So, since pleasure is clearly a function of relieved pain, i
I am going to have to come back later but this is clearly not a full and correct statement of the issue. Some pleasure is indeed such, but much pleasure is not. There is a long discussion of this in Gosling and Taylor and if I recall correctly they show that not even Plato eventually took that position (that all pleasure arises from loss of pain). There is a classic example of the smell of a rose when walking through a garden -- that is clearly pleasurable, but the pleasure does not arise from any kind of pain existing prior to smelling the rose.
As to the gas tank analogy, that arises from there being only two feelings -- the gas in the tank does not derive its essence from the "air" in the tank any more than the air derives its essence from the gas.