"Guest, S (Steve) " wrote:
> I'm looking at using VisAD for applications in astronomy and STP. FITS is a
> common format in both so I thought I'd try that first. However I'm having
> some problems.
>
> Firstly it appears to fail with a 'No data' exception when the file is not a
> multiple of 2880 bytes. This is permitted by the standard and should be
> treated as end-of-file.
Oops, sorry! I've got that fixed.
I'm only allowing short files if the padding is absent, and
throwing an exception if end-of-file is found where data is
expected. Does this sound correct?
> Secondly it doesn't appear to handle arrays within tables:
> Ignoring assumed 6-char String column
> FitsAdapter.buildBTRange: Column 0 was short 10 of 10 rows
> FitsAdapter.copyColumn: Punting on wide column ([I)
> FitsAdapter.buildBTRange: Column 6 was short 10 of 10 rows
> FitsAdapter.copyColumn: Punting on wide column ([I)
> FitsAdapter.buildBTRange: Column 7 was short 10 of 10 rows
>
> These columns have format 6A, 4J, 4J respectively.
>
> Are these problems known about? Is the FITS support still active?
The FITS adapter was my first VisAD project, and it only covers a
few data types. My current project is a generalized file adapter
interface and, as part of that, I'm revisiting the FITS adapter
and trying to make it a bit more robust.
However, I'll check out this problem. Any chance that I could
get a copy of your data file?
