Here are some examples of using
stilts tmatch1 matcher=sky values="RA2000 DE2000" params=20 \ action=keep0 in=crowded.vot out=sparse.vot
stilts tmatch1 matcher=skyerr values="RA2000 DE2000 RADIUS*4" params=40 \ action=keep0 in=crowded.vot out=sparse.vot
skyerrmatcher which determines the proximity threshold on a row-by-row basis from values in the table - in this case 4 times the value of the RADIUS column (this value must be in arc seconds). The
params=40value does not affect the result, but it gives the algorithm an idea of the rough scale of object separation.
stilts tmatch1 matcher=3d values="XPIX YPIX ZPIX" params=10 action=identify \ in=state.fit ocmd='select GroupSize>3' out=groups3+.fit
action=identifyparameter means that the input table is written out with the same rows, but with additional columns indicating which rows are associated with each other. One of these columns, "GroupSize" gives the number of objects in each group. The postprocessing filter
ocmd='select GroupSize>3'selects only those rows which are part of groups of three objects or larger; singletons and pairs are discarded before writing the output file.
stilts tmatch1 matcher=sky values="ra dec" params=3 action=wide2 \ ocmd='keepcols "id_1 ra_1 dec_1 id_2 ra_2 dec_2"' in=galaxy.fits out=binaries.txt ofmt=ascii
ocmdpost-processing filter discards all of the columns except the identifiers and sky positions for each object. The output is to a text file.