SEI plots and possibly others show data in Ready state when there is no active segment

[eagoetz]

See for example
https://ldas-jobs.ligo-la.caltech.edu/~evan.goetz/summary/gps/1357516818-1357520418/sei/ham7/#isolated
and
https://ldas-jobs.ligo-la.caltech.edu/~evan.goetz/summary/gps/1357516818-1357520418/sei/ham7/#ready

There is no active segment for this interval in the "Ready" state:

<SummaryState('Ready',
              known=[[1357516818 ... 1357520418)]
              active=[]
              description=None)>

but nonetheless, there is a plot with data shown.

Something seems amiss here.

To reproduce this example:

python -m pdb -m gwsumm gps --verbose --ifo L1 --on-segdb-error warn --on-datafind-error warn --output-dir . 1357516818 1357520418 --config-file ${CONFIG}/defaults.ini,${CONFIG}/common/global.ini,${CONFIG}/common/sei-ham.ini,${CONFIG}/l1/l1sei-ham.ini -t SEI-HAM7

[eagoetz]

There may be a related issue with the states for the "quiet" and "noisy" times on the network SEI pages: the "Quiet" state and "Noisy" states are simply the entire day because the known/active properties are set to be the entire day. They do not derive from the actual quiet and noisy times for H1 and L1. In some sense the "Quiet" and "Noisy" states should be meta-states, where the data shown for "Quiet" should be the H1 quiet times and L1 quiet times, which do not fully intersect. The same is true for Noisy.