Most enterprise-class database applications must be able to work with character data encoded in different character sets or provided as Unicode. AnyDAC provides seamless support for:

• different single byte client character sets and code pages, including standard ANSI;
• as well multi-byte strings such as Unicode, including UTF8, UTF16 and UCS2 encodings.

Note, that the complete Unicode support may be achieved only with the Unicode Delphi version usage - Delphi 2009 and higher. And application is free to use the Delphi native ANSI or Unicode strings. AnyDAC will perform transparent character set conversion between Delphi application and DBMS character set. 

In AnyDAC, UTF-16 encoded Unicode characters may be used for the following elements:

• SQL command text;
• SQL script text;
• metadata values;
• result set field values;
• parameter values;
• text data files;
• trace output.

AnyDAC's handling of Unicode depends on:

• the version of Delphi being used
• client character set - UTF8/UTF16 or ACP compatible
• DBMS driver.

Note: It is recommended to set the DBMS client character set to Unicode, for Delphi 2009 and higher, to avoid conversion loss. 

In general, to configure connection definitions for Unicode, it is required to set the DBMS client character set to Unicode. How this is done depends on the version of Delphi and the used DBMS driver:

Starting with Delphi 2009, TADQuery.SQL, TADCommand.CommandText, IADPhysCommand.CommandText, and other property values are Unicode encoded. The SQL command text preprocessor supports only UCS2 encoding, which means that surrogate pairs are not supported. 

Before sending SQL command text to the DBMS, AnyDAC converts it to:

• Active Code Page, if the client character set is SBCS (non Unicode). If the client character set is not compatible with ACP, conversion loss is possible.
• UTF8, if client character set is UTF8.
• UTF16, if client character set is UTF16.

To specify Unicode encoded SQL command text, just do the following: 

ADQuery1.SQL.Text := 'select ''Hello world !'' where Lang = ''RU'' and Val = ''Привет мир !''';

Starting with Delphi 2009, TADScript.SQLScripts collections can contain Unicode encoded SQL script texts. On prior Delphi versions TADScript can load Unicode encoded SQL script files, but they will be converted into ACP encoding before processing. Other than that, SQL script text processing is similar to SQL command text processing. 

To control SQL script file and log file encoding, use TADScript.ScriptOptions.FileEncoding property: 

ADScript1.ScriptOptions.FileEncoding := enUTF16;
ADScript1.SQLScriptFileName := 'u:\builddb.sql';
ADScript1.ExecuteAll;

Note: At the moment, AnyDAC does not support Unicode encoded column values in metadata result sets, returned by TADMetaInfoCommand, TADMetaInfoQuery and IADPhysMetaInfoCommand. 

Unicode encoded metadata text is supported starting from Delphi 2009. It includes:

• table names;
• resultset column names;
• stored procedure names;
• stored procedure parameter names;
• SQL command parameter names;
• SQL command macro names;
• user name;
• schema, catalog names.

To read/write Unicode string values programmatically use:

• TField.AsWideString or TField.Value properties;
• TField.AsString on Delphi 2009 and higher. On earlier Delphi versions reading / writing of AsString property may lead to conversion loss, since AsString returns ANSI strings there.

To read/write the widestring field on Delphi 2009: 

ADQuery1.Edit;
ADQuery1.Fields[0].AsWideString := ADQuery1.Fields[0].AsWideString + 'русский + english';
ADQuery1.Post;

To read/write widememo field on Delphi 2009: 

ADQuery1.Edit;
ADQuery1.FieldsByName('memo').Assign(Memo1.Lines);
ADQuery1.Post;

The Unicode encoded parameter value is converted to the character set supported by the DBMS Unicode and sent to the DBMS. This is not dependent on specified or default client character sets or Delphi versions. 

Firebird and Interbase cannot send Unicode character data to a server if the CharacterSet parameter is not UTF8. 

To read/write parameter value in Delphi 2009 do: 

ADQuery1.Params[0].AsWideString := 'русский + english';
// the same as following
ADQuery1.Params[0].Value := 'русский + english';
ADQuery1.Params[0].DataType := ftWideString;

Starting from Delphi 2009, TADDataMove can read/write Unicode encoded text data files without conversion loss. On prior Delphi versions TADDataMove can read/write Unicode encoded text data files, but internally handles all data as ACP encoded. So, conversion loss is possible. 

To control encoding of:

• text data file, use TADDataMove.TextFileEncoding property;
• log file, use TADScript.ScriptOptions.LogFileEncoding property.

To read Unicode data file and create ACP encoded log file do: 

ADDataMove1.TextFileName := 'u:\data.txt';
ADDataMove1.TextFileEncoding := enUTF8;
ADDataMove1.LogFileName := 'u:\log.txt';
ADDataMove1.TextFileEncoding := enANSI;
ADDataMove1.Destination := ADQuery1;
ADDataMove1.SourceKind := skText;
ADDataMove1.Execute;

To enable Unicode encoding for trace file output (MonitorBy=FlatFile), set TADMoniFlatFileClientLink.FileEncoding property value to ecUTF8 or ecUTF16. 

All AnyDAC utilities are compiled with Delphi XE and are 100% Unicode enabled.