Nested foreach with changing index size - r

Question

I'm trying to obtain the return of daily prices for each stock I have. The data is cross-sectionnal and very large. Thus I use doParallel and nested foreach.
Here is the code I've been using so far. (this is a reproduceable example)
and here is a reproduce-able example
stock_name <- as.data.frame(sample(x = 1:100, size = 250, replace=TRUE))
price <- as.data.frame(sample(x = 1:100, size = 1000,replace=TRUE))
## Calculating daily returns.
stock_list<-as.tbl(distinct(stock_name))
numStock<- as.integer(count(stock_list)) #150 #as.integer(count(stock_list))
nCPUcores = detectCores()
if (nCPUcores < 3) {
registerDoSEQ()
}else{
cl = makeCluster(nCPUcores - 1)
registerDoParallel(cl)
}
d_ret<-c()
foreach (stock=1:numStock, .packages = c("doParallel","foreach","data.table","plyr","dplyr")) %dopar%{
s<-as.integer(unlist(stock_list[stock,]))
stock_price <- as.matrix(price[which(stock_name[1,]==s),])
u<-nrow(stock_price)
d_ret<-foreach (p=2:u) %:%{
c(d_ret,(stock_price[p,]-stock_price[p-1,])/stock_price[p-1,])
}
}
stopCluster(cl)
##--
But the code doesn't work. After Florian Prive's remark, I checked the library and it seems that I should write nested foreach loops like this:
x <- foreach(b=bvec, .combine='cbind') %:%
foreach(a=avec, .combine='c') %dopar% {
sim(a, b)
}
So what I understand is I shouldn't be writing anything between %:% and the second foreach.
However, in my case, the second loop would change with the first foreach because there aren't the same number of prices for each stocks. Therefore I can't just write ' foreach(a=avec) '.
The second foreach would ideally depend on variable u
u<-nrow(stock_price)
Is this even possible with the foreach library?
Thank you for the help