PDF export of PDF images: don't preserve annotations in general

Regression from 33c9bc0225a92f26770f9ef20b252af47853e7b9 (PDF export of
PDF images: preserve hyperlinks, 2022-01-07), the problem was that we
want to preserve hyperlinks, but annotations are added by the PDF export
explicitly, so it isn't a good idea to "preserve" them as well.

Fix the problem by going back to the old behavior, except when the
annotation sub-type is /Link.

This keeps hyperlinks working but doesn't lead to duplicated comments
when re-exporting an image + adding comments explicitly.

Conflicts:
vcl/qa/cppunit/pdfexport/pdfexport.cxx

Change-Id: I910990da59bdc1150cc346f1a5471cb6da55dd2c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131261
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

3 files changed, 44 insertions, 2 deletions

diff --git a/vcl/qa/cppunit/pdfexport/data/pdf-image-annots.odg b/vcl/qa/cppunit/pdfexport/data/pdf-image-annots.odg
new file mode 100644
index 000000000000..6dee0145c536
--- /dev/null
+++ b/vcl/qa/cppunit/pdfexport/data/pdf-image-annots.odg
diff --git a/vcl/qa/cppunit/pdfexport/pdfexport.cxx b/vcl/qa/cppunit/pdfexport/pdfexport.cxx
index 8c1c81162ace..2206bd25d58a 100644
--- a/vcl/qa/cppunit/pdfexport/pdfexport.cxx
+++ b/vcl/qa/cppunit/pdfexport/pdfexport.cxx
@@ -2645,6 +2645,36 @@ CPPUNIT_TEST_FIXTURE(PdfExportTest, testPdfImageHyperlink)
     CPPUNIT_ASSERT_EQUAL(0.0012626264, rtl::math::round(aScale.getY(), 10));
 #endif
 }
+
+CPPUNIT_TEST_FIXTURE(PdfExportTest, testPdfImageAnnots)
+{
+    // Given a document with a PDF image that has 2 comments (popup, text) and a hyperlink:
+    OUString aURL = m_directories.getURLFromSrc(DATA_DIRECTORY) + "pdf-image-annots.odg";
+    mxComponent = loadFromDesktop(aURL);
+    CPPUNIT_ASSERT(mxComponent.is());
+
+    // When saving to PDF:
+    uno::Reference<frame::XStorable> xStorable(mxComponent, uno::UNO_QUERY);
+    utl::MediaDescriptor aMediaDescriptor;
+    aMediaDescriptor["FilterName"] <<= OUString("draw_pdf_Export");
+    xStorable->storeToURL(maTempFile.GetURL(), aMediaDescriptor.getAsConstPropertyValueList());
+
+    // Then make sure only the hyperlink is kept, since Draw itself has its own comments:
+    SvFileStream aFile(maTempFile.GetURL(), StreamMode::READ);
+    maMemory.WriteStream(aFile);
+    std::shared_ptr<vcl::pdf::PDFium> pPDFium = vcl::pdf::PDFiumLibrary::get();
+    std::unique_ptr<vcl::pdf::PDFiumDocument> pPdfDocument
+        = pPDFium->openDocument(maMemory.GetData(), maMemory.GetSize());
+    CPPUNIT_ASSERT(pPdfDocument);
+    CPPUNIT_ASSERT_EQUAL(1, pPdfDocument->getPageCount());
+    std::unique_ptr<vcl::pdf::PDFiumPage> pPdfPage = pPdfDocument->openPage(/*nIndex=*/0);
+    CPPUNIT_ASSERT(pPdfPage);
+    // Without the accompanying fix in place, this test would have failed with:
+    // - Expected: 1
+    // - Actual  : 3
+    // i.e. not only the hyperlink but also the 2 comments were exported, leading to duplication.
+    CPPUNIT_ASSERT_EQUAL(1, pPdfPage->getAnnotationCount());
+}
 }
 
 CPPUNIT_PLUGIN_IMPLEMENT();
diff --git a/vcl/source/gdi/pdfwriter_impl.cxx b/vcl/source/gdi/pdfwriter_impl.cxx
index 6aefaf2edabe..3e934a19825d 100644
--- a/vcl/source/gdi/pdfwriter_impl.cxx
+++ b/vcl/source/gdi/pdfwriter_impl.cxx
@@ -8511,7 +8511,7 @@ void PDFWriterImpl::writeReferenceXObject(ReferenceXObjectEmit& rEmit)
             return;
         }
 
-        // Merge page annotations (links, etc) from pPage to our page.
+        // Merge link annotations from pPage to our page.
         std::vector<filter::PDFObjectElement*> aAnnots;
         if (auto pArray = dynamic_cast<filter::PDFArrayElement*>(pPage->Lookup("Annots")))
         {
@@ -8529,7 +8529,19 @@ void PDFWriterImpl::writeReferenceXObject(ReferenceXObjectEmit& rEmit)
                     continue;
                 }
 
-                // Annotation refers to an object, remember it.
+                auto pType = dynamic_cast<filter::PDFNameElement*>(pObject->Lookup("Type"));
+                if (!pType || pType->GetValue() != "Annot")
+                {
+                    continue;
+                }
+
+                auto pSubtype = dynamic_cast<filter::PDFNameElement*>(pObject->Lookup("Subtype"));
+                if (!pSubtype || pSubtype->GetValue() != "Link")
+                {
+                    continue;
+                }
+
+                // Reference to a link annotation object, remember it.
                 aAnnots.push_back(pObject);
             }
         }