tdf#118047: RTF import: better deduplication for header/footer

Here are two related fixes in one patch:

1. By default current style index is 0, not -1. Due to this we
can assume usage of default "Normal" style during deduplication.
Seems there is no difference in case of no style mentioning in
run or mentioning "Normal" one (\s0), so value -1 looks
unnesesary.

2. During header/footer "substreams" processing StyleTableEntires
was not reused in new domain mapper impl instance. So deduplication
did not work well.

2a. To avoid copying of whole StyleTableEntries between instances
it is stored as pointer in domain mapper impl.

Change-Id: Id003d35554e9b43186238920d9a6373452095121
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131899
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
(cherry picked from commit 79b5ff92b17fde902c2f9db3e806f1c77a387dff)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131862
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>

5 files changed, 47 insertions, 18 deletions

diff --git a/sw/qa/extras/rtfexport/data/tdf118047.rtf b/sw/qa/extras/rtfexport/data/tdf118047.rtf
new file mode 100644
index 000000000000..258d99de9bef
--- /dev/null
+++ b/sw/qa/extras/rtfexport/data/tdf118047.rtf
@@ -0,0 +1,14 @@
+{\rtf1\ansi

+

+{\stylesheet

+{\fs72\sa3200 Normal;}

+}

+

+\sectd 

+{

+\headerr 

+Header\par

+}

+

+Text\par

+}

diff --git a/sw/qa/extras/rtfexport/rtfexport5.cxx b/sw/qa/extras/rtfexport/rtfexport5.cxx
index b9a7321d724b..36bf6c839b65 100644
--- a/sw/qa/extras/rtfexport/rtfexport5.cxx
+++ b/sw/qa/extras/rtfexport/rtfexport5.cxx
@@ -1324,6 +1324,21 @@ DECLARE_RTFEXPORT_TEST(testTdf131234, "tdf131234.rtf")
     CPPUNIT_ASSERT_EQUAL(awt::FontSlant_NONE, getProperty<awt::FontSlant>(xRun, "CharPosture"));
 }
 
+DECLARE_RTFEXPORT_TEST(testTdf118047, "tdf118047.rtf")
+{
+    uno::Reference<text::XTextRange> xPara = getParagraph(1);
+
+    // Ensure that default "Normal" style properties are not applied to text:
+    // text remains with fontsize 12pt and no huge margin below
+    CPPUNIT_ASSERT_EQUAL(12.f, getProperty<float>(getRun(xPara, 1), "CharHeight"));
+    CPPUNIT_ASSERT_EQUAL(sal_Int32(0), getProperty<sal_Int32>(getParagraph(1), "ParaBottomMargin"));
+
+    // Same for header, it should not derive props from "Normal" style
+    CPPUNIT_ASSERT_EQUAL(OUString("Header"), parseDump("/root/page[1]/header/txt/text()"));
+    sal_Int32 nHeight = parseDump("/root/page[1]/header/infos/bounds", "height").toInt32();
+    CPPUNIT_ASSERT_MESSAGE("Header is too large", 1000 > nHeight);
+}
+
 CPPUNIT_PLUGIN_IMPLEMENT();
 
 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/writerfilter/source/rtftok/rtfdispatchflag.cxx b/writerfilter/source/rtftok/rtfdispatchflag.cxx
index 2acb4d59f977..83c5a0966259 100644
--- a/writerfilter/source/rtftok/rtfdispatchflag.cxx
+++ b/writerfilter/source/rtftok/rtfdispatchflag.cxx
@@ -520,12 +520,8 @@ RTFError RTFDocumentImpl::dispatchFlag(RTFKeyword nKeyword)
                 {
                     m_aStates.top().getParagraphSprms().set(NS_ooxml::LN_CT_PPrBase_pStyle,
                                                             new RTFValue(aName));
-                    m_aStates.top().setCurrentStyleIndex(0);
-                }
-                else
-                {
-                    m_aStates.top().setCurrentStyleIndex(-1);
                 }
+                m_aStates.top().setCurrentStyleIndex(0);
             }
             // Need to send paragraph properties again, if there will be any.
             m_bNeedPap = true;
diff --git a/writerfilter/source/rtftok/rtfdocumentimpl.cxx b/writerfilter/source/rtftok/rtfdocumentimpl.cxx
index fb331f4b265a..072430ebe15f 100644
--- a/writerfilter/source/rtftok/rtfdocumentimpl.cxx
+++ b/writerfilter/source/rtftok/rtfdocumentimpl.cxx
@@ -324,6 +324,8 @@ RTFDocumentImpl::RTFDocumentImpl(uno::Reference<uno::XComponentContext> const& x
 
     m_pTokenizer = new RTFTokenizer(*this, m_pInStream.get(), m_xStatusIndicator);
     m_pSdrImport = new RTFSdrImport(*this, m_xDstDoc);
+
+    m_pStyleTableEntries = std::make_shared<RTFReferenceTable::Entries_t>();
 }
 
 RTFDocumentImpl::~RTFDocumentImpl() = default;
@@ -363,6 +365,7 @@ void RTFDocumentImpl::resolveSubstream(std::size_t nPos, Id nId, OUString const&
         m_aAuthorInitials.clear();
     }
     pImpl->m_nDefaultFontIndex = m_nDefaultFontIndex;
+    pImpl->m_pStyleTableEntries = m_pStyleTableEntries;
     pImpl->Strm().Seek(nPos);
     SAL_INFO("writerfilter.rtf", "substream start");
     Mapper().substream(nId, pImpl);
@@ -497,16 +500,16 @@ RTFDocumentImpl::getProperties(const RTFSprms& rAttributes, RTFSprms const& rSpr
     int nStyle = 0;
     if (!m_aStates.empty())
         nStyle = m_aStates.top().getCurrentStyleIndex();
-    auto it = m_aStyleTableEntries.find(nStyle);
-    if (it != m_aStyleTableEntries.end())
+    auto it = m_pStyleTableEntries->find(nStyle);
+    if (it != m_pStyleTableEntries->end())
     {
         // cloneAndDeduplicate() wants to know about only a single "style", so
         // let's merge paragraph and character style properties here.
-        auto itChar = m_aStyleTableEntries.end();
+        auto itChar = m_pStyleTableEntries->end();
         if (!m_aStates.empty())
         {
             int nCharStyle = m_aStates.top().getCurrentCharacterStyleIndex();
-            itChar = m_aStyleTableEntries.find(nCharStyle);
+            itChar = m_pStyleTableEntries->find(nCharStyle);
         }
 
         RTFSprms aStyleSprms;
@@ -517,7 +520,7 @@ RTFDocumentImpl::getProperties(const RTFSprms& rAttributes, RTFSprms const& rSpr
         RTFReferenceProperties& rProps = *static_cast<RTFReferenceProperties*>(it->second.get());
         lcl_copyFlatten(rProps, aStyleAttributes, aStyleSprms);
 
-        if (itChar != m_aStyleTableEntries.end())
+        if (itChar != m_pStyleTableEntries->end())
         {
             // Found active character style, then update aStyleSprms/Attributes.
             if (!nStyleType || nStyleType == NS_ooxml::LN_Value_ST_StyleType_character)
@@ -1402,7 +1405,7 @@ void RTFDocumentImpl::text(OUString& rString)
 
                             writerfilter::Reference<Properties>::Pointer_t const pProp(
                                 createStyleProperties());
-                            m_aStyleTableEntries.insert(
+                            m_pStyleTableEntries->insert(
                                 std::make_pair(m_nCurrentStyleIndex, pProp));
                         }
                         else
@@ -2091,7 +2094,7 @@ writerfilter::Reference<Properties>::Pointer_t RTFDocumentImpl::createStylePrope
 /** 2 different representations of the styles are needed:
 
     1) flat content, as read from the input file:
-       stored in m_aStyleTableEntries, used as reference input for
+       stored in m_pStyleTableEntries, used as reference input for
        deduplication both here and for hard formatting in getProperties()
 
     2) real content, with proper override of sprms/attributes where it differs
@@ -2100,7 +2103,7 @@ writerfilter::Reference<Properties>::Pointer_t RTFDocumentImpl::createStylePrope
 RTFReferenceTable::Entries_t RTFDocumentImpl::deduplicateStyleTable()
 {
     RTFReferenceTable::Entries_t ret;
-    for (auto const& it : m_aStyleTableEntries)
+    for (auto const& it : *m_pStyleTableEntries)
     {
         auto pStyle = it.second;
         ret[it.first] = pStyle;
@@ -2116,8 +2119,8 @@ RTFReferenceTable::Entries_t RTFDocumentImpl::deduplicateStyleTable()
             if (it.first == nBasedOn)
                 continue;
 
-            auto const itParent(m_aStyleTableEntries.find(nBasedOn)); // definition as read!
-            if (itParent != m_aStyleTableEntries.end())
+            auto const itParent(m_pStyleTableEntries->find(nBasedOn)); // definition as read!
+            if (itParent != m_pStyleTableEntries->end())
             {
                 auto const pStyleType(
                     static_cast<RTFReferenceProperties&>(*pStyle).getAttributes().find(
@@ -2143,7 +2146,7 @@ RTFReferenceTable::Entries_t RTFDocumentImpl::deduplicateStyleTable()
             }
         }
     }
-    assert(ret.size() == m_aStyleTableEntries.size());
+    assert(ret.size() == m_pStyleTableEntries->size());
     return ret;
 }
 
@@ -3668,7 +3671,7 @@ RTFParserState::RTFParserState(RTFDocumentImpl* pDocumentImpl)
     , m_nHour(0)
     , m_nMinute(0)
     , m_pCurrentDestinationText(nullptr)
-    , m_nCurrentStyleIndex(-1)
+    , m_nCurrentStyleIndex(0)
     , m_nCurrentCharacterStyleIndex(-1)
     , m_pCurrentBuffer(nullptr)
     , m_bInListpicture(false)
diff --git a/writerfilter/source/rtftok/rtfdocumentimpl.hxx b/writerfilter/source/rtftok/rtfdocumentimpl.hxx
index fc5c8802b5d7..66e27a509be5 100644
--- a/writerfilter/source/rtftok/rtfdocumentimpl.hxx
+++ b/writerfilter/source/rtftok/rtfdocumentimpl.hxx
@@ -933,7 +933,8 @@ private:
     /// Raw default font index, use getFont() on it to get a real one.
     int m_nDefaultFontIndex;
 
-    RTFReferenceTable::Entries_t m_aStyleTableEntries;
+    /// To avoid copying entries between DomainMapper instances it is stored as pointer
+    std::shared_ptr<RTFReferenceTable::Entries_t> m_pStyleTableEntries;
     int m_nCurrentStyleIndex;
     bool m_bFormField;
     /// For the INCLUDEPICTURE field's argument.