On 7/26/05 11:18 PM, "Peter N. Lundblad" <peter@famlundblad.se> sneezed:
> On Tue, 26 Jul 2005, señior ¿tyrtle? wrote:
>
>> On 7/26/05 2:22 AM, "Julian Foad" <julianfoad@btopenworld.com> sneezed:
>>
>
>
>> The code is currently using an 'xml:endcoding="base64"' attribute (again XML
>> 1.1?), which I think is fairly known, is it worth using something less
>> common just because it might be "more readable"?
>>
> Excuse my ignorance (honest), but where is xml:encoding defined? I didn't
> find it in the XML 1.1 spec. That prefix is reserved, so we have to be
> careful with what we use. An alternative would be "encoding", but if
> "xml:encoding" is well-defined in some spec it might be OK and even a good
> thing to use that.
Your ignorance is also mine...notice the question mark after 1.1.
I found a couple of posts on xml:encoding being rejected for the 1.0 spec,
but that it had a good chance of getting in 1.1 (I don't even know if 1.1 is
final yet).
I saw it being used in a Perl module and thought it would be foolish to roll
my own with this. I have no preference either way.
>> Why is base64 so bad, the speed, the size?
>>
> My only concern is human readability and ease of debugging. I think base64
> is fine where the data can't be represented in XML as "plain text".
I agree, this was more in a question to Julian's statement "We probably
wouldn't want to base-64-encode all properties except 'svn:*'," which I
actually mis-read/interpreted.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Wed Jul 27 00:57:13 2005