W3C home > Mailing lists > Public > public-qa-dev@w3.org > May 2010

Re: Unicorn and network access

From: Thomas Gambet <tgambet@w3.org>
Date: Tue, 18 May 2010 18:58:24 +0200
Message-ID: <4BF2C730.6050300@w3.org>
To: Jean-Guilhem Rouel <jean-gui@w3.org>
CC: Dan Connolly <connolly@w3.org>, public-qa-dev <public-qa-dev@w3.org>
Le 18/05/2010 18:13, Jean-Guilhem Rouel a écrit :
> Le 18/05/2010 17:41, Dan Connolly a écrit :
>> The Unicorn module I just wrote does its own network access;
>> i.e. unicorn gives me a URI, I fetch it, and look at the contents.
>> Is that as designed?
>> i.e. if there are 5 observers, does the page get fetched 5 times?
> Yes, one time per observer (or more if the observer decides to). 
> Actually, I think Unicorn also fetches it one time, to see if the 
> document can be reached to avoid calling many observers if that's not 
> necessary.
> That may seem a lot, but I think that's the only way unless we used 
> some complicated caching mechanism on Unicorn's side (which could lead 
> to a loss of information btw).
> Jean-Gui

You're right, I never realised it but Unicorn does fetch the document 
although it is useless because we only need the response code. I'll fix it.

Received on Tuesday, 18 May 2010 16:58:35 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:36:28 UTC