The dependency graph told us that the data package had lots of users. The test coverage told us that it was also one of the biggest packages we had. We can, therefore, infer that if we wanted to make improvements, this might be the right place to start.

As you may remember from earlier chapters, the data package uses functions and a global singleton pool, both of which caused us inconvenience. So, let's see if we can use JIT injection to get rid of these pain points.