From caab1b0e1642bc6ee0194c7ab4c4a4de2ab22f2c Mon Sep 17 00:00:00 2001
From: Jay Berkenbilt <ejb@ql.org>
Date: Fri, 14 Nov 2014 13:43:34 -0500
Subject: [PATCH] Handle pages with no /Contents from getPageContents()

---
 ChangeLog                           | 8 ++++++++
 libqpdf/QPDFObjectHandle.cc         | 2 +-
 qpdf/qtest/qpdf.test                | 7 ++++++-
 qpdf/qtest/qpdf/page-no-content.out | 8 ++++++++
 qpdf/qtest/qpdf/page-no-content.pdf | Bin 0 -> 1362 bytes
 5 files changed, 23 insertions(+), 2 deletions(-)
 create mode 100644 qpdf/qtest/qpdf/page-no-content.out
 create mode 100644 qpdf/qtest/qpdf/page-no-content.pdf

diff --git a/ChangeLog b/ChangeLog
index 8816d6c..77ab74b 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,11 @@
+2014-11-14  Jay Berkenbilt  <ejb@ql.org>
+
+	* Bug fix: QPDFObjectHandle::getPageContents() no longer throws an
+	exception when called on a page that has no /Contents key in its
+	dictionary. This is allowed by the spec, and some software
+	packages generate files like this for pages that are blank in the
+	original.
+
 2014-06-07  Jay Berkenbilt  <ejb@ql.org>
 
 	* 5.1.2: release
diff --git a/libqpdf/QPDFObjectHandle.cc b/libqpdf/QPDFObjectHandle.cc
index ef92e3a..eec4fae 100644
--- a/libqpdf/QPDFObjectHandle.cc
+++ b/libqpdf/QPDFObjectHandle.cc
@@ -633,7 +633,7 @@ QPDFObjectHandle::getPageContents()
     {
 	result.push_back(contents);
     }
-    else
+    else if (! contents.isNull())
     {
 	throw std::runtime_error("unknown object type inspecting /Contents "
 				 "key in page dictionary");
diff --git a/qpdf/qtest/qpdf.test b/qpdf/qtest/qpdf.test
index 8e69784..b2bf06a 100644
--- a/qpdf/qtest/qpdf.test
+++ b/qpdf/qtest/qpdf.test
@@ -199,7 +199,7 @@ $td->runtest("remove page we don't have",
 show_ntests();
 # ----------
 $td->notify("--- Miscellaneous Tests ---");
-$n_tests += 72;
+$n_tests += 73;
 
 $td->runtest("qpdf version",
 	     {$td->COMMAND => "qpdf --version"},
@@ -554,6 +554,11 @@ $td->runtest("check file",
              {$td->FILE => "a.pdf"},
              {$td->FILE => "indirect-decode-parms-out.pdf"});
 
+$td->runtest("handle page no with contents",
+             {$td->COMMAND => "qpdf --show-pages page-no-content.pdf"},
+             {$td->FILE => "page-no-content.out", $td->EXIT_STATUS => 0},
+             $td->NORMALIZE_NEWLINES);
+
 show_ntests();
 # ----------
 $td->notify("--- Numeric range parsing tests ---");
diff --git a/qpdf/qtest/qpdf/page-no-content.out b/qpdf/qtest/qpdf/page-no-content.out
new file mode 100644
index 0000000..15df72d
--- /dev/null
+++ b/qpdf/qtest/qpdf/page-no-content.out
@@ -0,0 +1,8 @@
+page 1: 3 0 R
+  content:
+    6 0 R
+page 2: 4 0 R
+  content:
+page 3: 5 0 R
+  content:
+    9 0 R
diff --git a/qpdf/qtest/qpdf/page-no-content.pdf b/qpdf/qtest/qpdf/page-no-content.pdf
new file mode 100644
index 0000000..77b7832
Binary files /dev/null and b/qpdf/qtest/qpdf/page-no-content.pdf differ
--
libgit2 0.21.4