Fix some static analysis warnings #1958
Conversation
| @@ -65,7 +65,7 @@ ASInterfaceState ASInterfaceStateForDisplayNode(ASDisplayNode *displayNode, UIWi | |||
| // if not already, about to be set to invisible as it is not possible for an element to be visible | |||
| // while outside of a window. | |||
| ASInterfaceState interfaceState = displayNode.pendingInterfaceState; | |||
| return (window == nil ? (interfaceState &= (~ASInterfaceStateVisible)) : interfaceState); | |||
| return (window == nil ? (interfaceState & (~ASInterfaceStateVisible)) : interfaceState); | |||
There was a problem hiding this comment.
while Xcode may flag this, I think the comment above indicates its intentional.
There was a problem hiding this comment.
I don't believe it is intentional. The stuff about directly clearing the interface bit would be satisfied by interfaceState & (~ASInterfaceStateVisible). The &= is also leaving interfaceState different after the expression is run, but it doesn't matter because interfaceState is never used again. This code is only working because &= also returns the value of the newly mutated variable. You couldn't write this code in Swift, because I believe &= doesn't return anything by default.
There was a problem hiding this comment.
regardless, since I have no idea, but that seems more in-depth thinking is required than the other one. I think splitting this PR up is best.
For this change, I think I'd want unit tests that ensure the existing behavior persists. I've no idea how much work that would be w/o an expert weighing in on this one.
There was a problem hiding this comment.
Opened a new PR for this. If you still want a unit test after reading my description, I'm happy to add one :)
| @@ -2385,7 +2385,7 @@ static void ASTextDrawRun(ASTextLine *line, CTRunRef run, CGContextRef context, | |||
| if (mode) { // CJK glyph, need rotated | |||
| CGFloat ofs = (ascent - descent) * 0.5; | |||
| CGFloat w = glyphAdvances[g].width * 0.5; | |||
| CGFloat x = x = line.position.x + verticalOffset + glyphPositions[g].y + (ofs - w); | |||
There was a problem hiding this comment.
This seems good. small diffs are terrific. Would you mind splitting this up? Thanks.
ZevEisenberg
deleted the
chore/zeveisenberg/fix-some-static-analysis-warnings
branch
Didn't fix all of them, just a couple that were easy.