Side note: if you're using IEEE-754 doubles (not a great choice for decimalized prices), use NaN to indicate no price, missing price, etc. instead of using an "

Side note: if you're using IEEE-754 doubles (not a great choice for decimalized prices), use NaN to indicate no price, missing price, etc. instead of using an "impossible" value like -1.0. It's always possible the system will be expanded to some domain where -1.0 is valid, requiring lots of on-the-fly rewriting of live systems literally handling multiple billions of dollars per day. Better yet, use proper algebraic data types.