search

rahuldave / appsem

js server and ui for semflow
4 stars 1 forks source link

Time/exposure range facets #13

Open DougBurke opened 13 years ago

DougBurke commented 13 years ago

These facets do not seem to follow the available data - e.g. we have Chandra data that is > 10 ks and papers after 2010 that (probably, have not checked) are included in the search results. The facets only seem to be applied when one (or both) of the ends is moved.

DougBurke commented 13 years ago

I have relaxed the publication year range to 1978-2011 and the exposure length to 0-500 ks (I have not checked to see if this misses anything). Really these should be data driven.

DougBurke commented 13 years ago

An initial attempt at improving the user experience of the publication-year facet is provided in

https://github.com/DougBurke/appsem/commit/32ed40bdf8d9fa6989700d930e03a0ee4a628d1a

DougBurke commented 13 years ago

The commits leading up to 8e4c190 introduce a common scheme for the pubyear, ra, dec and exptime facets. If a filter has been applied using the facet then we use those limits for the display even if the actual range from the search is different.

Note that for ra/dec/exptime facets, the data range can exceed the selection on that facet if there are multiple values of the field for the selected documents: if you select dec=0 to +90 but a paper also includes an observation with dec=-20 then do we want the display to reflect this or not? Initial testing by Doug suggests not but this needs thinking about.