BEWARE: signature of cond() operator changed: ezp 3.2-1 -> 3.2-2

Next topic

Author	Message
Volker Lenz	Friday 17 October 2003 4:29:36 am Why do you alter the signature of a core control operator like this without taking a little care about backward compatibility ? I have to redesign a vast array of templates now, just for nothing.
Jan Borsodi	Friday 17 October 2003 5:28:36 am The signature hasn't changed however a bug was fixed. The old code did not multiply the counter by two so you got strange results. The syntax is: cond(MATCH,RESULT[,MATCH,RESULT]...[,DEFAULTRESULT]) -- Amos Documentation: http://ez.no/ez_publish/documentation FAQ: http://ez.no/ez_publish/documentation/faq
Volker Lenz	Friday 17 October 2003 5:40:29 am Yep, you are right. I apologize. The signature is still the same. However, the old 'unfixed' code actually worked in the most used case of a TRIPLE of parameters (fork mode), while the 'fixed' code does no more. I found the new bug & fixed it. Shall post this as new bug report here. Kind regards Volker

Author

Message

Friday 17 October 2003 4:29:36 am

Why do you alter the signature of a core control operator like this without taking a little care about backward compatibility ?

I have to redesign a vast array of templates now, just for nothing.

Jan Borsodi

Friday 17 October 2003 5:28:36 am

The signature hasn't changed however a bug was fixed. The old code did not multiply the counter by two so you got strange results. The syntax is:

cond(MATCH,RESULT[,MATCH,RESULT]...[,DEFAULTRESULT])

--
Amos

Documentation: http://ez.no/ez_publish/documentation
FAQ: http://ez.no/ez_publish/documentation/faq

Volker Lenz

Friday 17 October 2003 5:40:29 am

Yep, you are right. I apologize. The signature is still the same. However, the old 'unfixed' code actually worked in the most used case of a TRIPLE of parameters (fork mode), while the 'fixed' code does no more. I found the new bug & fixed it. Shall post this as new bug report here.

Kind regards

Volker