Add QPDFObjectHandle::shallowCopy()

9 files changed, 208 insertions, 10 deletions

diff --git a/ChangeLog b/ChangeLog
index 3b164e56..d67a82e6 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,7 @@
 2012-06-21  Jay Berkenbilt  <ejb@ql.org>
 
+	* QPDFObjectHandle: add shallowCopy() method
+
 	* QPDF: add new APIs for adding and removing pages.  This includes
 	addPage(), addPageAt(), and removePage().  Also a method
 	updateAllPagesCache() is now available to force update of the
diff --git a/include/qpdf/QPDFObjectHandle.hh b/include/qpdf/QPDFObjectHandle.hh
index 3a4bff13..e431d8a7 100644
--- a/include/qpdf/QPDFObjectHandle.hh
+++ b/include/qpdf/QPDFObjectHandle.hh
@@ -190,6 +190,13 @@ class QPDFObjectHandle
     QPDF_DLL
     bool isOrHasName(std::string const&);
 
+    // Create a shallow copy of an object as a direct object.  Since
+    // this is a shallow copy, for dictionaries and arrays, any keys
+    // or items that were indirect objects will still be indirect
+    // objects that point to the same place.
+    QPDF_DLL
+    QPDFObjectHandle shallowCopy();
+
     // Mutator methods.  Use with caution.
 
     // Recursively copy this object, making it direct.  Throws an
diff --git a/libqpdf/QPDFObjectHandle.cc b/libqpdf/QPDFObjectHandle.cc
index 53d9414c..4e28c405 100644
--- a/libqpdf/QPDFObjectHandle.cc
+++ b/libqpdf/QPDFObjectHandle.cc
@@ -663,6 +663,38 @@ QPDFObjectHandle::newStream(QPDF* qpdf, PointerHolder<Buffer> data)
     return result;
 }
 
+QPDFObjectHandle
+QPDFObjectHandle::shallowCopy()
+{
+    assertInitialized();
+
+    if (isStream())
+    {
+	QTC::TC("qpdf", "QPDFObjectHandle ERR shallow copy stream");
+	throw std::runtime_error(
+	    "attempt to make a shallow copy of a stream");
+    }
+
+    QPDFObjectHandle new_obj;
+    if (isArray())
+    {
+	QTC::TC("qpdf", "QPDFObjectHandle shallow copy array");
+	new_obj = newArray(getArrayAsVector());
+    }
+    else if (isDictionary())
+    {
+	QTC::TC("qpdf", "QPDFObjectHandle shallow copy dictionary");
+        new_obj = newDictionary(getDictAsMap());
+    }
+    else
+    {
+	QTC::TC("qpdf", "QPDFObjectHandle shallow copy scalar");
+        new_obj = *this;
+    }
+
+    return new_obj;
+}
+
 void
 QPDFObjectHandle::makeDirectInternal(std::set<int>& visited)
 {
diff --git a/qpdf/qpdf.testcov b/qpdf/qpdf.testcov
index c7207bc8..5c40037e 100644
--- a/qpdf/qpdf.testcov
+++ b/qpdf/qpdf.testcov
@@ -209,3 +209,7 @@ QPDF insert page 2
 QPDF updateAllPagesCache 0
 QPDF insert non-indirect page 0
 QPDF insert indirect page 0
+QPDFObjectHandle ERR shallow copy stream 0
+QPDFObjectHandle shallow copy array 0
+QPDFObjectHandle shallow copy dictionary 0
+QPDFObjectHandle shallow copy scalar 0
diff --git a/qpdf/qtest/qpdf.test b/qpdf/qtest/qpdf.test
index 053fb3b2..a1a76ce3 100644
--- a/qpdf/qtest/qpdf.test
+++ b/qpdf/qtest/qpdf.test
@@ -144,7 +144,7 @@ $td->runtest("duplicate page",
 	     $td->NORMALIZE_NEWLINES);
 # ----------
 $td->notify("--- Miscellaneous Tests ---");
-$n_tests += 37;
+$n_tests += 40;
 
 $td->runtest("qpdf version",
 	     {$td->COMMAND => "qpdf --version"},
@@ -338,6 +338,18 @@ $td->runtest("check output",
 	     {$td->FILE => "c-info-out.pdf"});
 unlink "a.pdf" or die;
 
+$td->runtest("shallow copy an array",
+	     {$td->COMMAND => "test_driver 20 shallow_array.pdf"},
+	     {$td->STRING => "test 20 done\n", $td->EXIT_STATUS => 0},
+	     $td->NORMALIZE_NEWLINES);
+$td->runtest("check output",
+	     {$td->FILE => "a.pdf"},
+	     {$td->FILE => "shallow_array-out.pdf"});
+$td->runtest("shallow copy a stream",
+	     {$td->COMMAND => "test_driver 21 shallow_array.pdf"},
+	     {$td->FILE => "shallow_stream.out", $td->EXIT_STATUS => 2},
+	     $td->NORMALIZE_NEWLINES);
+
 show_ntests();
 # ----------
 $td->notify("--- Error Condition Tests ---");
diff --git a/qpdf/qtest/qpdf/shallow_array-out.pdf b/qpdf/qtest/qpdf/shallow_array-out.pdf
new file mode 100644
index 00000000..31d86ecc
--- /dev/null
+++ b/qpdf/qtest/qpdf/shallow_array-out.pdf
@@ -0,0 +1,40 @@
+%PDF-1.3
+%¿÷¢þ
+1 0 obj
+<< /Pages 2 0 R /Type /Catalog >>
+endobj
+2 0 obj
+<< /Count 1 /Kids [ 3 0 R ] /Type /Pages >>
+endobj
+3 0 obj
+<< /Contents 4 0 R /MediaBox [ 0 0 612 792 ] /Parent 2 0 R /Resources << /Font << /F1 5 0 R >> /ProcSet 6 0 R >> /Type /Page >>
+endobj
+4 0 obj
+<< /Length 44 >>
+stream
+BT
+  /F1 24 Tf
+  72 720 Td
+  (Potato) Tj
+ET
+endstream
+endobj
+5 0 obj
+<< /BaseFont /Helvetica /Encoding /WinAnsiEncoding /Name /F1 /Subtype /Type1 /Type /Font >>
+endobj
+6 0 obj
+[ /PDF /Text ]
+endobj
+xref
+0 7
+0000000000 65535 f 
+0000000015 00000 n 
+0000000064 00000 n 
+0000000123 00000 n 
+0000000266 00000 n 
+0000000359 00000 n 
+0000000466 00000 n 
+trailer << /QTest [ /A 1 0 R /B ] /QTest2 [ /A 1 0 R /B 7 ] /Root 1 0 R /Size 7 /ID [<31415926535897932384626433832795><31415926535897932384626433832795>] >>
+startxref
+496
+%%EOF
diff --git a/qpdf/qtest/qpdf/shallow_array.pdf b/qpdf/qtest/qpdf/shallow_array.pdf
new file mode 100644
index 00000000..63c39b09
--- /dev/null
+++ b/qpdf/qtest/qpdf/shallow_array.pdf
@@ -0,0 +1,80 @@
+%PDF-1.3
+1 0 obj
+<<
+  /Type /Catalog
+  /Pages 2 0 R
+>>
+endobj
+
+2 0 obj
+<<
+  /Type /Pages
+  /Kids [
+    3 0 R
+  ]
+  /Count 1
+>>
+endobj
+
+3 0 obj
+<<
+  /Type /Page
+  /Parent 2 0 R
+  /MediaBox [0 0 612 792]
+  /Contents 4 0 R
+  /Resources <<
+    /ProcSet 5 0 R
+    /Font <<
+      /F1 6 0 R
+    >>
+  >>
+>>
+endobj
+
+4 0 obj
+<<
+  /Length 44
+>>
+stream
+BT
+  /F1 24 Tf
+  72 720 Td
+  (Potato) Tj
+ET
+endstream
+endobj
+
+5 0 obj
+[
+  /PDF
+  /Text
+]
+endobj
+
+6 0 obj
+<<
+  /Type /Font
+  /Subtype /Type1
+  /Name /F1
+  /BaseFont /Helvetica
+  /Encoding /WinAnsiEncoding
+>>
+endobj
+
+xref
+0 7
+0000000000 65535 f 
+0000000009 00000 n 
+0000000063 00000 n 
+0000000135 00000 n 
+0000000307 00000 n 
+0000000403 00000 n 
+0000000438 00000 n 
+trailer <<
+  /Size 7
+  /Root 1 0 R
+  /QTest [ /A 1 0 R /B ]
+>>
+startxref
+556
+%%EOF
diff --git a/qpdf/qtest/qpdf/shallow_stream.out b/qpdf/qtest/qpdf/shallow_stream.out
new file mode 100644
index 00000000..fec1aa83
--- /dev/null
+++ b/qpdf/qtest/qpdf/shallow_stream.out
@@ -0,0 +1 @@
+attempt to make a shallow copy of a stream
diff --git a/qpdf/test_driver.cc b/qpdf/test_driver.cc
index b2ceca5c..d2869317 100644
--- a/qpdf/test_driver.cc
+++ b/qpdf/test_driver.cc
@@ -691,17 +691,15 @@ void runtest(int n, char const* filename)
         // dictionary and modify it.  Using the results of
         // getDictAsMap to create a new dictionary effectively creates
         // a shallow copy.
-        std::map<std::string, QPDFObjectHandle> page_dict_keys =
-            QPDFObjectHandle(pages[0]).getDictAsMap();
+        QPDFObjectHandle page_tempate = pages[0];
         std::vector<QPDFObjectHandle> new_pages;
         for (std::vector<QPDFObjectHandle>::iterator iter = contents.begin();
              iter != contents.end(); ++iter)
         {
             // We will retain indirect object references to other
             // indirect objects other than page content.
-            page_dict_keys["/Contents"] = *iter;
-            QPDFObjectHandle page =
-                QPDFObjectHandle::newDictionary(page_dict_keys);
+            QPDFObjectHandle page = page_tempate.shallowCopy();
+            page.replaceKey("/Contents", *iter);
             if (iter == contents.begin())
             {
                 // leave direct
@@ -745,12 +743,10 @@ void runtest(int n, char const* filename)
         std::vector<QPDFObjectHandle> const& all_pages = pdf.getAllPages();
 
         QPDFObjectHandle contents = createPageContents(pdf, "New page 10");
-        std::map<std::string, QPDFObjectHandle> page_dict_keys =
-            QPDFObjectHandle(all_pages[0]).getDictAsMap();
-        page_dict_keys["/Contents"] = contents;
         QPDFObjectHandle page =
             pdf.makeIndirectObject(
-                QPDFObjectHandle::newDictionary(page_dict_keys));
+                QPDFObjectHandle(all_pages[0]).shallowCopy());
+        page.replaceKey("/Contents", contents);
 
         // Insert the page manually.
 	QPDFObjectHandle root = pdf.getRoot();
@@ -807,6 +803,30 @@ void runtest(int n, char const* filename)
         pdf.addPage(pages[5], false);
         std::cout << "you can't see this" << std::endl;
     }
+    else if (n == 20)
+    {
+        // Shallow copy an array
+	QPDFObjectHandle trailer = pdf.getTrailer();
+	QPDFObjectHandle qtest = trailer.getKey("/QTest");
+        QPDFObjectHandle copy = qtest.shallowCopy();
+        // Append shallow copy of a scalar
+        copy.appendItem(trailer.getKey("/Size").shallowCopy());
+        trailer.replaceKey("/QTest2", copy);
+
+	QPDFWriter w(pdf, "a.pdf");
+	w.setStaticID(true);
+	w.setStreamDataMode(qpdf_s_preserve);
+	w.write();
+    }
+    else if (n == 21)
+    {
+        // Try to shallow copy a stream
+        std::vector<QPDFObjectHandle> const& pages = pdf.getAllPages();
+        QPDFObjectHandle page = pages[0];
+        QPDFObjectHandle contents = page.getKey("/Contents");
+        contents.shallowCopy();
+        std::cout << "you can't see this" << std::endl;
+    }
     else
     {
 	throw std::runtime_error(std::string("invalid test ") +