[xquery-talk] Collections - family relationships

Ihe Onwuka ihe.onwuka at gmail.com
Sat Jan 18 19:28:42 PST 2014


On Sun, Jan 19, 2014 at 2:06 AM, Joe Wicentowski <joewiz at gmail.com> wrote:
> In all seriousness now, my satire was meant to help us reflect on and
> acknowledge a pattern of abrasive interactions that has dominated this list
> as of late, to raise it for discussion if needed, and hopefully to bring it
> to a close by letting everyone see its absurdity.
>
> Discussion can be civil but productive, direct and even heated but
> respectful.
>
> When you start a topic, you bear a special responsibility to be
> comprehensive up front, to respect the time and effort everyone takes to
> respond in good faith, and to guide your thread to a productive, respectful
> close.  Acknowledge failures in this responsibility readily.  Neglect it at
> your own peril.
>
> Productive discussions, started and guided by responsible participants, has
> always been the norm on xquery-talk in the time I have been a member.  I
> look forward to its return to that state.
>

Sometimes people have a sense of humour  that may not necessarily
translate like .... of course I say nothing of the sort when raising
it on eXist.............

yeah I was thinking that but y'know how it is when you go in all guns
blazing and claiming you've found a bug...... but now that you have
emboldened my keyboard I'm thinking of a subject heading like t will
be Bug. Fix it you b******ds.

On Thu, Jan 2, 2014 at 9:18 AM, Joe Wicentowski <joewiz at gmail.com> wrote:
I think you may have found a bug in eXist-db.  Your code can be
reduced to the following tests:

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Sometimes you need a sense of humour when your morning is dissipating
in the face of a simple query that turns into a runaway task

When the scheduler says terminating does that me............... or
rather is it meant to  stop saying terminating.


On Tue, Jan 14, 2014 at 11:49 AM, Ihe Onwuka <ihe.onwuka at gmail.com> wrote:
> So I killed it..... and the bastard is now taking an age to die.
>
> On Tue, Jan 14, 2014 at 11:31 AM, Ihe Onwuka <ihe.onwuka at gmail.com> wrote:
>> Killing it would leave me with a half-updated database would it
>> not,...... as long as it finishes.
>

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

I found alot of things just in the course of doing stuff.

That eXist doesn't let you do let after group by.
That it doesn't support the count clause in 3.0,
That there is something wrong with it's implementation of the
position() function because it sometimes gives 0.
That the order of it's arguments to HOF's are not spec-compliant

and you think - yeah this stuff is not a problem for me because I've
solved it, or you think I'm not going to use this thing ...say for
example if position() doesn't give me the nth out of m then it doesn't
do what it says on the tin and you don't want to use it

Of course you could avoid being cast as a villain, by just doing your
thing and keeping quiet. Sometimes my attitude is look there is
something wrong here - I have moved on but I'm doing my bit for the
community by reporting it.

________________________________________________________________________________________________________

It doesn't seem that you can pose a question here without somebody
trying to force feed it into something that can be answered with code
or reference to the spec. This thread is a prime but not sole example.
The thread divides into 2 strands. Those that do the former and those
that see the issue. You give the thing a title - Collections - Family
relationships. Your original post says you are asking whether and why
document elements are or are not siblings and somehow that gets parsed
into - I don't know how to group these things and count them.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

and wobetide you if you ask why something is the way it is  - because
the spec says so, you don't know anything about the language,  if you
don't like it go off and write your own.


More information about the talk mailing list