0xFFF9 is a better choice for CH_TXTATR_INWORD than 0x0002.

a) the default properties for the code point make it not split a word it
appears in into two different words in any break mode we have. Which is what we
want from a CH_TXTATR_INWORD

b) unicode TR#20 gives for the interlinear annotation anchor: "What to do if
detected: In a proxy context or browser context, remove U+FFF9", so when we
need to strip it from text to run that text through e.g. the spellchecker or
word counting then there's a solid precedent for stripping it

In addition I *do* want the footnote placeholder to break the word it appears
in, that gives the desired wordcount and cursor travelling behaviour

The BREAKWORD and other *random* selection of CH_TXTATR are still odd choices,
and there's way too many of them.

Change-Id: I930ff8ff806af448829bc1a1ae6cb92053e9a284

11 files changed, 48 insertions, 31 deletions

diff --git a/i18npool/qa/cppunit/test_breakiterator.cxx b/i18npool/qa/cppunit/test_breakiterator.cxx
index 56f117e4aae9..b72deda6ebc3 100644
--- a/i18npool/qa/cppunit/test_breakiterator.cxx
+++ b/i18npool/qa/cppunit/test_breakiterator.cxx
@@ -252,7 +252,7 @@ void TestBreakIterator::testWordBoundaries()
         }
     }
 
-    sal_Unicode aJoinTests[] = { 'X', 0x200C, 0x200D, 0x2060, 0xFEFF, 0xFFF9 };
+    sal_Unicode aJoinTests[] = { 'X', 0x200C, 0x200D, 0x2060, 0xFEFF, 0xFFF9, 0xFFFA, 0xFFFB };
     for (int mode = i18n::WordType::ANY_WORD; mode <= i18n::WordType::WORD_COUNT; ++mode)
     {
         //make sure that in all cases isBeginWord and isEndWord matches getWordBoundary
diff --git a/sw/inc/hintids.hxx b/sw/inc/hintids.hxx
index 9428072fa53a..9ea4bc4ca16f 100644
--- a/sw/inc/hintids.hxx
+++ b/sw/inc/hintids.hxx
@@ -35,7 +35,7 @@
 // For SwTxtHints without end index the following char is added:
 
 #define CH_TXTATR_BREAKWORD     ((sal_Unicode)0x01)
-#define CH_TXTATR_INWORD        ((sal_Unicode)0x02)
+#define CH_TXTATR_INWORD        ((sal_Unicode)0xFFF9)
 #define CH_TXTATR_TAB           ((sal_Unicode)'\t')
 #define CH_TXTATR_NEWLINE       ((sal_Unicode)'\n')
 #define CH_TXT_ATR_FIELDSTART ((sal_Unicode)0x04)
diff --git a/sw/qa/core/swdoc-test.cxx b/sw/qa/core/swdoc-test.cxx
index 49b0930189c2..8c06e252bdc5 100644
--- a/sw/qa/core/swdoc-test.cxx
+++ b/sw/qa/core/swdoc-test.cxx
@@ -60,6 +60,7 @@
 #include "swtypes.hxx"
 #include "fmtftn.hxx"
 #include "fmtrfmrk.hxx"
+#include "fmtfld.hxx"
 
 SO2_DECL_REF(SwDocShell)
 SO2_IMPL_REF(SwDocShell)
@@ -366,6 +367,25 @@ void SwDocTest::testSwScanner()
         pTxtNode->CountWords(aDocStat, 0, pTxtNode->Len());
         CPPUNIT_ASSERT(aDocStat.nWord == 1);
         CPPUNIT_ASSERT_MESSAGE("refmark anchor should not be counted", aDocStat.nChar == 11);
+
+        m_pDoc->AppendTxtNode(*aPaM.GetPoint());
+        m_pDoc->InsertString(aPaM, rtl::OUString("Apple"));
+
+        DateTime aDate(DateTime::SYSTEM);
+        SwPostItField aPostIt(
+            (SwPostItFieldType*)m_pDoc->GetSysFldType(RES_POSTITFLD), rtl::OUString("An Author"),
+            rtl::OUString("Some Text"), rtl::OUString("WhatEver"), aDate );
+        m_pDoc->InsertPoolItem(aPaM, SwFmtFld(aPostIt), 0);
+
+        m_pDoc->InsertString(aPaM, rtl::OUString("Apple"));
+        pTxtNode = aPaM.GetNode()->GetTxtNode();
+        aDocStat.Reset();
+        pTxtNode->CountWords(aDocStat, 0, pTxtNode->Len());
+        CPPUNIT_ASSERT(aDocStat.nWord == 1);
+        CPPUNIT_ASSERT_MESSAGE("postit anchor should effectively not exist", aDocStat.nChar == 10);
+        CPPUNIT_ASSERT(pTxtNode->Len() == 11);
+
+        aDocStat.Reset();
     }
 
     //See https://bugs.freedesktop.org/show_bug.cgi?id=46757
diff --git a/sw/source/core/crsr/crsrsh.cxx b/sw/source/core/crsr/crsrsh.cxx
index 54d0bb9c16b5..31bd6dadb3ac 100644
--- a/sw/source/core/crsr/crsrsh.cxx
+++ b/sw/source/core/crsr/crsrsh.cxx
@@ -3369,9 +3369,9 @@ void SwCrsrShell::GetSmartTagTerm( const Point& rPt, SwRect& rSelectRect,
             Pop(sal_False);
 
             // make sure the selection build later from the data below does not
-            // include footnotes and other "in word" character to the left and
-            // right in order to preserve those. Therefore count those "in
-            // words" in order to modify the selection accordingly.
+            // include "in word" character to the left and right in order to
+            // preserve those. Therefore count those "in words" in order to
+            // modify the selection accordingly.
             const sal_Unicode* pChar = aText.GetBuffer();
             xub_StrLen nLeft = 0;
             while (pChar && *pChar++ == CH_TXTATR_INWORD)
diff --git a/sw/source/core/edit/edlingu.cxx b/sw/source/core/edit/edlingu.cxx
index a5e6f9fb6b92..ffe27d3c3236 100644
--- a/sw/source/core/edit/edlingu.cxx
+++ b/sw/source/core/edit/edlingu.cxx
@@ -1049,11 +1049,10 @@ uno::Reference< XSpellAlternatives >
                 xub_StrLen nLineEnd = GetCrsr()->GetPoint()->nContent.GetIndex();
                 Pop(sal_False);
 
-                // make sure the selection build later from the
-                // data below does not include footnotes and other
-                // "in word" character to the left and right in order
-                // to preserve those. Therefore count those "in words"
-                // in order to modify the selection accordingly.
+                // make sure the selection build later from the data below does
+                // not "in word" character to the left and right in order to
+                // preserve those. Therefore count those "in words" in order to
+                // modify the selection accordingly.
                 const sal_Unicode* pChar = aText.GetBuffer();
                 xub_StrLen nLeft = 0;
                 while (pChar && *pChar++ == CH_TXTATR_INWORD)
@@ -1179,11 +1178,10 @@ bool SwEditShell::GetGrammarCorrection(
                 xub_StrLen nLineEnd = GetCrsr()->GetPoint()->nContent.GetIndex();
                 Pop(sal_False);
 
-                // make sure the selection build later from the
-                // data below does not include footnotes and other
-                // "in word" character to the left and right in order
-                // to preserve those. Therefore count those "in words"
-                // in order to modify the selection accordingly.
+                // make sure the selection build later from the data below does
+                // not include "in word" character to the left and right in
+                // order to preserve those. Therefore count those "in words" in
+                // order to modify the selection accordingly.
                 const sal_Unicode* pChar = aText.GetBuffer();
                 xub_StrLen nLeft = 0;
                 while (pChar && *pChar++ == CH_TXTATR_INWORD)
diff --git a/sw/source/core/txtnode/ndtxt.cxx b/sw/source/core/txtnode/ndtxt.cxx
index 85336a8e42a6..0f24ecf87ae7 100644
--- a/sw/source/core/txtnode/ndtxt.cxx
+++ b/sw/source/core/txtnode/ndtxt.cxx
@@ -1377,7 +1377,7 @@ void lcl_CopyHint( const sal_uInt16 nWhich, const SwTxtAttr * const pHt,
         case RES_TXTATR_META:
         case RES_TXTATR_METAFIELD:
             OSL_ENSURE(pNewHt, "copying Meta should not fail!");
-            OSL_ENSURE(pDest && (CH_TXTATR_INWORD ==
+            OSL_ENSURE(pDest && (CH_TXTATR_BREAKWORD ==
                                 pDest->GetTxt().GetChar(*pNewHt->GetStart())),
                    "missing CH_TXTATR?");
             break;
diff --git a/sw/source/core/txtnode/thints.cxx b/sw/source/core/txtnode/thints.cxx
index fd419ef9d7f6..a1afcde4cfa0 100644
--- a/sw/source/core/txtnode/thints.cxx
+++ b/sw/source/core/txtnode/thints.cxx
@@ -3013,16 +3013,16 @@ sal_Unicode GetCharOfTxtAttr( const SwTxtAttr& rAttr )
     sal_Unicode cRet = CH_TXTATR_BREAKWORD;
     switch ( rAttr.Which() )
     {
-        case RES_TXTATR_FTN:
         case RES_TXTATR_REFMARK:
         case RES_TXTATR_TOXMARK:
-        case RES_TXTATR_META:
-        case RES_TXTATR_METAFIELD:
             cRet = CH_TXTATR_INWORD;
         break;
 
         case RES_TXTATR_FIELD:
         case RES_TXTATR_FLYCNT:
+        case RES_TXTATR_FTN:
+        case RES_TXTATR_META:
+        case RES_TXTATR_METAFIELD:
         {
             cRet = CH_TXTATR_BREAKWORD;
 
diff --git a/sw/source/core/txtnode/txtedt.cxx b/sw/source/core/txtnode/txtedt.cxx
index 26716a7cf79c..7da33e915e77 100644
--- a/sw/source/core/txtnode/txtedt.cxx
+++ b/sw/source/core/txtnode/txtedt.cxx
@@ -942,11 +942,11 @@ sal_uInt16 SwTxtNode::Spell(SwSpellArgs* pArgs)
                     }
                     else
                     {
-                        // make sure the selection build later from the
-                        // data below does not include footnotes and other
-                        // "in word" character to the left and right in order
-                        // to preserve those. Therefore count those "in words"
-                        // in order to modify the selection accordingly.
+                        // make sure the selection build later from the data
+                        // below does not include "in word" character to the
+                        // left and right in order to preserve those. Therefore
+                        // count those "in words" in order to modify the
+                        // selection accordingly.
                         const sal_Unicode* pChar = rWord.GetBuffer();
                         xub_StrLen nLeft = 0;
                         while (pChar && *pChar++ == CH_TXTATR_INWORD)
diff --git a/sw/source/filter/rtf/swparrtf.cxx b/sw/source/filter/rtf/swparrtf.cxx
index a7ca88635a33..8b53d6db4daf 100644
--- a/sw/source/filter/rtf/swparrtf.cxx
+++ b/sw/source/filter/rtf/swparrtf.cxx
@@ -3566,7 +3566,7 @@ void SwRTFParser::ReadHeaderFooter( int nToken, SwPageDesc* pPageDesc )
                 pPam->GetPoint()->nContent--;
                 nPos--;
                 aFtnNote.SetNumStr(rtl::OUString(pTxtNd->GetTxt().GetChar(nPos)));
-                ((String&)pTxtNd->GetTxt()).SetChar( nPos, CH_TXTATR_INWORD );
+                ((String&)pTxtNd->GetTxt()).SetChar( nPos, CH_TXTATR_BREAKWORD );
                 bDelFirstChar = sal_True;
             }
 
diff --git a/sw/source/filter/ww8/ww8par.cxx b/sw/source/filter/ww8/ww8par.cxx
index 992089013cb5..6acf843cc003 100644
--- a/sw/source/filter/ww8/ww8par.cxx
+++ b/sw/source/filter/ww8/ww8par.cxx
@@ -3159,7 +3159,7 @@ bool SwWW8ImplReader::ReadChar(long nPosCp, long nCpOfs)
             break;
         case 0x2: // Auto-Footnote-Number, should be replaced by SwWW8ImplReader::End_Ftn later
             if (!maFtnStack.empty())
-                cInsert = CH_TXTATR_INWORD;
+                cInsert = 0x2;
             break;
 #if OSL_DEBUG_LEVEL > 1
         default:
diff --git a/sw/source/ui/uiview/viewling.cxx b/sw/source/ui/uiview/viewling.cxx
index 1722998e6499..ae8c23453035 100644
--- a/sw/source/ui/uiview/viewling.cxx
+++ b/sw/source/ui/uiview/viewling.cxx
@@ -533,11 +533,10 @@ void SwView::InsertThesaurusSynonym( const String &rSynonmText, const String &rL
 
         pWrtShell->SelWrd();
 
-        // make sure the selection build later from the
-        // data below does not include footnotes and other
-        // "in word" character to the left and right in order
-        // to preserve those. Therefore count those "in words"
-        // in order to modify the selection accordingly.
+        // make sure the selection build later from the data below does not
+        // include "in word" character to the left and right in order to
+        // preserve those. Therefore count those "in words" in order to modify
+        // the selection accordingly.
         const sal_Unicode* pChar = rLookUpText.GetBuffer();
         xub_StrLen nLeft = 0;
         while (pChar && *pChar++ == CH_TXTATR_INWORD)