16690 breaks `make check` [WAS Re: AUDIT: r16690 - gnucash/trunk/src/engine - Daniel Harding's update to Afghani currency. closes #504257]

[Andrew Sackville-West]

On Mon, Dec 24, 2007 at 03:04:00PM +0430, Daniel Harding wrote:
> Andrew Sackville-West wrote:
>> On Sun, Dec 23, 2007 at 11:19:42PM -0800, Andrew Sackville-West wrote:
>>   
>>> Well 16690 breaks make check. 
>>> meanwhile I discovered that src/engine/iso-currencies-to-c isn't
>>> getting run. At least it looks that way because its supposed to update
>>> iso-4217-currencies.c and it doesn't. That doesn't seem to be the
>>> source of the failure but is something to note anyway. I did manually
>>> run iso-currencies-to-c, saw that it appeared to update
>>> iso-4217-currencies.c properly and make check still failed. :(
>>>     
>>
>> okay, ignore that. sorry, it does run properly.
>>   
> I'm on Windows, and when I was working on that patch, it appeared that at 
> least some times iso-4217-currencies.c was not getting regenerated 
> properly.  I ended up just deleting iso-4217-currencies.c before rebuilding 
> any time I changed iso-4217-currencies.scm.  So there may still be 
> something going on here.

fair enough.

>>> But interestingly, Derek reverted 16619 and that causes make check to
>>> succeed. Seems odd to me that adjusting a currency would cause 16619
>>> to cause a test to break. So maybe there is more going on here than
>>> meets the eye.     
>>
>> Somehow, the two lines in 16690 cause make check to fails. If you
>> delete either line (or jsut comment the out), it passes. Leave them
>> both in, it fails. I can't see what the heck would cause it.
>>   
> Can you give more details about which test(s) are failing? 

Lots. part of the engine tests.

> Being on 
> Windows, make check fails for me all over the place (I'm willing to help 
> improve this, but won't be able to tackle it right away).  However, I have 
> been able to successfully run some individual tests, so if I know where to 
> look, I might be able to help investigate.  Does the tests also fail if you 
> remove/comment out both currencies?

it works fine in any combination other than two Afghani currencies. I
even tried tweaking some of the values in the two lines to no avail,
though I only tried a little. I didn't try moving one of the Afghani
lines farther into the file in case it is somehow a problem with
having two of the same currencies first. shrug. 

I think it's a case of this change exposing problems elsewhere, maybe
even in the tests themselves.

Oh, and can you comment at all on the discussion about whether there
should be two Afghani currencies or just one? 

A
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://lists.gnucash.org/pipermail/gnucash-devel/attachments/20071224/46ab7930/attachment.bin