UIGestureRecognizer 是定義基本手勢的抽象類,常見的子類包括:
UITapGestureRecognizer:任意次數的點按UIPinchGestureRecognizer:捏合,用於縮放UIPanGestureRecognizer:拖曳UISwipeGestureRecognizer:任意方向的快速掃動UIRotationGestureRecognizer:手指反向移動造成的旋轉UILongPressGestureRecognizer:長按本文主要用到拖拽手勢 UIPanGestureRecognizer,目標是讓被拖動的元件不會跑出父視覺範圍。
思路是每次回呼都計算「新的中心點」,再用 MAX / MIN 把它夾在合法範圍內,最後把 translation 歸零以便下次增量計算:
- (void)doMoveAction:(UIPanGestureRecognizer *)recognizer {
// 計算使用者試圖把 view 拖到哪裡
CGPoint translation = [recognizer translationInView:self.view];
CGPoint newCenter = CGPointMake(recognizer.view.center.x + translation.x,
recognizer.view.center.y + translation.y);
// 限制在螢幕範圍內
newCenter.y = MAX(recognizer.view.frame.size.height / 2, newCenter.y);
newCenter.y = MIN(self.view.frame.size.height - recognizer.view.frame.size.height / 2, newCenter.y);
newCenter.x = MAX(recognizer.view.frame.size.width / 2, newCenter.x);
newCenter.x = MIN(self.view.frame.size.width - recognizer.view.frame.size.width / 2, newCenter.x);
recognizer.view.center = newCenter;
[recognizer setTranslation:CGPointZero inView:self.view];
}這裡的關鍵是 setTranslation:CGPointZero——歸零之後,下次回呼拿到的才是相對於上一次的增量位移,否則位移會不斷累加。
手指接觸螢幕時,無論單點還是多點,事件週期就開始了,直到所有手指離開螢幕為止。期間所有 UITouch 物件都封裝在 UIEvent 中,由程式分發給處理者,並記錄這個週期內每個觸摸物件狀態的變化。
做法是在 touchesBegan 中記錄「按下的起點」並標記是否命中目標視圖,再在 touchesMoved 中把手指位置換算成元件中心點,同樣用邊界判斷夾住:
BOOL isMove;
CGPoint legend_point;
- (void)touchesBegan:(NSSet *)touches withEvent:(UIEvent *)event {
[super touchesBegan:touches withEvent:event];
isMove = NO;
UITouch *touch = [touches anyObject];
CGPoint point = [touch locationInView:self.view];
if (CGRectContainsPoint(V2.frame, point)) {
legend_point = [touch locationInView:V2];
isMove = YES;
}
}
- (void)touchesMoved:(NSSet *)touches withEvent:(UIEvent *)event {
[super touchesMoved:touches withEvent:event];
if (!isMove) return;
UITouch *touch = [touches anyObject];
CGPoint point = [touch locationInView:self.view];
// 轉換成相對中心
point.x += V2.frame.size.width / 2.0f - legend_point.x;
point.y += V2.frame.size.height / 2.0f - legend_point.y;
// 邊界限制
if (point.x < V2.frame.size.width / 2.0f) {
point.x = V2.frame.size.width / 2.0f;
}
if (point.y < V2.frame.size.height / 2.0f) {
point.y = V2.frame.size.height / 2.0f;
}
if (point.x > self.view.frame.size.width - V2.frame.size.width / 2.0f) {
point.x = self.view.frame.size.width - V2.frame.size.width / 2.0f;
}
if (point.y > self.view.frame.size.height - V2.frame.size.height / 2.0f) {
point.y = self.view.frame.size.height - V2.frame.size.height / 2.0f;
}
V2.center = point;
}
如果只是單純的拖拽,用 UIPanGestureRecognizer 更省事,它天然處理了多指、取消等邊界情況;如果需要更精細地控制「從哪裡按下才算命中」,或者要與其他觸摸邏輯深度耦合,直接用 UITouch 反而更靈活。兩種方式的邊界計算本質相同,差別只在事件來源。
Demo:https://github.com/sinceresk/MoveView
本文整理自 neilbee 的簡書。