Akim Demaille wrote:
I don’t see what it buys to have two instead of one call only, have I missed something?
If every call to the function f will be of the form f(A, g(B), C) where A, B, and C are expressions, it's generally cleaner to push the g call into f's body, and change f's API so that calls to it look like f(A, B, C) instead. That's all I was doing there. It was more of a clarity than an efficiency thing (the form I used is more efficient than what you're proposing, but efficiency doesn't really matter here).
