lines=scan(file = "emaillists", what = character(0), sep = "\n")
# alternative way to open a file for reading
# con <- file("emaillists", "r", blocking = FALSE)
# lines=readLines(con)
# close(con)
emails=list() # list to hold email lists and subscribers
allemails=character(0) # vector of unique email addresses
currlist=''
for(i in lines) {
#     look for text string 'Members' - this line holds list name
#     grep returns either a 1 or an empty integer vector
#     using length I can change the value to 1 or 0
    if(length(grep("Members",i))) {
#         strsplit returns a list - the second element of the first list holds the list name
#         save the current list name to use for the addresses that will follow
        currlist=strsplit(i,"'")[[1]][2]
#         create an empty character vector to hold each email address
        emails[[currlist]]=character(0)
    }
#     any line with a '@' contains an email address
    if(length(grep("@",i))) {
        emails[[currlist]]=append(emails[[currlist]],i)
        allemails=append(allemails,i)
    }
}
allemails=unique(allemails[order(allemails)]) # remove duplicate emails
# create a data frame where rows are subscribers and columns are email lists
emailtable=as.data.frame(matrix(data=0,nrow=length(allemails),ncol=length(emails)))
names(emailtable)=names(emails) # set up column names - email lists
row.names(emailtable)=allemails # set up row names - subscribers
# loop through each email list
for(i in names(emails)) {
#     set value to '1' where email address is part of a email list
    emailtable[emails[[i]],i]=1
}
# save my new output file as a CSV
write.csv(emailtable,'emaillists.csv')