This routine fits the main vertex in several steps :
FRCREJ : bitted word. If bit i is set, then all tracks with true history code i are rejected from the vertex fit. The special value -1 means that all non-vertex tracks are rejected. The value 0 means that the routine ZAURST will try, using the meassured momenta, to find tracks from K_O and Lambda decays as well as from photon conversions and nuclear interactions in the detector material.
ERROR : error-flag 0 : OK -1 : Not possible to get vertex with requested probability. Beam-spot returned in XYZ, and beam-spot covariance matrix in VCOV. <-1: OK, but measured covariance matrix not calculated, VCOVM = VCOV, SFACT = 1.0 Hence, testing ERROR .LE. 0 would be OK, but if ERROR = -1 errors are larger than otherwise. They are statistically correct, however.
From /ZAEVT/ : SFACT, error scale factor; NSGINI, # of sigmas for sfact definition; PCUT, prob-cut to accept vertex; MTRK, total # of seen tracks. From /ZASTE/ : DOWNUP, vertex-fit method : include tracks until Xi^2 becomes bad (.TRUE.), or exclude tracks until it becomes good (.FALSE.). From /ZATRS/ : PAR and WGT, perigee parameters with weight-matrix; INVTX, the track is in the main vertex; INSV, the track is in a sec. vertex (initialised here);
To /ZAEVT/ : XYZ, fitted vertex pos; VCOV, calculated vertex cov mat; VCOVM, measured vertex cov mat; CHI2VX, chi^2 of vertex; NVTRK, # of track in vertex;
(To do Vertex Fit 2.)
(To Reject STrange particle decay prods.)