How to sort the state rows within a region by the first occurrence of rainfall as opposed to by state name??
I'm writing in C# using VS2008 unpatched and Infragistics 9.1.20091.2056.
Let's pretend I'm using a WinGrid to display rainfall amounts for states within regions over a changing date range.
The regions - such as Western, Northwest, Eastern, Southeast, etc. - are the categories in the Group By. A hidden category column allows me to group by category while still displaying a "header" row for each region in the grid.
Sorting is turned off for all columns other than the hidden category column which means Category is the only sort button in the sorting band across the top, no additional columns can be dragged there and all additional sorting needs to be handled programmatically.
The states - such as California, Nevada, etc. - belong to a region and are categorized under their region based on the region they are attached to when the rows are added to the grid.
The columns - other than the column containing the region heading and state names - are dates and appear in the grid based on there having been rainfall during the specified date range. Sorting by the user is not allowed.
If a date column appears then it indicates that at least one state in one region has a rainfall value for that date.
My confusion:
How to sort the state names within a region by the first occurrence of rainfall as opposed to by state name??
For example, if California had rain on 10/20/2009 and Nevada had rain on 10/19/2009 then Nevada should appear before California in its region group. How to accomplish this?
BTW I've been reading the posts regarding sorting WinGrids closely and repeatedly but am still confused so please give it to me simply and with examples if possible. :-)
Thanks.
Allen
Hi Allen,
In order to group the row, the grid must first sort the rows by region. This is neccessary and you can't change that, because the grid needs to sort the rows by region in order to loop through and group rows with the same values.
But rows that have the same region are placed in no particular order or in the original order in which they appeared in the grid.
One easy way to handle this would be to enable sorting on the rainfail column and sort it. But, you obviously don't want to do that.
So what you need to do is set it up so that the Region columns sorts by both the region (first) and the rainfall (second). As long as the Region sorting takes precedence, this won't interfere with the grouping. So all you have to do is use a SortComparer on the Region column that sorts based on two columns instead of one.
I don't know as we are on the same page Mike.
Yes it would be nice to sort the regions too by geographical area instead of by name but I'm assuming I can't do that in a group by situation.
There are a variable number of rows depending on which states had rainfall in the variable date range. There will be a row for any state that has a date in the range being selected for. Likewise the columns will change based on how many dates have states with rainfall within the range being selected for.
There is not one column that will have the rainfall. For example, if one state has rain on 1/1 and one state has rain on 1/2 and one state has rain on 1/3 then there will be three columns (one for each date) and three rows (one for each state) and I want to sort the rows on 1/1 followed by 1/2 followed by 1/3 regardless of state name (but within region of course).
Are we on the same page? How am I going to do that with one SortComparer sorting based on Region and only one date field?
I do realize I can sort before populating the grid. I'm interested in knowing how to manipulate the grid in code as if I were moving the rows around by hand.
Perhaps it is not possible in code? Perhaps I need to create a sample program to better describe my issue? In real life I'm working with patient data, not rainfall amounts.
Thank you,