Any way to avoid this on iOS?

[DeluxeOwl]

Screen.Recording.2023-11-12.at.21.05.51.mov

I'm using flexbox to position the button at the end (flex-end, flex: 1). Is there any way to avoid this delayed repositioning of the button on iOS?

[rlemasquerier]

Hello @DeluxeOwl and thank you for reporting this behaviour.

I'd be happy to help and have a look, but first could you share the code of your screen ? So that I can understand better the positioning in your scenario.

(you could share either the piece of code of the screen you're showing in the example, or a reproduction of the behaviour in the example app)

[landabaso]

This is a promising package.
I also have minor issues dealing with Flex, though.

<KeyboardScrollView
      keyboardShouldPersistTaps="handled"
      contentContainerStyle={{
        flexGrow: 1,
        justifyContent: 'center'
      }}
    >
      {your stuff...}
    </KeyboardScrollView>

Try setting some content shorter than the screen height (just one line of text and the TextInput for example). The contentContainerStyleabove will vertically-center the content in the ScrollView. This is quite typical. However, when KeyboardScrollView adds the padding, then the whole ScrollView is scrolled up even if unnecessary.

Related to this, there are other contentContainerStyle configurations which do not work. For example:

    <KeyboardScrollView
      keyboardShouldPersistTaps="handled"
      contentContainerStyle={{
        marginTop: 300
      }}
    >

combined with short inner content will not scroll to the TextInput. This package will be lit after a few tweaks. Thanks for your work @rlemasquerier !

[landabaso]

Hi again @rlemasquerier , I've been testing your code. This change will help with the cases reported above:

      contentContainerStyle={[
        contentContainerStyle
        //{ paddingBottom: additionalPadding }
      ]}
      contentInset={{ bottom: additionalPadding }}

[rlemasquerier]

Hello @landabaso thank you for your investigation.

I have tried your change, did not notice any regression so I guess we can release this 🙂 - does it solve your issues using flex?

[landabaso]

Yeah, contentInset worked better in my tests.

[rlemasquerier]

@landabaso OK, just released this change then, in 0.2.1.

Thank you!