Add QPDFObjectHandle::makeDirect(bool allow_streams)

Jay Berkenbilt
1 parent 573b6eb8
Showing 9 changed files with 390 additions and 13 deletions
ChangeLog
TODO
include/qpdf/QPDFObjectHandle.hh
libqpdf/QPDFObjectHandle.cc
qpdf/qpdf.testcov
qpdf/qtest/qpdf.test
qpdf/qtest/qpdf/test4-5.pdf
qpdf/qtest/qpdf/test4-5.qdf
qpdf/test_driver.cc
+2020-12-22  Jay Berkenbilt  <ejb@ql.org>
+
+	* Add QPDFObjectHandle::makeDirect(bool allow_streams) -- if
+	allow_streams is true, preserve indirect references to streams
+	rather than throwing an exception. This allows the object to be
+	made as direct as possible while preserving stream references.
+
 2020-12-20  Jay Berkenbilt  <ejb@ql.org>
  
 	* Add qpdf_register_progress_reporter method to C API,
@@ -155,6 +155,8 @@ ABI Changes
 This is a list of changes to make next time there is an ABI change.
 Comments appear in the code prefixed by "ABI"
  
+* Merge two versions of QPDFObjectHandle::makeDirect per comment
+
 Page splitting/merging
 ======================
  
@@ -721,8 +721,20 @@ class QPDFObjectHandle
  
     // Mutator methods.  Use with caution.
  
-    // Recursively copy this object, making it direct.  Throws an
-    // exception if a loop is detected or any sub-object is a stream.
+    // Recursively copy this object, making it direct. An exception is
+    // thrown if a loop is detected. With allow_streams true, keep
+    // indirect object references to streams. Otherwise, throw an
+    // exception if any sub-object is a stream. Note that, when
+    // allow_streams is true and a stream is found, the resulting
+    // object is still associated with the containing qpdf. When
+    // allow_streams is false, the object will no longer be connected
+    // to the original QPDF object after this call completes
+    // successfully.
+    QPDF_DLL
+    void makeDirect(bool allow_streams);
+    // Zero-arg version is equivalent to makeDirect(false).
+    // ABI: delete zero-arg version of makeDirect, and make
+    // allow_streams default to false.
     QPDF_DLL
     void makeDirect();
  
@@ -1121,7 +1133,7 @@ class QPDFObjectHandle
     void assertType(char const* type_name, bool istype);
     void dereference();
     void copyObject(std::set<QPDFObjGen>& visited, bool cross_indirect,
-                    bool first_level_only);
+                    bool first_level_only, bool stop_at_streams);
     void shallowCopyInternal(QPDFObjectHandle& oh, bool first_level_only);
     void releaseResolved();
     static void setObjectDescriptionFromInput(
@@ -2605,18 +2605,24 @@ QPDFObjectHandle::shallowCopyInternal(QPDFObjectHandle&amp; new_obj,
     }
  
     std::set<QPDFObjGen> visited;
-    new_obj.copyObject(visited, false, first_level_only);
+    new_obj.copyObject(visited, false, first_level_only, false);
 }
  
 void
 QPDFObjectHandle::copyObject(std::set<QPDFObjGen>& visited,
-                             bool cross_indirect, bool first_level_only)
+                             bool cross_indirect, bool first_level_only,
+                             bool stop_at_streams)
 {
     assertInitialized();
  
     if (isStream())
     {
-	QTC::TC("qpdf", "QPDFObjectHandle ERR clone stream");
+	QTC::TC("qpdf", "QPDFObjectHandle copy stream",
+                stop_at_streams ? 0 : 1);
+        if (stop_at_streams)
+        {
+            return;
+        }
 	throw std::runtime_error(
 	    "attempt to make a stream into a direct object");
     }
@@ -2690,7 +2696,8 @@ QPDFObjectHandle::copyObject(std::set&lt;QPDFObjGen&gt;&amp; visited,
                 (cross_indirect || (! items.back().isIndirect())))
             {
                 items.back().copyObject(
-                    visited, cross_indirect, first_level_only);
+                    visited, cross_indirect,
+                    first_level_only, stop_at_streams);
             }
 	}
 	new_obj = new QPDF_Array(items);
@@ -2708,7 +2715,8 @@ QPDFObjectHandle::copyObject(std::set&lt;QPDFObjGen&gt;&amp; visited,
                 (cross_indirect || (! items[*iter].isIndirect())))
             {
                 items[*iter].copyObject(
-                    visited, cross_indirect, first_level_only);
+                    visited, cross_indirect,
+                    first_level_only, stop_at_streams);
             }
 	}
 	new_obj = new QPDF_Dictionary(items);
@@ -2730,8 +2738,14 @@ QPDFObjectHandle::copyObject(std::set&lt;QPDFObjGen&gt;&amp; visited,
 void
 QPDFObjectHandle::makeDirect()
 {
+    makeDirect(false);
+}
+
+void
+QPDFObjectHandle::makeDirect(bool allow_streams)
+{
     std::set<QPDFObjGen> visited;
-    copyObject(visited, true, false);
+    copyObject(visited, true, false, allow_streams);
 }
  
 void
@@ -79,7 +79,7 @@ QPDFObjectHandle clone string 0
 QPDFObjectHandle clone array 0
 QPDFObjectHandle clone dictionary 0
 QPDFObjectHandle makeDirect loop 0
-QPDFObjectHandle ERR clone stream 0
+QPDFObjectHandle copy stream 1
 QPDF default for xref stream field 0 0
 QPDF prev key in xref stream dictionary 0
 QPDF prev key in trailer dictionary 0
@@ -2963,7 +2963,7 @@ $td-&gt;runtest(&quot;check output&quot;,
 show_ntests();
 # ----------
 $td->notify("--- Mutability Tests ---");
-$n_tests += 4;
+$n_tests += 5;
  
 $td->runtest("no normalization",
 	     {$td->COMMAND => "test_driver 4 test4-1.pdf"},
@@ -2975,13 +2975,18 @@ $td-&gt;runtest(&quot;object ordering&quot;,
 	     {$td->FILE => "test4-4.qdf",
 	      $td->EXIT_STATUS => 0});
  
-$td->runtest("loop detected",
+$td->runtest("make direct with allow_streams",
+	     {$td->COMMAND => "test_driver 4 test4-5.pdf"},
+	     {$td->FILE => "test4-5.qdf",
+	      $td->EXIT_STATUS => 0});
+
+$td->runtest("stream detected",
 	     {$td->COMMAND => "test_driver 4 test4-2.pdf"},
 	     {$td->FILE => "test4-2.out",
 	      $td->EXIT_STATUS => 2},
 	     $td->NORMALIZE_NEWLINES);
  
-$td->runtest("stream detected",
+$td->runtest("loop detected",
 	     {$td->COMMAND => "test_driver 4 test4-3.pdf"},
 	     {$td->FILE => "test4-3.out",
 	      $td->EXIT_STATUS => 2},
+%PDF-1.3
+%¿÷¢þ
+%QDF-1.0
+
+1 0 obj
+<<
+  /Pages 3 0 R
+  /Type /Catalog
+>>
+endobj
+
+2 0 obj
+<<
+  /A 4 0 R
+  /B 5 0 R
+  /Subject (Subject)
+  /Title (Some Title Is Here)
+>>
+endobj
+
+3 0 obj
+<<
+  /Count 1
+  /Kids [
+    6 0 R
+  ]
+  /Type /Pages
+>>
+endobj
+
+4 0 obj
+[
+  100
+  2
+  3
+]
+endobj
+
+5 0 obj
+<<
+  /A 4 0 R
+  /B (B)
+>>
+endobj
+
+%% Page 1
+6 0 obj
+<<
+  /Contents 7 0 R
+  /MediaBox [
+    0
+    0
+    612
+    792
+  ]
+  /Parent 3 0 R
+  /Resources <<
+    /Font <<
+      /F1 9 0 R
+    >>
+    /ProcSet 10 0 R
+  >>
+  /Type /Page
+>>
+endobj
+
+%% Contents for page 1
+7 0 obj
+<<
+  /Length 8 0 R
+>>
+stream
+BT
+  /F1 24 Tf
+  72 720 Td
+  (Potato) Tj
+ET
+endstream
+endobj
+
+8 0 obj
+44
+endobj
+
+9 0 obj
+<<
+  /BaseFont /Helvetica
+  /Encoding /WinAnsiEncoding
+  /Name /F1
+  /Subtype /Type1
+  /Type /Font
+>>
+endobj
+
+10 0 obj
+[
+  /PDF
+  /Text
+]
+endobj
+
+11 0 obj
+<<
+  /A 12 0 R
+  /C (potato)
+>>
+endobj
+
+12 0 obj
+<< /B 13 0 R >>
+endobj
+
+13 0 obj
+<<
+  /Length 14 0 R
+>>
+stream
+salad
+endstream
+endobj
+
+14 0 obj
+6
+endobj
+
+xref
+0 15
+0000000000 65535 f 
+0000000025 00000 n 
+0000000079 00000 n 
+0000000174 00000 n 
+0000000246 00000 n 
+0000000280 00000 n 
+0000000332 00000 n 
+0000000548 00000 n 
+0000000647 00000 n 
+0000000666 00000 n 
+0000000784 00000 n 
+0000000820 00000 n 
+0000000869 00000 n 
+0000000902 00000 n 
+0000000965 00000 n 
+trailer <<
+  /QTest 2 0 R
+  /QTest2 11 0 R
+  /Root 1 0 R
+  /Size 15
+  /ID [<c61bd35bada064f61e0a56aa9588064e><c893e7330be149468080ad6518819868>]
+>>
+startxref
+984
+%%EOF
+%PDF-1.3
+%¿÷¢þ
+%QDF-1.0
+
+%% Original object ID: 1 0
+1 0 obj
+<<
+  /Pages 6 0 R
+  /Type /Catalog
+>>
+endobj
+
+%% Original object ID: 15 0
+2 0 obj
+<<
+  /A [
+    14
+    15
+    9
+  ]
+  /Author (Mr. Potato Head)
+  /B <<
+    /A [
+      100
+      2
+      3
+    ]
+    /B (B)
+  >>
+  /Title (Some Title Is Here)
+>>
+endobj
+
+%% Original object ID: 2 0
+3 0 obj
+<<
+  /A 7 0 R
+  /B 8 0 R
+  /Subject (Subject)
+  /Title (Some Title Is Here)
+>>
+endobj
+
+%% Original object ID: 13 0
+4 0 obj
+<<
+  /Length 5 0 R
+>>
+stream
+salad
+endstream
+endobj
+
+5 0 obj
+6
+endobj
+
+%% Original object ID: 3 0
+6 0 obj
+<<
+  /Count 1
+  /Kids [
+    9 0 R
+  ]
+  /Type /Pages
+>>
+endobj
+
+%% Original object ID: 4 0
+7 0 obj
+[
+  100
+  2
+  3
+]
+endobj
+
+%% Original object ID: 5 0
+8 0 obj
+<<
+  /A 7 0 R
+  /B (B)
+>>
+endobj
+
+%% Page 1
+%% Original object ID: 6 0
+9 0 obj
+<<
+  /Contents 10 0 R
+  /MediaBox [
+    0
+    0
+    612
+    792
+  ]
+  /Parent 6 0 R
+  /Resources <<
+    /Font <<
+      /F1 12 0 R
+    >>
+    /ProcSet 13 0 R
+  >>
+  /Type /Page
+>>
+endobj
+
+%% Contents for page 1
+%% Original object ID: 7 0
+10 0 obj
+<<
+  /Length 11 0 R
+>>
+stream
+BT
+  /F1 24 Tf
+  72 720 Td
+  (Potato) Tj
+ET
+endstream
+endobj
+
+11 0 obj
+44
+endobj
+
+%% Original object ID: 9 0
+12 0 obj
+<<
+  /BaseFont /Helvetica
+  /Encoding /WinAnsiEncoding
+  /Name /F1
+  /Subtype /Type1
+  /Type /Font
+>>
+endobj
+
+%% Original object ID: 10 0
+13 0 obj
+[
+  /PDF
+  /Text
+]
+endobj
+
+xref
+0 14
+0000000000 65535 f 
+0000000052 00000 n 
+0000000134 00000 n 
+0000000337 00000 n 
+0000000460 00000 n 
+0000000521 00000 n 
+0000000566 00000 n 
+0000000665 00000 n 
+0000000726 00000 n 
+0000000805 00000 n 
+0000001050 00000 n 
+0000001151 00000 n 
+0000001198 00000 n 
+0000001345 00000 n 
+trailer <<
+  /Info 2 0 R
+  /QTest 3 0 R
+  /QTest2 <<
+    /A <<
+      /B 4 0 R
+    >>
+    /C (potato)
+  >>
+  /Root 1 0 R
+  /Size 14
+  /ID [<c61bd35bada064f61e0a56aa9588064e><31415926535897932384626433832795>]
+>>
+startxref
+1381
+%%EOF
@@ -485,6 +485,14 @@ void runtest(int n, char const* filename1, char const* arg2)
             A.setArrayFromVector(items);
         }
  
+        QPDFObjectHandle qtest2 = trailer.getKey("/QTest2");
+        if (! qtest2.isNull())
+        {
+            // Test allow_streams=true
+            qtest2.makeDirect(true);
+            trailer.replaceKey("/QTest2", qtest2);
+        }
+
 	trailer.replaceKey("/Info", pdf.makeIndirectObject(qtest));
 	QPDFWriter w(pdf, 0);
 	w.setQDFMode(true);