Yes they were all green before loading the ODBC for Squeak package.  I have also verified with previous versions of Squeak (3.7 up to 3.10) and they were also green for all Squeak versions BEFORE loading the ODBC for "Squeak package".  In other words, it's not a Pharo related bug, it affects all versions of Squeak and it's caused by the ODBC package...

A very quick look at it shows that the ODBC package has changed the way we print TimeStamp (an extra comma to separate the date from the time) and that the #date method in class Time returns the receiver #asDate, not the "plain" instance variable.  Not rocket science to fix but I wonder if it won't break everything on the ODBC side.  Besides, one might ask if it makes sense to convert a <Time> into a TimeStamp (as I used to say to Smalltalk newcomers, "yes you can use a chair as a table and vice-versa but does it really make sense" ?).  Well, that's what this method is trying to do.  I would have to take a closer look at the actual implication and effects of possible fixes for this method and/or odbc-related methods.

I will enter 3 bugs into the tracker, ODBC tests failing, the find string tool and the email bug report in the debugger.

I know, I complain all the time!  ;) 

I should have been a QA tester!!!  lol


 
-----------------
Benoit St-Jean
Yahoo! Messenger: bstjean
Blog: lamneth.wordpress.com
A standpoint is an intellectual horizon of radius zero.
(Albert Einstein)


From: St�phane Ducasse <stephane.ducasse@inria.fr>
To: Pharo Development <Pharo-project@lists.gforge.inria.fr>
Sent: Wednesday, December 10, 2008 3:01:23 PM
Subject: Re: [Pharo-project] ODBC package bugs

Hi benoit


On Dec 10, 2008, at 8:53 PM, Benoit St-Jean wrote:

> Hello everyone,
>
> Since I am not familiar with the Pharo process for reporting/fixing 
> bugs, here's a quick email to let you know what I have found by 
> loading (in order) FFI, ODBC and ODBCEnh packages from SqueakMap
> FYI :
>
> Loading the "ODBC for Squeak" package from SqueakMap into Pharo 
> breaks 8 unit tests (4 failures, 4 errors), namely:

Were these tests green  before loading ODBC?
Then do you think that this is because ODBC extend/change TimeStamp?

> 1) TimeStampTest>>#testPrint
> 2) TimeStampTest>>#testPrinting
> 3) TimeStampTest>>#testStoreOn
> 4) TimeTest>>#testTimeStamp
> 5) DateAndTimeLeapTest>>#testAsTimeStamp
> 6) TimeStampTest>>#testInstanceCreation
> 7) TimeStampTest>>#testReadFromA1
> 8) TimespanTest>>#testAsTimeStamp
>
Thanks can you enter a bug entry on the bug tracker.

> On a different topic, doing a string search in the code (menu item 
> "method source with it") creates a walkback because :
>
> AbstractObjectsAsMethod(Object)>>doesNotUnderstand: #getSourceFor:in:
>
> Also, I have noticed that if I want to mail a bug report from the 
> debugger, the address the email would be sent to is squeak-dev@lists.squeakfoundation.org
>  instead of a "Pharo" one.
>
> Hope this helps.

Sure it does
>
>
> Thank you.
>
> -----------------
> Benoit St-Jean
> Yahoo! Messenger: bstjean
> Blog: lamneth.wordpress.com
> A standpoint is an intellectual horizon of radius zero.
> (Albert Einstein)
>
>
> _______________________________________________
> Pharo-project mailing list
> Pharo-project@lists.gforge.inria.fr
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project


_______________________________________________
Pharo-project mailing list
Pharo-project@lists.gforge.inria.fr
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project