为什么rep（）与这个简单的R例子行为不一致？(why does rep() behave inconsistently with this simple R example?)

 我正在构建代码来运行和管理可以在三个站点队列之一的站点上的抽样事件的模拟。 我使用rep() ）使用以下代码分配同类群组标识符（1,2或3）： 
 
cohort <- rep(1:n.cohorts, n.sites) 
 
 我已经把关键行放在首位，虽然为了重现我的问题，你需要运行以下行，这些行分配队列之间的站点总数以便呈现给rep()调用。 
 
n.cohorts <- 3
s <- 10 # total available sites in this example

# different proportions of the total can be allocated to each cohort, for example 
prop.control <- 0.4 ; prop.int <- 0.4 ; prop.ref <- 1-(prop.int+prop.control)
n.control <- prop.control * s; n.int <- prop.int * s; n.ref <- prop.ref * s 
n.sites <- c(n.control, n.int, n.ref)  
 
 现在， n.sites本身就会回归 
 
 
 
 [1] 4 4 2 
 
 
 因此当我运行我的cohort <- rep(1:n.cohorts, n.sites)再次呼叫时，我希望cohort <- rep(1:n.cohorts, n.sites) cohort成为10个项目的列表，如下所示： [1] 1 1 1 1 2 2 2 2 3 3 。 然而，我得到的只有9： 
 
> cohort
[1] 1 1 1 1 2 2 2 2 3    
 
 如果我运行相同的代码，其中n.sites直接定义如下： n.sites <- c(4, 4, 2) ，我得到了我期望的10个项目。 我已经多次重做这件事来说服自己，在两种情况下， n.sites本身都会产生相同的结果。 
 
 谁能解释为什么会这样？ 提前谢谢了。 
 
 大卫 

I am building code to run and manage simulations of sampling events at sites that can be in one of three site cohorts. I use rep() to assign the cohort identifier (1,2, or 3) using the code below:
 
cohort <- rep(1:n.cohorts, n.sites) 
 
I have put the key line first, although to reproduce my problem you need to run the following lines, which allocate a total number of sites between cohorts for presentation to the rep() call.
 
n.cohorts <- 3
s <- 10 # total available sites in this example

# different proportions of the total can be allocated to each cohort, for example 
prop.control <- 0.4 ; prop.int <- 0.4 ; prop.ref <- 1-(prop.int+prop.control)
n.control <- prop.control * s; n.int <- prop.int * s; n.ref <- prop.ref * s 
n.sites <- c(n.control, n.int, n.ref)  
 
now, n.sites by itself returns
 
 
 
[1] 4 4 2
 
 
so when I run my cohort <- rep(1:n.cohorts, n.sites) call again I expect cohort to be a list of 10 items, like this: [1] 1 1 1 1 2 2 2 2 3 3. What i get, however, is only 9:
 
> cohort
[1] 1 1 1 1 2 2 2 2 3    
 
If i run the same code where n.sites is defined directly like so: n.sites <- c(4, 4, 2), I get the 10 items i expect. I have redone this several times to convince myself that under both scenarios n.sites by itself produces identical results. 
 
Can anyone explain why this happens? Many thanks in advance.
 
David

原文：https://stackoverflow.com/questions/21398586