Replaced /'s with \'s as BCC requires \'s for path names

[wxWidgets.git] / src / html / htmlpars.cpp

diff --git a/src/html/htmlpars.cpp b/src/html/htmlpars.cpp
index 6a69f9e0b928e6a8568f0d31a1df98f571bd9717..57a7406d032463b5e827531475c00ff58fc023db 100644 (file)
--- a/src/html/htmlpars.cpp
+++ b/src/html/htmlpars.cpp
@@ -32,8 +32,33 @@
 #include "wx/fontmap.h"
 #include "wx/html/htmldefs.h"
 #include "wx/html/htmlpars.h"
 #include "wx/fontmap.h"
 #include "wx/html/htmldefs.h"
 #include "wx/html/htmlpars.h"

+#include "wx/dynarray.h"
+#include "wx/arrimpl.cpp"

 
 

+//-----------------------------------------------------------------------------
+// wxHtmlParser helpers
+//-----------------------------------------------------------------------------
+
+class wxHtmlTextPiece
+{
+public:
+    wxHtmlTextPiece(int pos, int lng) : m_pos(pos), m_lng(lng) {}
+    int m_pos, m_lng;
+};
+
+WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces);
+WX_DEFINE_OBJARRAY(wxHtmlTextPieces);

 
 

+class wxHtmlParserState
+{
+public:
+    wxHtmlTag         *m_curTag;
+    wxHtmlTag         *m_tags;
+    wxHtmlTextPieces  *m_textPieces;
+    int                m_curTextPiece;
+    wxString           m_source;
+    wxHtmlParserState *m_nextState;
+};

 
 //-----------------------------------------------------------------------------
 // wxHtmlParser
 
 //-----------------------------------------------------------------------------
 // wxHtmlParser


@@ -42,14 +67,21 @@
 IMPLEMENT_ABSTRACT_CLASS(wxHtmlParser,wxObject)
 
 wxHtmlParser::wxHtmlParser()
 IMPLEMENT_ABSTRACT_CLASS(wxHtmlParser,wxObject)
 
 wxHtmlParser::wxHtmlParser()

-    : wxObject(), m_Cache(NULL), m_HandlersHash(wxKEY_STRING),
+    : wxObject(), m_HandlersHash(wxKEY_STRING),

       m_FS(NULL), m_HandlersStack(NULL)
 {
     m_entitiesParser = new wxHtmlEntitiesParser;
       m_FS(NULL), m_HandlersStack(NULL)
 {
     m_entitiesParser = new wxHtmlEntitiesParser;

+    m_Tags = NULL;
+    m_CurTag = NULL;
+    m_TextPieces = NULL;
+    m_CurTextPiece = 0;
+    m_SavedStates = NULL;

 }
 
 wxHtmlParser::~wxHtmlParser()
 {
 }
 
 wxHtmlParser::~wxHtmlParser()
 {

+    while (RestoreState())
+        DestroyDOMTree();

     delete m_HandlersStack;
     m_HandlersHash.Clear();
     m_HandlersList.DeleteContents(TRUE);
     delete m_HandlersStack;
     m_HandlersHash.Clear();
     m_HandlersList.DeleteContents(TRUE);


@@ -75,62 +107,193 @@ void wxHtmlParser::InitParser(const wxString& source)
 
 void wxHtmlParser::DoneParser()
 {
 
 void wxHtmlParser::DoneParser()
 {

-    delete m_Cache;
-    m_Cache = NULL;
+    DestroyDOMTree();

 }
 
 void wxHtmlParser::SetSource(const wxString& src)
 {
 }
 
 void wxHtmlParser::SetSource(const wxString& src)
 {

+    DestroyDOMTree();

     m_Source = src;
     m_Source = src;

-    delete m_Cache;
-    m_Cache = new wxHtmlTagsCache(m_Source);
+    CreateDOMTree();
+    m_CurTag = NULL;
+    m_CurTextPiece = 0;

 }
 
 }
 

-void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
+void wxHtmlParser::CreateDOMTree()

 {
 {

-    if (end_pos <= begin_pos) return;
+    wxHtmlTagsCache cache(m_Source);
+    m_TextPieces = new wxHtmlTextPieces;
+    CreateDOMSubTree(NULL, 0, m_Source.Length(), &cache);
+    m_CurTextPiece = 0;
+}

 
 

-    char c;
-    char *temp = new char[end_pos - begin_pos + 1];
-    int i;
-    int templen;
+void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
+                                    int begin_pos, int end_pos,
+                                    wxHtmlTagsCache *cache)
+{
+    if (end_pos <= begin_pos) return;

 
 

-    templen = 0;
-    i = begin_pos;
+    wxChar c;
+    int i = begin_pos;
+    int textBeginning = begin_pos;

 
     while (i < end_pos)
     {
 
     while (i < end_pos)
     {

-        c = m_Source[(unsigned int) i];
-
-        // continue building word:
-        if (c != '<')
-           {
-            temp[templen++] = c;
-            i++;
-        }
+        c = m_Source.GetChar(i);

 
 

-        else if (c == '<')
-           {
-            wxHtmlTag tag(m_Source, i, end_pos, m_Cache, m_entitiesParser);
+        if (c == wxT('<'))
+        {
+            // add text to m_TextPieces:
+            if (i - textBeginning > 0)
+                m_TextPieces->Add(
+                    wxHtmlTextPiece(textBeginning, i - textBeginning));
+
+            // if it is a comment, skip it:
+            if (i < end_pos-6 && m_Source.GetChar(i+1) == wxT('!') &&
+                                 m_Source.GetChar(i+2) == wxT('-') &&
+                                 m_Source.GetChar(i+3) == wxT('-'))
+            {
+                // Comments begin with "<!--" and end with "--[ \t\r\n]*>"
+                // according to HTML 4.0
+                int dashes = 0;
+                i += 4;
+                while (i < end_pos)
+                {
+                    c = m_Source.GetChar(i++);
+                    if ((c == wxT(' ') || c == wxT('\n') ||
+                        c == wxT('\r') || c == wxT('\t')) && dashes >= 2) {}
+                    else if (c == wxT('>') && dashes >= 2)
+                    {
+                        textBeginning = i;
+                        break;
+                    }
+                    else if (c == wxT('-'))
+                        dashes++;
+                    else
+                        dashes = 0;
+                }
+            }

 
 

-            if (templen)
+            // add another tag to the tree:
+            else if (i < end_pos-1 && m_Source.GetChar(i+1) != wxT('/'))

                {
                {

-                temp[templen] = 0;
-                AddText(temp);
-                templen = 0;
+                wxHtmlTag *chd;
+                if (cur)
+                    chd = new wxHtmlTag(cur, m_Source,
+                                        i, end_pos, cache, m_entitiesParser);
+                else
+                {
+                    chd = new wxHtmlTag(NULL, m_Source,
+                                        i, end_pos, cache, m_entitiesParser);
+                    if (!m_Tags)
+                    {
+                        // if this is the first tag to be created make the root
+                        // m_Tags point to it:
+                        m_Tags = chd;
+                    }
+                    else
+                    {
+                        // if there is already a root tag add this tag as
+                        // the last sibling:
+                        chd->m_Prev = m_Tags->GetLastSibling();
+                        chd->m_Prev->m_Next = chd;
+                    }
+                }
+
+                if (chd->HasEnding())
+                {
+                    CreateDOMSubTree(chd,
+                                     chd->GetBeginPos(), chd->GetEndPos1(),
+                                     cache);
+                    i = chd->GetEndPos2();
+                }
+                else
+                    i = chd->GetBeginPos();
+                textBeginning = i;
+            }
+
+            // ... or skip ending tag:
+            else
+            {
+                while (i < end_pos && m_Source.GetChar(i) != wxT('>')) i++;
+                textBeginning = i+1;

             }
             }

-            AddTag(tag);
-            if (tag.HasEnding()) i = tag.GetEndPos2();
-            else i = tag.GetBeginPos();

         }
         }

+        else i++;

     }
 
     }
 

-    if (templen)
-    { // last word of block :-(
-        temp[templen] = 0;
-        AddText(temp);
+    // add remaining text to m_TextPieces:
+    if (end_pos - textBeginning > 0)
+        m_TextPieces->Add(
+            wxHtmlTextPiece(textBeginning, end_pos - textBeginning));
+}
+
+void wxHtmlParser::DestroyDOMTree()
+{
+    wxHtmlTag *t1, *t2;
+    t1 = m_Tags;
+    while (t1)
+    {
+        t2 = t1->GetNextSibling();
+        delete t1;
+        t1 = t2;
+    }
+    m_Tags = m_CurTag = NULL;
+
+    delete m_TextPieces;
+    m_TextPieces = NULL;
+}
+
+void wxHtmlParser::DoParsing()
+{
+    m_CurTag = m_Tags;
+    m_CurTextPiece = 0;
+    DoParsing(0, m_Source.Length());
+}
+
+void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
+{
+    if (end_pos <= begin_pos) return;
+
+    wxHtmlTextPieces& pieces = *m_TextPieces;
+    size_t piecesCnt = pieces.GetCount();
+
+    while (begin_pos < end_pos)
+    {
+        while (m_CurTag && m_CurTag->GetBeginPos() < begin_pos)
+            m_CurTag = m_CurTag->GetNextTag();
+        while (m_CurTextPiece < piecesCnt &&
+               pieces[m_CurTextPiece].m_pos < begin_pos)
+            m_CurTextPiece++;
+
+        if (m_CurTextPiece < piecesCnt &&
+            (!m_CurTag ||
+             pieces[m_CurTextPiece].m_pos < m_CurTag->GetBeginPos()))
+        {
+            // Add text:
+            AddText(GetEntitiesParser()->Parse(
+                       m_Source.Mid(pieces[m_CurTextPiece].m_pos,
+                                    pieces[m_CurTextPiece].m_lng)));
+            begin_pos = pieces[m_CurTextPiece].m_pos +
+                        pieces[m_CurTextPiece].m_lng;
+            m_CurTextPiece++;
+        }
+        else if (m_CurTag)
+        {
+            // Add tag:
+            if (m_CurTag)
+            {
+                if (m_CurTag->HasEnding())
+                    begin_pos = m_CurTag->GetEndPos2();
+                else
+                    begin_pos = m_CurTag->GetBeginPos();
+            }
+            wxHtmlTag *t = m_CurTag;
+            m_CurTag = m_CurTag->GetNextTag();
+            AddTag(*t);
+        }
+        else break;

     }
     }

-    delete[] temp;

 }
 
 void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 }
 
 void wxHtmlParser::AddTag(const wxHtmlTag& tag)


@@ -151,7 +314,7 @@ void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
 {
     wxString s(handler->GetSupportedTags());
 void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
 {
     wxString s(handler->GetSupportedTags());

-    wxStringTokenizer tokenizer(s, ", ");
+    wxStringTokenizer tokenizer(s, wxT(", "));

 
     while (tokenizer.HasMoreTokens())
         m_HandlersHash.Put(tokenizer.NextToken(), handler);
 
     while (tokenizer.HasMoreTokens())
         m_HandlersHash.Put(tokenizer.NextToken(), handler);


@@ -164,7 +327,7 @@ void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
 
 void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
 {
 
 void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
 {

-    wxStringTokenizer tokenizer(tags, ", ");
+    wxStringTokenizer tokenizer(tags, wxT(", "));

     wxString key;
 
     if (m_HandlersStack == NULL)
     wxString key;
 
     if (m_HandlersStack == NULL)


@@ -197,6 +360,45 @@ void wxHtmlParser::PopTagHandler()
     m_HandlersStack->DeleteNode(first);
 }
 
     m_HandlersStack->DeleteNode(first);
 }
 

+void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
+{
+    wxHtmlParserState *s = new wxHtmlParserState;
+
+    s->m_curTag = m_CurTag;
+    s->m_tags = m_Tags;
+    s->m_textPieces = m_TextPieces;
+    s->m_curTextPiece = m_CurTextPiece;
+    s->m_source = m_Source;
+
+    s->m_nextState = m_SavedStates;
+    m_SavedStates = s;
+
+    m_CurTag = NULL;
+    m_Tags = NULL;
+    m_TextPieces = NULL;
+    m_CurTextPiece = 0;
+    m_Source = wxEmptyString;
+
+    SetSource(src);
+}
+
+bool wxHtmlParser::RestoreState()
+{
+    if (!m_SavedStates) return FALSE;
+
+    wxHtmlParserState *s = m_SavedStates;
+    m_SavedStates = s->m_nextState;
+
+    m_CurTag = s->m_curTag;
+    m_Tags = s->m_tags;
+    m_TextPieces = s->m_textPieces;
+    m_CurTextPiece = s->m_curTextPiece;
+    m_Source = s->m_source;
+
+    delete s;
+    return TRUE;
+}
+

 //-----------------------------------------------------------------------------
 // wxHtmlTagHandler
 //-----------------------------------------------------------------------------
 //-----------------------------------------------------------------------------
 // wxHtmlTagHandler
 //-----------------------------------------------------------------------------


@@ -219,7 +421,9 @@ wxHtmlEntitiesParser::wxHtmlEntitiesParser()
 
 wxHtmlEntitiesParser::~wxHtmlEntitiesParser()
 {
 
 wxHtmlEntitiesParser::~wxHtmlEntitiesParser()
 {

+#if wxUSE_WCHAR_T && !wxUSE_UNICODE

     delete m_conv;
     delete m_conv;

+#endif

 }
 
 void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
 }
 
 void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)


@@ -231,6 +435,8 @@ void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
     m_encoding = encoding;
     if (m_encoding != wxFONTENCODING_SYSTEM)
         m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));
     m_encoding = encoding;
     if (m_encoding != wxFONTENCODING_SYSTEM)
         m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));

+#else
+    (void) encoding;

 #endif
 }
 
 #endif
 }
 


@@ -254,9 +460,9 @@ wxString wxHtmlEntitiesParser::Parse(const wxString& input)
                    (*c >= wxT('0') && *c <= wxT('9')) ||
                    *c == wxT('_') || *c == wxT('#'); c++) {}
             entity.append(ent_s, c - ent_s);
                    (*c >= wxT('0') && *c <= wxT('9')) ||
                    *c == wxT('_') || *c == wxT('#'); c++) {}
             entity.append(ent_s, c - ent_s);

-            if (*c == wxT(';')) c++;
+            if (*c != wxT(';')) c--;
+            last = c+1;

             output << GetEntityChar(entity);
             output << GetEntityChar(entity);

-            last = c;

         }
     }
     if (*last != wxT('\0'))
         }
     }
     if (*last != wxT('\0'))


@@ -270,7 +476,7 @@ struct wxHtmlEntityInfo
     unsigned code;
 };
 
     unsigned code;
 };
 

-static int compar_entity(const void *key, const void *item)
+static int LINKAGEMODE compar_entity(const void *key, const void *item)

 {
     return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
 }
 {
     return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
 }


@@ -285,7 +491,7 @@ wxChar wxHtmlEntitiesParser::GetCharForCode(unsigned code)
     wbuf[0] = (wchar_t)code;
     wbuf[1] = 0;
     wxMBConv *conv = m_conv ? m_conv : &wxConvLocal;
     wbuf[0] = (wchar_t)code;
     wbuf[1] = 0;
     wxMBConv *conv = m_conv ? m_conv : &wxConvLocal;

-    if (conv->WC2MB(buf, wbuf, 1) == (size_t)-1)
+    if (conv->WC2MB(buf, wbuf, 2) == (size_t)-1)

         return '?';
     return buf[0];
 #else
         return '?';
     return buf[0];
 #else