X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/435595e02f50afafdaf324e4e52b229c46f174d5..dafb7332056d68f0daa76a77ca6bbcb28e1a5a00:/src/common/string.cpp?ds=inline diff --git a/src/common/string.cpp b/src/common/string.cpp index 539b631152..89cc7772f1 100644 --- a/src/common/string.cpp +++ b/src/common/string.cpp @@ -259,11 +259,11 @@ wxString::wxString(const char *psz, wxMBConv& conv, size_t nLength) size_t nLen = psz ? conv.MB2WC((wchar_t *) NULL, psz, 0) : 0; // nLength is number of *Unicode* characters here! - if (nLen > nLength) + if ((nLen != (size_t)-1) && (nLen > nLength)) nLen = nLength; // empty? - if ( nLen != 0 ) { + if ( (nLen != 0) && (nLen != (size_t)-1) ) { AllocBuffer(nLen); conv.MB2WC(m_pchData, psz, nLen); } @@ -274,6 +274,7 @@ wxString::wxString(const char *psz, wxMBConv& conv, size_t nLength) #else +#if wxUSE_WCHAR_T // from wide string wxString::wxString(const wchar_t *pwz) { @@ -281,7 +282,7 @@ wxString::wxString(const wchar_t *pwz) size_t nLen = pwz ? wxWC2MB((char *) NULL, pwz, 0) : 0; // empty? - if ( nLen != 0 ) { + if ( (nLen != 0) && (nLen != (size_t)-1) ) { AllocBuffer(nLen); wxWC2MB(m_pchData, pwz, nLen); } @@ -289,6 +290,7 @@ wxString::wxString(const wchar_t *pwz) Init(); } } +#endif #endif @@ -495,12 +497,14 @@ wxString& wxString::operator=(const unsigned char* psz) return *this; } +#if wxUSE_WCHAR_T wxString& wxString::operator=(const wchar_t *pwz) { wxString str(pwz); *this = str; return *this; } +#endif #endif @@ -1164,7 +1168,7 @@ int wxString::PrintfV(const wxChar* pszFormat, va_list argptr) char *val = va_arg(argptr, char *); #if wxUSE_UNICODE // ASCII->Unicode constructor handles max_width right - wxString s(val, wxConv_libc, max_width); + wxString s(val, wxConvLibc, max_width); #else size_t len = wxSTRING_MAXLEN; if (val) { @@ -1185,8 +1189,8 @@ int wxString::PrintfV(const wxChar* pszFormat, va_list argptr) if (s.Len() < min_width) s.Pad(min_width - s.Len(), _T(' '), adj_left); *this += s; - done = TRUE; } + done = TRUE; break; case _T('n'): if (ilen == 0) { @@ -1406,7 +1410,7 @@ size_t wxString::rfind(const wxString& str, size_t nStart) const wxASSERT( str.GetStringData()->IsValid() ); wxASSERT( nStart <= Len() ); - // # could be quicker than that + // TODO could be made much quicker than that const wxChar *p = c_str() + (nStart == npos ? Len() : nStart); while ( p >= c_str() + str.Len() ) { if ( wxStrncmp(p - str.Len(), str, str.Len()) == 0 ) @@ -1421,19 +1425,131 @@ size_t wxString::rfind(const wxString& str, size_t nStart) const #if !defined(__VISUALC__) || defined(__WIN32__) size_t wxString::rfind(const wxChar* sz, size_t nStart, size_t n) const { - return rfind(wxString(sz, n == npos ? 0 : n), nStart); + return rfind(wxString(sz, n == npos ? 0 : n), nStart); } size_t wxString::rfind(wxChar ch, size_t nStart) const { - wxASSERT( nStart <= Len() ); + if ( nStart == npos ) + { + nStart = Len(); + } + else + { + wxASSERT( nStart <= Len() ); + } - const wxChar *p = wxStrrchr(c_str() + nStart, ch); + const wxChar *p = wxStrrchr(c_str(), ch); - return p == NULL ? npos : p - c_str(); + if ( p == NULL ) + return npos; + + size_t result = p - c_str(); + return ( result > nStart ) ? npos : result; } #endif // VC++ 1.5 +size_t wxString::find_first_of(const wxChar* sz, size_t nStart) const +{ + const wxChar *start = c_str() + nStart; + const wxChar *firstOf = wxStrpbrk(start, sz); + if ( firstOf ) + return firstOf - start; + else + return npos; +} + +size_t wxString::find_last_of(const wxChar* sz, size_t nStart) const +{ + if ( nStart == npos ) + { + nStart = Len(); + } + else + { + wxASSERT( nStart <= Len() ); + } + + for ( const wxChar *p = c_str() + length() - 1; p >= c_str(); p-- ) + { + if ( wxStrchr(sz, *p) ) + return p - c_str(); + } + + return npos; +} + +size_t wxString::find_first_not_of(const wxChar* sz, size_t nStart) const +{ + if ( nStart == npos ) + { + nStart = Len(); + } + else + { + wxASSERT( nStart <= Len() ); + } + + size_t nAccept = wxStrspn(c_str() + nStart, sz); + if ( nAccept >= length() - nStart ) + return npos; + else + return nAccept; +} + +size_t wxString::find_first_not_of(wxChar ch, size_t nStart) const +{ + wxASSERT( nStart <= Len() ); + + for ( const wxChar *p = c_str() + nStart; *p; p++ ) + { + if ( *p != ch ) + return p - c_str(); + } + + return npos; +} + +size_t wxString::find_last_not_of(const wxChar* sz, size_t nStart) const +{ + if ( nStart == npos ) + { + nStart = Len(); + } + else + { + wxASSERT( nStart <= Len() ); + } + + for ( const wxChar *p = c_str() + nStart - 1; p >= c_str(); p-- ) + { + if ( !wxStrchr(sz, *p) ) + return p - c_str(); + } + + return npos; +} + +size_t wxString::find_last_not_of(wxChar ch, size_t nStart) const +{ + if ( nStart == npos ) + { + nStart = Len(); + } + else + { + wxASSERT( nStart <= Len() ); + } + + for ( const wxChar *p = c_str() + nStart - 1; p >= c_str(); p-- ) + { + if ( *p != ch ) + return p - c_str(); + } + + return npos; +} + wxString wxString::substr(size_t nStart, size_t nLen) const { // npos means 'take all' @@ -1801,13 +1917,18 @@ void wxArrayString::DoSort() // MBConv // ============================================================================ -WXDLLEXPORT_DATA(wxMBConv *) wxConv_current = &wxConv_libc; +WXDLLEXPORT_DATA(wxMBConv *) wxConvCurrent = &wxConvLibc; +#if !wxUSE_WCHAR_T +WXDLLEXPORT_DATA(wxMBConv) wxConvLibc, wxConvFile; +#endif + +#if wxUSE_WCHAR_T // ---------------------------------------------------------------------------- // standard libc conversion // ---------------------------------------------------------------------------- -WXDLLEXPORT_DATA(wxMBConv) wxConv_libc; +WXDLLEXPORT_DATA(wxMBConv) wxConvLibc; size_t wxMBConv::MB2WC(wchar_t *buf, const char *psz, size_t n) const { @@ -1823,15 +1944,15 @@ size_t wxMBConv::WC2MB(char *buf, const wchar_t *psz, size_t n) const // standard file conversion // ---------------------------------------------------------------------------- -WXDLLEXPORT_DATA(wxMBConv_file) wxConv_file; +WXDLLEXPORT_DATA(wxMBConvFile) wxConvFile; // just use the libc conversion for now -size_t wxMBConv_file::MB2WC(wchar_t *buf, const char *psz, size_t n) const +size_t wxMBConvFile::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 +size_t wxMBConvFile::WC2MB(char *buf, const wchar_t *psz, size_t n) const { return wxWC2MB(buf, psz, n); } @@ -1840,12 +1961,12 @@ size_t wxMBConv_file::WC2MB(char *buf, const wchar_t *psz, size_t n) const // standard gdk conversion // ---------------------------------------------------------------------------- -#if defined(__WXGTK__) && (GTK_MINOR_VERSION > 0) -WXDLLEXPORT_DATA(wxMBConv_gdk) wxConv_gdk; +#ifdef __WXGTK12__ +WXDLLEXPORT_DATA(wxMBConvGdk) wxConvGdk; #include -size_t wxMBConv_gdk::MB2WC(wchar_t *buf, const char *psz, size_t n) const +size_t wxMBConvGdk::MB2WC(wchar_t *buf, const char *psz, size_t n) const { if (buf) { return gdk_mbstowcs((GdkWChar *)buf, psz, n); @@ -1857,7 +1978,7 @@ size_t wxMBConv_gdk::MB2WC(wchar_t *buf, const char *psz, size_t n) const } } -size_t wxMBConv_gdk::WC2MB(char *buf, const wchar_t *psz, size_t n) const +size_t wxMBConvGdk::WC2MB(char *buf, const wchar_t *psz, size_t n) const { char *mbstr = gdk_wcstombs((GdkWChar *)psz); size_t len = mbstr ? strlen(mbstr) : 0; @@ -1874,15 +1995,29 @@ size_t wxMBConv_gdk::WC2MB(char *buf, const wchar_t *psz, size_t n) const // UTF-7 // ---------------------------------------------------------------------------- -WXDLLEXPORT_DATA(wxMBConv_UTF7) wxConv_UTF7; +WXDLLEXPORT_DATA(wxMBConvUTF7) wxConvUTF7; + +#if 0 +static char utf7_setD[]="ABCDEFGHIJKLMNOPQRSTUVWXYZ" + "abcdefghijklmnopqrstuvwxyz" + "0123456789'(),-./:?"; +static char utf7_setO[]="!\"#$%&*;<=>@[]^_`{|}"; +static char utf7_setB[]="ABCDEFGHIJKLMNOPQRSTUVWXYZ" + "abcdefghijklmnopqrstuvwxyz" + "0123456789+/"; +#endif // TODO: write actual implementations of UTF-7 here -size_t wxMBConv_UTF7::MB2WC(wchar_t *buf, const char *psz, size_t n) const +size_t wxMBConvUTF7::MB2WC(wchar_t * WXUNUSED(buf), + const char * WXUNUSED(psz), + size_t WXUNUSED(n)) const { return 0; } -size_t wxMBConv_UTF7::WC2MB(char *buf, const wchar_t *psz, size_t n) const +size_t wxMBConvUTF7::WC2MB(char * WXUNUSED(buf), + const wchar_t * WXUNUSED(psz), + size_t WXUNUSED(n)) const { return 0; } @@ -1891,17 +2026,74 @@ size_t wxMBConv_UTF7::WC2MB(char *buf, const wchar_t *psz, size_t n) const // UTF-8 // ---------------------------------------------------------------------------- -WXDLLEXPORT_DATA(wxMBConv_UTF8) wxConv_UTF8; +WXDLLEXPORT_DATA(wxMBConvUTF8) wxConvUTF8; + +static unsigned long utf8_max[]={0x7f,0x7ff,0xffff,0x1fffff,0x3ffffff,0x7fffffff,0xffffffff}; -// TODO: write actual implementations of UTF-8 here -size_t wxMBConv_UTF8::MB2WC(wchar_t *buf, const char *psz, size_t n) const +size_t wxMBConvUTF8::MB2WC(wchar_t *buf, const char *psz, size_t n) const { - return wxMB2WC(buf, psz, n); + size_t len = 0; + + while (*psz && ((!buf) || (len>cnt); + while (cnt--) { + cc = *psz++; + if ((cc&0xC0)!=0x80) { + // invalid UTF-8 sequence + return (size_t)-1; + } + res=(res<<6)|(cc&0x3f); + } + if (res<=utf8_max[ocnt]) { + // illegal UTF-8 encoding + return (size_t)-1; + } + if (buf) *buf++=res; + len++; + } + } + } + if (buf && (lenutf8_max[cnt]; cnt++); + if (!cnt) { + // plain ASCII char + if (buf) *buf++=cc; + len++; + } else { + len+=cnt+1; + if (buf) { + *buf++=(-128>>cnt)|((cc>>(cnt*6))&(0x3f>>cnt)); + while (cnt--) + *buf++=0x80|((cc>>(cnt*6))&0x3f); + } + } + } + if (buf && (lennames.Add(name); - } + if (token.GetNextToken() == _T("alias")) + cset->names.Add(token.GetNextToken()); } else if (cmd == _T("")) cset->names.Add(token.GetNextToken()); @@ -1993,7 +2182,7 @@ static void wxLoadCharacterSets(void) // skip whitespace again while (wxIsEmpty(uni) && token.HasMoreTokens()) uni = token.GetNextToken(); - if ((hex.GetChar(0) == escchar) && (hex.GetChar(1) == _T('x')) && + if ((hex.Len() > 2) && (hex.GetChar(0) == escchar) && (hex.GetChar(1) == _T('x')) && (uni.Left(2) == _T("LoadNow(); // discard constness if (buf) { if (m_cset) { - for (size_t c=0; c<=n; c++) - buf[c] = m_cset->data[psz[c]]; + for (size_t c=0; cdata[(unsigned char)(psz[c])]; } else { // latin-1 (direct) - for (size_t c=0; c<=n; c++) - buf[c] = psz[c]; + for (size_t c=0; cLoadNow(); // discard constness if (buf) { if (m_cset) { - for (size_t c=0; c<=n; c++) { + for (size_t c=0; cdata[n] != psz[c]); n++); buf[c] = (n>0xff) ? '?' : n; } } else { // latin-1 (direct) - for (size_t c=0; c<=n; c++) + for (size_t c=0; c0xff) ? '?' : psz[c]; } + return n; } - return n; + return wcslen(psz); } +#endif//wxUSE_WCHAR_T + +#if wxUSE_WCHAR_T +const wxWCharBuffer wxMBConv::cMB2WC(const char *psz) const + { + if (psz) { + size_t nLen = MB2WC((wchar_t *) NULL, psz, 0); + wxWCharBuffer buf(nLen); + MB2WC(WCSTRINGCAST buf, psz, nLen); + return buf; + } else return wxWCharBuffer((wchar_t *) NULL); + } + +const wxCharBuffer wxMBConv::cWC2MB(const wchar_t *psz) const + { + if (psz) { + size_t nLen = WC2MB((char *) NULL, psz, 0); + wxCharBuffer buf(nLen); + WC2MB(MBSTRINGCAST buf, psz, nLen); + return buf; + } else return wxCharBuffer((char *) NULL); + } + +#endif//wxUSE_WCHAR_T +