[Comm2011] code to get the 32 intersting raws when imaging some standard stars on all the cjips

[Jeffrey Bout]

Hi all,
Here some code to determine for a bunch of rawscience frames for
'putting some standard stars on all the 32 chips', which of the
rawscience frames are the ones that image the standard stars:

1) get some raws...

awe> context.set_project('COMM2011')
awe> raws = (RawScienceFrame.OBJECT=='SA110')&(RawScienceFrame.is_valid<>0)&(RawScienceFrame.DATE_OBS>datetime.datetime(2011,5,9))&(RawScienceFrame.filter.name=='OCAM_g_SDSS')&(RawScienceFrame.astrom.CRVAL1<>280.458333)
# not 280.458333 which is not part of the 32 pointings

2) separate the different exposures

awe> from astro.experimental import jtools
awe> rg = jtools.group_list(raws, 'DATE_OBS')

3) simulate the the boundingbox of each corr. coadd.

awe> coadds = [jtools.BoundingBox({'method':'union',
'objects':frames}) for frames in rg.values()]

4) get the overlap of all these coadds, this is the interesting area

awe> overlap = jtools.BoundingBox({'method':'intersection', 'objects':coadds})

5) optional: a nice plot?

awe> import pylab as pl
awe> pl.clf()
awe> for c in coadds: c.plot()
awe> overlap.plot('g', cross=True)

6) get the raws that overlap more than 90% with the 'coadds-overlap'

awe> overlapping_raws = dict([(key, jtools.get_value(filter(lambda
frame: jtools.BoundingBox(frame).get_overlap(overlap, fraction=True) >
0.9, frames))) for key, frames in rg.items()])