ofz#5566 sw: HTML import: ignore <DIV> in table structure elements

Looking at the HTML4 DTD https://www.w3.org/TR/html4/sgml/dtd.html,
inside TABLE only various elements defining the structure of the table
allowed, except inside cells (TD and TH elements).

DIV in a table but outside cells may cause cursor positions to go
off the rails, so better ignore such invalid DIV tags.

Reviewed-on: https://gerrit.libreoffice.org/48359
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
(cherry picked from commit 8b1a83bffe35ae0e71735569512c1586bcb37b25)

Reviewed-on: https://gerrit.libreoffice.org/48526
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Eike Rathke <erack@redhat.com>
(cherry picked from commit 7283bdaf9cdc81dbec55c915ecd8c2571478bc4d)

Change-Id: Ia6195d80670631669c252d572242874b13642b74

3 files changed, 25 insertions, 13 deletions

diff --git a/sw/source/filter/html/htmltab.cxx b/sw/source/filter/html/htmltab.cxx
index d5d5736b6cd4..be6ea7d0bcba 100644
--- a/sw/source/filter/html/htmltab.cxx
+++ b/sw/source/filter/html/htmltab.cxx
@@ -18,6 +18,7 @@
  */
 
 #include "hintids.hxx"
+#include <comphelper/flagguard.hxx>
 #include <vcl/svapp.hxx>
 #include <vcl/wrkwin.hxx>
 #include <editeng/boxitem.hxx>
@@ -3398,6 +3399,7 @@ void SwHTMLParser::BuildTableCell( HTMLTable *pCurTable, bool bReadOptions,
     if( !IsParserWorking() && !m_pPendStack )
         return;
 
+    ::comphelper::FlagRestorationGuard g(m_isInTableStructure, false);
     CellSaveStruct* pSaveStruct;
 
     int nToken = 0;
@@ -5120,6 +5122,7 @@ HTMLTable *SwHTMLParser::BuildTable( SvxAdjust eParentAdjust,
     if( !IsParserWorking() && !m_pPendStack )
         return nullptr;
 
+    ::comphelper::FlagRestorationGuard g(m_isInTableStructure, true);
     int nToken = 0;
     bool bPending = false;
     TableSaveStruct* pSaveStruct;
diff --git a/sw/source/filter/html/swhtml.cxx b/sw/source/filter/html/swhtml.cxx
index 058c39ad0b89..fb2a16ed1bff 100644
--- a/sw/source/filter/html/swhtml.cxx
+++ b/sw/source/filter/html/swhtml.cxx
@@ -296,6 +296,7 @@ SwHTMLParser::SwHTMLParser( SwDoc* pD, SwPaM& rCursor, SvStream& rIn,
     m_bInFootEndNoteSymbol( false ),
     m_bIgnoreHTMLComments( bNoHTMLComments ),
     m_bRemoveHidden( false ),
+    m_isInTableStructure(false),
     m_pTempViewFrame(nullptr)
 {
     m_nEventId = nullptr;
@@ -1543,26 +1544,32 @@ void SwHTMLParser::NextToken( int nToken )
     // divisions
     case HTML_DIVISION_ON:
     case HTML_CENTER_ON:
-        if( m_nOpenParaToken )
-        {
-            if( IsReadPRE() )
-                m_nOpenParaToken = 0;
-            else
-                EndPara();
+        if (!m_isInTableStructure)
+            {
+            if( m_nOpenParaToken )
+            {
+                if( IsReadPRE() )
+                    m_nOpenParaToken = 0;
+                else
+                    EndPara();
+            }
+            NewDivision( nToken );
         }
-        NewDivision( nToken );
         break;
 
     case HTML_DIVISION_OFF:
     case HTML_CENTER_OFF:
-        if( m_nOpenParaToken )
+        if (!m_isInTableStructure)
         {
-            if( IsReadPRE() )
-                m_nOpenParaToken = 0;
-            else
-                EndPara();
+            if( m_nOpenParaToken )
+            {
+                if( IsReadPRE() )
+                    m_nOpenParaToken = 0;
+                else
+                    EndPara();
+            }
+            EndDivision( nToken );
         }
-        EndDivision( nToken );
         break;
 
     case HTML_MULTICOL_ON:
diff --git a/sw/source/filter/html/swhtml.hxx b/sw/source/filter/html/swhtml.hxx
index 6ef39ce7dd88..53111f6ef36d 100644
--- a/sw/source/filter/html/swhtml.hxx
+++ b/sw/source/filter/html/swhtml.hxx
@@ -477,6 +477,8 @@ class SwHTMLParser : public SfxHTMLParser, public SwClient
     bool m_bIgnoreHTMLComments : 1;
     bool m_bRemoveHidden : 1; // the filter implementation might set the hidden flag
 
+    bool m_isInTableStructure;
+
     /// the names corresponding to the DOCINFO field subtypes INFO[1-4]
     OUString m_InfoNames[4];