Currently I want to update row items in a QStandardItemModel without losing the sort order and row selection in the respective QTableView.
I have tested two approaches:
Clearing the model by clear() and re-adding the rows "destroys" everything including headers.
Removing and re-adding all rows keeps the headers, but still "destroys" selection and sort order.
I could try to manually to a) remove all rows no longer required and then b) update the items of the changed rows. But is there no easier way?
I don't know, how you sorted your data before the update, but please take a look at the QTableView's sortByColumn() function.
As for the selection, if it still keeps disappearing, you can manually put back your selection, where it should be by:
// You access the selected index when the editing starts
QModelIndex index = table->selectionModel()->currentIndex();
//Later when you finished editing, you can select it again
table->selectionModel()->select(index, QItemSelectionModel::Select);
More about this:
QItemSelectionModel
QTableView
You can either remove the rows no longer required and add the new ones or manually remember the sort order and which items were selected before clearing. I.e. assign a unique ID (one can use setData() with a custom role for that), retrieve that before clearing from the selected items, and reapply the selection after recreating the items.
Related
I have a self created Qt-Model derieved by QAbstractTableModel. The data behind the model contains multiple QUuid-columns, whose cell-data I need to pass around the application. Due to design-reasons I don't want to show the QUuid-columns to the user, but keep them in the background to always guarantee access to the needed id-columns.
The data is bound to a Qtitan TableView Grid, where I can hide the column, but not totally remove it from the view. I can always reenable the visability which is not what I want.
So my question is if there are any options from the Qt-Model-side to hide a column or to avoid binding it to the view and just keep the data in the background.
You can subtract those columns from the visible columns by returning the column respectively in columnCount.
This would require to either move them to the end, or map the user visible column count to the underlying columns in your data() implementation.
It is probably a bit simpler to move those invisible columns to the end to avoid the mapping, but you can also do the mapping if you like.
int MyModel::columnCount(const QModelIndex& parent) const
{
return allColumns - columnsToHide;
}
I want to know how to change the order of QTableView rows by dragging, and store the order changes to the model?
I use QTableView as view and QSqlTableModel as model. I am using Qt 5.15.
I set:
ui->table_view->setSelectionMode(QAbstractItemView::SingleSelection);
ui->table_view->setSelectionBehavior(QAbstractItemView::SelectRows);
ui->table_view->setEditTriggers(QAbstractItemView::NoEditTriggers);
Methods I tried:
1.
ui->table_view->setSectionsMovable(true);
ui->table_view->setDragEnabled(true);
ui->table_view->setDragDropMode(QAbstractItemView::InternalMove);
ui->table_view->setAcceptDrops(true);
It doesn't take effect. The row cannot be dragged.
2.
ui->table_view->verticalHeader()->setSectionsMovable(true);
ui->table_view->verticalHeader()->setDragEnabled(true);
ui->table_view->verticalHeader()->setDragDropMode(QAbstractItemView::InternalMove);
ui->table_view->verticalHeader()->setAcceptDrops(true);
The row can be dragged by vertical header. But the order of changes will not affect the model.
As far as I know the QTableView does not call moveRows method of the model by itself. Instead drag and drop actions call *mimeData methods.
So, one option would be the following:
In your model you need to implement mimeData and dropMimeData functions. When you drag and drop a row the view will ask for mimeData() of the given row. If you are sure that you need only internal moves you can encode the selected row indices to QMimeData. Then in the dropMimeData() you decode the indices that were selected and use them to call your moveRows() implementation. Return false from this function to prevent removing of the moved out rows.
Another option can be to override the QTableView methods such as dropEvent() in a way, that it calls the model moveRows() method directly.
I'm trying to implement multiple record selection feature on a grid.
It is very similar to http://www.tek-tips.com/faqs.cfm?fid=3831
It adds an extra column with check boxes. I want those check boxes!!
But it depends on a extra logical field in the underlying table. It need to create a class clscheck which inherits CHECKBOX. I'm not sure why this CLICK procedure is needed for the checkbox.
PROCEDURE CLICK
IF DODEFAULT()
KEYBOARD '{DNARROW}'
ENDIF
ENDPROC
When I removed it, row selection did not work correctly as expected. Why this?
Here is my requirement:
1) I don't want to add an extra logical field in the underlying table.
2) To work with controls in the grid, I think AllowCellSelection must be .T. I want AllowCellSelection = .F. because I don't need to work with any control in the grid except the check boxes. I need to work only with check boxes. The other columns will be read-only.
3) Can I have selected list without the logical field in the underlying table?
4) Can I remove the usage of KEYBOARD '{DNARROW}'?
In fact, I have a grid which is AllowCellSelection = .F., but it only provides single selection.
I need to enhance it with multiple selection, thus, I just want to add an extra column with check boxes so that user can know he can select multiple records.
No need Shift+Click or Ctrl+Click which is not familiar with idiot users.
I have found this - http://www.tek-tips.com/faqs.cfm?fid=433
It also depends on an extra logical field and it depends Shift+Click and Ctrl+Click.
What you are seeing is quite common for multi-select grids. I've used them SIMILAR to this in the past. However, you are afraid of the extra column in the underlying table. That may/not be true. You don't always have to update the ORIGINAL table, but a temporary CURSOR you are presenting to the user. Ex: If you want to display a list of employees in a table. No, you don't want to keep adding this column to the original employee table as then anyone else trying to do multi-select could falsely get your selection. However, if you pulled into your own local cursor and presented to the user, then no problem. Example...
Thisform.YourGrid.RecordSource = "Employees"
(bound directly to your employee table -- not necessarily the right thing)
vs
use in select( "C_MultiPickEmployees" )
select ;
.F. as IsChosen, ;
E.* ;
from ;
Employees E;
into ;
cursor C_MultiPickEmployees READWRITE
Thisform.YourGrid.RecordSource = "C_MultiPickEmployees"
NOW, you have your extra column without dealing with issues to the underlying table. If you wanted to further filter what you were showing -- such as employees for a certain division/department, then just add that to a WHERE clause, add an Order By if so needed and you are good to go.
As for the "Allow Cell Selection", I've never had to deal with that. I just add a "checkbox" to the first column and set
Thisform.YourGrid.Column[1].CurrentControl = "CheckBoxControl"
(based on the name it is added to the column).
Then, set the column 1's "ControlSource" = "C_MultiPickEmployees.IsChosen" and you should mostly be done.
As for the "CLICK" event trying to force the down arrow. This is more for automatically scrolling to the next record so you can just click, click, click for multiple entries.
Hope this helps clarify things for you.
I have a list of values in a SQL Table which are used to popluate a DropDownList, having a unique Integer as the value of each item and a String as the visible text (via SqlDataSource). There is also a third field in the database which is a flag to indicate whether the list item is active or not (inactive items are not shown in the DropDownList)
Selections made in the dropdown are stored in the database as their integer value (as part of a dataset making up the overall record), not the text value.
Over time, the items in the DropDownList may be removed by marking the items as inactive. However, there is still a need to open old records which may have had a now-inactive item as part of it's data...
My question is, what's the best way to ensure that the missing value included in the dropdown for the old record?
The two methods that spring to mind are to either:
Populate DropDownList with only the currently active items and, when loading a record, catch when the app tries to select a value that doesn't exist, go back to the db to see what it should be (the text value) and manually add it into the dropdown.
or...
Populate DropDownList with all list items (both active and inactive), load the record and then programatically remove all the inactive items (execpt for any that are now selected).
Neither of these seem particularly efficient, so I was wondering whether there is a best practice for this kind of thing?
there are so many optimum ways to do that sort of things, i am defining here a couple of them, use any of following if your Drop down list items count is less than 200 , lets say drop down list is of Products
1)
i) Load all Products records in drop down list and hide the inactive ones by setting visible=false
i) When you load a user record than look for its drop down list value if its visible than select it and enjoy, if its not visible than make it visible by setting its property visible=true and select it and also set its index or id in a flag to change its visibility(visible=false) again after your/users required operation performed.
2)
i) load only active Product records in drop down list ii) while loading a user record also load its product details(name, id, inactive_status) using Joins in sql.
iii) check in that user record if item is inactive then add its record in drop down list as you have all its details now with user details record else just select it.
IMPORTANT NOTE: if you drop down list has items in millions than use ADVANCE SEARCH techniques
The first thing I would do is question your business logic - should you be able to make an item inactive if it is being used as a foreign key in an active row elsewhere? If you make it inactive should it not remove all foreign keys as well?
To answer your question though I would go with a variation on the second idea but filtering in the page like that is probably slower than doing directly with SQL so I guess you have something like this at the moment to populate the dropdown
SELECT * FROM Table WHERE Active = 1
You should already have your record and the foreign key value so I would change it to this
SELECT * FROM Table WHERE Active = 1 OR PrimaryKey = [YourForeignKey]
Then you will always have the selected item but should also be fairly efficient.
I am using QTableWidget and when the user clicks on a particular row, I use that index to perform some operation. I have also enabled sorting using setSortEnable, so when I sort it by clicking on a header, all row index change but I want previous index so what should I do?
I'm guessing you want the unsorted index so you can map that back to some external list of data? If that is the case, then you should use a QTableView() which will allow you to setup a sort proxy and map indexes back to your original data (model).
This question has some info that may help: Can't use itemFromIndex in clicked signal of QTreeView with QSortFilterProxyModel