Subject: Shared software and algorithms

Dear friends,

Taking yesterday off erased my memory completely so my role as
'discussion initiator' comes with less preparation than planned.
My intention was to go through the ADD to spot where similar
functionality appears.

I'm sending this out quickly - I'll make additions if ideas
will pop up.

/NJ

--------------

Shared software and algorithms.

Basic level:
We all use saving of FITS tables with results of the processing.
Some of us save images. Both with the help of DAL routines.
There is a lot of keyword copying, defining structures etc.

Is there something to gain by defining a function with a lot of
arguments so that this process can be done in a single call
(something like DAL4) ?

Higher level:
Example with background: I calculate the predicted background
shadowgram using Sami's input maps and tables. Stefan uses his
own background estimate (right ??), but Peter needs something
for the non-imaging binning.
Could we benefit from code sharing here?
Are there other examples ?

----------
Niels J. Westergaard
Danish Space Research Institute
Juliane Maries Vej 30, DK 2100 Copenhagen O
Phone: +45 35 32 57 05; FAX:  +45 35 36 24 75


Subject: Re: Shared software and algorithms
Sender: pkretsch@mbx.unige.ch
To: Niels Joergen Westergaard <njw@dsri.dk>, Carol Anne Oxborrow <oxborrow@dsri.dk>, Juhani Huovelin <Juhani.Huovelin@Helsinki.fi>, Niels Lund <nl@dsri.dk>, Niels Jørgen Westergaard <njw@dsri.dk>, Søren Brandt <sb@dsri.dk>, Stefan Larsson <stefan@astro.su.se>, Sami Maisala <Sami.Maisala@astro.Helsinki.fi>, Carl Budtz-Jørgensen <carl@dsri.dk>, Jerome Chenevez <jerome@dsri.dk>, Silvia Martínez Núñez <silvia.martinez@uv.es>
Message-id: <400E5EC0.DFF71807@obs.unige.ch>
Organization: ISDC
MIME-version: 1.0
X-Mailer: Mozilla 4.79 [en] (X11; U; SunOS 5.8 sun4u)
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 8BIT
X-Accept-Language: de,fr
References: <20040121100919.2AB693E020C@uhuru.dsri.dk>
X-Comment: This message was scanned against viruses by mbx.unige.ch.


  Thanks for starting this Niels Jørgen!

  From my POV I see the following areas where we could profit from
shared development:

* Data input beyond the event data, i.e. for things like GTIs,
  the greyfilter & deadtime history but also instrument model data
  and background information.

  Partially we have already exchanged functions for these, but
  this could be streamlined and cleaned up. See also my mail from
  Dec 5 last year, where I proposed wrt AI 26.10 the two functions:

> 
> JEMXgetGTI(group,
>            instrument,                       [JMX1 or JMX2]
>            gtiNamesUsed,                     [usually "MERGED"]
>            *numGTI,
>            **gtiStartOBT,
>            **gtiEndOBT,
>            **gtiStartIJD,
>            **gtiEndIJD,
>            chatter,status)
> 
>   would take an ISDC group (SWG, OG or Index of SWG)
>   loop over all ScWs in there, allocate buffers of
>   sufficient size and fill them with GTI information in 
>   OBT and IJD. numGTI tells you how many there are in total.
> 
> 
> JEMXgoodFraction(numIntervals,*intervalStart,*intervalEnd,
>                  numGTI,*gtiStartIJD,*gtiEndIJD,
>                  **goodFraction,
>                  chatter,status)
> 
>    would loop over a list of time intervals in IJD, and calculate
>    for each the fraction (0.0-1.0 covered by the given GTIs).


* Data output, i.e. clear writing functions for images, spectra and
  lightcurves at least. So I agree with Niels Jørgen, but would like  
  to extend that.

* In my opinion there should also be a core set of functions to calculate
  illuminated areas, vignetting effects etc. This may be the most difficult
  to agree on and to implement, but it would probably help us a lot if
  we had those.

  
  'See' you soon,
  Peter