XSF Discussion - 2018-06-14

Is this semantically equivalent?

<foo xmlns="mynamespace"><bar/></foo> <x:foo xmlns:x="mynamespace"><bar/></x:foo>

is the 'bar' element in the same namespace?

I believe the answer is "yes", but never trust anyone giving XML answers off the top of their head :)

Guus, no, it’s not equivalent

<bar/> is in the default namespace in both examples. but in the second example, the default namespace was rebound by <foo xmlns="…"/>, while in the second example, the default namespace is undefined (i.e. no namespace; unless there is outer content which defines the default namespace)

Guus, ``` >>> t1 = lxml.etree.fromstring('<foo xmlns="mynamespace"><bar/></foo>') >>> t2 = lxml.etree.fromstring('<x:foo xmlns:x="mynamespace"><bar/></x:foo>') >>> t1[0].tag '{mynamespace}bar' >>> t2[0].tag 'bar' ``` (this is clarks notation, i.e. {namespace}local-name)

Thanks Jonas, Kev

I said not to trust me :)

also, I meant to say "but in the *first* example", not "but in the second example"

No, but you confirmed my suspicion that I had to check 🙂

jonasw, I assumed as much

I didn't get that clarks notation, btw

Guus, http://www.jclark.com/xml/xmlns.htm

it’s a way to write namespaced elements, independently of prefixes. it’s very useful for debugging and even for an internal data representaiton

it’s sometimes used on the mailing list. the name of the element <foo xmlns="bar"/> would be {bar}foo in clarks notation

{jabber:client}message, etc.

yupp

Prosody uses that notation in its simplified xpath thing

it should be used everywhere :)

ping

Is it Board Time?

ralphm, Guus, nyco

I can't make it, sorry

About to be in a car again

np

Hey, bandwidth issue on mobile

Sorry

Ok, I think we're skipping this week then :)