[AstroPy] Opening file in PyFits, works in v2.4, not in v3.0.6

Erik Bray embray@stsci....
Tue Apr 3 12:32:02 CDT 2012


On 04/03/2012 01:20 PM, Erik Bray wrote:
> Part of me agrees with this in principle, but in practice it just
> wouldn't work.  It's like how part of me wishes that whenever a web
> browser hits some malformed or invalid HTML it would just throw up an
> error and refuse to render the page.
>
> But it's a good thing that web browsers *don't* do that, as such
> behavior would have severely hindered the adoption and growth of the
> web.  I think that FITS, for better or for worse, needs to be treated a
> little like HTML in that respect.
>
> In particular, it shouldn't throw up errors when it's fairly unambiguous
> what was *meant*.  There's absolutely nothing about the "spaces for
> padding" requirement that affects parsing of the header--as long as the
> header ends on the right alignment that's all that matters.
>
> I already have a fix in git (and subversion as soon as a I `git svn
> dcommit`) for this issue.  So I'll try to slip that in to the 3.0.7
> release, which I hope to get done today or tomorrow.
>
> Incidentally, PyFITS has no problem with the invalid apostrophe in the
> origin keyword.  Maybe it should say something about that and consider
> fixing it...

Though I also just found that PyFITS does have some problems with the 
malformatted float values in the header that might be worth investigating.

Erik



More information about the AstroPy mailing list