X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/f5595af61ed87fec9bd5dd7f62c66089bd90bc57..e10caf88b6cbf09bb38dea47dc258eb638237559:/src/html/winpars.cpp

diff --git a/src/html/winpars.cpp b/src/html/winpars.cpp
index 1769f56c26..ac32fb27bf 100644
--- a/src/html/winpars.cpp
+++ b/src/html/winpars.cpp
@@ -47,11 +47,14 @@ wxHtmlWinParser::wxHtmlWinParser(wxHtmlWindowInterface *wndIface)
     m_CharHeight = m_CharWidth = 0;
     m_UseLink = false;
 #if !wxUSE_UNICODE
+    m_nbsp = 0;
     m_EncConv = NULL;
     m_InputEnc = wxFONTENCODING_ISO8859_1;
     m_OutputEnc = wxFONTENCODING_DEFAULT;
 #endif
+    m_whitespaceMode = Whitespace_Normal;
     m_lastWordCell = NULL;
+    m_posColumn = 0;
 
     {
         int i, j, k, l, m;
@@ -342,97 +345,180 @@ wxFSFile *wxHtmlWinParser::OpenURL(wxHtmlURLType type,
     return GetFS()->OpenFile(myurl, flags);
 }
 
+#define NBSP_UNICODE_VALUE  (wxChar(160))
+#if !wxUSE_UNICODE
+    #define CUR_NBSP_VALUE m_nbsp
+#else
+    #define CUR_NBSP_VALUE NBSP_UNICODE_VALUE
+#endif
+
 void wxHtmlWinParser::AddText(const wxString& txt)
 {
-    register wxChar d;
-    int templen = 0;
-    wxChar nbsp = GetEntitiesParser()->GetCharForCode(160 /* nbsp */);
+#if !wxUSE_UNICODE
+    if ( m_nbsp == 0 )
+        m_nbsp = GetEntitiesParser()->GetCharForCode(NBSP_UNICODE_VALUE);
+#endif
 
-    size_t lng = txt.length();
-    if (lng+1 > m_tmpStrBufSize)
+    if ( m_whitespaceMode == Whitespace_Normal )
     {
-        delete[] m_tmpStrBuf;
-        m_tmpStrBuf = new wxChar[lng+1];
-        m_tmpStrBufSize = lng+1;
-    }
-    wxChar *temp = m_tmpStrBuf;
-
-    wxString::const_iterator i = txt.begin();
-    wxString::const_iterator end = txt.end();
+        int templen = 0;
 
-    if (m_tmpLastWasSpace)
-    {
-        while ( (i < end) &&
-                (*i == wxT('\n') || *i == wxT('\r') || *i == wxT(' ') ||
-                 *i == wxT('\t')) )
+        size_t lng = txt.length();
+        if (lng+1 > m_tmpStrBufSize)
         {
-            ++i;
+            delete[] m_tmpStrBuf;
+            m_tmpStrBuf = new wxChar[lng+1];
+            m_tmpStrBufSize = lng+1;
         }
-    }
+        wxChar *temp = m_tmpStrBuf;
 
-    while (i < end)
-    {
-        size_t x = 0;
-        d = temp[templen++] = *i;
-        if ((d == wxT('\n')) || (d == wxT('\r')) || (d == wxT(' ')) || (d == wxT('\t')))
+        wxString::const_iterator i = txt.begin();
+        const wxString::const_iterator end = txt.end();
+
+        if (m_tmpLastWasSpace)
         {
-            ++i, ++x;
             while ( (i < end) &&
-                    (*i == wxT('\n') || *i == wxT('\r') ||
-                     *i == wxT(' ') || *i == wxT('\t')) )
+                    (*i == wxT('\n') || *i == wxT('\r') || *i == wxT(' ') ||
+                     *i == wxT('\t')) )
             {
                 ++i;
-                ++x;
             }
         }
-        else
-            ++i;
 
-        if (x)
+        while (i < end)
         {
-            temp[templen-1] = wxT(' ');
-            DoAddText(temp, templen, nbsp);
-            m_tmpLastWasSpace = true;
+            size_t x = 0;
+            const wxChar d = temp[templen++] = *i;
+            if ((d == wxT('\n')) || (d == wxT('\r')) || (d == wxT(' ')) || (d == wxT('\t')))
+            {
+                ++i, ++x;
+                while ( (i < end) &&
+                        (*i == wxT('\n') || *i == wxT('\r') ||
+                         *i == wxT(' ') || *i == wxT('\t')) )
+                {
+                    ++i;
+                    ++x;
+                }
+            }
+            else
+            {
+                ++i;
+            }
+
+            if (x)
+            {
+                temp[templen-1] = wxT(' ');
+                FlushWordBuf(temp, templen);
+                m_tmpLastWasSpace = true;
+            }
         }
-    }
 
-    if (templen && (templen > 1 || temp[0] != wxT(' ')))
+        if (templen && (templen > 1 || temp[0] != wxT(' ')))
+        {
+            FlushWordBuf(temp, templen);
+            m_tmpLastWasSpace = false;
+        }
+    }
+    else // m_whitespaceMode == Whitespace_Pre
     {
-        DoAddText(temp, templen, nbsp);
+        if ( txt.find(CUR_NBSP_VALUE) != wxString::npos )
+        {
+            // we need to substitute spaces for &nbsp; here just like we
+            // did in the Whitespace_Normal branch above
+            wxString txt2(txt);
+            txt2.Replace(CUR_NBSP_VALUE, ' ');
+            AddPreBlock(txt2);
+        }
+        else
+        {
+            AddPreBlock(txt);
+        }
+
+        // don't eat any whitespace in <pre> block
         m_tmpLastWasSpace = false;
     }
 }
 
-void wxHtmlWinParser::DoAddText(wxChar *temp, int& templen, wxChar nbsp)
+void wxHtmlWinParser::FlushWordBuf(wxChar *buf, int& len)
 {
-    temp[templen] = 0;
-    templen = 0;
+    buf[len] = 0;
+
+    for ( int i = 0; i < len; i++ )
+    {
+        if ( buf[i] == CUR_NBSP_VALUE )
+            buf[i] = ' ';
+    }
+
 #if !wxUSE_UNICODE
     if (m_EncConv)
-        m_EncConv->Convert(temp);
+        m_EncConv->Convert(buf);
 #endif
-    size_t len = wxStrlen(temp);
-    for (size_t j = 0; j < len; j++)
-    {
-        if (temp[j] == nbsp)
-            temp[j] = wxT(' ');
-    }
 
-    wxHtmlCell *c = new wxHtmlWordCell(temp, *(GetDC()));
+    AddWord(wxString(buf, len));
 
-    ApplyStateToCell(c);
+    len = 0;
+}
 
-    m_Container->InsertCell(c);
-    ((wxHtmlWordCell*)c)->SetPreviousWord(m_lastWordCell);
-    m_lastWordCell = (wxHtmlWordCell*)c;
+void wxHtmlWinParser::AddWord(wxHtmlWordCell *word)
+{
+    ApplyStateToCell(word);
+
+    m_Container->InsertCell(word);
+    word->SetPreviousWord(m_lastWordCell);
+    m_lastWordCell = word;
 }
 
+void wxHtmlWinParser::AddPreBlock(const wxString& text)
+{
+    if ( text.find('\t') != wxString::npos )
+    {
+        wxString text2;
+        text2.reserve(text.length());
+
+        const wxString::const_iterator end = text.end();
+        wxString::const_iterator copyFrom = text.begin();
+        size_t pos = 0;
+        int posColumn = m_posColumn;
+        for ( wxString::const_iterator i = copyFrom; i != end; ++i, ++pos )
+        {
+            if ( *i == '\t' )
+            {
+                if ( copyFrom != i )
+                    text2.append(copyFrom, i);
+
+                const unsigned SPACES_PER_TAB = 8;
+                const size_t expandTo = SPACES_PER_TAB - posColumn % SPACES_PER_TAB;
+                text2.append(expandTo, ' ');
+
+                posColumn += expandTo;
+                copyFrom = i + 1;
+            }
+            else
+            {
+                ++posColumn;
+            }
+        }
+        if ( copyFrom != text.end() )
+            text2.append(copyFrom, text.end());
+
+        AddWord(new wxHtmlWordWithTabsCell(text2, text, m_posColumn, *(GetDC())));
+
+        m_posColumn = posColumn;
+    }
+    else
+    {
+        // no special formatting needed
+        AddWord(text);
+        m_posColumn += text.length();
+    }
+}
 
 
 wxHtmlContainerCell* wxHtmlWinParser::OpenContainer()
 {
     m_Container = new wxHtmlContainerCell(m_Container);
     m_Container->SetAlignHor(m_Align);
+    m_posColumn = 0;
     m_tmpLastWasSpace = true;
         /* to avoid space being first character in paragraph */
     return m_Container;
@@ -458,8 +544,10 @@ wxHtmlContainerCell* wxHtmlWinParser::CloseContainer()
 
 void wxHtmlWinParser::SetFontSize(int s)
 {
-    if (s < 1) s = 1;
-    else if (s > 7) s = 7;
+    if (s < 1)
+        s = 1;
+    else if (s > 7)
+        s = 7;
     m_FontSize = s;
 }
 
@@ -520,8 +608,10 @@ void wxHtmlWinParser::SetLink(const wxHtmlLinkInfo& link)
 
 void wxHtmlWinParser::SetFontFace(const wxString& face)
 {
-    if (GetFontFixed()) m_FontFaceFixed = face;
-    else m_FontFaceNormal = face;
+    if (GetFontFixed())
+        m_FontFaceFixed = face;
+    else
+        m_FontFaceNormal = face;
 
 #if !wxUSE_UNICODE
     if (m_InputEnc != wxFONTENCODING_DEFAULT)
@@ -543,6 +633,9 @@ void wxHtmlWinParser::ApplyStateToCell(wxHtmlCell *cell)
 #if !wxUSE_UNICODE
 void wxHtmlWinParser::SetInputEncoding(wxFontEncoding enc)
 {
+    // the character used for non-breakable space may change:
+    m_nbsp = 0;
+
     m_InputEnc = m_OutputEnc = wxFONTENCODING_DEFAULT;
     if (m_EncConv)
     {
@@ -550,29 +643,37 @@ void wxHtmlWinParser::SetInputEncoding(wxFontEncoding enc)
         m_EncConv = NULL;
     }
 
-    if (enc == wxFONTENCODING_DEFAULT) return;
+    if (enc == wxFONTENCODING_DEFAULT)
+        return;
 
     wxFontEncoding altfix, altnorm;
     bool availfix, availnorm;
 
-    // exact match?
     availnorm = wxFontMapper::Get()->IsEncodingAvailable(enc, m_FontFaceNormal);
     availfix = wxFontMapper::Get()->IsEncodingAvailable(enc, m_FontFaceFixed);
+
     if (availnorm && availfix)
+    {
+        // exact match?
         m_OutputEnc = enc;
+    }
 
-    // alternatives?
     else if (wxFontMapper::Get()->GetAltForEncoding(enc, &altnorm, m_FontFaceNormal, false) &&
              wxFontMapper::Get()->GetAltForEncoding(enc, &altfix, m_FontFaceFixed, false) &&
              altnorm == altfix)
+    {
+        // alternatives?
         m_OutputEnc = altnorm;
-
-    // at least normal face?
+    }
     else if (availnorm)
+    {
+        // at least normal face?
         m_OutputEnc = enc;
+    }
     else if (wxFontMapper::Get()->GetAltForEncoding(enc, &altnorm, m_FontFaceNormal, false))
+    {
         m_OutputEnc = altnorm;
-
+    }
     else
     {
 #ifndef __WXMAC__
@@ -585,11 +686,16 @@ void wxHtmlWinParser::SetInputEncoding(wxFontEncoding enc)
 
     m_InputEnc = enc;
     if (m_OutputEnc == wxFONTENCODING_DEFAULT)
+    {
         GetEntitiesParser()->SetEncoding(wxFONTENCODING_SYSTEM);
+    }
     else
+    {
         GetEntitiesParser()->SetEncoding(m_OutputEnc);
+    }
 
-    if (m_InputEnc == m_OutputEnc) return;
+    if (m_InputEnc == m_OutputEnc)
+        return;
 
     m_EncConv = new wxEncodingConverter();
     if (!m_EncConv->Init(m_InputEnc,