How to apply Bootstrap estimation with complex sampling technique in R?

I need to compute bootstrap estimation of the standard error of the mean of the main variable in my sample. I would like to use the package "boot". The problem is that the sample was taken with a 2 stage stratified sampling method (the stratification is in the first stage). For this reason in the bootstrap resampling I must use this particular sampling technique for having correct estimations. So an example of my problem:

data <- data.frame(NQUEST, index, strata, weights)

NQUEST     index     strata(1 stage)     weights
 1         1432         1                 423
 2          678         2                 749
 3         9284         1                 398
 4          445         1                  96
 5        10045         1                 324
 6         7784         3                 334 
[...]     [...]       [...]              [...]     
5000       3290         1                 932

-NQUEST are the units

-immagine we have 3 strata

I want to compute the standard error of the mean of "index" with bootstrap method, 1000 replications and with 2 stage resampling technique.