Re: T0006 Wide at CADC.

From: Kanoa Withington <kanoa_at_cfht.hawaii.edu>
Date: Mon, 14 Sep 2009 14:40:46 -1000

Hi John,

I'm pretty sure fitsverify does not have any sort of severity setting or
contextual awareness (ie: image files vs tables) and for that reason I
question its utility as in input filter. I am always an advocate for
quality control and consistency checking - I'm just not sure fitsverify
is the right tool for the job.

Has anyone heard of any alternatives? What would be perfect is something
that just checks the basic FITS structure (ie: padding is correct,
NEXTEND matches found extension, etc.) and critical errors that could
cause a file to be mis-interpreted (ie: datetime formatting). Anything
beyond that is really up to the data provider to format and check.

-Kanoa

John Ouellette wrote, On 09/14/2009 12:53 PM:
> Hi Kanoa -- going through the files that have been rejected due to
> fitsverify errors, it looks like the errors are rather benign: they seem
> to be failing due to a problem with the AUTHOR keyword.
>
> I'm loathe to just turn off the fits verification, and would rather that
> we determine why some files have the above errors and others do not. I'm
> not sure whether we can choose to ignore certain FITS errors while still
> catching fatal errors, but I will enquire with the developers here.
>
> J.
Received on Mon Sep 14 2009 - 14:40:50 HST