spam,Re: Testing reports

[Colin Scott]

Ah, gotcha.  Fully agreed!

Colin

-------- Original Message --------

*Subject:* RE: spam,Re: Testing reports
*From:* "John Layman" <john.layman at laymanandlayman.com>
*To:* <gnucash at double-bars.net>
*Date:* Wed, 11 Apr 2012 19:15:20 -0400

> Not quite sure what you are getting at here ...

I'm hazarding a guess as to why anyone would imagine a test should shrug off
case changes in output.  Unless you are trying to avoid side effects of
version change in a generator that's a black box, it's hard to imagine what
a "good reason" would be for ignoring even minor differences in output.
When you compose a software product using components you don't control, the
components can change out from under you.  But your product is a composite,
and the composite is what you have to test, even if the NIH parts sometimes
cause additional work for you.

-----Original Message-----
From: Colin Scott [mailto:gnucash at double-bars.net] 
Sent: Wednesday, April 11, 2012 4:46 PM
To: john.layman at laymanandlayman.com
Cc: gnucash at double-bars.net
Subject: RE: spam,Re: Testing reports


> Software is still deterministic, last time I checked.  

:-)

> Altering test oracles to reflect changes in low-level pieces/parts of 
> a product just comes with the territory where software composition is 
> concerned.  The work required to make those alterations is trivial 
> compared to the alternative.

Agreed.  Good reasons are all I need, but I am pretty insistent that they
should not only exist but that they should be thoroughly checked and tested!

> And, in an event, you ARE testing the composite, not just the part 
> invented here.

Not quite sure what you are getting at here ...

Colin

-------- Original Message --------

*Subject:* RE: spam,Re: Testing reports
*From:* "John Layman" <john.layman at laymanandlayman.com>
*To:* <gnucash at double-bars.net>
*Date:* Wed, 11 Apr 2012 15:38:03 -0400

Software is still deterministic, last time I checked.  Altering test oracles
to reflect changes in low-level pieces/parts of a product just comes with
the territory where software composition is concerned.  The work required to
make those alterations is trivial compared to the alternative.  And, in any
event, you ARE testing the composite, not just the part invented here.

-----Original Message-----
From: gnucash-user-bounces+john.layman=laymanandlayman.com at gnucash.org
[mailto:gnucash-user-bounces+john.layman=laymanandlayman.com at gnucash.org] On
Behalf Of Colin Scott
Sent: Wednesday, April 11, 2012 8:06 AM
To: warlord at MIT.EDU; gnucash at double-bars.net
Cc: gnucash-user at gnucash.org
Subject: Re: spam,Re: Testing reports


> I would argue that case inversion like this doesn't necessarily 
> require a good reason, but that's a minor detail.

Are you saying that you are happy for the output to change without that
change being the planned result of a deliberate, targetted, action?  I would
be *very* unhappy were any program of mine to behave that way!

Colin

-------- Original Message --------

*Subject:* spam,Re: Testing reports
*From:* Derek Atkins <warlord at MIT.EDU>
*To:* gnucash at double-bars.net
*CC:* jralls at ceridwen.us, yawar.amin at gmail.com, gnucash-user at gnucash.org
*Date:* Tue, 10 Apr 2012 13:48:31 -0400

"Colin Scott" <gnucash at double-bars.net> writes:

> Sorry for the delay - I've been away.
>
>> Define "html text"?  For example, would you consider this:
>> 
>>   <A HREF="foo">foo</A>
>> 
>> and this:
>> 
>>   <a href="foo">foo</a>
>
> For the purposes of the exercise in hand, I would consider them to be 
> different.  Working out whether the difference is significant is not 
> something I would want to leave to a robot - not because it isn't 
> perfectly possible to do robotically, but because it simply isn't 
> worth devoting that much time and effort to inventing a robot that 
> will do it reliably.  After all, even such an apparently minor change 
> as a case inversion shouldn't be made without a damn' good reason!

I would argue that case inversion like this doesn't necessarily require a
good reason, but that's a minor detail.

>>  What about:
>>   <a href="foo">foo</a>    and    <a href="bar">foo</a>
>> ?
>
> Different under *any* circumstances!  :-)

Yes, I expected this answer, and I said so in my message :) But you didn't
respond to my third example:

>    Similarly, I would want these two to result in a "match":
>
>   <p>This is a paragraph</p>
>
> and
>
>   <p>This
>   is
>   a parahraph</p>

What would do here?

> -derek

-derek

-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord at MIT.EDU                        PGP key available

_______________________________________________
gnucash-user mailing list
gnucash-user at gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-user
-----
Please remember to CC this list on all your replies.
You can do this by using Reply-To-List or Reply-All.