How to use LazyColumn's animateItemPlacement() without autoscrolling on changes? - android-jetpack-compose

I am using a LazyColumn in a checklist like style. The list shows all to-be-done items first and all done items last. Tapping on an item toggles whether it is done.
Here is an MWE of what I am doing:
data class TodoItem(val id: Int, val label: String, var isDone: Boolean)
#Composable
fun TodoCard(item: TodoItem, modifier: Modifier, onClick: () -> Unit) {
val imagePainterDone = rememberVectorPainter(Icons.Outlined.Done)
val imagePainterNotDone = rememberVectorPainter(Icons.Outlined.Add)
Card(
modifier
.padding(8.dp)
.fillMaxWidth()
.clickable {
onClick()
}) {
Row {
Image(
if (item.isDone) imagePainterDone else imagePainterNotDone,
null,
modifier = Modifier.size(80.dp)
)
Text(text = item.label)
}
}
}
#OptIn(ExperimentalFoundationApi::class)
#Composable
fun ExampleColumn() {
val todoItems = remember {
val list = mutableStateListOf<TodoItem>()
for (i in 0..20) {
list.add(TodoItem(i, "Todo $i", isDone = false))
}
list
}
val sortedTodoItems by remember {
derivedStateOf { todoItems.sortedWith(compareBy({ it.isDone }, { it.id })) }
}
LazyColumn {
items(sortedTodoItems, key = {it.label}) { item ->
TodoCard(item = item, modifier = Modifier.animateItemPlacement()) {
val index = todoItems.indexOfFirst { it.label == item.label }
if (index < 0) return#TodoCard
todoItems[index] = todoItems[index].copy(isDone = !todoItems[index].isDone)
}
}
}
}
This work well except for one side effect introduced with Modifier.animateItemPlacement(): When toggling the first currently visible list element, the LazyListState will scroll to follow the element.
Which is not what I want (I would prefer it to stay at the same index instead).
I found this workaround, suggesting to scroll back to the first element if it changes, but this only solves the problem if the first item of the column is the first one to be currently displayed. If one scrolls down such that the third element is the topmost visible and taps that element, the column will still scroll to follow it.
Is there any way to decouple automatic scrolling from item placement animations? Both seem to rely on the LazyColumn's key parameter?

I haven't tried to compile/run your code, but it looks like your'e having a similar issue like this, and it looks like because of this in LazyListState
/**
* When the user provided custom keys for the items we can try to detect when there were
* items added or removed before our current first visible item and keep this item
* as the first visible one even given that its index has been changed.
*/
internal fun updateScrollPositionIfTheFirstItemWasMoved(itemsProvider: LazyListItemsProvider) {
scrollPosition.updateScrollPositionIfTheFirstItemWasMoved(itemsProvider)
}
I suggested a possible work around where the first item's animation will be sacrificed by setting its key as its index instead of a unique identifier of the backing data.
itemsIndexed(
items = checkItems.sortedBy { it.checked.value },
key = { index, item -> if (index == 0) index else item.id }
) { index, entry ->
...
}
Though I haven't regressed any use-case with this kind of set-up of Lazy keys yet nor could solve your issue, but you can consider trying it.

Related

Generic Linked Lists find duplicates

Good day all,
I am studying Bsc-IT but am having problems.
With the current covid-19 situation we have been left to basically self-study and I need someone to put me in the right direction (not give me the answer) with my code.
I must write a program called appearsTwice that receives a linked list as parameter and return another list containing all the items from the parameter list that appears twice or more in the calling list.
My code so far(am I thinking in the right direction? What must I look at?)
public MyLinkedList appearsTwice(MyLinkedList paramList)
{
MyLinkedList<E> returnList = new MyLinkedList<E>(); // create an empty list
Node<E> ptrThis = this.head;//Used to traverse calling list
Node<E> ptrParam= paramList.head; //neither lists are empty
if (paramList.head == null) // parameter list is empty
return returnList;
if (ptrThis == null) //Calling list is empty
return returnList;
for (ptrThis = head; ptrThis != null; ptrThis = ptrThis.next)
{
if (ptrThis == ptrThis.element)
{
returnList.append(ptrThis.element);
}
}
Some issues:
Your code never iterates through the parameter list. The only node that is visited is its head node. You'll need to iterate over the parameter list for every value found in the calling list (assuming you are not allowed to use other data structures like hashsets).
if (ptrThis == ptrThis.element) makes little sense: it tries to compare a node with a node value. In practice this will never be true, nor is it useful. You want to compare ptrParam.element with ptrThis.element, provided that you have an iteration where ptrParam moves along the parameter list.
There is no return statement after the for loop...
You need a counter to address the requirement that a match must occur at least twice.
Here is some code you could use:
class MyLinkedList {
public MyLinkedList appearsTwice(MyLinkedList paramList) {
MyLinkedList<E> returnList = new MyLinkedList<E>();
if (paramList.head == null) return returnList; // shortcut
for (Node<E> ptrParam = paramList.head; ptrParam != null; ptrParam = ptrParam.next) {
// For each node in the param list, count the number of occurrences,
// starting from 0
int count = 0;
for (Node<E> ptrThis = head; ptrThis != null; ptrThis = ptrThis.next) {
// compare elements from both nodes
if (ptrThis.element == ptrParam.element) {
count++;
if (count >= 2) {
returnList.append(ptrParam.element);
// no need to look further in the calling list
// for this param element
break;
}
}
}
}
return returnList; // always return the return list
}
}

Add an "abstract" method to (an F#) record

I have created the following record in an attempt to translate a C# class to F#:
type Days = Days of int
type Value = Value of int
type Item = {
Name: string
Expires: Days
Value: Value
}
Thing is I also need every Item to have a... "way", to run another function, yet not defined, handleDevalue, which acts on the item itself to manipulate the item's Value value.
The handleDevalue function is dependent on the Expires property of the item and thus each item's implementation of it would be different, with the only common thread being the function's name and signature (Item -> Item).
On the C# code I'm translating this method was defined as abstract on the Item class and overriden on every item instantiated (where every item is a subclass inheriting from Item).
What I've tried, unsuccessfully till now:
Add an abstract method on the record: ...} with abstract handleDevalue: Item -> Item.
1.1 Reason for failure: IDE tells me "abstract can't be added here as an augmentation" (or something close to the same effect). (I'm not F#-savvy enough to even know what it means, but the compiler won't let it compile so... no).
Add handleDevalue as a function on the record: {... HandleDevalue: Item -> Item...}.
2.1. Reason for failure: this function is dependent on the Expires property. Apparently a record's fields are mutually independent of each other, and besides... how will the function "know" which item to act on (it's supposed to act on the item itself)? The this keyword is not allowed when implementing a function when "instantiating" a record (i.e. no {...handleDevalue = fun this -> <some implementation code here>).
I could remember to define the function on every item I create (I should anyway), but that's not using the type system to my advantage.
I want the compiler to force me to implement the function and remind me if I don't.
With these ways failing I'm out of ideas how to move forward.
Thanks for any advice in advance.
I'm not really sure what you are trying to accomplish here, but I'll give it a shot.
Why not do something similar to this, using a discriminated union instead of inheritance?
type Days = Days of int
type Value = Value of int
type Item = {
name: string
expires: Days
value: Value
}
type ItemType =
| FooItem of Item
| BarItem of Item
| BazItem of Item
// ...
let deValue item =
match item with
| FooItem i ->
{
name = i.name
expires = i.expires -1
value = i.value -1
} |> FooItem
| BarItem i ->
{
name = i.name
expires = i.expires -1
value = i.value -10
} |> BarItem
| // etc
Think about what actually happens in the C# program.
You have several different implementations of handleDevalue, and every item instance has one of those implementations associated with it. But what determines which one goes with which item? Well, this is determined by the specific descendant class that the item is. Ok, but what determines which descendant class gets instantiated?
Somewhere, at some point, there must be a place that picks a descendant class somehow. Let's assume it looks something like this:
class Item { public abstract int handleDevalue() { ... } }
class FooItem : Item { public override int handleDevalue() { ... } }
class BarItem : Item { public override int handleDevalue() { ... } }
public Item createItem(string name, Days expires, Value value) {
if (foo) return new FooItem(name, expires, value)
else return BarItem(name, expires, value)
}
So, look what's happening: ultimately, whoever creates the items is choosing which handleDevalue implementation ends up being used, and then that implementation gets attached to the item instance via the method table.
Now that we know this, we can do the same thing in F#. We'd just need to make attaching of the implementation explicit:
type Item = {
Name : string
Expires : Days
Value : Value
handleDevalue : Item -> Item
}
let handleDevalueFoo item = ...
let handleDevalueBar item = ...
let createItem name expires value = {
Name = name
Expires = expires
Value = value
handleDevalue = if foo then handleDevalueFoo else handleDevalueBar
}

How do you add an initial selection for Angular Material Table SelectionModel?

The Angular Material documentation gives a nice example for how to add selection to a table (Table Selection docs). They even provide a Stackblitz to try it out.
I found in the code for the SelectionModel constructor that the first argument is whether there can be multiple selections made (true) or not (false). The second argument is an array of initially selected values.
In the demo, they don't have any initially selected values, so the second argument in their constructor (line 36) is an empty array ([]).
I want to change it so that there is an initially selected value, so I changed line 36 to:
selection = new SelectionModel<PeriodicElement>(true, [{position: 2, name: 'Helium', weight: 4.0026, symbol: 'He'}]);
This changes the checkbox in the header to an indeterminate state (as expected), but does not cause the row in the table to be selected. Am I setting the initial value incorrectly, or what am I missing here? How can I set an initially selected value?
Tricky one. You need to initialize the selection by extracting that particular PeriodicElement object from your dataSource input, and passing it to the constructor.
In this particular case, you could code
selection = new SelectionModel<PeriodicElement>(true, [this.dataSource.data[1]);
It's because of the way SelectionModel checks for active selections.
In your table markup you have
<mat-checkbox ... [checked]="selection.isSelected(row)"></mat-checkbox>
You expect this binding to mark the corresponding row as checked. But the method isSelected(row) won't recognize the object passed in here as being selected, because this is not the object your selection received in its constructor.
"row" points to an object from the actual MatTableDataSource input:
dataSource = new MatTableDataSource<PeriodicElement>(ELEMENT_DATA);
But the selection initialization:
selection = new SelectionModel<PeriodicElement>(true, [{position: 2, name: 'Helium', weight: 4.0026, symbol: 'He'}]);
happens with a new object you create on the fly. Your selection remembers THIS object as a selected one.
When angular evaluates the bindings in the markup, SelectionModel internally checks for object identity. It's going to look for the object that "row" points to in the internal set of selected objects.
Compare to lines 99-101 and 16 from the SelectionModel source code:
isSelected(value: T): boolean {
return this._selection.has(value);
}
and
private _selection = new Set<T>();
I was facing the same issue, I used dataSource to set the initial value manually in ngOnInit()
ngOnInit() {
this.dataSource.data.forEach(row => {
if (row.symbol == "H") this.selection.select(row);
});
}
If you do the following, it works too
selection = new SelectionModel<PeriodicElement>(true, [ELEMENT_DATA[1]])
To select all you can do
selection = new SelectionModel<PeriodicElement>(true, [...ELEMENT_DATA])
I hope the answer is helpful
Or more dynamically if you have a set of values and you want to filter them before:
selection = new SelectionModel<PeriodicElement>(true, [
...this.dataSource.data.filter(row => row.weight >= 4.0026)
]);
This gets more tricky if you have data loading asynchronously from an api. Here is how I did it:
Firstly I have implemented the DataSource from "#angular/cdk/table". I also have an RxJS Subject that fires whenever data is loaded (first time or when user changes page in the pagination section)
export abstract class BaseTableDataSource<T> implements DataSource<T>{
private dataSubject = new BehaviorSubject<T[]>([]);
private loadingSubject = new BehaviorSubject<boolean>(false);
private totalRecordsSubject = new BehaviorSubject<number>(null);
public loading$ = this.loadingSubject.asObservable();
public dataLoaded$ = this.dataSubject.asObservable();
public totalRecords$ = this.totalRecordsSubject.asObservable().pipe(filter(v => v != null));
constructor(){}
connect(collectionViewer: CollectionViewer): Observable<T[]>{
return this.dataSubject.asObservable();
}
disconnect(collectionViewer: CollectionViewer): void {
this.dataSubject.complete();
this.loadingSubject.complete();
this.totalRecordsSubject.complete();
}
abstract fetchData(pageIndex, pageSize, ...params:any[]) : Observable<TableData<T>>;
abstract columnMetadata(): {[colName: string]: ColMetadataDescriptor };
loadData(pageIndex, pageSize, params?:any[]): void{
this.loadingSubject.next(true);
this.fetchData(pageIndex, pageSize, params).pipe(
finalize(() => this.loadingSubject.next(false))
)
.subscribe(data => {
this.totalRecordsSubject.next(data.totalNumberOfRecords);
this.dataSubject.next(data.records)
});
}
}
Now when I want to pre-select a row, I can write a function like this in my component which hosts a table that uses an implementation of the above mentioned data source
selectRow(rowSelectionFn: (key: string) => boolean){
this.dataSource.dataLoaded$.pipe(takeUntil(this.destroyed$))
.subscribe(data => {
const foundRecord = data.filter(rec => rowSelectionFn(rec));
if(foundRecord && foundRecord.length >= 0){
this.selection.toggle(foundRecord[0]);
}
});
}

jqTreeView - getting selected node's value in select event

Following is how I populate my jqTreeView.
View
#Html.Trirand().JQTreeView(
new JQTreeView
{
DataUrl = Url.Action("RenderTree"),
Height = Unit.Pixel(500),
Width = Unit.Pixel(150),
HoverOnMouseOver = false,
MultipleSelect = false,
ClientSideEvents = new TreeViewClientSideEvents()
{
Select="spawnTabAction"
}
},
"treeview"
)
<script>
function spawnTabAction(args, event) {
alert(args);
}
</script>
Controller
public JsonResult RenderTree()
{
var tree = new JQTreeView();
List<JQTreeNode> nodes = new List<JQTreeNode>();
nodes.Add(new LeafNode { Text = "Products", Value="Product/Product/Index" });
FolderNode fNode = new FolderNode { Text = "Customers" };
fNode.Nodes.Add(new LeafNode() { Text = "Today's Customers", Value = "Customer/Customer/Today" });
nodes.Add(fNode);
nodes.Add(new LeafNode { Text = "Suppliers", Value = "Supplier/Supplier/Index" });
nodes.Add(new LeafNode { Text = "Employees", Value = "Employee/Employee/Index" });
nodes.Add(new LeafNode { Text = "Orders", Value = "Order/Order/Index" });
return tree.DataBind(nodes);
}
What I want to do is spawn a tab based on the Value of the selected node. I tried a lot but couldn't get hold of the selected node's value.
Later I checked the DOM of rendered page and found that the value is nowhere added to the node but magically when I select the node the value appears in a hidden control by the name treeview_selectedState (treeview being the id of the control). I even traced all ajax calls but couldn't find anything.
Questions:
1) Where does it keep the Values of tree nodes?
2) How do I get the Selected Node's value in select event?
I even tried to get the treeview_selectedState control's value in select event but it returned [].
Then I added a button the view and hooked that onto a js function and found the value there. It makes me think that the value is not available in select event, am I right in thinking that?
I don't think getting selected node's value should be a this big deal? Am I missing something very obvious?
Thanks,
A
After trying so many things , I checked their demo and found the hints there (I shouldve done this as the first thing).
It was actually pretty straight forward
function spawnTabAction(args, event) {
alert($("#treeview").getTreeViewInstance().getNodeOptions(args).value);
}
Thanks,
A

Two related questions on jqGrid column header filters and the advanced filtering dialog

In developing my first ASP.NET MVC 3 app using the jqGrid to display some data, I'm using the column header filters and also allowing for the advanced filter toolbar filtering to be done. Independently these things work pretty well.
First question - Has anyone a solution for communicating the current column header filter settings to the advanced filters?
As an example, a user can filter on the "Ice Cream Name" column, entering a partial name, e.g., "Chocolate", and it'll filter down to "Chocolate Explosion", "Dark Chocolate", etc. - great. What would be nice would be to open the advanced filter and have that "contains 'Chocolate'" column filter automatically populated in the advanced filter. I recognize that the other direction (where someone could AND or OR two values for the same column, e.g. 'Chocolate' OR 'Caramel') becomes problematic but in the other direction, it seems like it might be possible. Perhaps this is just a setting of the grid I'm missing. Anyone solved this?
Second question - I currently can do some filtering with the column header filters, show some result set in the grid and then go into the advanced filter dialog and set up a different filter. That will display the correct results but the column header filters are not cleared, giving the impression that the filtering is not working. How can I reset those column header filters after the use clicks the "Find" button on the dialog?
I find your question very interesting, so I prepared the demo which demonstrate how one can combine Advanced Searching dialog and Toolbar Searching in one grid.
One important, but simple trick is the usage of recreateFilter: true. Per default the searching dialog will be created once and then will be only hide or show. As the result the postData.filters parameter will be not refreshed. After setting recreateFilter: true the problem with filling of the advanced searching dialog with the values from the searching toolbar will be solved. I personally set the default searching options as the following
$.extend(
$.jgrid.search,
{
multipleSearch: true,
multipleGroup: true,
recreateFilter: true,
overlay: 0
}
);
Now to more complex part of the solution is the function refreshSerchingToolbar which I wrote. The function is not so simple, but it's simply in use:
loadComplete: function () {
refreshSerchingToolbar($(this), 'cn');
}
The last parameter is the same parameter which you used as defaultSearch property of the searching toolbar method filterToolbar (the default value is 'bw', but I personally prefer to use 'cn' and set jqGrid parameter ignoreCase: true).
If you fill the advanced searching dialog of the demo with the following field
and click the "Find" button, you will have the following grid:
(I marked the 'Total' column as non-searchable with respect of search: false to show only that all works correctly in the case also)
One can see that all fields of the searching toolbar excepting "Amount" are filled with the values from the searching dialog. The field are not filled because we used "grater or equal" operation instead of "equal". The function refreshSerchingToolbar fills only the elements of the searching toolbar which can be produced by the
Just as a reminder I should mention that in case of the usage of Filter Toolbar it is very important to define searchoptions.sopt options of the colModel. For all non-string column contains (dates, numbers, selects, int, currency) it is extremely important to have 'eq' as the first element of the sopt array. See here and here for details.
If you change the filter of the Advanced Dialog to the following
you will have as expected
At the end I include the code of the refreshSerchingToolbar function:
var getColumnIndex = function (grid, columnIndex) {
var cm = grid.jqGrid('getGridParam', 'colModel'), i = 0, l = cm.length;
for (; i < l; i += 1) {
if ((cm[i].index || cm[i].name) === columnIndex) {
return i; // return the colModel index
}
}
return -1;
},
refreshSerchingToolbar = function ($grid, myDefaultSearch) {
var postData = $grid.jqGrid('getGridParam', 'postData'), filters, i, l,
rules, rule, iCol, cm = $grid.jqGrid('getGridParam', 'colModel'),
cmi, control, tagName;
for (i = 0, l = cm.length; i < l; i += 1) {
control = $("#gs_" + $.jgrid.jqID(cm[i].name));
if (control.length > 0) {
tagName = control[0].tagName.toUpperCase();
if (tagName === "SELECT") { // && cmi.stype === "select"
control.find("option[value='']")
.attr('selected', 'selected');
} else if (tagName === "INPUT") {
control.val('');
}
}
}
if (typeof (postData.filters) === "string" &&
typeof ($grid[0].ftoolbar) === "boolean" && $grid[0].ftoolbar) {
filters = $.parseJSON(postData.filters);
if (filters && filters.groupOp === "AND" && typeof (filters.groups) === "undefined") {
// only in case of advance searching without grouping we import filters in the
// searching toolbar
rules = filters.rules;
for (i = 0, l = rules.length; i < l; i += 1) {
rule = rules[i];
iCol = getColumnIndex($grid, rule.field);
cmi = cm[iCol];
control = $("#gs_" + $.jgrid.jqID(cmi.name));
if (iCol >= 0 && control.length > 0) {
tagName = control[0].tagName.toUpperCase();
if (((typeof (cmi.searchoptions) === "undefined" ||
typeof (cmi.searchoptions.sopt) === "undefined")
&& rule.op === myDefaultSearch) ||
(typeof (cmi.searchoptions) === "object" &&
$.isArray(cmi.searchoptions.sopt) &&
cmi.searchoptions.sopt[0] === rule.op)) {
if (tagName === "SELECT") { // && cmi.stype === "select"
control.find("option[value='" + $.jgrid.jqID(rule.data) + "']")
.attr('selected', 'selected');
} else if (tagName === "INPUT") {
control.val(rule.data);
}
}
}
}
}
}
};
UPDATED: The above code is no more needed in case of usage free jqGrid 4.13.1 or higher. It contains the new default option loadFilterDefaults: true of the filterToolbar, which refreshes the values of the filter toolbar and the filter operations (if searchOperators: true option of filterToolbar is ised) if postData.filters and search: true are set (the filter is applied). Free jqGrid refreshes the filter toolbar on jqGridAfterLoadComplete (if loadFilterDefaults: true are set) or if the event jqGridRefreshFilterValues are explicitly triggered.
I know it's an old post - but if you have multiple grids on the same page the above code can add the filter text to the wrong grid.
Changing this in the first loop in refreshSearchingToolbar, from
control = $("#gs_" + $.jgrid.jqID(cm[i].name));
to
control = $("#gview_"+$grid.attr('id')+" #gs_" + $.jgrid.jqID(cm[i].name));
and this in the second loop from
control = $("#gs_" + $.jgrid.jqID(cmi.name));
to
control = $("#gview_"+$grid.attr('id')+" #gs_" + $.jgrid.jqID(cmi.name));
should do the trick.
Kudos to Oleg

Resources