Ensure wxCharTypeBuffer data is NUL-terminated after extend() call.

[wxWidgets.git] / include / wx / tokenzr.h

diff --git a/include/wx/tokenzr.h b/include/wx/tokenzr.h
index 1668cd47ba6d48ee0bdc262e89100f56dd572ce8..0708ebff1334f8dcb8f061535a3fe91f5a633170 100644 (file)
--- a/include/wx/tokenzr.h
+++ b/include/wx/tokenzr.h
@@ -1,8 +1,8 @@
 /////////////////////////////////////////////////////////////////////////////
 /////////////////////////////////////////////////////////////////////////////

-// Name:        tokenzr.h
-// Purpose:     String tokenizer
+// Name:        wx/tokenzr.h
+// Purpose:     String tokenizer - a C++ replacement for strtok(3)

 // Author:      Guilhem Lavaux
 // Author:      Guilhem Lavaux

-// Modified by:
+// Modified by: (or rather rewritten by) Vadim Zeitlin

 // Created:     04/22/98
 // RCS-ID:      $Id$
 // Copyright:   (c) Guilhem Lavaux
 // Created:     04/22/98
 // RCS-ID:      $Id$
 // Copyright:   (c) Guilhem Lavaux


@@ -12,46 +12,138 @@
 #ifndef _WX_TOKENZRH
 #define _WX_TOKENZRH
 
 #ifndef _WX_TOKENZRH
 #define _WX_TOKENZRH
 

-#ifdef __GNUG__
-#pragma interface
-#endif
-

 #include "wx/object.h"
 #include "wx/string.h"
 #include "wx/object.h"
 #include "wx/string.h"

-#include "wx/filefn.h"
+#include "wx/arrstr.h"
+
+// ----------------------------------------------------------------------------
+// constants
+// ----------------------------------------------------------------------------
+
+// default: delimiters are usual white space characters
+#define wxDEFAULT_DELIMITERS (wxT(" \t\r\n"))
+
+// wxStringTokenizer mode flags which determine its behaviour
+enum wxStringTokenizerMode
+{
+    wxTOKEN_INVALID = -1,   // set by def ctor until SetString() is called
+    wxTOKEN_DEFAULT,        // strtok() for whitespace delims, RET_EMPTY else
+    wxTOKEN_RET_EMPTY,      // return empty token in the middle of the string
+    wxTOKEN_RET_EMPTY_ALL,  // return trailing empty tokens too
+    wxTOKEN_RET_DELIMS,     // return the delim with token (implies RET_EMPTY)
+    wxTOKEN_STRTOK          // behave exactly like strtok(3)
+};
+
+// ----------------------------------------------------------------------------
+// wxStringTokenizer: replaces infamous strtok() and has some other features
+// ----------------------------------------------------------------------------

 
 

-class WXDLLEXPORT wxStringTokenizer : public wxObject
+class WXDLLIMPEXP_BASE wxStringTokenizer : public wxObject

 {
 public:
 {
 public:

-    wxStringTokenizer(const wxString& to_tokenize,
-                      const wxString& delims = " \t\r\n",
-                      bool ret_delim = FALSE);
-    wxStringTokenizer() { m_retdelims = FALSE;}
-    virtual ~wxStringTokenizer();
+    // ctors and initializers
+        // default ctor, call SetString() later
+    wxStringTokenizer() { m_mode = wxTOKEN_INVALID; }
+        // ctor which gives us the string
+    wxStringTokenizer(const wxString& str,
+                      const wxString& delims = wxDEFAULT_DELIMITERS,
+                      wxStringTokenizerMode mode = wxTOKEN_DEFAULT);
+
+        // args are same as for the non default ctor above
+    void SetString(const wxString& str,
+                   const wxString& delims = wxDEFAULT_DELIMITERS,
+                   wxStringTokenizerMode mode = wxTOKEN_DEFAULT);
+
+        // reinitialize the tokenizer with the same delimiters/mode
+    void Reinit(const wxString& str);
+
+    // tokens access
+        // return the number of remaining tokens
+    size_t CountTokens() const;
+        // did we reach the end of the string?
+    bool HasMoreTokens() const;
+        // get the next token, will return empty string if !HasMoreTokens()
+    wxString GetNextToken();
+        // get the delimiter which terminated the token last retrieved by
+        // GetNextToken() or NUL if there had been no tokens yet or the last
+        // one wasn't terminated (but ran to the end of the string)
+    wxChar GetLastDelimiter() const { return m_lastDelim; }
+
+    // get current tokenizer state
+        // returns the part of the string which remains to tokenize (*not* the
+        // initial string)
+    wxString GetString() const { return wxString(m_pos, m_string.end()); }
+
+        // returns the current position (i.e. one index after the last
+        // returned token or 0 if GetNextToken() has never been called) in the
+        // original string
+    size_t GetPosition() const { return m_pos - m_string.begin(); }
+
+    // misc
+        // get the current mode - can be different from the one passed to the
+        // ctor if it was wxTOKEN_DEFAULT
+    wxStringTokenizerMode GetMode() const { return m_mode; }
+        // do we return empty tokens?
+    bool AllowEmpty() const { return m_mode != wxTOKEN_STRTOK; }

 
 

-    int CountTokens() const;
-    bool HasMoreTokens();

 
 

-    wxString NextToken();
-    wxString GetNextToken() { return NextToken(); };
+    // backwards compatibility section from now on
+    // -------------------------------------------

 
 

-    wxString GetString() const { return m_string; }
+    // for compatibility only, use GetNextToken() instead
+    wxString NextToken() { return GetNextToken(); }

 
 

+    // compatibility only, don't use

     void SetString(const wxString& to_tokenize,
     void SetString(const wxString& to_tokenize,

-                   const wxString& delims = " \t\r\n",
-                   bool ret_delim = FALSE)
+                   const wxString& delims,
+                   bool WXUNUSED(ret_delim))

     {
     {

-        m_string = to_tokenize;
-        m_delims = delims;
-        m_retdelims = ret_delim;
+        SetString(to_tokenize, delims, wxTOKEN_RET_DELIMS);
+    }
+
+    wxStringTokenizer(const wxString& to_tokenize,
+                      const wxString& delims,
+                      bool ret_delim)
+    {
+        SetString(to_tokenize, delims, ret_delim);

     }
 
 protected:
     }
 
 protected:

-    off_t FindDelims(const wxString& str, const wxString& delims) const;
-    void EatLeadingDelims();
+    bool IsOk() const { return m_mode != wxTOKEN_INVALID; }
+
+    bool DoHasMoreTokens() const;
+
+    enum MoreTokensState
+    {
+        MoreTokens_Unknown,
+        MoreTokens_Yes,
+        MoreTokens_No
+    };

 
 

-    wxString m_string, m_delims;
-    bool m_retdelims;
+    MoreTokensState m_hasMoreTokens;
+
+    wxString m_string;              // the string we tokenize
+    wxString::const_iterator m_stringEnd;
+    // FIXME-UTF8: use wxWcharBuffer
+    wxWxCharBuffer m_delims;        // all possible delimiters
+    size_t m_delimsLen;
+
+    wxString::const_iterator m_pos; // the current position in m_string
+
+    wxStringTokenizerMode m_mode;   // see wxTOKEN_XXX values
+
+    wxChar   m_lastDelim;           // delimiter after last token or '\0'

 };
 
 };
 

+// ----------------------------------------------------------------------------
+// convenience function which returns all tokens at once
+// ----------------------------------------------------------------------------
+
+// the function takes the same parameters as wxStringTokenizer ctor and returns
+// the array containing all tokens
+wxArrayString WXDLLIMPEXP_BASE
+wxStringTokenize(const wxString& str,
+                 const wxString& delims = wxDEFAULT_DELIMITERS,
+                 wxStringTokenizerMode mode = wxTOKEN_DEFAULT);
+

 #endif // _WX_TOKENZRH
 #endif // _WX_TOKENZRH