-// ----------------------------------------------------------------------------
-// standard file conversion
-// ----------------------------------------------------------------------------
-
-WXDLLEXPORT_DATA(wxMBConv_file) wxConv_file;
-
-// just use the libc conversion for now
-size_t wxMBConv_file::MB2WC(wchar_t *buf, const char *psz, size_t n) const
-{
- return wxMB2WC(buf, psz, n);
-}
-
-size_t wxMBConv_file::WC2MB(char *buf, const wchar_t *psz, size_t n) const
-{
- return wxWC2MB(buf, psz, n);
-}
-
-// ----------------------------------------------------------------------------
-// standard gdk conversion
-// ----------------------------------------------------------------------------
-
-#ifdef __WXGTK__
-WXDLLEXPORT_DATA(wxMBConv_gdk) wxConv_gdk;
-
-#include <gdk/gdk.h>
-
-size_t wxMBConv_gdk::MB2WC(wchar_t *buf, const char *psz, size_t n) const
-{
- if (buf) {
- return gdk_mbstowcs((GdkWChar *)buf, psz, n);
- } else {
- GdkWChar *nbuf = new GdkWChar[n=strlen(psz)];
- size_t len = gdk_mbstowcs(nbuf, psz, n);
- delete [] nbuf;
- return len;
- }
-}
-
-size_t wxMBConv_gdk::WC2MB(char *buf, const wchar_t *psz, size_t n) const
-{
- char *mbstr = gdk_wcstombs((GdkWChar *)psz);
- size_t len = mbstr ? strlen(mbstr) : 0;
- if (buf) {
- if (len > n) len = n;
- memcpy(buf, psz, len);
- if (len < n) buf[len] = 0;
- }
- return len;
-}
-#endif
-
-// ----------------------------------------------------------------------------
-// UTF-7
-// ----------------------------------------------------------------------------
-
-WXDLLEXPORT_DATA(wxMBConv_UTF7) wxConv_UTF7;
-
-// TODO: write actual implementations of UTF-7 here
-size_t wxMBConv_UTF7::MB2WC(wchar_t *buf, const char *psz, size_t n) const
-{
- return 0;
-}
-
-size_t wxMBConv_UTF7::WC2MB(char *buf, const wchar_t *psz, size_t n) const
-{
- return 0;
-}
-
-// ----------------------------------------------------------------------------
-// UTF-8
-// ----------------------------------------------------------------------------
-
-WXDLLEXPORT_DATA(wxMBConv_UTF8) wxConv_UTF8;
-
-// TODO: write actual implementations of UTF-8 here
-size_t wxMBConv_UTF8::MB2WC(wchar_t *buf, const char *psz, size_t n) const
-{
- return wxMB2WC(buf, psz, n);
-}
-
-size_t wxMBConv_UTF8::WC2MB(char *buf, const wchar_t *psz, size_t n) const
-{
- return wxWC2MB(buf, psz, n);
-}
-
-// ----------------------------------------------------------------------------
-// specified character set
-// ----------------------------------------------------------------------------
-
-class wxCharacterSet
-{
-public:
- wxArrayString names;
- wchar_t *data;
-};
-
-#ifndef WX_PRECOMP
- #include "wx/dynarray.h"
- #include "wx/filefn.h"
- #include "wx/textfile.h"
- #include "wx/tokenzr.h"
- #include "wx/utils.h"
-#endif
-
-WX_DECLARE_OBJARRAY(wxCharacterSet, wxCSArray);
-#include "wx/arrimpl.cpp"
-WX_DEFINE_OBJARRAY(wxCSArray);
-
-static wxCSArray wxCharsets;
-
-static void wxLoadCharacterSets(void)
-{
- static bool already_loaded = FALSE;
-
-#if defined(__UNIX__) && wxUSE_UNICODE
- // search through files in /usr/share/i18n/charmaps
- for (wxString fname = ::wxFindFirstFile(_T("/usr/share/i18n/charmaps/*"));
- !fname.IsEmpty();
- fname = ::wxFindNextFile()) {
- wxTextFile cmap(fname);
- if (cmap.Open()) {
- wxCharacterSet *cset = new wxCharacterSet;
- wxString comchar,escchar;
- bool in_charset = FALSE;
-
- wxPrintf(_T("yup, loaded %s\n"),fname.c_str());
-
- for (wxString line = cmap.GetFirstLine();
- !cmap.Eof();
- line = cmap.GetNextLine()) {
- wxPrintf(_T("line contents: %s\n"),line.c_str());
- wxStringTokenizer token(line);
- wxString cmd = token.GetNextToken();
- if (cmd == comchar) {
- if (token.GetNextToken() == _T("alias")) {
- wxStringTokenizer names(token.GetNextToken(),_T("/"));
- wxString name;
- while (!(name = names.GetNextToken()).IsEmpty())
- cset->names.Add(name);
- }
- }
- else if (cmd == _T("<code_set_name>"))
- cset->names.Add(token.GetNextToken());
- else if (cmd == _T("<comment_char>"))
- comchar = token.GetNextToken();
- else if (cmd == _T("<escape_char>"))
- escchar = token.GetNextToken();
- else if (cmd == _T("<mb_cur_min")) {
- delete cset;
- goto forget_it; // we don't support multibyte charsets ourselves (yet)
- }
- else if (cmd == _T("CHARMAP")) {
- cset->data = (wchar_t *)calloc(256, sizeof(wxChar));
- in_charset = TRUE;
- }
- else if (cmd == _T("END")) {
- if (token.GetNextToken() == _T("CHARMAP"))
- in_charset = FALSE;
- }
- else if (in_charset) {
- // format: <NUL> /x00 <U0000> NULL (NUL)
- wxString hex = token.GetNextToken();
- wxString uni = token.GetNextToken();
- // just assume that we've got the right format
- int pos = ::wxHexToDec(hex.Mid(2,2));
- unsigned long uni1 = ::wxHexToDec(uni.Mid(2,2));
- unsigned long uni2 = ::wxHexToDec(uni.Mid(4,2));
- cset->data[pos] = (uni1 << 16) | uni2;
- }
- }
- cset->names.Shrink();
- wxCharsets.Add(cset);
- forget_it:
- continue;
- }
- }
-#endif
- wxCharsets.Shrink();
- already_loaded = TRUE;
-}
-
-static wxCharacterSet *wxFindCharacterSet(const wxString& charset)
-{
- for (size_t n=0; n<wxCharsets.GetCount(); n++)
- if (wxCharsets[n].names.Index(charset) != wxNOT_FOUND)
- return &(wxCharsets[n]);
- return (wxCharacterSet *)NULL;
-}
-
-WXDLLEXPORT_DATA(wxCSConv) wxConv_local((const wxChar *)NULL);
-
-wxCSConv::wxCSConv(const wxChar *charset)
-{
- wxLoadCharacterSets();
- if (!charset) {
-#ifdef __UNIX__
- wxChar *lang = wxGetenv(_T("LANG"));
- wxChar *dot = wxStrchr(lang, _T('.'));
- if (dot) charset = dot+1;
-#endif
- }
- cset = (wxCharacterSet *) NULL;
-
-#ifdef __UNIX__
- // first, convert the character set name to standard form
- wxString codeset;
- if (wxString(charset,3) == _T("ISO")) {
- // make sure it's represented in the standard form: ISO_8859-1
- codeset = _T("ISO_");
- charset += 3;
- if ((*charset == _T('-')) || (*charset == _T('_'))) charset++;
- if (wxStrlen(charset)>4) {
- if (wxString(charset,4) == _T("8859")) {
- codeset << _T("8859-");
- if (*charset == _T('-')) charset++;
- }
- }
- }
- codeset << charset;
- codeset.MakeUpper();
- cset = wxFindCharacterSet(codeset);
-#endif
-}
-
-wxCSConv::~wxCSConv(void)
-{
-}
-
-size_t wxCSConv::MB2WC(wchar_t *buf, const char *psz, size_t n) const
-{
- if (buf) {
- if (cset) {
- for (size_t c=0; c<=n; c++)
- buf[c] = cset->data[psz[c]];
- } else {
- // latin-1 (direct)
- for (size_t c=0; c<=n; c++)
- buf[c] = psz[c];