LINQ and group by data nested within a structure

Question

I have a structure that roughly looks like this:

List<ProductLine> -> ID
                     Name
                     ...
                     List<LineOfBusiness> -> ID
                                             Name
                                             ...
                                             List<Watchlist> -> ID
                                                                Name
                                                                ReportCount

A Watchlist can exist under multiple LoBs but the ReportCount will only be for the count of reports that exist under that LoB for that WatchList. I need them in the structure this way because the count of how many reports exist within a LoB for a given Watchlist is important elsewhere.

What I need to do is get a list of the distinct WatchLists (grouped based on ID) and have the ReportCount be the SUM of that watchlist's ReportCount across all LoBs. I can't quite get the nested select logic to work right.

Answer 1

The technique to flatten a hierarchical structure is to use the SelectMany method. You need something like this:

var result = mainList.SelectMany(x => x.LineOfBusinessList)
                     .SelectMany(lob => lob.Watchlists)
                     .GroupBy(wl => wl.ID)
                     .Select(g => new { 
                            WatchlistID = g.Key,
                            WatchlistName = g.First().Name,
                            ReportCount = g.Sum(item => item.ReportCount)  
                     });

The first SelectMany call will transform the original list to sequence of all LineOfBusiness objects in all items. The second SelectMany call will transform a sequence of LineOfBusiness objects to a sequence containing all Watchlist objects it them. Then you group these Watchlists by they ID and perform the actual query on them.