wilcox_R: What the package does (short line)

rmaseq <-
function(x,est=onestep,alpha=.05,grp=NA,nboot=NA,...){
#
#   Using the percentile bootstrap method,
#   test hypothesis that all marginal distributions
#   among J dependent groups
#   have a common measure of location.
#   This is done by using a sequentially rejective method
#   of J-1 pairs of groups.
#   That is, compare group 1 to group 2, group 2 to group 3, etc.
#
#   By default, onestep M-estimator is used.
#
#   nboot is the bootstrap sample size. If not specified, a value will
#   be chosen depending on the number of groups
#
#   x can be an n by J matrix or it can have list mode
#   grp can be used to specify a subset of the groups for analysis
#
#   the argument ... can be used to specify options associated
#   with the argument est.
#
if(!is.list(x) && !is.matrix(x)){
stop("Data must be stored in a matrix or in list mode.")
}
if(is.list(x)){
# put the data in an n by J matrix
mat<-matrix(0,length(x[[1]]),length(x))
for (j in 1:length(x))mat[,j]<-x[[j]]
}
if(is.matrix(x))mat<-x
mat<-elimna(mat) # Remove rows with missing values.
J<-ncol(mat)
Jm<-J-1
con<-matrix(0,ncol=Jm,nrow=J)
for(j in 1:Jm){
jp<-j+1
for(k in j:jp){
con[j,j]<-1
con[jp,j]<-0-1
}}
rmmcp(x,est=est,alpha=alpha,con=con,nboot=nboot,...)
}