# #---------------------------------- # ### Testing eEvents function # For comparing floating-point numbers, an exact match cannot be expected. # For such test cases, the tolerance is set to 1e-6 (= 0.000001), # a sufficiently low value # #---------------------------------- test_that( desc = "test: checking number of events on a single arm source : helper.R", code = { nEv <- eEvents( lambda = 0.2, eta = 0.1, gamma = 1, R = 12, S = NULL, T = 24, Tfinal = NULL, minfup = 0 ) ## Compute # of events from independently coded validator. ## the sample size is taken as the one that is output from ## eEvents(). ## sample sizes are validated separately using nSurvival() vldEv <- expect_ev_arm( lambda = .2, drprate = 0.1, maxstudy = 24, accdur = 12, totalSS = nEv$n, rndprop = 1 ) expect_lte( object = abs(nEv$d - vldEv), expected = 1e-6 ) } )