How do I skip “#” sign without loop while extract elements?

Question

I want to get a new data.frame from this data set,but there are some description with\"#\" between some rows and some rows contain \"#\" sign, I can use \"for\" loop under the c

Answer 1

I am assuming you want to read the data from external table(it's unclear from your question), so accordingly i am answering your question,Use comment.char="#" in "read.table" option, it will ignore the lines starting with #.

See ?read.table.

So, your first line could be:

x <- read.table("comm.txt",comment.char="#"),

where "comm.txt" is file which contains data according to your given format.

You can then use following code to split columns based on delimeter "-"

library(reshape2)
LS <- lapply(seq_along(x), function(i){
    colsplit(x[, i], "-", paste0(colnames(x)[i], letters[1:3]))
    }
)

do.call('cbind', LS)

Hope this helps

Answer 2

First, remove the rows starting with # from your one-column data frame x:

vec <- grep("^[^#]", x[[1]], value = TRUE)

Then, create a new data frame based on the remaining data:

data.frame(V1 = gsub("(.*\\:[0-9]+) .*", "\\1", vec),
           V2 = gsub(".* ([0-9]+) [0-9]+ [0-9]+ *$", "\\1", vec))

#                    V1  V2
# 1 2013-08-27 16:00:00 200
# 2 2013-08-27 16:00:01 200
# 3 2013-08-27 16:00:02 200
# 4 2013-08-27 16:00:03 200