Question

please download the file in the
https://www.box.com/s/z3tldpdykhppmivfexjl

,save it as ‘/tmp/mydata’
and run the following code

datafile="/tmp/mydata"
totalsize=file.info(datafile)$size
lines=totalsize/32
con=file(datafile,"rb")
date<-c()
data<-c()
for (i in 1:lines){
    result1<-readBin(con,integer(),n=1,size=4,endian="little")
    date<-c(date,result1)
    result2<-round(readBin(con,double(),n=7,size=4,endian="little"),2)
    data<-c(data,result2)
}
date=data.frame(matrix(date,nrow=lines,ncol=1,byrow=TRUE))
data=data.frame(matrix(round(data,2),nrow=lines,ncol=7,byrow=TRUE))
data=cbind(date,data)
names(data)<-c("date","open","high","low","close","volume","amount","reserved")
data<-data[-c(1:2),]

when you finish ,please input data in R console,
you can see long digits,
https://www.box.com/s/lxkbs96vk02p6ekdflbe

why round(data,2) can not round 2 digits?
it is so strange ,when i write the data into a csv file,open the csv file,i found it has only 2 digits!

write.csv(file="/tmp/test",data)

Joris Meys say it is bad way to grow vector in for-loops,but i find maybe in my case it is good way to do that ,if not ,how to revise it?let me see the better codes.

Answer 1

On the issue about growing the loop, one option is to put everything into a list if you don’t know the length of each binary vector read in (as I don’t because I have not downloaded your data nor run your code!). Once in a list you can go to a compact form after if each element read in is of the same length, or in this case if all you want is a vector, just unlist() each list.

datafile <- "/tmp/mydata"
totalsize <- file.info(datafile)$size
lines <- totalsize / 32
con <- file(datafile, "rb")
## allocate storage
date <- data <- vector(mode = "list", length = lines)
## although actually growing lists is not inefficient
for (i in 1:lines){
    ## directly fill in the lists we created using index i to point
    ## to the ith component of each list
    date[[i]] <- readBin(con, integer(), n=1, size=4, endian="little")
    data[[i]] <- round(readBin(con,double(), n=7, size=4, endian="little"), 2)
}

At this point you have two lists and if you want them compact-ified into a vector do

date <- unlist(date)
data <- unlist(date)