Question

I have a data frame like this:

x

TicketID   Application     Impacted_Systems
1          Web             Online; PeopleSoft, Financials
etc

what I like to do is create another data.frame based on x$Impacted_Systems put each items separated by “;” to its own colum, then compine it with data frame x to graph it.

I have this so far:

data.frame(do.call('rbind', strsplit(as.character(x$Impacted_Systems),';'))) 

this creates the columns for each string separated by “;” twice:

X1      X2       X3       X4       X5   X6      X7
1   Online  PeopleSoft Financials   Online  PeopleSoft Financials           Online

any ideas what I am doing wrong? In this particular case there should be only 3 columns, not 7.

I tried this

p<-colsplit(x$Impacted_Systems, ";") 

this is expecting names option. The problem is number of impacted systems will vary, it is not fixed.

Answer 1

# some example data
df <- data.frame(Impacted_Systems = c("foo;bar", "foo;bar1;bar2", "foo;bar1;bar2;bar4;bar4"))

library(plyr)
library(reshape2)
split_names <- llply(as.list(df$Impacted_Systems), function(x){
                split_data <- data.frame(rbind(strsplit(x, ";")[[1]]))
                names(split_data) <- paste0("Impacted_Systems", 1:length(strsplit(x, ";")[[1]]))
                split_data
    })


> cbind(df, ldply(split_names))
         Impacted_Systems Impacted_Systems1 Impacted_Systems2
1                 foo;bar               foo               bar
2           foo;bar1;bar2               foo              bar1
3 foo;bar1;bar2;bar4;bar4               foo              bar1
  Impacted_Systems3 Impacted_Systems4 Impacted_Systems5
1              <NA>              <NA>              <NA>
2              bar2              <NA>              <NA>
3              bar2              bar4              bar4