[NeXus-committee] DECTRIS example summary

[Watts Benjamin]

To be precise, Ray, we just need a way of recording the order of the
data sections so that someone else's program can do the actual
recombination. In this case a simple ordered list would be enough, but
maybe we should be forward thinking and allow an n-dimensional array of
links for a more general solution.

Cheers,
Ben 

-----Original Message-----
From: Ray Osborn [mailto:ROsborn at anl.gov] 
Sent: Montag, 14. Januar 2013 16:26
To: Watts Benjamin
Cc: Discussions of the NeXus Advisory Committee
Subject: Re: [NeXus-committee] DECTRIS example summary

Sorry to be late to this discussion, but is there much difference
between storing multiple data sets in separate NXdata's (and having to
stitch them together later) and, for example, the SNS practice of
breaking their detector banks up into 100 or so NXdata's that are also
just a single data set. In the first case, it's because of performance
issues - in the second, it's because they later want to parallelize
analysis operations on separate banks.

In either case, we need a method of recombining them, although the SNS
ones will not necessarily have contiguous pixels.

Ray
P.S. I think this is resolved in earlier emails, but to reiterate there
should be at least one signal=1 array attribute in each NXdata group.

On Jan 14, 2013, at 8:51 AM, Watts Benjamin <Benjamin.Watts at psi.ch>
wrote:

> I think that everyone is looking at the Dectris data the wrong way. It

> is a single data array and is only broken up into separate files 
> because of performance issues. Therefore, ideally, all the data shoud 
> be written into a single NXdata group. Since we are forced to receive 
> it in separate files, linking is an obvious solution. However, links 
> in NeXus have always been discrete - mapping one location to one array

> in another location. Our situation here is different, with the one 
> array being separated into a number of pieces in different locations. 
> The master file needs some way of saying "my data is there, and then 
> continues there, and there and there...", which does not yet exist (to

> the best of my knowledge). Options for doing this include:
> 
> 1. The NXdata group of the master file contains a list of external 
> links, with the order of the list describing the order in which the 
> linked arrays should be concatenated.
> 2. The master NXdata group links to the first part of the array and 
> then the file containing that first part has some record pointing to 
> where the data is continued.
> 3. Both could be done together.
> 
> Option #1 seems best to me.
> 
> Cheers,
> Ben
> 
> -----Original Message-----
> From: nexus-committee-bounces at nexusformat.org
> [mailto:nexus-committee-bounces at nexusformat.org] On Behalf Of yayahjb
> Sent: Montag, 14. Januar 2013 15:18
> To: Koennecke Mark
> Cc: fcb at bernstein-plus-sons.com; Discussions of the NeXus Advisory 
> Committee
> Subject: Re: [NeXus-committee] DECTRIS example summary
> 
> The one fussy disagreement I have with this summary would be on 
> signal=1.  I am not sure this matters, but I think the current rule is

> that we are only supposed to have one dataset with "plottable" data, 
> i.e.
> with signal=1.  If that rule is important, then I we probably should 
> ask for signal=2, signal=3, etc. on the second, third, etc. dataset.
> I think this is actually consistent with an old NeXus rule when you 
> had multiple datasets.  The alternatives would be to change the rule 
> and say
> either:
> 
>   You can have multiple plotable datasets with signal=1 or
> 
>   The plotable datasets have signal=n, where n designates the order of

> plotting, starting with 1.
> 
> The second version would be consistent with this particular data.
> 
> Sorry to be fussy
> 
>   Regards,
>     Herbert
> 
> 
> On 1/14/13 8:38 AM, Mark Koennecke wrote:
>> Hi,
>> 
>> from your feedback and my own study I find that there are the 
>> following problems with the example file as is:
>> 
>> - Structure is wrong regarding linked data sets. Should be either:
>> 
>>  * entry:NXentry/data_00001:NXdata/data (Preferred ....)
>>  * entry:NXData/data:NXdata/data_0001
>> 
>> - On data, signal=1 attribute is missing
>> - On external link, NeXus attribute
>> NAPIMOUNT=nxfile://th02c_ps02_1_data_000001.h5#data
>>  missing
>> - In detector, flatfield, efficiency data is missing. This is not 
>> critical,  I just wonder.
>> 
>> Am I right? Is this what I tell DECTRIS tomorrow? Or do I miss
> something?
>> 
>> Also, what is the opinion, do we need another Telco on this one?
>> 
>> Regards,
>> 
>>             Mark
>> 
>> 
>> _______________________________________________
>> NeXus-committee mailing list
>> NeXus-committee at nexusformat.org
>> http://lists.nexusformat.org/mailman/listinfo/nexus-committee
>> 
> 
> _______________________________________________
> NeXus-committee mailing list
> NeXus-committee at nexusformat.org
> http://lists.nexusformat.org/mailman/listinfo/nexus-committee
> 
> _______________________________________________
> NeXus-committee mailing list
> NeXus-committee at nexusformat.org
> http://lists.nexusformat.org/mailman/listinfo/nexus-committee

--
Ray Osborn, Senior Scientist
Materials Science Division
Argonne National Laboratory
Argonne, IL 60439, USA
Phone: +1 (630) 252-9011
Email: ROsborn at anl.gov