Accept extraneous space after stream keyword (fixes #329)

ChangeLog

@@ -1,5 +1,8 @@
 2019-08-19  Jay Berkenbilt  <ejb@ql.org>
 
+	* Accept (and warn for) extraneous whitespace between the stream
+	keyword and newline. Fixes #329.
+
 	* Properly handle name tokens containing # not preceding two
 	hexadecimal digits. Such names are invalid in PDF >= 1.2 but valid
 	in PDF 1.0 and 1.1. Prior to this fix, qpdf's behavior was to

libqpdf/QPDF.cc

@@ -1477,8 +1477,13 @@ QPDF::readObject(PointerHolder<InputSource> input,
             // stream data.  However, some readers, including
             // Adobe reader, accept a carriage return by itself
             // when followed by a non-newline character, so that's
-            // what we do here.
+            // what we do here. We have also seen files that have
+            // extraneous whitespace between the stream keyword and
+            // the newline.
+            bool done = false;
+            while (! done)
             {
+                done = true;
                 char ch;
                 if (input->read(&ch, 1) == 0)
                 {
@@ -1519,14 +1524,21 @@ QPDF::readObject(PointerHolder<InputSource> input,
                         }
                     }
                 }
+                else if (QUtil::is_space(ch))
+                {
+                    warn(QPDFExc(
+                             qpdf_e_damaged_pdf,
+                             input->getName(),
+                             this->m->last_object_description,
+                             input->tell(),
+                             "stream keyword followed by"
+                             " extraneous whitespace"));
+                    done = false;
+                }
                 else
                 {
                     QTC::TC("qpdf", "QPDF stream without newline");
-                    if (! QUtil::is_space(ch))
+                    input->unreadCh(ch);
-                    {
-                        QTC::TC("qpdf", "QPDF stream with non-space");
-                        input->unreadCh(ch);
-                    }
                     warn(QPDFExc(qpdf_e_damaged_pdf, input->getName(),
                                  this->m->last_object_description,
                                  input->tell(),

qpdf/qpdf.testcov

@@ -275,7 +275,6 @@ QPDFObjectHandle no val for last key 0
 QPDF resolve failure to null 0
 QPDFWriter preserve unreferenced standard 0
 QPDFObjectHandle errors in parsecontent 0
-QPDF stream with non-space 0
 qpdf same file error 0
 qpdf read args from stdin 0
 qpdf read args from file 0

qpdf/qtest/qpdf/stream-line-enders.out

@@ -1,4 +1,6 @@
-WARNING: stream-line-enders.pdf (object 5 0, offset 384): stream keyword followed by carriage return only
+WARNING: stream-line-enders.pdf (object 5 0, offset 391): stream keyword followed by carriage return only
-WARNING: stream-line-enders.pdf (object 6 0, offset 443): stream keyword not followed by proper line terminator
+WARNING: stream-line-enders.pdf (object 6 0, offset 450): stream keyword followed by extraneous whitespace
-WARNING: stream-line-enders.pdf (object 7 0, offset 503): stream keyword not followed by proper line terminator
+WARNING: stream-line-enders.pdf (object 6 0, offset 450): stream keyword not followed by proper line terminator
+WARNING: stream-line-enders.pdf (object 7 0, offset 509): stream keyword not followed by proper line terminator
+WARNING: stream-line-enders.pdf (object 8 0, offset 567): stream keyword followed by extraneous whitespace
 qpdf: operation succeeded with warnings; resulting file may have some problems

qpdf/qtest/qpdf/stream-line-enders.pdf

@@ -7,7 +7,7 @@ endobj
 << /Count 1 /Kids [ 3 0 R ] /Type /Pages >>
 endobj
 3 0 obj
-<< /Contents [ 4 0 R 5 0 R 6 0 R 7 0 R ] /MediaBox [ 0 0 612 792 ] /Parent 2 0 R /Resources << /Font << /F1 8 0 R >> /ProcSet 9 0 R >> /Type /Page >>
+<< /Contents [ 4 0 R 5 0 R 6 0 R 7 0 R 8 0 R ] /MediaBox [ 0 0 612 792 ] /Parent 2 0 R /Resources << /Font << /F1 9 0 R >> /ProcSet 10 0 R >> /Type /Page >>
 endobj
 4 0 obj
 << /Length 14 >>
@@ -27,30 +27,37 @@ stream (Potato) Tj
 endstream
 endobj
 7 0 obj
-<< /Length 11 >>
+<< /Length 9 >>
 stream%comment
-ET
 endstream
 endobj
 8 0 obj
-<< /BaseFont /Helvetica /Encoding /WinAnsiEncoding /Name /F1 /Subtype /Type1 /Type /Font >>
+<< /Length 11 >>
+stream 
+%comment
+ET
+endstream
 endobj
 9 0 obj
+<< /BaseFont /Helvetica /Encoding /WinAnsiEncoding /Name /F1 /Subtype /Type1 /Type /Font >>
+endobj
+10 0 obj
 [ /PDF /Text ]
 endobj
 xref
-0 10
+0 11
 0000000000 65535 f 
 0000000015 00000 n 
 0000000064 00000 n 
 0000000123 00000 n 
-0000000288 00000 n 
+0000000295 00000 n 
-0000000352 00000 n 
+0000000359 00000 n 
-0000000411 00000 n 
+0000000418 00000 n 
-0000000472 00000 n 
+0000000479 00000 n 
-0000000532 00000 n 
+0000000535 00000 n 
-0000000639 00000 n 
+0000000598 00000 n 
-trailer << /Root 1 0 R /Size 10 /ID [<08aa98c73f8a7262d77c8328772c3989><5a35fde138b2e8251b9e079b560c0253>] >>
+0000000705 00000 n 
+trailer << /Root 1 0 R /Size 11 /ID [<08aa98c73f8a7262d77c8328772c3989><5a35fde138b2e8251b9e079b560c0253>] >>
 startxref
-669
+736
 %%EOF

qpdf/qtest/qpdf/stream-line-enders.qdf

@@ -30,6 +30,7 @@ endobj
     6 0 R
     8 0 R
     10 0 R
+    12 0 R
   ]
   /MediaBox [
     0
@@ -40,9 +41,9 @@ endobj
   /Parent 2 0 R
   /Resources <<
     /Font <<
-      /F1 12 0 R
+      /F1 14 0 R
     >>
-    /ProcSet 13 0 R
+    /ProcSet 15 0 R
   >>
   /Type /Page
 >>
@@ -102,17 +103,32 @@ endobj
 >>
 stream
 %comment
+endstream
+endobj
+
+11 0 obj
+9
+endobj
+
+%% Contents for page 1
+%% Original object ID: 8 0
+12 0 obj
+<<
+  /Length 13 0 R
+>>
+stream
+%comment
 ET
 endstream
 endobj
 
 %QDF: ignore_newline
-11 0 obj
+13 0 obj
 11
 endobj
 
-%% Original object ID: 8 0
+%% Original object ID: 9 0
-12 0 obj
+14 0 obj
 <<
   /BaseFont /Helvetica
   /Encoding /WinAnsiEncoding
@@ -122,8 +138,8 @@ endobj
 >>
 endobj
 
-%% Original object ID: 9 0
+%% Original object ID: 10 0
-13 0 obj
+15 0 obj
 [
   /PDF
   /Text
@@ -131,26 +147,28 @@ endobj
 endobj
 
 xref
-0 14
+0 16
 0000000000 65535 f 
 0000000052 00000 n 
 0000000133 00000 n 
 0000000242 00000 n 
-0000000527 00000 n 
+0000000538 00000 n 
-0000000596 00000 n 
+0000000607 00000 n 
-0000000665 00000 n 
+0000000676 00000 n 
-0000000730 00000 n 
+0000000741 00000 n 
-0000000799 00000 n 
+0000000810 00000 n 
-0000000866 00000 n 
+0000000877 00000 n 
-0000000935 00000 n 
+0000000946 00000 n 
-0000001025 00000 n 
+0000001012 00000 n 
-0000001072 00000 n 
+0000001081 00000 n 
+0000001171 00000 n 
 0000001218 00000 n 
+0000001365 00000 n 
 trailer <<
   /Root 1 0 R
-  /Size 14
+  /Size 16
   /ID [<08aa98c73f8a7262d77c8328772c3989><31415926535897932384626433832795>]
 >>
 startxref
-1254
+1401
 %%EOF