Implementation detail of #at:ifAbsent: in COM

Had a slight problem with Seaside this weekend which came down to
Com-ExtensionsToBase overriding Seaside's implementation of
Array>>at:ifAbsent: with its own that does not consider 0-index access as
"absent". So doing "Array new at: 0 ifAbsent: [nil]" COM's version would
return SubscriptOutOfBoundsError whereas the Seaside version would return
nil. Don't you hate these? One could argue for either depending on the mood,
so I just thought I'd throw it out there and see what people think in
general.

Cheers!

-Boris

--
+1.604.689.0322
DeepCove Labs Ltd.
4th floor 595 Howe Street
Vancouver, Canada V6C 2T5

[hidden email]

CONFIDENTIALITY NOTICE

This email is intended only for the persons named in the message
header. Unless otherwise indicated, it contains information that is
private and confidential. If you have received it in error, please
notify the sender and delete the entire message including any
attachments.

Thank you.

Boris Popov wrote:
I would never expect an at:ifAbsent: method to raise a
SubscriptOutOfBoundsError.  That's what the ifAbsent: is there to
prevent.  I think that the seaside version is the right one in this case.

David Buck

> Had a slight problem with Seaside this weekend which came down to
> Com-ExtensionsToBase overriding Seaside's implementation of
> Array>>at:ifAbsent: with its own that does not consider 0-index
> access as "absent". So doing "Array new at: 0 ifAbsent: [nil]"
> COM's version would return SubscriptOutOfBoundsError whereas the
> Seaside version would return nil. Don't you hate these? One could
> argue for either depending on the mood, so I just thought I'd throw it

> out there and see what people think in general.

FWIW, I'm the one who added #at:ifAbsent: (to SequenceableCollection)
to Seaside. I simply copy/pasted the method code from Squeak.
Seaside needs it in a couple of locations where, ironically, the index
value is always positive, i.e. COM's version would be OK.

Michel.

There was a discussion on the list a while back as to whether Array
should have at:ifAbsent: as part of their protocol, that is a valid
question. But if it is implemented then it should be the Squeak
implementation.

There is no point to the implementation in the COM package. Either the
index is in bounds in which case no need for the ifAbsent: block or the
index is out of bounds in which case it will throw an exception so no
need for the ifAbsent: block, you may as well use at:. The only thing
having at:ifAbsent:  buys you is the ability to be a bit sloppy about
the parameters to code that uses at:ifAbsent:. But this is a
pathological bit of polymorphism.

Joerg


-----
Joerg Beekmann
DeepCove Labs
4th floor 595 Howe Street
Vancouver, BC, V6C 2T5
voice +1.604.689.0322
fax   +1.604.689.0311
[hidden email]
 

CONFIDENTIALITY NOTICE
Unless otherwise indicated this email contains information that is
private
and confidential. If you have received it in error, please notify the
sender
and delete this message along with any attachments. it
>
> > out there and see what people think in general.
>
> FWIW, I'm the one who added #at:ifAbsent: (to SequenceableCollection)
> to Seaside. I simply copy/pasted the method code from Squeak.
> Seaside needs it in a couple of locations where, ironically, the index
> value is always positive, i.e. COM's version would be OK.
>
> Michel.