LCL Unicode Support

From Free Pascal wiki
Revision as of 12:21, 20 May 2006 by Borut (talk | contribs) (cosmetical changes)

Deutsch (de) English (en) español (es) français (fr) 日本語 (ja) 한국어 (ko) русский (ru) 中文(中国大陆)‎ (zh_CN) 中文(台灣)‎ (zh_TW)

Introduction

Lazarus support of the Unicode standard needs further development, mostly in regard to the Windows platform. Here are some basic information for those who would like to further develop the Lazarus Unicode support. Please correct, extend and update this page.

It will help if you have already heard for the Unicode standard and if you perhaps had some experience with WideStrings under Delphi. Previous use of non-(western)Latin scripts and their various character sets will help too.

Unicode essentials

Unicode standard maps integers from 0 to 10FFFF(h) to characters. Each such mapping is called a code point. In other words, Unicode characters are in principle defined for code points from U+000000 to U+10FFFF (0 to 1 114 111).

There are three schemes for representing Unicode code points as unique byte sequences. These schemes are called Unicode transformation formats: UTF-8, UTF-16 and UTF-32. The conversions between all of them are possible. Here are their basic properties:

                           UTF-8 UTF-16 UTF-32
Smallest code point [hex] 000000 000000 000000
Largest code point  [hex] 10FFFF 10FFFF 10FFFF
Code unit size [bits]          8     16     32
Minimal bytes/character        1      2      4
Maximal bytes/character        4      4      4

UTF-8 has several important and useful properties: It is interpreted as a sequence of bytes, so that the concept of lo- and hi-order byte does not exist. Unicode characters U+0000 to U+007F (ASCII) are encoded simply as bytes 00h to 7Fh (ASCII compatibility). This means that files and strings which contain only 7-bit ASCII characters have the same encoding under both ASCII and UTF-8. All characters >U+007F are encoded as a sequence of several bytes, each of which has the most significant bit set. No byte sequence of one character is contained within a longer byte sequence of another character. This allows easy search for substrings. The first byte of a multibyte sequence that represents a non-ASCII character is always in the range C0h to FDh and it indicates how many bytes follow for this character. All further bytes in a multibyte sequence are in the range 80h to BFh. This allows easy resynchronization and robustness.

UTF-16 has the following most important properties: It uses a single 16-bit word to encode any of the characters from U+0000 to U+FFFF, and a pair of 16-bit words to encode any of the remaining Unicode characters.

Finally, any Unicode character can be represented as a single 32-bit unit in UTF-32.

For more, see: Unicode FAQ - Basic questions, Unicode FAQ - UTF-8, UTF-16, UTF-32 & BOM, Wikipedia: UTF-8


Lazarus component library architecture essentials

(This part based on a mail by Marc Weustink) The LCL consists of two parts:

  1. A target platform independent part, which implements a class hierarchy analogous to Delphi VCL;
  2. "Interfaces" - a part that implements the interface to APIs of each target platform.

The communication between the two parts is done by an abstract class TWidgetset. Each widgetset is implemented by its own derived class form TWidgetset.

The GTK widgetset is the oldest. In this widgetset the string encoding is determined by the LANG environment var. If it is a UTF8 variant, all strings from and to native controls/widgets are UTF8 encoded. However utf8 may affect keyboard handling for gtk1. On gtk2 this problem is solved, but not implemented yet, the keyboard routines still rely on gtk1 code there.

The win32 interfaces is setup with ansii widgets, so it is currently not possible to use Unicode with win32.

For more, see: Internals of the LCL

Unicode-enabling the win32 interface

Requirements

The spirit of Lazarus is: "Write once, compile everywhere." This means that, ideally, an Unicode enabled application should have only one Unicode supporting source code version, without any conditional defines in respect to various target platforms.

The "interface" part of the LCL should support Unicode for the target platforms which support it themselves, concealing at the same time all peculiarities from the application programmer.

Windows platforms <=Win9x are based on ISO code page standards and do not support Unicode. Windows platforms starting with WinNT support Unicode. In doing that, these platforms offer two parallel sets of API functions: the old ANSII enabled *A and the new, Unicode enabled *W. *W functions accept wide strings, i.e. UTF-16 encoded strings, as parameters.

What concerns Lazarus, the internal string communication at the boundaries "Application code <--> LCL", as well as "LCL <--> Widgetsets" is based on the classical (byte oriented) strings. Logically, their contents should be encoded according to the UTF-8.

It is sound to assume that the existing WinXX application base internally does not use UTF-8 encoded strings, but the ISO code page based ones. Any Unicode enabling changes to LCL and widget sets for win32 must not break the existing application base. At the same time they should support applications which are internally based on the Unicode UTF-8 encoded strings, both on older Win9x platforms, as well as on Unicode based >=WinNT platforms.


A solution approach

Making progress