问题
I'm looking for a Win32 equivalent of the .Net Encoding.GetEncoding Method to be used in Delphi7.
What I would like to achieve is to convert a Codepage ID (ie.: 28592) to a Codepage name (iso-8859-2 in this case).
I've found a Win32 function called GetCPInfoEx, but that returns a long CodePage name, and I need the short one, like the ones listed on this page: (see Name column) http://msdn.microsoft.com/en-us/library/system.text.encoding.aspx
Thank you!
回答1:
Here is my look-up table with code, feel free to use it.
type
TCPData = record
CPID: Integer;
CPName: String;
end;
const
MaxEncodings = 140;
Encodings: Array[0..MaxEncodings - 1] of TCPData =
(
(CPID: 37; CPName: 'IBM037'),
(CPID: 437; CPName: 'IBM437'),
(CPID: 500; CPName: 'IBM500'),
(CPID: 708; CPName: 'ASMO-708'),
(CPID: 720; CPName: 'DOS-720'),
(CPID: 737; CPName: 'ibm737'),
(CPID: 775; CPName: 'ibm775'),
(CPID: 850; CPName: 'ibm850'),
(CPID: 852; CPName: 'ibm852'),
(CPID: 855; CPName: 'IBM855'),
(CPID: 857; CPName: 'ibm857'),
(CPID: 858; CPName: 'IBM00858'),
(CPID: 860; CPName: 'IBM860'),
(CPID: 861; CPName: 'ibm861'),
(CPID: 862; CPName: 'DOS-862'),
(CPID: 863; CPName: 'IBM863'),
(CPID: 864; CPName: 'IBM864'),
(CPID: 865; CPName: 'IBM865'),
(CPID: 866; CPName: 'cp866'),
(CPID: 869; CPName: 'ibm869'),
(CPID: 870; CPName: 'IBM870'),
(CPID: 874; CPName: 'windows-874'),
(CPID: 875; CPName: 'cp875'),
(CPID: 932; CPName: 'shift_jis'),
(CPID: 936; CPName: 'gb2312'),
(CPID: 949; CPName: 'ks_c_5601-1987'),
(CPID: 950; CPName: 'big5'),
(CPID: 1026; CPName: 'IBM1026'),
(CPID: 1047; CPName: 'IBM01047'),
(CPID: 1140; CPName: 'IBM01140'),
(CPID: 1141; CPName: 'IBM01141'),
(CPID: 1142; CPName: 'IBM01142'),
(CPID: 1143; CPName: 'IBM01143'),
(CPID: 1144; CPName: 'IBM01144'),
(CPID: 1145; CPName: 'IBM01145'),
(CPID: 1146; CPName: 'IBM01146'),
(CPID: 1147; CPName: 'IBM01147'),
(CPID: 1148; CPName: 'IBM01148'),
(CPID: 1149; CPName: 'IBM01149'),
(CPID: 1200; CPName: 'utf-16'),
(CPID: 1201; CPName: 'unicodeFFFE'),
(CPID: 1250; CPName: 'windows-1250'),
(CPID: 1251; CPName: 'windows-1251'),
(CPID: 1252; CPName: 'Windows-1252'),
(CPID: 1253; CPName: 'windows-1253'),
(CPID: 1254; CPName: 'windows-1254'),
(CPID: 1255; CPName: 'windows-1255'),
(CPID: 1256; CPName: 'windows-1256'),
(CPID: 1257; CPName: 'windows-1257'),
(CPID: 1258; CPName: 'windows-1258'),
(CPID: 1361; CPName: 'Johab'),
(CPID: 10000; CPName: 'macintosh'),
(CPID: 10001; CPName: 'x-mac-japanese'),
(CPID: 10002; CPName: 'x-mac-chinesetrad'),
(CPID: 10003; CPName: 'x-mac-korean'),
(CPID: 10004; CPName: 'x-mac-arabic'),
(CPID: 10005; CPName: 'x-mac-hebrew'),
(CPID: 10006; CPName: 'x-mac-greek'),
(CPID: 10007; CPName: 'x-mac-cyrillic'),
(CPID: 10008; CPName: 'x-mac-chinesesimp'),
(CPID: 10010; CPName: 'x-mac-romanian'),
(CPID: 10017; CPName: 'x-mac-ukrainian'),
(CPID: 10021; CPName: 'x-mac-thai'),
(CPID: 10029; CPName: 'x-mac-ce'),
(CPID: 10079; CPName: 'x-mac-icelandic'),
(CPID: 10081; CPName: 'x-mac-turkish'),
(CPID: 10082; CPName: 'x-mac-croatian'),
(CPID: 12000; CPName: 'utf-32'),
(CPID: 12001; CPName: 'utf-32BE'),
(CPID: 20000; CPName: 'x-Chinese-CNS'),
(CPID: 20001; CPName: 'x-cp20001'),
(CPID: 20002; CPName: 'x-Chinese-Eten'),
(CPID: 20003; CPName: 'x-cp20003'),
(CPID: 20004; CPName: 'x-cp20004'),
(CPID: 20005; CPName: 'x-cp20005'),
(CPID: 20105; CPName: 'x-IA5'),
(CPID: 20106; CPName: 'x-IA5-German'),
(CPID: 20107; CPName: 'x-IA5-Swedish'),
(CPID: 20108; CPName: 'x-IA5-Norwegian'),
(CPID: 20127; CPName: 'us-ascii'),
(CPID: 20261; CPName: 'x-cp20261'),
(CPID: 20269; CPName: 'x-cp20269'),
(CPID: 20273; CPName: 'IBM273'),
(CPID: 20277; CPName: 'IBM277'),
(CPID: 20278; CPName: 'IBM278'),
(CPID: 20280; CPName: 'IBM280'),
(CPID: 20284; CPName: 'IBM284'),
(CPID: 20285; CPName: 'IBM285'),
(CPID: 20290; CPName: 'IBM290'),
(CPID: 20297; CPName: 'IBM297'),
(CPID: 20420; CPName: 'IBM420'),
(CPID: 20423; CPName: 'IBM423'),
(CPID: 20424; CPName: 'IBM424'),
(CPID: 20833; CPName: 'x-EBCDIC-KoreanExtended'),
(CPID: 20838; CPName: 'IBM-Thai'),
(CPID: 20866; CPName: 'koi8-r'),
(CPID: 20871; CPName: 'IBM871'),
(CPID: 20880; CPName: 'IBM880'),
(CPID: 20905; CPName: 'IBM905'),
(CPID: 20924; CPName: 'IBM00924'),
(CPID: 20932; CPName: 'EUC-JP'),
(CPID: 20936; CPName: 'x-cp20936'),
(CPID: 20949; CPName: 'x-cp20949'),
(CPID: 21025; CPName: 'cp1025'),
(CPID: 21866; CPName: 'koi8-u'),
(CPID: 28591; CPName: 'iso-8859-1'),
(CPID: 28592; CPName: 'iso-8859-2'),
(CPID: 28593; CPName: 'iso-8859-3'),
(CPID: 28594; CPName: 'iso-8859-4'),
(CPID: 28595; CPName: 'iso-8859-5'),
(CPID: 28596; CPName: 'iso-8859-6'),
(CPID: 28597; CPName: 'iso-8859-7'),
(CPID: 28598; CPName: 'iso-8859-8'),
(CPID: 28599; CPName: 'iso-8859-9'),
(CPID: 28603; CPName: 'iso-8859-13'),
(CPID: 28605; CPName: 'iso-8859-15'),
(CPID: 29001; CPName: 'x-Europa'),
(CPID: 38598; CPName: 'iso-8859-8-i'),
(CPID: 50220; CPName: 'iso-2022-jp'),
(CPID: 50221; CPName: 'csISO2022JP'),
(CPID: 50222; CPName: 'iso-2022-jp'),
(CPID: 50225; CPName: 'iso-2022-kr'),
(CPID: 50227; CPName: 'x-cp50227'),
(CPID: 51932; CPName: 'euc-jp'),
(CPID: 51936; CPName: 'EUC-CN'),
(CPID: 51949; CPName: 'euc-kr'),
(CPID: 52936; CPName: 'hz-gb-2312'),
(CPID: 54936; CPName: 'GB18030'),
(CPID: 57002; CPName: 'x-iscii-de'),
(CPID: 57003; CPName: 'x-iscii-be'),
(CPID: 57004; CPName: 'x-iscii-ta'),
(CPID: 57005; CPName: 'x-iscii-te'),
(CPID: 57006; CPName: 'x-iscii-as'),
(CPID: 57007; CPName: 'x-iscii-or'),
(CPID: 57008; CPName: 'x-iscii-ka'),
(CPID: 57009; CPName: 'x-iscii-ma'),
(CPID: 57010; CPName: 'x-iscii-gu'),
(CPID: 57011; CPName: 'x-iscii-pa'),
(CPID: 65000; CPName: 'utf-7'),
(CPID: 65001; CPName: 'utf-8')
);
implementation
function GetEncoding(CPID: Integer): String;
var
I: Integer;
begin
Result := 'iso-8859-2'; //put the default encoding here
for I := 0 to MaxEncodings - 1 do
if Encodings[I].CPID = CPID then
begin
Result := Encodings[I].CPName;
break;
end;
end;
Thanks to everyone for the answers, but this proved to be the only usable solution in this case...
回答2:
Are you looking for the IANA official names? Since you want them correlated to Windows CP identifiers, I don't think you can do better than this table.
回答3:
You need to use IMultiLanguage::GetCodePageInfo, exported as part of MLang.dll. It's included with IE4 and above. You can create the import library yourself using Delphi's Import Component... command, or grab MultiLanguage_TLB.pas from Colin Wilson's low-level utilities package.
function CodePageToCharSet(ACodePage: Cardinal): string;
var
MimeCPInfo: TMimeCPInfo;
MultiLanguage: IMultiLanguage;
begin
if Succeeded(CoCreateInstance(CLASS_CMultiLanguage, nil, CLSCTX_INPROC_SERVER,
IID_IMultiLanguage, MultiLanguage)) and
Succeeded(MultiLanguage.GetCodePageInfo(ACodePage, MimeCPInfo)) then
Result := string(MimeCPInfo.wszWebCharset)
else
Result := 'US-ASCII';
end;
回答4:
I think you mean the LCIDToLocaleName function.
回答5:
type
TCPInfoEx = record
MaxCharSize: UINT;
DefaultChar: array[0..MAX_DEFAULTCHAR - 1] of Byte;
LeadByte: array[0..MAX_LEADBYTES - 1] of Byte;
UnicodeDefaultChar: WideChar;
CodePage: UINT;
CodePageName: array[0..MAX_PATH - 1] of Char;
end;
function GetCPInfoEx(CodePage: UINT; Flags: DWORD; var lpCPInfoEx: TCPInfoEx): BOOL;
stdcall; external 'kernel32.dll' name 'GetCPInfoExA';
procedure TForm1.Button1Click(Sender: TObject);
var
CPInfoEx: TCPInfoEx;
begin
if GetCPInfoEx(28592, 0, CPInfoEx) then
ShowMessage(CPInfoEx.CodePageName);
end;
回答6:
Use search, Luke. Just old good text search. In RegEdit. For any cp short name you'd need :-) and remember, those IDs are really Internet MIME ids
HKEY_CLASSES_ROOT\MIME\DataBase
http://msdn.microsoft.com/en-us/library/ms775147.aspx this page hints that probably - just probably - that key exists since MSIE4 - in other words since Windows 98
uses
{$IfDef MSWINDOWS}Registry, Windows, {$EndIf}
AnsiStrings;
/// todo - add implementation for non-Windows platforms
/// if anyone would need it :-)
/// probably via http://sourceforge.net/projects/natspec/
function CharSetByCodePage(const cp:Word): AnsiString;
{$IfDef MSWINDOWS }
var reg: TRegistry;
begin
// HKEY_CLASSES_ROOT\MIME\DataBase\Codepage
reg := TRegistry.Create(KEY_READ);
try
reg.RootKey := HKEY_CLASSES_ROOT;
if reg.OpenKeyReadOnly('MIME\DataBase\Codepage\' + IntToStr(cp)) then begin
Result := Trim(AnsiString(reg.ReadString('WebCharset'))); // This key prevails, see #1251 for example
if Result = '' then
Result := Trim(AnsiString(reg.ReadString('BodyCharset')));
if Result > '' then exit;
end;
finally
reg.Free;
end;
Raise EZXSaveException.Create('No charset (MIME id) found for codepage '+IntToStr(cp));
end;
{$Else}
begin
Raise EZXSaveException.Create('Cannot get charset by numeric codepage on this platform.');
//not implemented, perhaps http://sourceforge.net/projects/natspec/ ?
end;
{$EndIf}
来源：https://stackoverflow.com/questions/2574289/codepage-id-to-codepage-name-getencoding-equivalent-in-delphi