Mon, 02/23/2015 - 21:28
Forgive me, but I'd like to know if it might be possible, when saving APASS data, to have the 'NA' columns changed to 'NaN' at some future date. The official reason might be that NaN makes all the data numeric (except of course the first row), providing a uniformity that makes the data simpler to parse. The selfish reason is that functions, such as Matlab's csvread, already assume this uniformity and thus can be used w/o a preparser (e.g. I like to fart it). Thanks, Jesse
Hi Jesse,
The raw APASS catalog does not have "NA" in it, so are you referring to using Seqplot, or the APASS web query, or what?
Arne
Hi, I was refering to the APASS web query.
I've got Matlab code that reads a FITS header to determine the RA, Dec and approximate FOV then issues an http request to download the relevant CSV. I'd like to eliminate the step that translates the NAs to NaNs when I subsequently read the file e.g. for transform coeficient calculations (per my last post).
Thanks, Jesse
Hi Jesse,
Thanks for the clarification. Matt Templeton wrote the web interface; I'll talk to him. It may not be worth the effort, as APASS will be available through Vizier in about a month, and you will have much more flexibility through that portal.
Arne
Does the move to
Hi again,
Does the move to VizieR mean you'll drop the existing access? I hope not. "More flexibility" typically implies increased complexity and one only needs to go to the VizieR homepage for proof. I also noticed that commas were missing from the three seperated value download options they provide (viz. pipe, semi and tab).
Thanks,
Jesse