4 ms·

Interesting, so it's the android WebFrame that's the issue. Do you know what version of android you were using? Android 4.4 uses the same rendering and javasc

• ds206 12y ago
  I'm currently running 4.4.2 on a Verizon LG G2. I'm using the "HN - Hacker News Reader" app by Manuel Maly. Hope that helps.
• jdmichal 12y ago
  For what it's worth, I see the exact same behavior on the desktop in Firefox 31.0. If I pull the menu down a little bit and release, it goes by itself to the bottom and stops. If I pull it all the way down to the bottom myself, it just keeps going off the screen. (As in, I don't release the mouse button until the menu is completely locked to the bottom. Then, when I do release the button, it starts moving down and off the screen.) EDIT: Some debugging shows that "from" and "to" are both (0, 416). This is causing direction in Acceleration.onStart to be (0, 0), which causes acceleration to also be zero. However, velocity is not zero; it's some apparently random number. Also, the boundary bottom is set to Infinity, because it uses (to.y > from.y) instead of (to.y >= from.y), so it never enters a "bounce" state as it's impossible to go outside the boundary. TL;DR: I think you need to add tests for when "to" and "from" are the same location.
  • coderzach 12y ago
    Thanks a ton for looking into this, I made an issue at https://github.com/luster-io/impulse/issues/5 https://github.com/luster-io/impulse/issues/5