From 7d94fdeb489ae8ac18b8c38f011abf961b5fb45c Mon Sep 17 00:00:00 2001
From: Jonas Jenwald <jonas.jenwald@gmail.com>
Date: Fri, 6 Jan 2023 10:27:20 +0100
Subject: [PATCH] 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                 |  38 ++++++++++++++++++++++---------
 test/pdfs/.gitignore             |   1 +
 test/pdfs/issue15893_reduced.pdf | Bin 0 -> 1610 bytes
 test/test_manifest.json          |   7 ++++++
 4 files changed, 35 insertions(+), 11 deletions(-)
 create mode 100644 test/pdfs/issue15893_reduced.pdf

diff --git a/src/core/xref.js b/src/core/xref.js
index c55715463..b378e9476 100644
--- a/src/core/xref.js
+++ b/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.
diff --git a/test/pdfs/.gitignore b/test/pdfs/.gitignore
index 4439ffd3d..bb699a606 100644
--- a/test/pdfs/.gitignore
+++ b/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
diff --git a/test/pdfs/issue15893_reduced.pdf b/test/pdfs/issue15893_reduced.pdf
new file mode 100644
index 0000000000000000000000000000000000000000..1a4e66fbeeb88e5329ca3dd475e412861431275f
GIT binary patch
literal 1610
zcmcgsJ!lj`6t=LMMub@Vnj<+RnVJ1RIWKs*9K=Ku?oKU^Xy$I`&JDTUxVwjmRSH2(
zp^aEtSlDS}qsBte!k<D!un;Anji4yOM#aL}z1z(u7e$0PY_Ys~Z@%xn`Mya_6^es&
zNJpu6@1MU!6bLxASwPt=!c(ltJ&;7Ggm7hXo&z3XKAUrzD3?RrHp9$dKQ0-!Ob@DZ
zP%<6WgvTH};<RlaXtFfOi%X}N%SA#LuEag3?bZc;NTukAt<E0>C1fNhhz}p;rp5Bk
z0#u0zON4+*Hb4i+wl9wMK#d9`?!iehfjskgh=VIui+h75r^Rf9%k7#U_8v@tMok5Y
z2#?xz$F%Gw;NzBk(Dtk-knA!*jX@^3-Skg^VgwnU?{d~c`3j-|b*K~&F4FL?fd<Ar
zug!s~Mn+CUWfqNAP%!jP5k&J}aCrj}&=83iXsRl!(118b1&3gPy&luFo*0#S^CVRa
zX;ZDNCh~ljS##V);q8<hN8E8ZW?Z(;a={Ca6`-0m>k_Zk8wyi6(`$`PjS_<>b*3qd
z8dQ~XSuW{1H5*JZSw=DqT{1YYNe!N<aY{`t=cYR&W3C_etjYwXQlpbal-f=WN3!i2
zpd;pl{Qscg-$?VXrM*eF2?j}d{fmYpnfB_o3lhpZ(dAYeR^^0-XKt^~yiKpZ@^>zO
zeX@4<(8bH2o~=~XTUO!bnI&WBk^TMSjr|Xvmi6%qm#k~U=g$rA>yx&SGZLy?cS{Lm
z@eIMBoFNLrlQ3}a^@|T?-n!DiyOo<RoxZ<xeWCB?iKTCQzI^_cQ72BOg@LR&pSys|
zKC?~cnh1A2M<oN{V;~`X6bA0B;pK-v`VZ`VLyla1ys~S@rj<^NxN9*zwv#Z@f3;ZX
kKSC5uI4iQ^2HT1IbQkA+2!<!#>K+&B+vY-z7+n|s0u+F{_y7O^

literal 0
HcmV?d00001

diff --git a/test/test_manifest.json b/test/test_manifest.json
index 70f277edd..d126594d3 100644
--- a/test/test_manifest.json
+++ b/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",