Look in form XObjects when removing unreferenced resources (fixes #373)

If a page contains a form XObject, also filter the form XObject and
remove its unreferenced resources.

ChangeLog

@@ -1,5 +1,9 @@
 2020-03-31  Jay Berkenbilt  <ejb@ql.org>
 
+	* When detecting unreferenced images during page splitting, if any
+	XObjects are form XObjects, recursively descend into them and
+	remove any unreferenced objects from them too. Fixes #373.
+
 	* Add QPDFObjectHandle::filterAsContents, which filters a stream's
 	data as if it were page contents. This can be useful to filter
 	form XObjects the same way we would filter page contents.

libqpdf/QPDFPageObjectHelper.cc

@@ -568,6 +568,29 @@ QPDFPageObjectHelper::removeUnreferencedResourcesHelper(
             {
                 dict.removeKey(*k_iter);
             }
+            QPDFObjectHandle resource = dict.getKey(*k_iter);
+            if (resource.isStream() &&
+                resource.getDict().getKey("/Type").isName() &&
+                ("/XObject" == resource.getDict().getKey("/Type").getName()) &&
+                resource.getDict().getKey("/Subtype").isName() &&
+                ("/Form" == resource.getDict().getKey("/Subtype").getName()))
+            {
+                QTC::TC("qpdf", "QPDFPageObjectHelper filter form xobject");
+                removeUnreferencedResourcesHelper(
+                    resource.getDict(), seen,
+                    [&resource]() {
+                        auto result = resource.getDict().getKey("/Resources");
+                        if (result.isDictionary())
+                        {
+                            result = result.shallowCopy();
+                            resource.getDict().replaceKey("/Resources", result);
+                        }
+                        return result;
+                    },
+                    [&resource](QPDFObjectHandle::TokenFilter* f) {
+                        resource.filterAsContents(f);
+                    });
+            }
         }
     }
 }

qpdf/qpdf.testcov

@@ -449,3 +449,4 @@ QPDFObjectHandle duplicate dict key 0
 QPDFWriter no encryption sig contents 0
 QPDFPageObjectHelper colorspace lookup 0
 QPDFWriter ignore XRef in qdf mode 0
+QPDFPageObjectHelper filter form xobject 0

qpdf/qtest/qpdf.test

@@ -1686,7 +1686,8 @@ my @sp_cases = (
     [11, 'pdf extension', '', 'split-out.Pdf'],
     [4, 'fallback', '--pages 11-pages.pdf 1-3 minimal.pdf --', 'split-out'],
     );
-$n_tests += 23;
+$n_tests += 32;
+$n_compare_pdfs += 1;
 for (@sp_cases)
 {
     $n_tests += 1 + $_->[0];
@@ -1801,6 +1802,25 @@ $td->runtest("check output",
              {$td->FILE => "split-out-bad-token-1-2.pdf"},
              {$td->FILE => "coalesce-split-1-2.pdf"});
 
+$td->runtest("shared images in form xobject",
+             {$td->COMMAND => "qpdf --qdf --static-id --split-pages".
+                  " shared-form-images.pdf split-out-shared-form.pdf"},
+             {$td->STRING => "", $td->EXIT_STATUS => 0});
+foreach my $i (qw(1 2 3 4 5 6))
+{
+    $td->runtest("check output ($i)",
+                 {$td->FILE => "split-out-shared-form-$i.pdf"},
+                 {$td->FILE => "shared-form-split-$i.pdf"});
+}
+$td->runtest("merge for compare",
+             {$td->COMMAND => "qpdf --static-id --empty --pages" .
+                  " split-out-shared-form*.pdf -- a.pdf"},
+             {$td->STRING => "", $td->EXIT_STATUS => 0});
+$td->runtest("check output",
+             {$td->FILE => "a.pdf"},
+             {$td->FILE => "shared-form-images-merged.pdf"});
+compare_pdfs("shared-form-images.pdf", "a.pdf");
+
 show_ntests();
 # ----------
 $td->notify("--- Keep Files Open ---");