A user (thanks Steve U!) brought to our attention a problem with the GALEX survey in SkyView. SkyView uses the GALEX archive at MAST and only retrieves files when a user first requests data in a region. We keep files in a cache, so we never request the same file twice. Only about 1/2 of the GALEX GR6 data are cached, so user requests often require us to retrieve data from MAST. It turns out that MAST reprocessed a bit under 1% of the GR6 data after we got the URL locations, which meant that the URLs that we were using to point to the data were out-of-date. If we already had the image in our cache all was copacetic, but an attempt to download an updated file caused a complete failure for that survey.
Version 3.1.4 of SkyView has been released with updated survey description files for the two GALEX surveys and should address this problem. Users of the SkyView-in-a-Jar should download the latest jar if they want to retrieve GALEX data.
In testing this out, I noticed that SkyView was failing due to memory problems when trying to tile together a large number of images. This was due to a failure to deallocate memory when we were using 4-byte reals in the input images. [That was put in recently to allow us to handle larger input images.] That’s fixed too though we haven’t heard of anyone having problems with it.
Hello,
Are you making GALEX GR7 available anytime soon?
Thanks!
We’ll try to get around to in sometime over the next month or so. It’s unclear how much additional
coverage there is.
One thing I noticed was the Cygnus Loop only shows up in GR7. I was hoping to see it in SkyView so I could change the projection map and use it in my project. Its a pretty nice size relative to the moon and impressive looking. Also I think their were some added tiles or adendums that improved the Mira star trail like shown here: http://upload.wikimedia.org/wikipedia/commons/3/38/A-Mira-Full_down_sampled_and_cropped.jpg which I was also hoping to reproject. Looking forward to doing that with SkyView.
We’re looking at updating the GALEX survey but there seem to be some inconsistencies between MAST’s catalog of GALEX observations and the archive that we’re trying to track down. So it will not happen until next week at the earliest.