+1 on documenting existing implementation first
On Wed, Aug 28, 2013 at 12:15 PM, Shazron <[email protected]> wrote: > IMO first step would be that we find out what the existing implementations > actually do, and doc them > do our standard deprecation dance and implement the new shiny and correct > functionality > > > On Thu, Aug 29, 2013 at 12:04 AM, James Jong <[email protected]> wrote: > > > Several ways we could go here. One is to improve the documentation. iOS > > chooses the larger image size. I'm not sure if all the platforms do it > the > > same way. I'd be happy to update the doc if it's all the same. Second > is > > to modify the implementation to only require either W or H. Note that we > > would break backwards compatibility there. > > > > -James Jong > > > > On Aug 28, 2013, at 10:16 AM, Ray Camden <[email protected]> wrote: > > > > > As a user though, that doesn't necessarily make sense. You are saying, > > > "You must give me a H and W, but I'm going to maintain the aspect ratio > > no > > > matter what." Given that, which side is "corrected" if I pass a H and W > > > that do not maintain the aspect ratio? Do we document it? I've worked > on > > > another platform that provides a way to pass a H and W that act as a > > > bounding box, so if I use 150x150, my final result will be no bigger > than > > > 150x150, but possibly smaller in order to maintain aspect ratio. If > that > > > is what PG is doing, then the docs should clearly spell that out. Maybe > > it > > > is assumed by "Aspect ratio remains constant", but I think it could be > > > clearer. > > > > > > On 8/28/13 9:04 AM, "James Jong" <[email protected]> wrote: > > > > > >> You're right that it could be calculated based on one or the other. > The > > >> code expects both today. I think the point is to be clear that the > > >> aspect ratio is maintained, so that the user does not expect to be > able > > >> to arbitrarily set both. > > >> > > >> -James Jong > > >> > > >> On Aug 28, 2013, at 7:15 AM, John Wargo <[email protected]> wrote: > > >> > > >>> I've got another silly question. In looking at the Camera API, I see > > >>> the following: > > >>> > > >>> targetWidth: Width in pixels to scale image. Must be used with > > >>> targetHeight. Aspect ratio remains constant. (Number) > > >>> > > >>> targetHeight: Height in pixels to scale image. Must be used with > > >>> targetWidth. Aspect ratio remains constant. (Number) > > >>> > > >>> I'm not getting why targetWidth MUST be used with targetHeight (and > > >>> visa versa) when aspect ratio remains constant. If aspect ratio > remains > > >>> constant, then setting one automatically forces the other - that's > the > > >>> whole point of maintaining aspect ratio, right? If the API is > > >>> maintaining aspect ratio while sizing the image, then forcing the > > >>> developer to specify both parameters is simply wasted work. > > >>> > > >>> What am I missing here? > > >> > > > > > > > > -- Carlos Santana <[email protected]>
