On, Thu Aug 23, 2007, Brian Fisher wrote: > On 8/22/07, Marcus von Appen <mva@xxxxxxxxxxxx> wrote: > > Sure. And anything else, what might be added to the PixelArray later on, > > as well. As a result we would have a glorious > > can-do-anything-implements-all-python-interfaces class called Surface :-). > > > Smiley notwithstanding, your entire argument that pixelarray > functonality shouldn't be in surface is a Reductio ad absurdum > argument. It presumes that if you would add this one thing, everything > else would also be added. Using that logic, then we should never add > any functionality to Surface - ever - because it would be the > beginning of a landslide of kitchen sinks freely flowing into surface. No - separate purposes are best dealt with in separate, non-interfering cases. As you clearly state later on in your mail, PixelArray serves a complete own need with a completely own (currently only a minimum) function set. My short- to midterm goals are full mapping support (array[x1:x2, y1:y2] ...), maybe the one or other fast manipulation function and whatever else might fit. Putting all this functionality directly into the Surface does not seem appropriate to me. > The whole point of Lenard's question was that it seems like the > functionality of PixelArray may be appropriate for being part of > Surface, that it may be different from all the things we shouldn't > add. You didn't address why it would or wouldn't be different than all > those things. So basically I don't think your answer addressed his > question. Agreed. I must confess, that I was sure that people would know, what the PixelArray class shall be capable of and what side-effects it might cause if implemented in another way (esp. regarding the buffer interface for interoperability with PIL, Numeric and anything else, which supports direct buffer access). Sorry, if I kept myself unclear about it. [...PixelArray might just be redundant...] > ...however keeping PixelArray as a seperate object but hiding it as an > "interface" would still serve the same goals How would the interface look like in your opinion? [...] > So I think the answer to why it couldn't be part of surface is slicig > & automagic managing or surface locking... And the completely different funcitonality purpose ;-). Regards Marcus
Attachment:
pgpfrApcgs3vA.pgp
Description: PGP signature