Re: Testing IRI fragment identifiers

The difference between (2) and (3) is that:

In (2) the slash separator between the IDN and the pathname part is U 
+002F SOLIDUS (ie the usual slash)

In (3) the slash separator between the IDN and the pathname part is U 
+FF0F FULLWIDTH SOLIDUS (which is what I get when I type slash whilst  
in my chinese input method)

So (2) is using the (currently) valid slash separator but (3) is not.  
But I think it essential that U+FF0F FULLWIDTH SOLIDUS is also treated  
as a valid separator (or auto converted to U+002F SOLIDUS)

André

On 3 Jan 2009, at 11:58, Ori Idan wrote:

> I don't understand the difference between 2 and 3.
> I understand that chinese input mode does not have . and /
>
> -- 
> Ori Idan
>
>
> On Sat, Jan 3, 2009 at 1:38 PM, Andre Schappo  
> <A.Schappo@lboro.ac.uk> wrote:
>
>
> There is also the issue of the separators / slash and  . dot in the  
> browser address bar.
>
> In order for the rest of my explanation to make sense I should say  
> that I am using the alternate root server <http://cesidianroot.net/>  
> for these tests and the consequence of that is that the China  
> ccTLD   .中国 works
>
> (1) When I type in:  拉夫堡大学.中国/计算机科学  it works  
> just fine because I am using U+002E FULL STOP and U+002F SOLIDUS/Slash
>
> (2) When I type in :  拉夫堡大学。中国/计算机科学 that  
> also works fine because the browser automatically converts from U 
> +3002 IDEOGRAPHIC FULL STOP (which is what I get when I type dot  
> whilst in my Chinese Input Method) to U+002E FULL STOP
>
> (3) When I type in:  拉夫堡大学。中国／计算机科学 it  
> does NOT work because I am using U+FF0F FULLWIDTH SOLIDUS which is  
> what I get when I type slash whilst in my Chinese Input Method.
>
> When typing a Chinese web address I argue that it is very important  
> that one should be able to remain within a Chinese Input Method.  
> But ... For (3)  I have to come out of the chinese input method to  
> type the slash and then back into the chinese input method to type  
> the pathname part
>
> So I think that the browser should automatically convert the U+FF0F  
> FULLWIDTH SOLIDUS to U+002F SOLIDUS then one can remain in the  
> chinese input method
>
> This issue may well occur in other languages
>
> (1)-(3) Tested using Safari, Firefox and Opera on MacOSX
>
> André Schappo
>
>
> On 15 Dec 2008, at 21:16, Richard Ishida wrote:
>
>
> Hello Leif,
>
> I added a couple of tests to my revised set of IDN and IRI tests, to  
> be found (unannounced as yet) at
> http://www.w3.org/International/tests/html/test-resid-0.html
>
> Cheers,
> RI
>
> ============
> Richard Ishida
> Internationalization Lead
> W3C (World Wide Web Consortium)
>
> http://www.w3.org/International/
> http://rishida.net/
>
>
>
> -----Original Message-----
> From: Leif Halvard Silli [mailto:lhs@malform.no]
> Sent: 12 December 2008 20:45
> To: Richard Ishida
> Cc: www-international@w3.org
> Subject: Testing IRI fragment identifiers
>
> *I suggest that the I18N Resource identifiers tests at *
>
> * http://www.w3.org/International/tests/sec-idn-0*
>
> *gets updated with a test of how UAs reacts to URLs which contain an  
> IRI
> in the fragment part. The test URL could for example look like this:*
>
> * http://www.w3.org/International/articles/idn-and-iri/file#АБВГ
> *
>
> *In my te**sting, all current mayor brow**ser**s, except Opera,
> **support **IRI fragment ientifier**s in link**s**.
> *
>
> *I don't know the con**sideration**s of Wikipedia, but Wikipedia  
> doe**s
> u**se IRI**s in it**s** page URL**s. However, in fragment  
> identifier**s
> it doe**s not** (except when the fragment identifier**s are pure  
> ASCII).*
> *--
> leif halvard **silli*
>
>
>
> 
> http://loughborough.ac.uk/ComputerScience
> http://拉夫堡大学.cn/计算机科学
> http://ラフバラ大学.com/コンピュータサイエンス
> http://러프버러대학교.com/컴퓨터과학
> http://УниверситетЛафборо.com/ 
> ВычислительнаяНаука
> http://جامعةلوفبورا.com/
>
> Andre Schappo
> Dept. Computer Science
> Loughborough University
> Loughborough
> Leics       LE11 3TU
>
> http://小山.321.cn/ & http://www-staff.lboro.ac.uk/~coas/
> tel: 01509 222682
> email: A.Schappo@lboro.ac.uk
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>


http://loughborough.ac.uk/ComputerScience
http://拉夫堡大学.cn/计算机科学
http://ラフバラ大学.com/コンピュータサイエンス
http://러프버러대학교.com/컴퓨터과학
http://УниверситетЛафборо.com/ 
ВычислительнаяНаука
http://جامعةلوفبورا.com/

Andre Schappo
Dept. Computer Science
Loughborough University
Loughborough
Leics       LE11 3TU

http://小山.321.cn/ & http://www-staff.lboro.ac.uk/~coas/
tel: 01509 222682
email: A.Schappo@lboro.ac.uk