[daip] UVFIX on Macs

[Eric Greisen]

On 2/29/20 10:44 AM, Andy Biggs via Daip wrote:
> Hi,
> 
> I've been reducing some old VLA data and noticed something when running 
> UVFIX. As a dataset was observed using B1950 coordinates I ran UVFIX to 
> precess it to J2000. As far as I can tell, UVFIX with its inputs set to 
> their defaults will do this.
> 
> On Linux, plotting the data as a function of uv distance will look 
> almost identical before and after running UVFIX, as I would expect. On 
> my Mac, the max uv point moves from ~170 kl to 40 Ml. See the attached 
> screenshot.
> 
> I've put the data in my ftp area if you'd like to take a look:
> 
> ftp://ftp.eso.org/pub/general/abiggs/DATA.FITS
> 
> My Mac is a MacBook Pro running an up-to-date Mojave. I tried it in 
> 31DEC19 and 31DEC20 - same results.

This bug came to my attention recently.  It affects all computers and, 
in fact, there was also a subtle bug as well as the gross 1950->2000 bug
in the computation of the differential aberation correction.

It has been fixed in 31DEC19 and 31DEC20 - the gross bug was not in 31DEC18.

Thanks

Eric Greisen