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)

Change-Id: Ia6195d80670631669c252d572242874b13642b74
Reviewed-on: https://gerrit.libreoffice.org/48381
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

3 files changed, 23 insertions, 12 deletions

diff --git a/sw/source/filter/html/htmltab.cxx b/sw/source/filter/html/htmltab.cxx
index b6e1d5db18bf..b3f7ca1e663f 100644
--- a/sw/source/filter/html/htmltab.cxx
+++ b/sw/source/filter/html/htmltab.cxx
@@ -19,6 +19,7 @@
 
 #include <memory>
 #include <hintids.hxx>
+#include <comphelper/flagguard.hxx>
 #include <vcl/svapp.hxx>
 #include <vcl/wrkwin.hxx>
 #include <editeng/boxitem.hxx>
@@ -3307,6 +3308,7 @@ void SwHTMLParser::BuildTableCell( HTMLTable *pCurTable, bool bReadOptions,
     if( !IsParserWorking() && !m_pPendStack )
         return;
 
+    ::comphelper::FlagRestorationGuard g(m_isInTableStructure, false);
     CellSaveStruct* pSaveStruct;
 
     HtmlTokenId nToken = HtmlTokenId::NONE;
@@ -4972,6 +4974,7 @@ HTMLTable *SwHTMLParser::BuildTable( SvxAdjust eParentAdjust,
     if( !IsParserWorking() && !m_pPendStack )
         return nullptr;
 
+    ::comphelper::FlagRestorationGuard g(m_isInTableStructure, true);
     HtmlTokenId nToken = HtmlTokenId::NONE;
     bool bPending = false;
     TableSaveStruct* pSaveStruct;
diff --git a/sw/source/filter/html/swhtml.cxx b/sw/source/filter/html/swhtml.cxx
index 3b18ef33328a..e483f926ee04 100644
--- a/sw/source/filter/html/swhtml.cxx
+++ b/sw/source/filter/html/swhtml.cxx
@@ -298,6 +298,7 @@ SwHTMLParser::SwHTMLParser( SwDoc* pD, SwPaM& rCursor, SvStream& rIn,
     m_bRemoveHidden( false ),
     m_bBodySeen( false ),
     m_bReadingHeaderOrFooter( false ),
+    m_isInTableStructure(false),
     m_pTempViewFrame(nullptr)
 {
     m_nEventId = nullptr;
@@ -1544,26 +1545,32 @@ void SwHTMLParser::NextToken( HtmlTokenId nToken )
     // divisions
     case HtmlTokenId::DIVISION_ON:
     case HtmlTokenId::CENTER_ON:
-        if( m_nOpenParaToken != HtmlTokenId::NONE )
+        if (!m_isInTableStructure)
         {
-            if( IsReadPRE() )
-                m_nOpenParaToken = HtmlTokenId::NONE;
-            else
-                EndPara();
+            if (m_nOpenParaToken != HtmlTokenId::NONE)
+            {
+                if (IsReadPRE())
+                    m_nOpenParaToken = HtmlTokenId::NONE;
+                else
+                    EndPara();
+            }
+            NewDivision( nToken );
         }
-        NewDivision( nToken );
         break;
 
     case HtmlTokenId::DIVISION_OFF:
     case HtmlTokenId::CENTER_OFF:
-        if( m_nOpenParaToken != HtmlTokenId::NONE )
+        if (!m_isInTableStructure)
         {
-            if( IsReadPRE() )
-                m_nOpenParaToken = HtmlTokenId::NONE;
-            else
-                EndPara();
+            if (m_nOpenParaToken != HtmlTokenId::NONE)
+            {
+                if (IsReadPRE())
+                    m_nOpenParaToken = HtmlTokenId::NONE;
+                else
+                    EndPara();
+            }
+            EndDivision();
         }
-        EndDivision();
         break;
 
     case HtmlTokenId::MULTICOL_ON:
diff --git a/sw/source/filter/html/swhtml.hxx b/sw/source/filter/html/swhtml.hxx
index 450f7ddbde56..6e84380cdfe1 100644
--- a/sw/source/filter/html/swhtml.hxx
+++ b/sw/source/filter/html/swhtml.hxx
@@ -480,6 +480,7 @@ class SwHTMLParser : public SfxHTMLParser, public SwClient
 
     bool m_bBodySeen : 1;
     bool m_bReadingHeaderOrFooter : 1;
+    bool m_isInTableStructure;
 
     /// the names corresponding to the DOCINFO field subtypes INFO[1-4]
     OUString m_InfoNames[4];