R Under development (unstable) (2019-08-23 r77061) -- "Unsuffered Consequences"
Copyright (C) 2019 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> options(na.action=na.exclude) # preserve missings
> options(contrasts=c('contr.treatment', 'contr.poly')) #ensure constrast type
> library(survival)
> 
> #
> # Test out the revised model.matrix code
> #
> test1 <- data.frame(time=  c(9, 3,1,1,6,6,8),
+                     status=c(1,NA,1,0,1,1,0),
+                     x=     c(0, 2,1,1,1,0,0),
+                     z= factor(c('a', 'a', 'b', 'b', 'c', 'c', 'a')),
+                     stringsAsFactors=FALSE)
> 
> fit1 <- coxph(Surv(time, status) ~ z, test1, iter=1)
> fit2 <- coxph(Surv(time, status) ~z, test1, x=T, iter=1)
> all.equal(model.matrix(fit1), fit2$x)
[1] TRUE
> 
> # This has no level 'b', make sure dummies recode properly
> test2 <- data.frame(time=  c(9, 3,1,1,6,6,8),
+                     status=c(1,NA,1,0,1,1,0),
+                     x=     c(0, 2,1,1,1,0,0),
+                     z= factor(c('a', 'a', 'a', 'a', 'c', 'c', 'a')),
+                     stringsAsFactors=FALSE)
> 
> ftest <- model.frame(fit1, data=test2)
> all.equal(levels(ftest$z), levels(test1$z))
[1] TRUE
> 
> # xtest will have one more row than the others, since it does not delete
> #  the observation with a missing value for status
> xtest <- model.matrix(fit1, data=test2)
> dummy <- fit2$x
> dummy[,1] <- 0
> all.equal(xtest[-2,], dummy, check.attributes=FALSE)
[1] TRUE
> 
> # The case of a strata by factor interaction
> #  Use iter=0 since there are too many covariates and it won't converge
> test1$x2 <- factor(rep(1:2, length=7))
> fit3 <- coxph(Surv(time, status) ~ strata(x2)*z, test1, iter=0)
> xx <- model.matrix(fit3)
> all.equal(attr(xx, "assign"), c(2,2,3,3))
[1] TRUE
> all.equal(colnames(xx), c("zb", "zc", "strata(x2)2:zb",
+                           "strata(x2)2:zc"))
[1] TRUE
> all.equal(attr(xx, "contrasts"), 
+           list("strata(x2)"= "contr.treatment", z="contr.treatment"))
[1] TRUE
> 
> fit3b <-   coxph(Surv(time, status) ~ strata(x2)*z, test1, iter=0, x=TRUE)
> all.equal(fit3b$x, xx)
[1] TRUE
> 
> 
> # A model with  a tt term
> fit4 <- coxph(Surv(time, status) ~ tt(x) + x, test1, iter=0,
+               tt = function(x, t, ...) x*t)
> ff <- model.frame(fit4)
> # There is 1 subject in the final risk set, 4 at risk at time 6, 6 at time 1
> # The .strata. variable numbers from last time point to first
> all.equal(ff$.strata., rep(1:3, c(1, 4,6)))
[1] TRUE
> all.equal(ff[["tt(x)"]], ff$x* c(9,6,1)[ff$.strata.])
[1] TRUE
> 
> xx <- model.matrix(fit4)
> all.equal(xx[,1], ff[[2]], check.attributes=FALSE)
[1] TRUE
> 
> 
> proc.time()
   user  system elapsed 
  0.746   0.041   0.781