Workflows with RT

Bryce Mecum

2020-04-21

The rt package was built to make it easier for our own team to work with RT in an automated fashion. As a result, most functions in the package have return values suitable for passing into other functions in the rt package or functions from other packages. Notably, functions are pipeable when possible.

Below are some example workflows you might use this package for.

Creating tickets

You can always email your RT installations associated email address to create a ticket but you can also create one programmatically:

rt_ticket_create("General", "me@example.com", "Hey, a quick question...")

You could also create a set of tickets from a database dump of some kind, such as an old ticket tracking system.

First, let’s create some example data to work with:

old_tickets <- data.frame(
  queue = "General",
  subject = c("I need some help", "about those TPS reports...", "hello!"),
  requestor = c("someone@example.com", "boss@initech.com", "user@example.com"),
  stringsAsFactors = FALSE
)

With some example data, creating tickets for each of those in on go is only a few lines of code:

lapply(seq_len(nrow(old_tickets)), function (row) {
  do.call(rt_ticket_create, as.list(old_tickets[row,]))
})

We can then check our work:

library(dplyr)

rt_ticket_search('Queue="General"') %>%
  select(id, Subject)
# A tibble: 3 x 2
  id    Subject
  <chr> <chr>
1 1     I need some help
2 2     about those TPS reports...
3 3     hello!

Editing tickets

Each ticket has a set of properties associated with it and rt makes it quick to update those programmatically.

For example, we write R code to resolve a ticket. For example, say we’re ready to resolve ticket number 6:

rt_ticket_edit(6, status = "resolved")

What if we wanted to re-assign a set of tickets? We could go about this the slow way: By looking up ticket IDs in our web browser and resolving them one by one. Instead, we can make rt do the work for us:

result <- rt_ticket_search("Owner = 'manager'", format = "i")
sapply(result, function (id) {
  rt_ticket_edit(id, owner = "intern")
})

Comment and Replying

If you’re already using RT, you’re probably already familiar with commenting and replying to tickets. You’ve also possibly pasted text into your web browser from another source. rt can help automate some of this.

For example, what if we were calculating some statistics for a ticket and wanted to put that information in the ticket as a comment?

rt_ticket_history_comment(6, paste(capture.output({summary(iris)}), collapse = "\n"))

Above, we used the handy capture.output function along with paste but there are many other ways to get information from R into RT.

We can look at the comment we just made like so:

cat(rt_ticket_history(3)$body)
...
# 3/3 (id/47/total)

id: 47
Ticket: 3
TimeTaken: 0
Type: Comment
Field:
OldValue:
NewValue:
Data: No Subject
Description: Comments added by root

Content: Sepal.Length    Sepal.Width     Petal.Length    Petal.Width          Species
         Min.   :4.300   Min.   :2.000   Min.   :1.000   Min.   :0.100   setosa    :50
         1st Qu.:5.100   1st Qu.:2.800   1st Qu.:1.600   1st Qu.:0.300   versicolor:50
         Median :5.800   Median :3.000   Median :4.350   Median :1.300   virginica :50
         Mean   :5.843   Mean   :3.057   Mean   :3.758   Mean   :1.199
         3rd Qu.:6.400   3rd Qu.:3.300   3rd Qu.:5.100   3rd Qu.:1.800
         Max.   :7.900   Max.   :4.400   Max.   :6.900   Max.   :2.500

Creator: root
Created: 2020-03-03 01:07:45
...

Piping

Many functions in rt return values suitable for passing into other functions, such as with the %>% operator from the magrittr package that’s commonly used in the tidyverse. Here’s an example of how this can be used:

rt_ticket_create("General") %>%
  rt_ticket_edit(owner = "some_user") %>%
  rt_ticket_history_comment("Hey, this is just a comment...")