Re: Error handling in URIs

Anne van Kesteren wrote:
> On Tue, 24 Jun 2008 14:04:05 +0200, Julian Reschke 
> <julian.reschke@gmx.de> wrote:
>> Or that the definition needs to be moved into a standalone spec.
> 
> I guess I don't really see how that's different from fixing the URI spec.

There is nothing that needs to be fixed in the URI spec.

If other specs accept invalid URIs, then those specs also need to define 
how to do that.

>>>>> The second is with IRIs and character encodings other than UTF-8. 
>>>>> While browsers reliably encode non-ASCII characters in the path 
>>>>> using UTF-8, non-ASCII characters in the query component are 
>>>>> encoded using the document's character encoding, and not UTF-8, 
>>>>> which is incompatible with how the IRI spec defines things.
>>>>
>>>> Could you please be more specific? Any URI is a IRI, so a query 
>>>> component based on an encoding other than UTF-8 still is a legal IRI.
>>>  It's also transmitted as another encoding than UTF-8 (while the path 
>>> component _is_ transmitted as UTF-8).
>>
>> Yes. It's still a legal URI, thus a legal IRI.
> 
> I think the problem is that currently no specification says how to 
> construct a URI from a bunch of Unicode characters while taking into 
> account that the path component always needs to be in UTF-8 and the 
> query component in the document encoding.

But again, that's not a problem with URI or IRI, right?

BR, Julian

Received on Tuesday, 24 June 2008 13:54:55 UTC