Why can't I evaluate this bivariate definite integral?

I want to directly use the x function to find the exact value of the following bivariate definite integral:

reg = ImplicitRegion[x^2 + y^2 <= 1 && x >= 0, {x, y}];
(* the answer should be π/2*Log[2] *)
Integrate[(1 + x*y)/(1 + x^2 + y^2), Element[{x, y}, reg]]


This bivariate integral is not complicated, but the above formula returns as it is. I want to know where the problem is and how I should modify it.

NIntegrate[(1 + x*y)/(1 + x^2 + y^2), Element[{x, y}, reg]]
(*1.08879304515*)

• This is worth reporting to Support. – J. M.'s torpor Aug 16 '20 at 3:52
• Despite your words, this is not a simple integral in the cartesian coordinates: both Integrate[(1 + x*y)/(1 + x^2 + y^2), {x, 0, 1}, {y, -Sqrt[1 - x^2], Sqrt[1 - x^2]}] and Integrate[(1 + x*y)/(1 + x^2 + y^2), {y, -1, 1}, {x, 0, Sqrt[1 - y^2]}] fail. – user64494 Aug 16 '20 at 4:05

Integrate[(1 + x*y)/(1 + x^2 + y^2), {x, y} ∈ Disk[{0, 0}, 1, {-π/2, π/2}]]

• Ah, I forgot that you can use Disk[] to make sectors. – J. M.'s torpor Aug 16 '20 at 4:14

One could just use an alternative region specification:

reg = RegionIntersection[Disk[], HalfPlane[{0, 0}, {0, 1}, {1, 0}]];
Integrate[(1 + x y)/(1 + x^2 + y^2), {x, y} ∈ reg]
1/2 π Log[2]


or switch to polar coordinates:

Simplify[((1 + x y)/(1 + x^2 + y^2) /. Thread[{x, y} -> r AngleVector[θ]])
Det[D[r AngleVector[θ], {{r, θ}}]]]
(r + r^3 Cos[θ] Sin[θ])/(1 + r^2)

Integrate[(r + r^3 Cos[θ] Sin[θ])/(1 + r^2), {r, 0, 1}, {θ, -π/2, π/2}]
1/2 π Log[2]

• Thank you very much. I want to know why we can't solve this problem directly. If you can, you can tag this post with bugs to attract professional attention. – A little mouse on the pampas Aug 16 '20 at 3:58
• @Monte, it's more of a "missing feature" than a bug. As you are already aware, Mathematica echoes the input if it doesn't know what to do with it, which is the case here. – J. M.'s torpor Aug 16 '20 at 4:06