I always use a script for requesting charts that invokes the VSX API. This script uses a request of the form:
Showing an example for star Z Cam, a chart 510 minutes wide, north up and east left.
This form works OK since for a long time , but in the last few days I noticed the strange phenomenon that it does not work anymore for just R Cam.
When trying to get a chart for just that star, using the same request form:
I get the following error:
Oops! Looks like something went wrong on this page. AAVSO staff has been notified of the problem.
I wonder if I am doing something wrong in just this case, or if the problem has its origin on the server side.
Thankful if someone chan ckeck this and comment back.
After posting the previous post, I realised that the plotting routine seemingly fails when the plot includes ("goes over") the pole. It is indeed possible to plot R Cam or any other star close to the pole if the chart is small enough not to contain the pole itself or extend "to the other side". It is not a bug. It seems reasonable from a cartographic point of view. Problem explained.
Good work identifying a work-around!
The problem happens in the json encoding; it doesn't happen for the same request if format=xml
Why that is related to including the pole?? Good question.
I'll bump this problem up the ladder.
Thanks for your answer, George. You are correct that posting the same request but asking for xml encoding instead of json does not lead to a problem. I may be wrong, but as I see the XML does not retrieve coordinates, I then believe it is apparently not involving coordinate calculations beyond finding stars located within some radius. However, this problem appears when one plots an actual map because one needs to calculate the coordinates of the corners. As the pole is, cartographically speaking, a singular point, I suspect the problem appears when some of the corners of the map have to be calculated to include points that may not fit in a simple scheme of a "min X, min Y: max X, max Y" square. I do not know what happens behind the VSX/VSP scenes, but from my experience plotting maps I know that one needs to calculate some special cases when one plots maps that include the pole. Perhaps the calculations behind VSX/VSP are not made to manage that complexity. This is of course not a problem, but is good to know that it may happen anyway.
Cheers and thanks for the good work in this tool!