W3C home > Mailing lists > Public > public-webapi@w3.org > February 2007

Re: [XMLHttpRequest] Comments on Feb 13 draft

From: Anne van Kesteren <annevk@opera.com>
Date: Sat, 17 Feb 2007 20:46:51 +0100
To: "Julian Reschke" <julian.reschke@gmx.de>
Cc: public-webapi@w3.org
Message-ID: <op.tnwrwddy64w2qv@id-c0020>

On Sat, 17 Feb 2007 20:13:52 +0100, Julian Reschke <julian.reschke@gmx.de>  
>>> "Otherwise, if the nominated request header field already has a value,  
>>> the new value MUST be combined with the existing value (section 4.2,  
>>> [RFC2616])."
>>> That's a bit misleading. What does "combine" mean precisely? Is the  
>>> intent to require implementations to assume that the header format is  
>>> a comma-separated list?
>>  Yes.
> So please clarify.

Section 4.2 already talks about this quite explicitly. What do you want it  
to say?

> If an XHR object is passed around between several parts of the code,  
> it's not trivial to ensure that a header is only set once. Some headers  
> have list semantics, some do not.
> It would be great if a script could say:
>    xhr.removeRequestHeader("xyz");
>    xhr.serRequestHeader("xyz", "bar");
> to make sure that the header value actually *is* "bar", and not "foo,  
> bar", just because some other part of the code set it before.

I added this to the future version wishlist. You probably also want  
getRequestHeader I suppose to see what has actually been set so far.

> "...MUST have the content-encoding automatically decoded"?


>>> "Note: For HEAD requests this attribute will always be null (it  
>>> remains unchanged). Similar to the responseXML attribute."
>>> I don't like special cases. So will it be null for any response  
>>> without entity body, or just for HEAD?
>>  This is just a note to inform people. Then again, I guess if a server  
>> gives you a request body with a HEAD it should prolly be filled...  
>> Suggestions?
> Drop it, or say it will be null for responses without a body, such as a  
> conforming HEAD response.

Clarified this for both responseText and responseXML.

> "HTTP requests sent from multiple different XMLHttpRequest should use a  
> shared HTTP connection."?



Anne van Kesteren
Received on Saturday, 17 February 2007 19:46:56 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:16:23 UTC