Hi
On Mon, 5 Feb 2001, Hetz B.H wrote:
> Hi all,
>
> This might not be news to people who are compiling KDE from CVS or the daily
> snapshot's or use Bero's daily RPM's, but Lars from Trolltech has fixed the
> numbers problem, so you can read Globes and other financial news more
> normally, but there is one problem which I would like input from people who
> either know the BiDi algoritm or are web designers...
>
There was a legthy discussion about this exact subject that spread accross
the ivrix-discuss and gtk-i18n lists. See
http://ivrix.org.il/mailing-lists/ivrix-discuss/2000/11/
(The workaround with "soft-hyphen" that is mentioned there does not work,
BTW)
Conclusion of this discussion:
unicode has seperate symbols for hyphen (makaf) and for minus sign. ascii
doesn't. And this is indeed a problem, if you want to stick to ascii (a
minus sign should be a LTR character, whereas a hyphen should be a neutral
character).
> This problem goes like this (and I use currently Outlook Express to type
> this email for showing the problem). Read the following sentence:
>
> &he;&bet;&vav;&resh;&samekh;&he; &yod;&resh;&dalet;&he; &bet;-3%
>
> or in "hebrew english"
>
> Haboorsa yarda b-3%
>
> Now. According to lars, there isn't a specific paragraph (at least thats
> what I understood from him) that says where the "makaf" should be, so Lars
> has left it like this situation:
>
> -3%&he;&bet;&vav;&resh;&samekh;&he; &yod;&resh;&dalet;&he; &bet;
>
> or in "hebrew english"
>
> Haboorsa yarda b-3%
>
> >From my experience, opening word documents on Windows 95/98 and Windows NT -
> the "makaf" was on different places, which left you to fix it manually. I
> never tested it on Windows 2000.
I think that the behaviour of different windows versions differ at this
point. But I'm not sure.
>
> According to lars - the way that MSIE shows the "makaf" is a bug and the
> financial and other web sites uses this bug to show the numbers and "makaf"
> correctly.
>
> So, the big question is - how is it on the windows world? how web designers
> are dealing with it? what Lars should do? ignore the spec and implement this
> "bug"?
I don't remember what they decided to do in gtk...
--
Tzafrir Cohen
mailto:[EMAIL PROTECTED]
http://www.technion.ac.il/~tzafrir
ннннннннннннннннннннни unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
