extract the subpattern of the perl-based regular expression

extract_subpattern.R

extract_subpattern <- function(text, pattern) {
	gregexpr.result <- gregexpr(pattern=pattern, text=text, perl=TRUE)
	retval <- list()
	for(i in 1:length(gregexpr.result)) {
		result <- gregexpr.result[[i]]
		retval[[i]] <- list()
		for(name in attr(result, "capture.names")) {
			start <- attr(result, "capture.start")[,name]
			len <- attr(result, "capture.length")[,name]
			retval[[i]][[name]] <- substr(text[i], start=start, stop = start + len - 1)
		}
	}
	retval
}