We have requirement to rearrange slick grid rows.
I have up/down icon in each row, I want to achieve animation like, clicking 'up' icon should move rows up by '1' position, but cursor must not shift its focus from current row. And other rows should rearrange their position in respect to the 'clicked row'.
Please suggest solution to this problem. Thanks
You essentially want a sorting function. The method I'd choose depends a bit on how the back end database stores the sort order, if you are persisting to a database.
If you are not persisting the data,
create an extra column SortOrder, sort by whatever other columns you want to be the default sort order and set SortOrder to 1, 2, 3, ... , n
create the buttons using a formatter, and make the up/down buttons call javascript that
switches the sort number with the row before/after
resorts and rerenders the grid
sets focus (if necessary) to the desired row using grid.setActiveCell() and grid.focus()
Pesonally, I store SortOrder as an int database column, and set it to all zeroes initially so that secondary default sort columns can be used if no custom sort is desired. However the sort index algorithm is significantly more complex.
Related
I am struggling to find information on a lot of small issues I'm starting to come across since recently integrating the ObjectListView into my application. I must admit, I love the control and it takes a lot of work away from me by having the control pretty much do everything that I need.
However, amongst a few issues that I've come across thus far, one is the fact that if I turn off grouping for individual columns it no longer allows me to sort these columns.
For example, my ListView control has an ID, Summary, Location, etc., etc. I currently run grouping on the ID because I attach a unique footer against every ID and it looks good. But, when I click on Summary, or Location it then groups all items by Summary or Location and therefore I end up with the group header but no footer anymore because the footer no longer corresponds with the ID.
What I want to achieve is to allow the user to click on any other column, but still maintain the group by ID and thus keep the footer against each item. I only want the items to sort, not group, but as mentioned, when I turn off the grouping against each column, it turns off the sort function.
UPDATE
Basically, looking at the screenshot below, I want to be able to click on the Priority Column Header (for example) to sort the list of object by Priority, but still maintain the Group By Ticket (with the Group Footer details). And basically be able to click on any Column header to sort the list but keeping the grouping by Ticket.
I have intercepted the OLV_BeforeCreatingGroups event to add:
e.Parameters.GroupByColumn = Ticket_ID
This now seems to maintain my grouping by Ticket no matter what Column header I click on, but I now loose the sort functionality.
Try setting AlwaysGroupByColumn to be the TicketId column. That will do the trick :)
Second attempt:
Thanks for the screenshot. I think I understand what you want. You are hoping that clicking on the header will sort the groups themselves. That's not the default behaviour, which is to sort the rows within each group.
To do what you are asking:
Listen for BeforeCreatingGroups event and fill in the GroupComparer property on the event argument block. This comparer controls the ordering of the groups.
The default group order is based on OLVGroup.SortValue. So you could install a GroupFormatter delegate and change the SortValue for each group so they are ordered in the manner you want.
In either case, you will need to use PrimarySort and PrimarySortOrder to know which column the user last clicked.
I am using an .rldc file to define the layout of the reports from my program. The problem is, it is to be used for incremental printing. That means the paper will be used over and over as newer rows need to be printed. I'm attempting to approach it this way:
List all corresponding data on the report view.
Make the older rows invisible and only show the latest row.
Print.
That way, the last row is already properly placed. The problem is, I don't know how to implement this. Can anyone help me out?
You could create an IIF(condition,true,false) statement in your report definition on the row visibility variable.
The best way i guess is to define in your data source something of a rank column.
example :
select col1,col2,col3,RANK() OVER (ORDER BY col3 DESC) AS 'rank' from table1
Then in your table or matrix, you click on the row or/and column that you want to make the borders and text white based on a expression.
Go to the properties and dropdown on bordercolor
choose expression and type in (based on my example query)
=IIf(rank.value <> max(rank.value),White,Black)
That will not remove the rows only make the borders white ( unvisible)
The same you can do with Font Color property.
I think this is your best shot at this issue.
Other solution I could think of is to just hide unneccesary rows (which also replaces the visible row)
Then to move the table down by using a expression with a formula like nr of rows hidden before the actual row * height of 1 row, only I m not sure if this is applicable without programming an RDL extension..
Good Luck !
I have a semi-dynamically created window ( and use PowerBuilder 10.5 ). Now there are a couple of columns which can have different colours and I want to see those colours when selecting a row. However I don't know how to deselect these columns and have the first couple of columns remain selected.
The highlight function in our application just does a dw.selectrow( x, true ).
Regards,
Marinus
I don't think you'll get what you want using selectrow. If you don't need multiple selections you could change the background of the current row with an expression in the datawindow. If you want some columns to stay normal you could do that with a rectangle behind the ones you want to highlight instead of changing the row color. If you need multiple rows highlighted you will have to simulate multi-select by adding a dummy column and using that to control the background, and of course you need to handle the selecting and deselecting in the clicked event. If you've got code that deals with selected rows you'll have to change them to use the new scheme, for example by checking the value in your dummy column.
If you want to use SelectRow(), maybe use of SetRowFocusIndicator() would help. IIRC (it's been pretty close to decade*s* since I've used it), it disables the row colouring in favour of the new method.
The other way that comes to mind is setting an expression for background colour that uses GetRow() and CurrentRow(). This wouldn't be my first choice, as it doesn't let users with vision impairment choose their colours through standard Windows colour selection to something they can deal with, but if you're dead set on colour indications on selective columns, this would be the way to go.
Good luck,
Terry.
I am using a dojox.grid.DataGrid to display data and allow the user to select one or more rows to highlight items to do other actions to.
My problem is that the grid uses indexes to remember the selected rows. If I have turned clientSort on, sorting the rows makes the rows change order but the selection changes to the rows that now occupy the selection indexes before sorting.
To combat this I get the items for the selected rows on each selection change and then when onSort event happens I use getItemIndex(item) on each of the items in my selected items collection to get the new row indexes and I re-select using those indexes.
Now, my problem is that the DataGrid loads data a certain number of rows at a time (it uses the term page to define a set number of rows) to improve performance. When a sort occurs it reloads the data in the new sort order but only up to the rowsPerPage value (defaults at 40). When I try and use getItemIndex for an item whose new row has not been loaded into the grid it returns and index of -1. The getItemIndex() function only returns the index once the user scrolls down and the row is loaded.
I cannot find an event that fires when the page loading occurs to try and restore the selection. I have looked through the 1.3.2 source code and can not find anything. I have tried connecting to a number of private functions but none work.
Has anyone come across this and found a fix?
Thanks in advance.
I found the answer. The dojox.grid.DataGrid has a private event (_onFetchComplete) which fires each time the next page is requesting. I use setTimeout to run my restore function after a duration of 300.
Job done!
Based on several report parameters in SQL Server 2005 reporting services, I would like to automatically generate one or several chart(s) for each row in the return result and paginate or space them out. How do I go about that?
If the number of charts will vary for each row, but the variations are known (e.g. it's either just chart 1, or chart 1 and 3, or charts 1 2 and 3) then it's simple enough using a table.
In the default detail row add any normal fields you need. Now insert a new detail row for each chart you might need. Lastly set the visibility of each chart row based on your rules, noting that the rule will hide the row if your expression evaluates to true. Make sure you select the row using the area to the left of the left-most cell, if you got it right you'll see that it's a row in the properties grid.
To get the layout you want you can merge cells for the charts to go in, or use a single cell and put a Rectangle in it, then in the Rectangle lay out your other controls.
Any rows that are hidden will be collapsed, so you wont get big empty sections like you can if you simply toggle the visibility of the charts themselves.
What you can do is place a List control on the page, set List grouping by record unique key (ID, or several fields if composite), and place a charts on the List. Next, set items visibility expressions to control it with report parameters.