[Abook-devel] 0.5.2pre2

[Jaakko H. <jhe...@us...>]

Download links:

	http://abook.sourceforge.net/devel/abook-0.5.2pre2.tar.gz
	http://abook.sourceforge.net/devel/abook-0.5.2pre2.tar.gz.asc

Pre2 adds an experimental support for wide characters. I haven't tested
it with truly _wide_ characters. I have tested it with UTF-8 locale
using characters which takes multible bytes when represented as
multibyte string but which are only one character wide on the screen.

There are currently several issues:

- Addressbook files must have the same encoding as you use with abook.
  If you have used abook previously with ISO-8859-1 encoding you can
  convert the addressbook files with iconv(1). For example:

  $ iconv -f ISO-8859-1 -t UTF-8 < ~/.abook/addressbook.old > \
  ~/.abook/addressbook.new

- Filters will output strings using multibyte representation of the used
  locale. This is incorrect for (most of) filters. You can again use
  iconv to convert between encodings. Same goes with input filters. If
  you use UTF-8 charset input is expected to be UTF-8 encoded.

- In some places strlen is used to calculate the amount of space which a
  string will take on a screen. This causes some display errors. This
  shouldn't be too hard to fix.

To enable the support use --enable-widec option for configure. You need
to have ncursesw library and some wide character functions in C
libarary. (I have tested only with GNU libc 2.3.2.)

Please report any problems or success.

-- 
Jaakko