/////////////////////////////////////////////////////////////////////////////
-// Name: tokenzr.h
-// Purpose: String tokenizer
+// Name: wx/tokenzr.h
+// Purpose: String tokenizer - a C++ replacement for strtok(3)
// Author: Guilhem Lavaux
-// Modified by:
+// Modified by: (or rather rewritten by) Vadim Zeitlin
// Created: 04/22/98
// RCS-ID: $Id$
// Copyright: (c) Guilhem Lavaux
#ifndef _WX_TOKENZRH
#define _WX_TOKENZRH
-#ifdef __GNUG__
-#pragma interface
+#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
+ #pragma interface "tokenzr.h"
#endif
#include "wx/object.h"
#include "wx/string.h"
-#include "wx/filefn.h"
+#include "wx/arrstr.h"
-class WXDLLEXPORT wxStringTokenizer : public wxObject
+// ----------------------------------------------------------------------------
+// constants
+// ----------------------------------------------------------------------------
+
+// default: delimiters are usual white space characters
+#define wxDEFAULT_DELIMITERS (_T(" \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 WXDLLIMPEXP_BASE wxStringTokenizer : public wxObject
{
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);
- int CountTokens() count;
- bool HasMoreTokens();
+ // reinitialize the tokenizer with the same delimiters/mode
+ void Reinit(const wxString& str);
- wxString NextToken();
- wxString GetNextToken() { return NextToken(); };
+ // tokens access
+ // count them
+ 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 current tokenizer state
+ // returns the part of the string which remains to tokenize (*not* the
+ // initial string)
wxString GetString() const { return m_string; }
+ // 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; }
+
+ // 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; }
+
+ // backwards compatibility section from now on
+ // -------------------------------------------
+
+ // for compatibility only, use GetNextToken() instead
+ wxString NextToken() { return GetNextToken(); }
+
+ // compatibility only, don't use
void SetString(const wxString& to_tokenize,
- const wxString& delims = " \t\r\n",
- bool ret_delim = FALSE)
+ const wxString& delims,
+ bool WXUNUSED(ret_delim))
+ {
+ SetString(to_tokenize, delims, wxTOKEN_RET_DELIMS);
+ }
+
+ wxStringTokenizer(const wxString& to_tokenize,
+ const wxString& delims,
+ bool ret_delim)
{
- m_string = to_tokenize;
- m_delims = delims;
- m_retdelims = ret_delim;
+ SetString(to_tokenize, delims, ret_delim);
}
protected:
- off_t FindDelims(const wxString& str, const wxString& delims);
- void EatLeadingDelims();
+ bool IsOk() const { return m_mode != wxTOKEN_INVALID; }
- wxString m_string, m_delims;
- bool m_retdelims;
+ wxString m_string, // the (rest of) string to tokenize
+ m_delims; // all delimiters
+
+ size_t m_pos; // the position in the original string
+
+ wxStringTokenizerMode m_mode; // see wxTOKEN_XXX values
+
+ bool m_hasMore; // do we have more (possible empty) tokens?
};
+// ----------------------------------------------------------------------------
+// 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