Changed the display position of EmptySearchResultBody to the center vertically.

[Yamaton0827]

Issue

• close #ISSUE_NUMBER

Overview (Required)

• Empty view of the search screen is not aligned center vertically.
  In figma, UFO image and description is centered in whole screen (including app top bar and search filters bar).

figma	emulated device

Links

• Related issues
  • Empty favorite screen's UI is slightly aligned towards the bottom #853

Screenshot (Optional if screenshot test is present or unrelated to UI)

Before	After	After: show layout bounds mode ON

Movie (Optional)

Before	After

[github-actions]

Detekt check failed. Please run ./gradlew detekt --auto-correct to fix the issues.

[github-actions]

Snapshot diff report

File name	Image
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter categor y chip click - when click category App A rchitecture en - it should selected cate gory App Architectur e en]_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter day chi p click - when click conference day 2 - it should selected d ay 2]_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter languag e chip click - it sh ould show drop down menu]_compare.png
SearchScreenTest[Sea rchScreen - when dev ice is tablet - it s hould no timetable i tems are displayed]_ compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter categor y chip click - when click category Other en - it should sele cted category Other en]_compare.png
SearchTextFieldAppBa rPreview_FilledSearc hWord_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter languag e chip click - when click language MIXED - it should selecte d language MIXED]_co mpare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter categor y chip click - it sh ould show drop down menu]_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter languag e chip click - when click language ENGLI SH - it should selec ted language ENGLISH ]_compare.png
EmptySearchResultBod yPreview_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - input search word t o TextField - it sho uld show search word and filtered items] _compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter categor y chip click - when click category Jetpa ck Compose en - it s hould selected categ ory Jetpack Compose en]_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter day chi p click - when click conference day 1 - it should selected d ay 1]_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter day chi p click - it should show drop down menu] _compare.png
SearchTextFieldAppBa rPreview_EmptySearch Word_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - it should no timeta ble items are displa yed]_compare.png
SearchScreenTest[Sea rchScreen - when dev ice is tablet - inpu t search word to Tex tField - it should s how search word and filtered items]_comp are.png
EmptySearchScreenPre view_compare.png
SearchScreenTest[Sea rchScreen - when ser ver is operational - when filter languag e chip click - when click language JAPAN ESE - it should sele cted language JAPANE SE]_compare.png

[takahirom]

Thank you for this valuable contribution. I believe we need this implementation to comply with the design. The problem is that we need to modify the globalPositioned each time we change the structure of a composable or add a new composable. This could make our development process difficult. I'm considering whether it might be better not to proceed with this pull request. 🙇

[naoele]

@Yamaton0827 -san @takahirom -san
Sorry to butt in, but this can be fixed with a much smaller change.
Rather than changing globalPositioned,
It would be better to get the size of the top element and padding it accordingly.

    var favoriteFiltersSize by remember { mutableStateOf(0.dp) }

    Column(modifier = modifier.fillMaxSize()) {
        FavoriteFilters(
            allFilterSelected = uiState.isAllFilterSelected,
            day1FilterSelected = uiState.isDay1FilterSelected,
            day2FilterSelected = uiState.isDay2FilterSelected,
            backgroundColor = filterBackgroundColor,
            onAllFilterChipClick = onAllFilterChipClick,
            onDay1FilterChipClick = onDay1FilterChipClick,
            onDay2FilterChipClick = onDay2FilterChipClick,
            modifier = Modifier
                .fillMaxWidth()
                .background(favoriteFiltersBackgroundColor)
                .onSizeChanged { size ->
                    favoriteFiltersSize = size.height.dp
                },
        )

        when (uiState) {
            is FavoritesSheetUiState.Empty -> {
                EmptyView(
                    modifier = Modifier.wrapContentSize(Alignment.Center)
                        .padding(bottom = WindowInsets.statusBars.asPaddingValues().calculateTopPadding() + favoriteFiltersSize)
                )
            }

Screenshot

It's a little bit off from Figma, but what do you think of this change?

Before	After	Figma

I don't mind if you use this code if you like.

By the way, it would be helpful if you could run detekt and organize the code.

[takahirom]

Thank you. That seems better, but I think it's not very good to add processing every time a component is added. However, if we handle it as an inset, I think it's relatively acceptable.

[naoele]

Sorry, I didn't understand what you meant by treating it as an inset.
I suggested that this would be fine if no elements will be added to FavoritesScreen in the future.

[takahirom]

I think this is a somewhat difficult problem. We also have this issue, and this PR could complicate things further.
#520

[naoele]

@takahirom -san, @Yamaton0827 -san
I understand. If that's the situation, there's no need to consider my suggestion.
Thank you for detailed explanation!

[Yamaton0827]

@takahirom
Thank you for your review and consideration. The content of the discussion was a little difficult for me, so I will try to understand it from now on.

@naoele
Thank you for the suggestion, it was helpful.

[Yamaton0827]

This pull request is closed because of a fundamental problem. Thank you for your review.