Ignore unused field in xref entry, avoiding range error (fixes #482)

Jay Berkenbilt
1 parent 3e5aaa29
Showing 6 changed files with 45 additions and 3 deletions
ChangeLog
libqpdf/QPDF.cc
qpdf/qtest/qpdf.test
qpdf/qtest/qpdf/issue-150.out
qpdf/qtest/qpdf/xref-range.out
qpdf/qtest/qpdf/xref-range.pdf
+2020-11-04  Jay Berkenbilt  <ejb@ql.org>
+
+	* Ignore the value of the offset/generation field in an xref entry
+	for a deleted object. Also attempt file recovery on lower-level
+	exceptions thrown while reading the xref table. Fixes #482.
+
 2020-10-31  Jay Berkenbilt  <ejb@ql.org>
  
 	* 10.0.3: release
@@ -425,7 +425,20 @@ QPDF::parse(char const* password)
 	    throw QPDFExc(qpdf_e_damaged_pdf, this->m->file->getName(), "", 0,
 			  "can't find startxref");
 	}
-	read_xref(xref_offset);
+        try
+        {
+            read_xref(xref_offset);
+        }
+        catch (QPDFExc&)
+        {
+            throw;
+        }
+        catch (std::exception& e)
+        {
+	    throw QPDFExc(qpdf_e_damaged_pdf, this->m->file->getName(), "", 0,
+			  std::string("error reading xref: ") + e.what());
+
+        }
     }
     catch (QPDFExc& e)
     {
@@ -1250,6 +1263,14 @@ QPDF::processXRefStream(qpdf_offset_t xref_offset, QPDFObjectHandle&amp; xref_obj)
 	    // This is needed by checkLinearization()
 	    this->m->first_xref_item_offset = xref_offset;
 	}
+        if (fields[0] == 0)
+        {
+            // Ignore fields[2], which we don't care about in this
+            // case. This works around the issue of some PDF files
+            // that put invalid values, like -1, here for deleted
+            // objects.
+            fields[2] = 0;
+        }
 	insertXrefEntry(obj, toI(fields[0]),
                         fields[1], toI(fields[2]));
     }
@@ -1496,7 +1496,7 @@ $td-&gt;runtest(&quot;detect loops in pages structure&quot;,
 show_ntests();
 # ----------
 $td->notify("--- Xref ---");
-$n_tests += 5;
+$n_tests += 6;
  
 # Handle file with invalid xref table and object 0 as a regular object
 # (bug 3159950).
@@ -1538,6 +1538,12 @@ $td-&gt;runtest(&quot;show number of pages&quot;,
              {$td->STRING => "20\n", $td->EXIT_STATUS => 0},
              $td->NORMALIZE_NEWLINES);
  
+# Issue 482 -- don't range check fields[2] for xref entry type 0.
+$td->runtest("out of range in deleted object",
+             {$td->COMMAND => "qpdf --check xref-range.pdf"},
+             {$td->FILE => "xref-range.out", $td->EXIT_STATUS => 0},
+             $td->NORMALIZE_NEWLINES);
+
 show_ntests();
 # ----------
 $td->notify("--- Overwrite self ---");
 WARNING: issue-150.pdf: can't find PDF header
-overflow/underflow converting 9900000000000000000 to 64-bit integer
+WARNING: issue-150.pdf: file is damaged
+WARNING: issue-150.pdf: error reading xref: overflow/underflow converting 9900000000000000000 to 64-bit integer
+WARNING: issue-150.pdf: Attempting to reconstruct cross-reference table
+issue-150.pdf: unable to find trailer dictionary while recovering damaged file
+checking xref-range.pdf
+PDF Version: 1.5
+File is not encrypted
+File is not linearized
+No syntax or stream encoding errors found; the file may still contain
+errors that qpdf cannot detect