representing play in relational db

I run a project that deals with开发者_StackOverflow社区 plays, mostly Shakespeare. Right now, for storage, we just parse them out into a JSON array formatted like this:

[0: {"title": "Hamlet", "author": ["Shakespeare", "William"], "noActs": 5}, | info 
 1: [null, ["Scene 1 Setting", "Stage direction", ["character", ["char line 1", "2"]] | act 1
 ...]

and save them to a file.

We'd like to now move to a relational database for a variety of reasons (foremost currently search) but have no idea how to represent things.

I'm looking for an outline of the best way to do things?

---

the topic is 'normalization'

begin by identifying your main classifications, such as PLAY, AUTHOR, ACT, SCENE

then add attributes - specifically, a primary key like play_id, act_id, etc.

then add still more attributes like NAME or other identifying information.

then finally, add some relationships between these object bby creating more tables like PLAY_AUTHOR which includes PLAY_ID and AUTHOR_ID.

---

Not much of a theater goer myself, but this may give you some ideas should you choose a relational DB.