Hey all,
> I'll be linking the code here as soon as there is something substantial
to look at.
We're now 6 years later and I finally have something to show!
A quick recap: writing result formats for SMW is a big pain due to a number
of serious design issues. And testing is (almost??) impossible.
Hey,
I figured that I do not have enough time to do this nicely integrated with
SMW. Both the query result objects and the result printers are best
rewritten from ground up to get rid of all the big issues. And while doing
this, future interoperability with Wikibase could be achieved. Out of scope
Hi,
+1
This last time I had to write a unit test for the table format, I had
to bit my hand off to get some sane results.
If you can make it in time, don't show them "old" interface as this
one is not going to last and it doesn't improve one's understanding in
how to interact with a query format
Hi,
Well, you make a strong case. And no, I think doing a result format makes a
lot of sense. Good luck!
-Yaron
On Thu, Oct 24, 2013 at 10:17 PM, Jeroen De Dauw wrote:
> Hey,
>
> The point of this workshop is not to teach people SMW interfaces. The
> amount of ground that can be covered in a o
Hey,
The point of this workshop is not to teach people SMW interfaces. The
amount of ground that can be covered in a one hour workshop aimed at
beginners is rather small. So however this is done, people will still not
know nearly all interfaces. So even if it was the case that this was likely
to c
Well, yes, that's what I meant - though I guess it depends on (a) how
likely this new code is to stay in SMW in its current form, (b) when it
would be released as part of a stable version, and (c) how different this
new interface is from the current one. But my vote would still be to teach
the old
This will exist when I give the tutorial. Are you saying that it should be
in a released SMW version first?
Sent from my HTC one X.
--
October Webinars: Code for Performance
Free Intel webinars can help you accelerate app
Hi,
It's great that you're working to improve the code, but isn't it weird to
teach people how to use an interface that doesn't exist yet? If they go to
create a new result format after the tutorial, won't their newfound
knowledge be (for the moment) useless?
-Yaron
On Thu, Oct 24, 2013 at 6:30
Hey,
While working on my dev workshop for SMWCon, I noticed it is currently
rather impossible to write a sane unit test for a result format. The
reasons for this are that the result object, SMWQueryResult and the objects
it is composed of, contain a ton of stuff not relevant to those tests,
includ
