Added some missing tests; replaced scrolwin.cpp with scrlwng.cpp

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

diff --git a/src/html/htmltag.cpp b/src/html/htmltag.cpp
index 386bb98a59cd64eef03b96a0eada5ef49c3900e3..52e0c1b8f8c7f692d912ba7439189390c2bf0869 100644 (file)
--- a/src/html/htmltag.cpp
+++ b/src/html/htmltag.cpp
@@ -12,7 +12,7 @@
 #pragma implementation
 #endif
 
 #pragma implementation
 #endif
 

-#include <wx/wxprec.h>
+#include "wx/wxprec.h"

 
 #include "wx/defs.h"
 #if wxUSE_HTML
 
 #include "wx/defs.h"
 #if wxUSE_HTML


@@ -22,20 +22,37 @@
 #endif
 
 #ifndef WXPRECOMP
 #endif
 
 #ifndef WXPRECOMP

-#include <wx/wx.h>
+#include "wx/wx.h"

 #endif
 
 #include "wx/html/htmltag.h"
 #endif
 
 #include "wx/html/htmltag.h"

+#include "wx/html/htmlpars.h"

 #include <stdio.h> // for vsscanf
 #include <stdarg.h>
 
 
 #include <stdio.h> // for vsscanf
 #include <stdarg.h>
 
 

-
-

 //-----------------------------------------------------------------------------
 // wxHtmlTagsCache
 //-----------------------------------------------------------------------------
 
 //-----------------------------------------------------------------------------
 // wxHtmlTagsCache
 //-----------------------------------------------------------------------------
 

+struct wxHtmlCacheItem
+{
+    // this is "pos" value passed to wxHtmlTag's constructor.
+    // it is position of '<' character of the tag
+    int Key;
+
+    // end positions for the tag:
+    // end1 is '<' of ending tag,
+    // end2 is '>' or both are
+    // -1 if there is no ending tag for this one...
+    // or -2 if this is ending tag  </...>
+    int End1, End2;
+
+    // name of this tag
+    wxChar *Name;
+};
+
+

 IMPLEMENT_CLASS(wxHtmlTagsCache,wxObject)
 
 #define CACHE_INCREMENT  64
 IMPLEMENT_CLASS(wxHtmlTagsCache,wxObject)
 
 #define CACHE_INCREMENT  64


@@ -52,14 +69,20 @@ wxHtmlTagsCache::wxHtmlTagsCache(const wxString& source)
     m_CachePos = 0;
 
     pos = 0;
     m_CachePos = 0;
 
     pos = 0;

-    while (pos < lng) {
-        if (src[pos] == wxT('<')) {  // tag found:
+    while (pos < lng) 
+    {
+        if (src[pos] == wxT('<'))   // tag found:
+        {

             if (m_CacheSize % CACHE_INCREMENT == 0)
             if (m_CacheSize % CACHE_INCREMENT == 0)

-                m_Cache = (sCacheItem*) realloc(m_Cache, (m_CacheSize + CACHE_INCREMENT) * sizeof(sCacheItem));
+                m_Cache = (wxHtmlCacheItem*) realloc(m_Cache, (m_CacheSize + CACHE_INCREMENT) * sizeof(wxHtmlCacheItem));

             tg = m_CacheSize++;
             m_Cache[tg].Key = stpos = pos++;
             dummy[0] = 0; i = 0;
             tg = m_CacheSize++;
             m_Cache[tg].Key = stpos = pos++;
             dummy[0] = 0; i = 0;

-            while ((src[pos] != wxT('>')) && (src[pos] != wxT(' '))) {
+            while (pos < lng && 
+                   src[pos] != wxT('>') &&
+                   src[pos] != wxT(' ') && src[pos] != wxT('\r') && 
+                   src[pos] != wxT('\n') && src[pos] != wxT('\t')) 
+            {

                 dummy[i] = src[pos++];
                 if ((dummy[i] >= wxT('a')) && (dummy[i] <= wxT('z'))) dummy[i] -= (wxT('a') - wxT('A'));
                 i++;
                 dummy[i] = src[pos++];
                 if ((dummy[i] >= wxT('a')) && (dummy[i] <= wxT('z'))) dummy[i] -= (wxT('a') - wxT('A'));
                 i++;


@@ -68,19 +91,22 @@ wxHtmlTagsCache::wxHtmlTagsCache(const wxString& source)
             m_Cache[tg].Name = new wxChar[i+1];
             memcpy(m_Cache[tg].Name, dummy, (i+1)*sizeof(wxChar));
 
             m_Cache[tg].Name = new wxChar[i+1];
             memcpy(m_Cache[tg].Name, dummy, (i+1)*sizeof(wxChar));
 

-            while (src[pos] != wxT('>')) pos++;
+            while (pos < lng && src[pos] != wxT('>')) pos++;

 
 

-            if (src[stpos+1] == wxT('/')) { // ending tag:
+            if (src[stpos+1] == wxT('/')) // ending tag:
+            {

                 m_Cache[tg].End1 = m_Cache[tg].End2 = -2;
                 // find matching begin tag:
                 for (i = tg; i >= 0; i--)
                 m_Cache[tg].End1 = m_Cache[tg].End2 = -2;
                 // find matching begin tag:
                 for (i = tg; i >= 0; i--)

-                    if ((m_Cache[i].End1 == -1) && (wxStrcmp(m_Cache[i].Name, dummy+1) == 0)) {
+                    if ((m_Cache[i].End1 == -1) && (wxStrcmp(m_Cache[i].Name, dummy+1) == 0)) 
+                    {

                         m_Cache[i].End1 = stpos;
                         m_Cache[i].End2 = pos + 1;
                         break;
                     }
             }
                         m_Cache[i].End1 = stpos;
                         m_Cache[i].End2 = pos + 1;
                         break;
                     }
             }

-            else {
+            else 
+            {

                 m_Cache[tg].End1 = m_Cache[tg].End2 = -1;
             }
         }
                 m_Cache[tg].End1 = m_Cache[tg].End2 = -1;
             }
         }


@@ -89,20 +115,24 @@ wxHtmlTagsCache::wxHtmlTagsCache(const wxString& source)
     }
 
     // ok, we're done, now we'll free .Name members of cache - we don't need it anymore:
     }
 
     // ok, we're done, now we'll free .Name members of cache - we don't need it anymore:

-    for (i = 0; i < m_CacheSize; i++) {
+    for (i = 0; i < m_CacheSize; i++) 
+    {

         delete[] m_Cache[i].Name;
         m_Cache[i].Name = NULL;
     }
 }
 
         delete[] m_Cache[i].Name;
         m_Cache[i].Name = NULL;
     }
 }
 

-
-

 void wxHtmlTagsCache::QueryTag(int at, int* end1, int* end2)
 {
     if (m_Cache == NULL) return;
 void wxHtmlTagsCache::QueryTag(int at, int* end1, int* end2)
 {
     if (m_Cache == NULL) return;

-    if (m_Cache[m_CachePos].Key != at) {
+    if (m_Cache[m_CachePos].Key != at) 
+    {

         int delta = (at < m_Cache[m_CachePos].Key) ? -1 : 1;
         int delta = (at < m_Cache[m_CachePos].Key) ? -1 : 1;

-        do {m_CachePos += delta;} while (m_Cache[m_CachePos].Key != at);
+        do 
+        { 
+            m_CachePos += delta; 
+        }
+        while (m_Cache[m_CachePos].Key != at);

     }
     *end1 = m_Cache[m_CachePos].End1;
     *end2 = m_Cache[m_CachePos].End2;
     }
     *end1 = m_Cache[m_CachePos].End1;
     *end2 = m_Cache[m_CachePos].End2;


@@ -117,133 +147,236 @@ void wxHtmlTagsCache::QueryTag(int at, int* end1, int* end2)
 
 IMPLEMENT_CLASS(wxHtmlTag,wxObject)
 
 
 IMPLEMENT_CLASS(wxHtmlTag,wxObject)
 

-wxHtmlTag::wxHtmlTag(const wxString& source, int pos, int end_pos, wxHtmlTagsCache* cache) : wxObject()
+wxHtmlTag::wxHtmlTag(const wxString& source, int pos, int end_pos, 
+                     wxHtmlTagsCache *cache,
+                     wxHtmlEntitiesParser *entParser) : wxObject()

 {
     int i;
 {
     int i;

-    char c;
+    wxChar c;

 
     // fill-in name, params and begin pos:
 
     // fill-in name, params and begin pos:

-    m_Name = m_Params = wxEmptyString;

     i = pos+1;
     i = pos+1;

-    if (source[i] == '/') {m_Ending = TRUE; i++;}
-    else m_Ending = FALSE;
-
-    while ((i < end_pos) && ((c = source[i++]) != ' ') && (c != '>')) {
-        if ((c >= 'a') && (c <= 'z')) c -= ('a' - 'A');
-        m_Name += c;
+    if (source[i] == wxT('/')) 
+        { m_Ending = TRUE; i++; }
+    else 
+        m_Ending = FALSE;
+
+    // find tag's name and convert it to uppercase:
+    while ((i < end_pos) && 
+           ((c = source[i++]) != wxT(' ') && c != wxT('\r') && 
+             c != wxT('\n') && c != wxT('\t') &&
+             c != wxT('>'))) 
+    {
+        if ((c >= wxT('a')) && (c <= wxT('z'))) 
+            c -= (wxT('a') - wxT('A'));
+        m_Name << c;

     }
 
     }
 

-    if (source[i-1] != '>')
-        while ((i < end_pos) && ((c = source[i++]) != '>')) {
-            if ((c >= 'a') && (c <= 'z')) c -= ('a' - 'A');
-            m_Params += c;
-            if (c == '"') {
-                while ((i < end_pos) && ((c = source[i++]) != '"')) m_Params += c;
-                m_Params += c;
+    // if the tag has parameters, read them and "normalize" them,
+    // i.e. convert to uppercase, replace whitespaces by spaces and 
+    // remove whitespaces around '=':
+    if (source[i-1] != wxT('>'))
+    {
+        #define IS_WHITE(c) (c == wxT(' ') || c == wxT('\r') || \
+                             c == wxT('\n') || c == wxT('\t'))
+        wxString pname, pvalue;
+        wxChar quote;
+        enum 
+        {
+            ST_BEFORE_NAME = 1, 
+            ST_NAME,
+            ST_BEFORE_EQ,
+            ST_BEFORE_VALUE,
+            ST_VALUE
+        } state;
+    
+        quote = 0;
+        state = ST_BEFORE_NAME;
+        while (i < end_pos)
+        {
+            c = source[i++];
+
+            if (c == wxT('>') && !(state == ST_VALUE && quote != 0)) 
+            {
+                if (state == ST_BEFORE_EQ || state == ST_NAME)
+                {
+                    m_ParamNames.Add(pname);
+                    m_ParamValues.Add(wxEmptyString);
+                }
+                else if (state == ST_VALUE && quote == 0)
+                {
+                    m_ParamNames.Add(pname);
+                    if (entParser)
+                        m_ParamValues.Add(entParser->Parse(pvalue));
+                    else
+                        m_ParamValues.Add(pvalue);
+                }
+                break;

             }
             }

-            else if (c == '\'') {
-                while ((i < end_pos) && ((c = source[i++]) != '\'')) m_Params += c;
-                m_Params += c;
+            switch (state)
+            {
+                case ST_BEFORE_NAME:
+                    if (!IS_WHITE(c))
+                    {
+                        pname = c;
+                        state = ST_NAME;
+                    }
+                    break;
+                case ST_NAME:
+                    if (IS_WHITE(c))
+                        state = ST_BEFORE_EQ;
+                    else if (c == wxT('='))
+                        state = ST_BEFORE_VALUE;
+                    else
+                        pname << c;
+                    break;
+                case ST_BEFORE_EQ:
+                    if (c == wxT('='))
+                        state = ST_BEFORE_VALUE;
+                    else if (!IS_WHITE(c))
+                    {
+                        m_ParamNames.Add(pname);
+                        m_ParamValues.Add(wxEmptyString);
+                        pname = c;
+                        state = ST_NAME;
+                    }
+                    break;
+                case ST_BEFORE_VALUE:
+                    if (!IS_WHITE(c))
+                    {
+                        if (c == wxT('"') || c == wxT('\''))
+                            quote = c, pvalue = wxEmptyString;
+                        else
+                            quote = 0, pvalue = c;
+                        state = ST_VALUE;
+                    }
+                    break;
+                case ST_VALUE:
+                    if ((quote != 0 && c == quote) ||
+                        (quote == 0 && IS_WHITE(c)))
+                    {
+                        m_ParamNames.Add(pname);
+                        if (quote == 0)
+                        {
+                            // VS: backward compatibility, no real reason,
+                            //     but wxHTML code relies on this... :(
+                            pvalue.MakeUpper();
+                        }
+                        if (entParser)
+                            m_ParamValues.Add(entParser->Parse(pvalue));
+                        else
+                            m_ParamValues.Add(pvalue);
+                        state = ST_BEFORE_NAME;
+                    }
+                    else
+                        pvalue << c;
+                    break;

             }
         }
             }
         }

+        
+        #undef IS_WHITE
+   }

    m_Begin = i;
 
    m_Begin = i;
 

-   cache -> QueryTag(pos, &m_End1, &m_End2);
+   cache->QueryTag(pos, &m_End1, &m_End2);

    if (m_End1 > end_pos) m_End1 = end_pos;
    if (m_End2 > end_pos) m_End2 = end_pos;
 }
 
    if (m_End1 > end_pos) m_End1 = end_pos;
    if (m_End2 > end_pos) m_End2 = end_pos;
 }
 

-    
-

 bool wxHtmlTag::HasParam(const wxString& par) const
 {
 bool wxHtmlTag::HasParam(const wxString& par) const
 {

-    const wxChar *st = m_Params, *p = par;
-    const wxChar *st2, *p2;
-
-    if (*st == 0) return FALSE;
-    if (*p == 0) return FALSE;
-    for (st2 = st, p2 = p; ; st2++) {
-        if (*p2 == 0) return TRUE;
-        if (*st2 == 0) return FALSE;
-        if (*p2 != *st2) p2 = p;
-        if (*p2 == *st2) p2++;
-        if (*st2 == ' ') p2 = p;
-        else if (*st2 == '=') {
-            p2 = p;
-            while (*st2 != ' ') {
-                if (*st2 == '"') {
-                    st2++;
-                    while (*st2 != '"') st2++;
-                }
-                st2++;
-                if (*st2 == 0) return FALSE;
-            }
-        }
-    }
+    return (m_ParamNames.Index(par, FALSE) != wxNOT_FOUND);

 }
 
 }
 

-
-

 wxString wxHtmlTag::GetParam(const wxString& par, bool with_commas) const
 {
 wxString wxHtmlTag::GetParam(const wxString& par, bool with_commas) const
 {

-    const wxChar *st = m_Params, *p = par;
-    const wxChar *st2, *p2;
-    bool comma;
-    char comma_char;
-
-    if (*st == 0) return "";
-    if (*p == 0) return "";
-    for (st2 = st, p2 = p; ; st2++) {
-        if (*p2 == 0) { // found
-            wxString fnd = "";
-            st2++; // '=' character
-            comma = FALSE;
-           comma_char = '\0';
-            if (!with_commas && (*(st2) == '"')) {
-               st2++;
-               comma = TRUE; 
-               comma_char = '"';
-           }
-           else if (!with_commas && (*(st2) == '\'')) {
-               st2++; 
-               comma = TRUE;
-               comma_char = '\'';
-           }
-            while (*st2 != 0) {
-                if (comma && *st2 == comma_char) comma = FALSE;
-                else if ((*st2 == ' ') && (!comma)) break;
-                fnd += (*(st2++));
-            }
-            if (!with_commas && (*(st2-1) == comma_char)) fnd.RemoveLast();
-            return fnd;
-        }
-        if (*st2 == 0) return "";
-        if (*p2 != *st2) p2 = p;
-        if (*p2 == *st2) p2++;
-        if (*st2 == ' ') p2 = p;
-        else if (*st2 == '=') {
-            p2 = p;
-            while (*st2 != ' ') {
-                if (*st2 == '"') {
-                    st2++;
-                    while (*st2 != '"') st2++;
-                }
-                else if (*st2 == '\'') {
-                    st2++;
-                    while (*st2 != '\'') st2++;
-                }
-                st2++;
-            }
-        }
+    int index = m_ParamNames.Index(par, FALSE);
+    if (index == wxNOT_FOUND)
+        return wxEmptyString;
+    if (with_commas)
+    {
+        // VS: backward compatibility, seems to be never used by wxHTML...
+        wxString s;
+        s << wxT('"') << m_ParamValues[index] << wxT('"');
+        return s;

     }
     }

+    else
+        return m_ParamValues[index];

 }
 
 }
 

-
-

 int wxHtmlTag::ScanParam(const wxString& par, wxChar *format, void *param) const
 {
     wxString parval = GetParam(par);
 int wxHtmlTag::ScanParam(const wxString& par, wxChar *format, void *param) const
 {
     wxString parval = GetParam(par);

-    return wxSscanf((const wxChar*)parval, format, param);
+    return wxSscanf(parval, format, param);
+}
+
+bool wxHtmlTag::GetParamAsColour(const wxString& par, wxColour *clr) const
+{
+    wxString str = GetParam(par);
+    
+    if (str.IsEmpty()) return FALSE;
+    if (str.GetChar(0) == wxT('#'))
+    {
+        unsigned long tmp;
+        if (ScanParam(par, wxT("#%lX"), &tmp) != 1)
+            return FALSE;
+        *clr = wxColour((unsigned char)((tmp & 0xFF0000) >> 16),
+                                           (unsigned char)((tmp & 0x00FF00) >> 8),
+                                           (unsigned char)(tmp & 0x0000FF));
+        return TRUE;
+    }
+    else
+    {
+        // Handle colours defined in HTML 4.0:
+        #define HTML_COLOUR(name,r,g,b)                 \
+            if (str.IsSameAs(wxT(name), FALSE))         \
+                { *clr = wxColour(r,g,b); return TRUE; }
+        HTML_COLOUR("black",   0x00,0x00,0x00)
+        HTML_COLOUR("silver",  0xC0,0xC0,0xC0)
+        HTML_COLOUR("gray",    0x80,0x80,0x80)
+        HTML_COLOUR("white",   0xFF,0xFF,0xFF)
+        HTML_COLOUR("maroon",  0x80,0x00,0x00)
+        HTML_COLOUR("red",     0xFF,0x00,0x00)
+        HTML_COLOUR("purple",  0x80,0x00,0x80)
+        HTML_COLOUR("fuchsia", 0xFF,0x00,0xFF)
+        HTML_COLOUR("green",   0x00,0x80,0x00)
+        HTML_COLOUR("lime",    0x00,0xFF,0x00)
+        HTML_COLOUR("olive",   0x80,0x80,0x00)
+        HTML_COLOUR("yellow",  0xFF,0xFF,0x00)
+        HTML_COLOUR("navy",    0x00,0x00,0x80)
+        HTML_COLOUR("blue",    0x00,0x00,0xFF)
+        HTML_COLOUR("teal",    0x00,0x80,0x80)
+        HTML_COLOUR("aqua",    0x00,0xFF,0xFF)
+        #undef HTML_COLOUR
+        return FALSE;
+    }
+}
+
+bool wxHtmlTag::GetParamAsInt(const wxString& par, int *clr) const
+{
+    if (!HasParam(par)) return FALSE;
+    long i;
+    bool succ = GetParam(par).ToLong(&i);
+    *clr = (int)i;
+    return succ;
+}
+
+wxString wxHtmlTag::GetAllParams() const
+{
+    // VS: this function is for backward compatiblity only, 
+    //     never used by wxHTML
+    wxString s;
+    size_t cnt = m_ParamNames.GetCount();
+    for (size_t i = 0; i < cnt; i++)
+    {
+        s << m_ParamNames[i];
+        s << wxT('=');
+        if (m_ParamValues[i].Find(wxT('"')) != wxNOT_FOUND)
+            s << wxT('\'') << m_ParamValues[i] << wxT('\'');
+        else
+            s << wxT('"') << m_ParamValues[i] << wxT('"');
+    }
+    return s;

 }
 
 #endif
 }
 
 #endif