# 1
x <- readLines('http://www.stat.cmu.edu/~cshalizi/statcomp/labs/09/dates.txt')

#a
pat <- "<[[:alpha:]]+[[:space:]]+[[:digit:]]+>"

#b
gregexpr(pat, x)

#c
i <- gregexpr(pat, x)[[1]]
substring(x, i + 1, i + attr(i , "match.length") - 2)

# 2
sj <- readLines('http://www.stat.cmu.edu/~cshalizi/statcomp/labs/09/stevejobs.txt')

#a
#
# sj is a character vector. Its length is given by length(sj).
# Each element of sj is a string corresponding to a line in the file stevejobs.txt. 

#b
sj.long <- paste(sj, collapse = ' ')
sj.sentences <- strsplit(sj.long, '[[:space:]]*[.!?]{1}[[:space:]]*')[[1]]

#c
i <- grep('life|death', sj.sentences, ignore.case = T)
cat(sj.sentences[i], sep = '\n')

#d
sj.words <- strsplit(sj.long, '[[:space:]]*[.!?[:space:]]+[[:space:]]*')
# or
sj.words <- strsplit(sj.sentences, '[[:space:]]+')
sj.words <- do.call(c, sj.words)

sj.words <- tolower(sj.words)

#e
sj.words <- gsub('^[[:punct:]]|[[:punct:]]$', '', sj.words)

#f
wc <- table(sj.words)