I have this data.frame and I want to identify which cells from sample1$domain have "www", replace that with "" and strsplit the corresponding sample1$suffix. The data looks like this:
domain suffix
1 wbx2 com
2 redhat com
3 something com
4 gstatic com
5 www googleapis.com
6 smartfilter com
I have managed to tackle this as shown below but it changes the position of the row(s) (I would like it to stay at position 5) and given that it will run for million of cases, I don't think this is the most efficient way to do it.:
library("stringr")
sample1$domain <- ifelse(sample1$domain == "www", "", sample1$domain)
sample1[sample1$domain == "", c("domain", "suffix")] <- sample1[sample1$domain == "", c("suffix", "domain")]
y <- sample1$domain[sample1$suffix == ""]
z <- as.data.frame(unlist(str_split_fixed(y, "[.]", 2)))
colnames(z) <- c("domain", "suffix")
sample1 <- rbind(sample1, z)
sample1 <- subset(sample1, sample1$suffix != "")
rownames(sample1) <- NULL
sample1
# domain suffix
#1 wbx2 com
#2 redhat com
#3 something com
#4 gstatic com
#5 smartfilter com
#6 googleapis com
DATA
sample1 <- structure(list(domain = c("wbx2", "redhat", "something",
"gstatic", "www", "smartfilter"), suffix = c("com", "com", "com",
"com", "googleapis.com", "com")), .Names = c("domain", "suffix"
), row.names = c(NA, 6L), class = "data.frame")
We can create an index for values with "www". Then use that index to replace the site name and lastly the site suffix:
ind <- sample1$domain == "www"
sample1$domain[ind] <- sub("^(.*)\\..*", "\\1", sample1$suffix[ind])
sample1$suffix[ind] <- sub(".*\\.(.*)", "\\1", sample1$suffix[ind])
sample1
# domain suffix
# 1 wbx2 com
# 2 redhat com
# 3 something com
# 4 gstatic com
# 5 googleapis com
# 6 smartfilter com