Fedora directory users - Nov 2005 - question about required fields and I18N issues

How can I enter non-ascii data in the attribute,
especially for dn, last name, first name, etc, and
still can use the native language for searching?

For example, if I want to enter greeks or some eastern
european characters, how can I do that?

How do I configure the server to support i18n and have
the proper collation? 

How many languages does the console support, i.e. have
the proper translation and display correctly?

I added an entry using ldapmodify, which contains
non-ascii in the DN. It seems to get it correctly, as
shown in the title bar of the property editor in the
console. But the property editor and the console does
not display correctly. The title bar displays
correctly, 
though.

And the result from ldapsearch just shows a bunch of
garbage characters.

All helps appreciated.

sz



	
		
__________________________________ 
Yahoo! Mail - PC Magazine Editors'' Choice 2005 
http://mail.yahoo.com

speedy zinc wrote:
>How can I enter non-ascii data in the attribute,
>especially for dn, last name, first name, etc, and
>still can use the native language for searching?
>  
>
Firstly, the data must be encoded in utf8.  There are usually system 
utilities available to do native charset -> utf8 conversion - see "man 
iconv".  Secondly, you must use language tags for your attributes if you 
want them to be properly sorted/collated e.g.
cn: Celine Andre
cn;lang-fr: \de\55\85\44line Andre

My utf8 encoding is not correct, but you should get the general drift.
>For example, if I want to enter greeks or some eastern
>european characters, how can I do that?
>
>How do I configure the server to support i18n and have
>the proper collation?
>  
>
You shouldn''t have to do anything.  As long as you make sure all data
is
utf8 encoded, the server should be able to handle it.  We use ICU 2.4 
which supports about 40 languages.
>How many languages does the console support, i.e. have
>the proper translation and display correctly?
>  
>
You mean, for how many languages has the console been localized for?

The console itself is written in Java, which has all of the unicode 
stuff built in, so it can handle the native charset -> utf8 conversion 
properly.  This assumes you have your LOCALE set up correctly with all 
of the fonts you need.
>I added an entry using ldapmodify, which contains
>non-ascii in the DN. It seems to get it correctly, as
>shown in the title bar of the property editor in the
>console. But the property editor and the console does
>not display correctly. The title bar displays
>correctly, 
>though.
>
>And the result from ldapsearch just shows a bunch of
>garbage characters.
>  
>
ldapsearch will display the data base64 encoded.  Just decode the base64 
to see your utf8 chars.
>All helps appreciated.
>
>sz
>
>
>
>	
>		
>__________________________________ 
>Yahoo! Mail - PC Magazine Editors'' Choice 2005 
>http://mail.yahoo.com
>
>--
>Fedora-directory-users mailing list
>Fedora-directory-users@redhat.com
>https://www.redhat.com/mailman/listinfo/fedora-directory-users
>  
>

We are working on a school project to build a 
"universal" directory service to support a global
village (:) on which  everyone can logon using their
native language. People can talk to each other
using their native language, but it gets translated
in real-time (don''t expect too much, just a school
project). And we use FDS as the underlying service
for user authentication, user profile, etc.

We want to allow user to register themselves,
in their own language. So, username etc, should be in
the native language.

--- Richard Megginson <rmeggins@redhat.com> wrote:
> speedy zinc wrote:
> 
> >How can I enter non-ascii data in the attribute,
> >especially for dn, last name, first name, etc, and
> >still can use the native language for searching?
> >  
> >
> Firstly, the data must be encoded in utf8.  There
> are usually system 
> utilities available to do native charset -> utf8
> conversion - see "man 
> iconv".  Secondly, you must use language tags for
> your attributes if you 
> want them to be properly sorted/collated e.g.
> cn: Celine Andre
> cn;lang-fr: \de\55\85\44line Andre
> 
Does that mean I can not enter native language (even
if my system is using UTF8 encoding) directly in
the console?
> 
> >For example, if I want to enter greeks or some
> eastern
> >european characters, how can I do that?
> >
> >How do I configure the server to support i18n and
> have
> >the proper collation?
> >  
> >
> You shouldn''t have to do anything.  As long as you
> make sure all data is 
> utf8 encoded, the server should be able to handle
> it.  We use ICU 2.4 
> which supports about 40 languages.
> 
> >How many languages does the console support, i.e.
> have
> >the proper translation and display correctly?
> >  
> >
> You mean, for how many languages has the console
> been localized for?
> 
Yeah, since it is in java, if I change my environment,
shouldn''t the console displayed in the right language?

thanks

chris p.



		
__________________________________ 
Yahoo! FareChase: Search multiple travel sites in one click.
http://farechase.yahoo.com

speedy zinc wrote:
>We are working on a school project to build a 
>"universal" directory service to support a global
>village (:) on which  everyone can logon using their
>native language. People can talk to each other
>using their native language, but it gets translated
>in real-time (don''t expect too much, just a school
>project). And we use FDS as the underlying service
>for user authentication, user profile, etc.
>
>We want to allow user to register themselves,
>in their own language. So, username etc, should be in
>the native language.
>  
>
Sure.  This is also quite common for large global enterprises who want 
to provide self service or locally administered access to the directory 
server.  The logic to convert from the local charset to utf8 must be 
done in the application - LDAP only provides for utf8 data.  What is 
registration application?  Is it open source?  What language is it 
written in?  For C apps, iconv is provided by most *nix OSes.  There is 
a way to do this in Windows - I can''t remember, but there is some code 
that the ldapsearch, ldapmodify commands use.  I have no idea about 
Mac.  It''s very easy to do this in Java - strings are stored in Unicode
internally, and the conversion code is built into the String class.
>--- Richard Megginson <rmeggins@redhat.com> wrote:
>
>  
>
>>speedy zinc wrote:
>>
>>    
>>
>>>How can I enter non-ascii data in the attribute,
>>>especially for dn, last name, first name, etc, and
>>>still can use the native language for searching?
>>> 
>>>
>>>      
>>>
>>Firstly, the data must be encoded in utf8.  There
>>are usually system 
>>utilities available to do native charset -> utf8
>>conversion - see "man 
>>iconv".  Secondly, you must use language tags for
>>your attributes if you 
>>want them to be properly sorted/collated e.g.
>>cn: Celine Andre
>>cn;lang-fr: \de\55\85\44line Andre
>>
>>    
>>
>
>Does that mean I can not enter native language (even
>if my system is using UTF8 encoding) directly in
>the console?
>
>  
>
>>>For example, if I want to enter greeks or some
>>>      
>>>
>>eastern
>>    
>>
>>>european characters, how can I do that?
>>>
>>>How do I configure the server to support i18n and
>>>      
>>>
>>have
>>    
>>
>>>the proper collation?
>>> 
>>>
>>>      
>>>
>>You shouldn''t have to do anything.  As long as you
>>make sure all data is 
>>utf8 encoded, the server should be able to handle
>>it.  We use ICU 2.4 
>>which supports about 40 languages.
>>
>>    
>>
>>>How many languages does the console support, i.e.
>>>      
>>>
>>have
>>    
>>
>>>the proper translation and display correctly?
>>> 
>>>
>>>      
>>>
>>You mean, for how many languages has the console
>>been localized for?
>>
>>    
>>
>
>Yeah, since it is in java, if I change my environment,
>shouldn''t the console displayed in the right language?
>
>thanks
>
>chris p.
>
>
>
>		
>__________________________________ 
>Yahoo! FareChase: Search multiple travel sites in one click.
>http://farechase.yahoo.com
>
>--
>Fedora-directory-users mailing list
>Fedora-directory-users@redhat.com
>https://www.redhat.com/mailman/listinfo/fedora-directory-users
>  
>

--- Richard Megginson <rmeggins@redhat.com> wrote:
> speedy zinc wrote:
> 
> >We are working on a school project to build a 
> >"universal" directory service to support a global
> >village (:) on which  everyone can logon using
> their
> >native language. People can talk to each other
> >using their native language, but it gets translated
> >in real-time (don''t expect too much, just a school
> >project). And we use FDS as the underlying service
> >for user authentication, user profile, etc.
> >
> >We want to allow user to register themselves,
> >in their own language. So, username etc, should be
> in
> >the native language.
> >  
> >
> Sure.  This is also quite common for large global
> enterprises who want 
> to provide self service or locally administered
> access to the directory 
> server.  The logic to convert from the local charset
> to utf8 must be 
> done in the application - LDAP only provides for
> utf8 data.  What is 
> registration application?  Is it open source?  What
> language is it 
> written in?  For C apps, iconv is provided by most
> *nix OSes.  There is 
> a way to do this in Windows - I can''t remember, but
> there is some code 
> that the ldapsearch, ldapmodify commands use.  I
> have no idea about 
> Mac.  It''s very easy to do this in Java - strings
> are stored in Unicode 
> internally, and the conversion code is built into
> the String class.
> 
But the console does not even display the content 
"correctly". We use the java sdk to get the data, and
it is correct. 

We are a team of 5, with 5 different lanaguages. We
aall
develop on Linux, using utf8 environment. We can add
entries using native characters, but despite setting
our environment to the right locale, the console just
displays some garbage characters.

regards

sz


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

speedy zinc wrote:
>--- Richard Megginson <rmeggins@redhat.com> wrote:
>
>  
>
>>speedy zinc wrote:
>>
>>    
>>
>>>We are working on a school project to build a 
>>>"universal" directory service to support a global
>>>village (:) on which  everyone can logon using
>>>      
>>>
>>their
>>    
>>
>>>native language. People can talk to each other
>>>using their native language, but it gets translated
>>>in real-time (don''t expect too much, just a school
>>>project). And we use FDS as the underlying service
>>>for user authentication, user profile, etc.
>>>
>>>We want to allow user to register themselves,
>>>in their own language. So, username etc, should be
>>>      
>>>
>>in
>>    
>>
>>>the native language.
>>> 
>>>
>>>      
>>>
>>Sure.  This is also quite common for large global
>>enterprises who want 
>>to provide self service or locally administered
>>access to the directory 
>>server.  The logic to convert from the local charset
>>to utf8 must be 
>>done in the application - LDAP only provides for
>>utf8 data.  What is 
>>registration application?  Is it open source?  What
>>language is it 
>>written in?  For C apps, iconv is provided by most
>>*nix OSes.  There is 
>>a way to do this in Windows - I can''t remember, but
>>there is some code 
>>that the ldapsearch, ldapmodify commands use.  I
>>have no idea about 
>>Mac.  It''s very easy to do this in Java - strings
>>are stored in Unicode 
>>internally, and the conversion code is built into
>>the String class.
>>
>>    
>>
>
>But the console does not even display the content 
>"correctly". We use the java sdk to get the data, and
>it is correct. 
>
>We are a team of 5, with 5 different lanaguages. We
>aall
>develop on Linux, using utf8 environment. We can add
>entries using native characters, but despite setting
>our environment to the right locale, the console just
>displays some garbage characters.
>  
>
I''m not sure then.  I know it''s tricky to get the console to
know how to
display the local charset correctly.  Hopefully one of our i18n experts 
will chime in here.
>regards
>
>sz
>
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around 
>http://mail.yahoo.com 
>
>--
>Fedora-directory-users mailing list
>Fedora-directory-users@redhat.com
>https://www.redhat.com/mailman/listinfo/fedora-directory-users
>  
>

speedy zinc wrote:
> 
> 
> But the console does not even display the content 
> "correctly". We use the java sdk to get the data, and
> it is correct. 
> 
> We are a team of 5, with 5 different lanaguages. We
> aall
> develop on Linux, using utf8 environment. We can add
> entries using native characters, but despite setting
> our environment to the right locale, the console just
> displays some garbage characters.
> 
The contents in the database has nothing to do with wether
the console can display or not though. If you see
"garbage" characters, it probably means the console actually
got the right data, except that it does not know how to
display correctly. Probably lack of the right font?

You can change the display font to see if it''s right, go to
Edit -> Preferences -> Font, and select a font that can
display your native language.

And I don''t think the console has localization for that many
languages either. It does not have localization for chinese
(maybe I''m wrong, but starting with "-l zh" or "-l
zh_CN"
does not work).

The annoying thing we had with FDS is that the uid must be in
7-bit ascii (duh!). I don''t know what does the LDAP specs said
(I haven''t looked up yet), but it makes it impossible to have
user login name in chinese. That''s a shame, coz directory server
is such an important piece of software.

To get around this, we add an attribute called "loginname", and
use that as the input to generate a hash, and use the hash as
the uid. According to the description of your project, you might
have to do this too :)

Hope that helps.

csp
-- 
Chen Shaopeng
http://www.idsignet.com

Chen Shaopeng wrote:
>speedy zinc wrote:
>  
>
>>But the console does not even display the content 
>>"correctly". We use the java sdk to get the data, and
>>it is correct. 
>>
>>We are a team of 5, with 5 different lanaguages. We
>>aall
>>develop on Linux, using utf8 environment. We can add
>>entries using native characters, but despite setting
>>our environment to the right locale, the console just
>>displays some garbage characters.
>>
>>    
>>
>
>The contents in the database has nothing to do with wether
>the console can display or not though. If you see
>"garbage" characters, it probably means the console actually
>got the right data, except that it does not know how to
>display correctly. Probably lack of the right font?
>
>You can change the display font to see if it''s right, go to
>Edit -> Preferences -> Font, and select a font that can
>display your native language.
>
>And I don''t think the console has localization for that many
>languages either. It does not have localization for chinese
>(maybe I''m wrong, but starting with "-l zh" or "-l
zh_CN"
>does not work).
>
>The annoying thing we had with FDS is that the uid must be in
>7-bit ascii (duh!). I don''t know what does the LDAP specs said
>(I haven''t looked up yet), but it makes it impossible to have
>user login name in chinese. That''s a shame, coz directory server
>is such an important piece of software.
>  
>
The default configuration for the 7-bit-clean-checking plugin is for 
uid, mail and userPassword attributes.  You can adjust it in dse.ldif or 
via plugin configuration screen in the Console.
>To get around this, we add an attribute called "loginname", and
>use that as the input to generate a hash, and use the hash as
>the uid. According to the description of your project, you might
>have to do this too :)
>
>Hope that helps.
>
>csp
>  
>
>------------------------------------------------------------------------
>
>--
>Fedora-directory-users mailing list
>Fedora-directory-users@redhat.com
>https://www.redhat.com/mailman/listinfo/fedora-directory-users
>  
>

uffe@loop.to wrote:
>>
> The default configuration for the 7-bit-clean-checking plugin is for
> uid, mail and userPassword attributes.  You can adjust it in dse.ldif or
> via plugin configuration screen in the Console.
> 
Cool, that''s something we never checked though, it might work.
I have to check that, later.

thx,

csp
-- 
Chen Shaopeng
http://www.idsignet.com

Chen Shaopeng wrote:
>speedy zinc wrote:
>  
>
>>But the console does not even display the content 
>>"correctly". We use the java sdk to get the data, and
>>it is correct. 
>>
>>We are a team of 5, with 5 different lanaguages. We
>>aall
>>develop on Linux, using utf8 environment. We can add
>>entries using native characters, but despite setting
>>our environment to the right locale, the console just
>>displays some garbage characters.
>>
>>    
>>
>
>The contents in the database has nothing to do with wether
>the console can display or not though. If you see
>"garbage" characters, it probably means the console actually
>got the right data, except that it does not know how to
>display correctly. Probably lack of the right font?
>
>You can change the display font to see if it''s right, go to
>Edit -> Preferences -> Font, and select a font that can
>display your native language.
>
>And I don''t think the console has localization for that many
>languages either. It does not have localization for chinese
>(maybe I''m wrong, but starting with "-l zh" or "-l
zh_CN"
>does not work).
>
>The annoying thing we had with FDS is that the uid must be in
>7-bit ascii (duh!). I don''t know what does the LDAP specs said
>(I haven''t looked up yet), but it makes it impossible to have
>user login name in chinese. That''s a shame, coz directory server
>is such an important piece of software.
>  
>
This is historical, due to the fact that in the old days, the uid 
attribute was also the unix login name, which was restricted by most 
OSes to be 7bit ASCII.  There is a 7 bit checking plugin which may be on 
by default - you can probably safely disable it for all except 
userPassword (and even then, your systems may be able to accept 8bit 
passwords).
>To get around this, we add an attribute called "loginname", and
>use that as the input to generate a hash, and use the hash as
>the uid. According to the description of your project, you might
>have to do this too :)
>
>Hope that helps.
>
>csp
>  
>
>------------------------------------------------------------------------
>
>--
>Fedora-directory-users mailing list
>Fedora-directory-users@redhat.com
>https://www.redhat.com/mailman/listinfo/fedora-directory-users
>  
>

Richard Megginson wrote:
>>
> This is historical, due to the fact that in the old days, the uid
> attribute was also the unix login name, which was restricted by most
> OSes to be 7bit ASCII.  There is a 7 bit checking plugin which may be on
> by default - you can probably safely disable it for all except
> userPassword (and even then, your systems may be able to accept 8bit
> passwords).
> 
Thanks, I''ll check the plugin more often. My apologies.

I see the argument list of the 7-bit plugin now. But is there
a reason why the base entry (dc=idsignet,dc=com) is also on
that list?

Thanks

csp
-- 
Chen Shaopeng
http://www.idsignet.com

Chen Shaopeng wrote:
>Richard Megginson wrote:
>  
>
>>This is historical, due to the fact that in the old days, the uid
>>attribute was also the unix login name, which was restricted by most
>>OSes to be 7bit ASCII.  There is a 7 bit checking plugin which may be on
>>by default - you can probably safely disable it for all except
>>userPassword (and even then, your systems may be able to accept 8bit
>>passwords).
>>
>>    
>>
>
>Thanks, I''ll check the plugin more often. My apologies.
>
>I see the argument list of the 7-bit plugin now. But is there
>a reason why the base entry (dc=idsignet,dc=com) is also on
>that list?
>  
>
You can also specify the suffixes to which the checking applies.  If the 
suffix is not in that list, it should not be checked for 7bit attrs.
>Thanks
>
>csp
>  
>
>------------------------------------------------------------------------
>
>--
>Fedora-directory-users mailing list
>Fedora-directory-users@redhat.com
>https://www.redhat.com/mailman/listinfo/fedora-directory-users
>  
>

--- Chen Shaopeng <chen_shaopeng@idsignet.com> wrote:
> You can change the display font to see if it''s
> right, go to
> Edit -> Preferences -> Font, and select a font that
> can
> display your native language.
> 
hah, that''s an easy one :) thanks.

The only thing thouhg, there is no easy way to switch
font quickly. If I want to view greek contents, I have
to change to a different font. If I want to view
chinese
contents, I have to switch to a font that can handle
chinese charset.... man, that''s not very productive.

regards

chris s.p.


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

speedy zinc wrote:
>--- Chen Shaopeng <chen_shaopeng@idsignet.com> wrote:
>
>  
>
>>You can change the display font to see if it''s
>>right, go to
>>Edit -> Preferences -> Font, and select a font that
>>can
>>display your native language.
>>
>>    
>>
>hah, that''s an easy one :) thanks.
>
>The only thing thouhg, there is no easy way to switch
>font quickly. If I want to view greek contents, I have
>to change to a different font. If I want to view
>chinese
>contents, I have to switch to a font that can handle
>chinese charset.... man, that''s not very productive.
>  
>
That''s an interesting problem.  Is there an app that can do that?
>regards
>
>chris s.p.
>
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around 
>http://mail.yahoo.com 
>
>--
>Fedora-directory-users mailing list
>Fedora-directory-users@redhat.com
>https://www.redhat.com/mailman/listinfo/fedora-directory-users
>  
>

Richard Megginson wrote:
speedy
zinc wrote:

  --- Chen Shaopeng
 wrote:

    You can change the display font to see if
it''s

right, go to

Edit -> Preferences -> Font, and select a font that

can

display your native language.

hah, that''s an easy one :) thanks.

The only thing thouhg, there is no easy way to switch

font quickly. If I want to view greek contents, I have

to change to a different font. If I want to view

chinese

contents, I have to switch to a font that can handle

chinese charset.... man, that''s not very productive.

That''s an interesting problem.  Is there an app that can do that?

I haven''t tried it for years, but there used to be a Unicode font set
on the net and we were using it.  That way you don''t have to switch
from one to another...

--noriko

  regards

chris s.p.

__________________________________________________

Do You Yahoo!?

Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com 

--

Fedora-directory-users mailing list

Fedora-directory-users@redhat.com

https://www.redhat.com/mailman/listinfo/fedora-directory-users

--
Fedora-directory-users mailing list
Fedora-directory-users@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-directory-users

Richard Megginson wrote:
> speedy zinc wrote:
> 
>> --- Chen Shaopeng <chen_shaopeng@idsignet.com> wrote:
>>
>>  
>>
>>> You can change the display font to see if it''s
>>> right, go to
>>> Edit -> Preferences -> Font, and select a font that
>>> can
>>> display your native language.
>>>
>>>   
>>
>> hah, that''s an easy one :) thanks.
>>
>> The only thing thouhg, there is no easy way to switch
>> font quickly. If I want to view greek contents, I have
>> to change to a different font. If I want to view
>> chinese
>> contents, I have to switch to a font that can handle
>> chinese charset.... man, that''s not very productive.
>>  
>>
> That''s an interesting problem.  Is there an app that can do that?
> 
I see this as two problems:

1. Most fonts can only handle one charset, some can handle two, e.g.
some chinese fonts. The annoying thing, I can choose a font that
looks nice in chinese, but way ugly in english (most chinese fonts
are like that). Or I can choose a font that looks nice in english,
but almost unreadable in chinese (some microsoft chinese fonts are
like that). It would be really cool if a font can handle charset
of the major languages, and look decent too. And if the application
is in utf8, that should display properly for most.Then again, I''m
not a font designer, I can only wish.

2. The application can provide the UI that is a little easier
for switching charset encoding. For example, Thunderbird and
Firefox. Provide a preference dialog, where I can set my
preferred fonts for my preferred charset. And from the menu
View -> Character Encoding -> list of prefered encodings.
When user selects an encoding, teh app switches instantly. IE
has this feature too. The auto detect kinda works too, but
not always.

I read emails and online news in 3 different languages: chinese,
english and french. Without an easy switch between encodings, that would
kill me :)

It would be even better if I can set my preferences for the
whole desktop, and apply to all applications. But that''s a
different issue.

rgds,

csp
-- 
Chen Shaopeng
http://www.idsignet.com

fontconfig can do a lot of what you''re describing here.  i.e. pick
different fonts based on the charset that you''re using or the character
that you''re displaying.  The latest versions of Gtk2 use fontconfig for
font selection.

--Chris

On Wed, 2005-11-09 at 09:57 +0800, Chen Shaopeng wrote:
> Richard Megginson wrote:
> > speedy zinc wrote:
> > 
> >> --- Chen Shaopeng <chen_shaopeng@idsignet.com> wrote:
> >>
> >>  
> >>
> >>> You can change the display font to see if it''s
> >>> right, go to
> >>> Edit -> Preferences -> Font, and select a font that
> >>> can
> >>> display your native language.
> >>>
> >>>   
> >>
> >> hah, that''s an easy one :) thanks.
> >>
> >> The only thing thouhg, there is no easy way to switch
> >> font quickly. If I want to view greek contents, I have
> >> to change to a different font. If I want to view
> >> chinese
> >> contents, I have to switch to a font that can handle
> >> chinese charset.... man, that''s not very productive.
> >>  
> >>
> > That''s an interesting problem.  Is there an app that can do
that?
> > 
> 
> I see this as two problems:
> 
> 1. Most fonts can only handle one charset, some can handle two, e.g.
> some chinese fonts. The annoying thing, I can choose a font that
> looks nice in chinese, but way ugly in english (most chinese fonts
> are like that). Or I can choose a font that looks nice in english,
> but almost unreadable in chinese (some microsoft chinese fonts are
> like that). It would be really cool if a font can handle charset
> of the major languages, and look decent too. And if the application
> is in utf8, that should display properly for most.Then again, I''m
> not a font designer, I can only wish.
> 
> 2. The application can provide the UI that is a little easier
> for switching charset encoding. For example, Thunderbird and
> Firefox. Provide a preference dialog, where I can set my
> preferred fonts for my preferred charset. And from the menu
> View -> Character Encoding -> list of prefered encodings.
> When user selects an encoding, teh app switches instantly. IE
> has this feature too. The auto detect kinda works too, but
> not always.
> 
> I read emails and online news in 3 different languages: chinese,
> english and french. Without an easy switch between encodings, that would
> kill me :)
> 
> It would be even better if I can set my preferences for the
> whole desktop, and apply to all applications. But that''s a
> different issue.
> 
> rgds,
> 
> csp
> --
> Fedora-directory-users mailing list
> Fedora-directory-users@redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-directory-users