rounds": 1,
+       "type": "eq",
+       "password": "test"
+    },
     {  "id": "bug1727053",
        "file": "pdfs/bug1727053.pdf",
        "md5": "8ed1e52da64000f9fdcb8b732f5a58f8",