We have access to two methods that we almost always want to implement, no matter what. One gives us access to the underlying chunk of data and allows us to perform the transformation on it. We implement this one with the _transform method. It takes three arguments: the chunk of data that we are working on, the encoding, and a callback to let the underlying system know that we are ready to process more information.

One special thing about the callback function that differs from the _write callback for a Writable stream is that we can pass data to it to emit data on the readable side of the Transform stream, or we can pass nothing to it to signal that we want to process more data. This allows us to only send out data events when we want to instead of almost always needing to pass them out.

The other method is the _flush method. This allows us to finish any processing of any data that we may still be holding. Or, it will allow us to output once all of the data that has been sent into the stream. This is what we will implement with our string hashing function.