regex: handle zero-length matches, fdo#60259 related

Also in backward search ignore all zero-length matches except the text
end single $ anchor search. The anchor search is a valid match, treat it
as such in Writer.

This still doesn't solve the backward $ backward search, the convoluted
Writer code in that place apparently never worked, someone more familiar
with those internals should straighten out the mess.

Change-Id: I2bc83cdcfe6425f9bebf06e23e879bc79b82daa9

2 files changed, 28 insertions, 5 deletions

diff --git a/i18npool/source/search/textsearch.cxx b/i18npool/source/search/textsearch.cxx
index 72e1d60d4507..14ecdcc1059d 100644
--- a/i18npool/source/search/textsearch.cxx
+++ b/i18npool/source/search/textsearch.cxx
@@ -797,17 +797,35 @@ SearchResult TextSearch::RESrchBkwrd( const OUString& searchStr,
     // find the last match
     int nLastPos = 0;
     int nFoundEnd = 0;
+    int nGoodPos = 0, nGoodEnd = 0;
+    bool bFirst = true;
     do {
         nLastPos = pRegexMatcher->start( nIcuErr);
         nFoundEnd = pRegexMatcher->end( nIcuErr);
+        if (nLastPos < nFoundEnd)
+        {
+            // remember last non-zero-length match
+            nGoodPos = nLastPos;
+            nGoodEnd = nFoundEnd;
+        }
         if( nFoundEnd >= startPos)
             break;
+        bFirst = false;
         if( nFoundEnd == nLastPos)
             ++nFoundEnd;
     } while( pRegexMatcher->find( nFoundEnd, nIcuErr));
 
+    // Ignore all zero-length matches except "$" anchor on first match.
+    if (nGoodPos == nGoodEnd)
+    {
+        if (bFirst && nLastPos == startPos)
+            nGoodPos = nLastPos;
+        else
+            return aRet;
+    }
+
     // find last match again to get its details
-    pRegexMatcher->find( nLastPos, nIcuErr);
+    pRegexMatcher->find( nGoodPos, nIcuErr);
 
     // fill in the details of the last match
     const int nGroupCount = pRegexMatcher->groupCount();
diff --git a/sw/source/core/crsr/findtxt.cxx b/sw/source/core/crsr/findtxt.cxx
index b2fbea85104b..e1c9358c0741 100644
--- a/sw/source/core/crsr/findtxt.cxx
+++ b/sw/source/core/crsr/findtxt.cxx
@@ -450,8 +450,9 @@ bool SwPaM::DoSearch( const SearchOptions& rSearchOpt, utl::TextSearch& rSTxt,
     }
 
     xub_StrLen nStringEnd = nEnd;
-    while ( (bSrchForward && nStart < nStringEnd) ||
-            (! bSrchForward && nStart > nStringEnd) )
+    bool bZeroMatch = false;    // zero-length match, i.e. only $ anchor as regex
+    while ( ((bSrchForward && nStart < nStringEnd) ||
+            (! bSrchForward && nStart > nStringEnd)) && !bZeroMatch )
     {
         // SearchAlgorithms_APPROXIMATE works on a per word base so we have to
         // provide the text searcher with the correct locale, because it uses
@@ -479,7 +480,8 @@ bool SwPaM::DoSearch( const SearchOptions& rSearchOpt, utl::TextSearch& rSTxt,
         }
 
         if( nSearchScript == nCurrScript &&
-            (rSTxt.*fnMove->fnSearch)( sCleanStr, &nStart, &nEnd, 0 ))
+                (rSTxt.*fnMove->fnSearch)( sCleanStr, &nStart, &nEnd, 0 ) &&
+                !(bZeroMatch = (nStart == nEnd)))
         {
             // set section correctly
             *GetPoint() = *pPam->GetPoint();
@@ -522,11 +524,14 @@ bool SwPaM::DoSearch( const SearchOptions& rSearchOpt, utl::TextSearch& rSTxt,
 
     if ( bFound )
         return true;
-    else if( ( bChkEmptyPara && !nStart && !nTxtLen ) || bChkParaEnd )
+    else if( ( bChkEmptyPara && !nStart && !nTxtLen ) || (bChkParaEnd && bZeroMatch && nEnd == nTxtLen))
     {
         *GetPoint() = *pPam->GetPoint();
         GetPoint()->nContent = bChkParaEnd ? nTxtLen : 0;
         SetMark();
+        /* FIXME: this condition does not work for !bSrchForward backward
+         * search, it probably never did. (pSttNd != &rNdIdx.GetNode())
+         * is never true in this case. */
         if( (bSrchForward || pSttNd != &rNdIdx.GetNode()) &&
             Move( fnMoveForward, fnGoCntnt ) &&
             (!bSrchForward || pSttNd != &GetPoint()->nNode.GetNode()) &&