Challenge: Modifying execution approaches

In RavenDB, we had this piece of code:

internal T[] LoadInternal<T>(string[] ids, string[] includes)
        {
            if(ids.Length == 0)
                return new T[0];

            IncrementRequestCount();
            Debug.WriteLine(string.Format("Bulk loading ids [{0}] from {1}", string.Join(", ", ids), StoreIdentifier));
            MultiLoadResult multiLoadResult;
            JsonDocument[] includeResults;
            JsonDocument[] results;
#if !SILVERLIGHT
            var sp = Stopwatch.StartNew();
#else
            var startTime = DateTime.Now;
#endif
            bool firstRequest = true;
            do
            {
                IDisposable disposable = null;
                if (firstRequest == false) // if this is a repeated request, we mustn't use the cached result, but have to re-query the server
                    disposable = DatabaseCommands.DisableAllCaching();
                using (disposable)
                    multiLoadResult = DatabaseCommands.Get(ids, includes);

                firstRequest = false;
                includeResults = SerializationHelper.RavenJObjectsToJsonDocuments(multiLoadResult.Includes).ToArray();
                results = SerializationHelper.RavenJObjectsToJsonDocuments(multiLoadResult.Results).ToArray();
            } while (
                AllowNonAuthoritiveInformation == false &&
                results.Any(x => x.NonAuthoritiveInformation ?? false) &&
#if !SILVERLIGHT
                sp.Elapsed < NonAuthoritiveInformationTimeout
#else 
                (DateTime.Now - startTime) < NonAuthoritiveInformationTimeout
#endif
                );

            foreach (var include in includeResults)
            {
                TrackEntity<object>(include);
            }

            return results
                .Select(TrackEntity<T>)
                .ToArray();
        }

And we needed to take this same piece of code and execute it in:

• Async fashion
• As part of a batch of queries (sending multiple requests to RavenDB in a single HTTP call).

Everything else is the same, but in each case the marked line is completely different.

When we had only one additional option, I choose the direct approach, and implement it using;

public Task<T[]> LoadAsync<T>(string[] ids)
{
    IncrementRequestCount();
    return AsyncDatabaseCommands.MultiGetAsync(ids)
        .ContinueWith(task => task.Result.Select(TrackEntity<T>).ToArray());
}

You might notice a few differences between those approaches. The implementation behave, most of the time, the same, but all the behavior for edge cases is wrong. The reason for that, by the way, is that initially the Load and LoadAsync impl was functionality the same, but the Load behavior kept getting more sophisticated, and I kept forgetting to also update the LoadAsync behavior.

When I started building support for batches, this really stumped me. The last thing that I wanted to do is to either try to maintain complex logic in three different location or have different behaviors depending if you were using a direct call, a batch or async call. Just trying to document that gave me a headache.

How would you approach solving this problem?