Static typing

fluent-codegen has been designed to work well static type checks to ensure that you are passing appropriate objects to different methods. In general, static type errors will indicate a misuse of the API, and it is assumed that you will be using a static type checker. If you pass the wrong kind of object, such as passing a string where an Expression is expected, often you will only get a runtime error later in the process – at the point where you are trying to convert to AST or Python source.

In terms of the Python code you are generating, fluent-codegen provides little-to-no protection for type errors, or name errors resulting from attributes that don’t exist. For example, you will get no warnings if you do:

mod.assign("x", codegen.Number(1).e + codegen.String("hello").e)  # x = 1 + "hello"

Or:

_, math_lib = mod.create_import('math')

math_lib.e.my_math_func(1, 2)   # No such function in `math`

So you will need other ways to ensure and test that your output is correct.

In addition to unit tests for your code, a useful technique is to run a static type checker on the output. ty is a fast type checker that can be useful for this purpose.