You need to log in before you can comment on or make changes to this bug.
Darren's Comment: Issue noticed during cwc2 1.0.3 training course. > 10. Changing projections and then going to full extent gives wrong > (i.e., not full) extent. > When changing projections using the projection selector the full extent image is not displayed fully. I realize that the shape of the full extent will have changed from a square to longer rectangle eg. lcc to epsg 4326 Lat/Long. Only half of Canada can be displayed. The map view should adjust for this by creating grey/white space above and below the current square view of Canada, to allowing a wide screen effect. When it is under a lat/long projection 1/2 ratio. I am sure this has been observed and would be difficult to resolve, but it is logically a bug and should be revisited.
This is actually not a bug, it is a conscious decision we made during CWC2 development. There are essentially two ways of handling reprojection and each has benefits and drawbacks: 1. reproject bounding box. This ensures that the entire visible area in one projection will also be visible in the new projection (pro) but often results in a smaller scale (larger scale number) and repeated changing between two projections means that you are effectively zooming out (con). 2. reproject centerpoint and preserve scale. This ensures that switching between projections does not zoom out because ground scale is maintained, but it does not guarantee that you are going to see the same information when you switch projections (as in the case noted here). Neither one is right nor wrong. Both are equally valid. We (the developers) have chosen option 2 which is, I admit, contrary to previous implementations in other software we have developed. But I am not going to change it unless there is a compeling reason to do so.
Also note that option #1 above used to cause major problems with some projections where the extents would eventually end up going beyond the area of validity of a given projection and result in mathematical errors and could mean losing the map completely or being teleported to a random location (a bit similar to the theory of improbability which is the fastest way to travel across the universe). That was the main motivation behind our choice to go with #2
