Seems to me you are right. But it does not matter either way. The result is the same.
Bishop endgame analysis

From what I've figured out, black shouldn't leave the d1-h5 diagonal until absolutely forced to do so. Then black should do everything possible to take white's g pawn. In your example I think black is better off playing ...Be2 than ...Bf7. If white advances the g pawn black's bishop can just capture it with a drawn game. I've done a bit of practice with this position and I see that it's a forced win. It just doesn't seem remotely close to as easy as Silman's example makes it look. I thought I must be overlooking something but I can't find what it is.
I'm working through Silman's "Complete Endgame Course" and I've bumped into an example I'm having some trouble with. I understand the theory of the following position which is basically: Black can not exchange bishops because white will win the K+2P vs. K endgame. White can not allow the black bishop any pawn captures that leaves white with only an h-file pawn as that is a drawn endgame. The problem I have is with Silman's example. I can't make much sense out of it and thought I'd ask you folks to take a look at it. Here is the position and the moves given in the example.
I think 1... Bg4 makes things more difficult for white. The white king will be forced to return to the pawns. I think the immediate 1. Bf5 is a stronger move for white than 1. Ke5. Am I missing something?