I would like to import a .ics file into R, however, when I try to do so like...
sneak_cal <- read.delim("iCal-TribeEvents.ics", sep = ":", header=FALSE, stringsAsFactors = FALSE, strip.white = TRUE, na.strings = "")
...I end up splitting the character strings of website (belonging to the X-ORIGINAL-URL or the UID field) too, which is undesirable
ie https and //www.kicksonfire.com
The ultimate goal is to get the data into a tidy format where each row represents a single VEVENT, which I think would be represented by a unique UID, without any loss of information (such as the URL)
Is there another approach that is recommended, such as pre-defining the fields that are expected as the key and matching the value or empty space to that key? Since the .ics file has the same expected fields each time, it seems like it might make sense to use those fields as a template to read in the data, but I can not figure out how to do it.
Here's an example
x <- readLines("https://www.kicksonfire.com/releases/?ical=1&tribe_display=list", warn = FALSE)
stopifnot(!any(grepl("^\\s+", x))) # disregarding value fields that have linefeeds for the sake of simplicity
keyval <- do.call(rbind, regmatches(x, regexpr(":", x, fixed = TRUE), invert = TRUE))
keyval <- keyval[which.max(keyval[,1]=="BEGIN" & keyval[,2]=="VEVENT"):tail(which(keyval[,1]=="END" & keyval[,2]=="VEVENT"), 1),]
keyval <- cbind.data.frame(keyval, id=cumsum(keyval[,1]=="BEGIN" & keyval[,2]=="VEVENT"))
df <- reshape(keyval, timevar="1", idvar="id", direction = "wide")
head(df[,c(3,4,9)])
# 2.DTSTART;VALUE=DATE 2.DTEND;VALUE=DATE 2.SUMMARY
# 1 20170422 20170423 Air Jordan 11 Low GS Blue Moon
# 14 20170422 20170423 Air Jordan 5 Premium Pure Platinum
# 27 20170427 20170428 Nike Air VaporMax Asphalt
# 40 20170427 20170428 Nike Air VaporMax Oreo
# 53 20170427 20170428 Nike WMNS Air VaporMax White Ice Blue
# 66 20170427 20170428 wings+horns x adidas NMD R2 Light Grey