Back from a brief break, we pick up this week by finishing up the “modifying sequence operations” with some gems like RandomShuffle and go through the “partitions” category with functions like Partition and IsPartitioned. These are all solid algorithms with a lot of potential uses, so read on to see how to use them with iterators and for the source code that implements them!
Posts Tagged list
Continuing the series this week we’ll delve into the iterator functions that modify the sequence. This includes handy tools like Copy, SwapRanges, and Transform. Of course this is all done without creating any garbage! Read on to see how and for the full source code.
Last week’s article introduced the concept of iterators as an alternative to the GC-heavy IEnumerable. Today’s article expands the iterator library to include a bunch of more functions to make it useful. Think of these like the extension functions in System.Linq: Any, IndexOf, etc. These have all been tailored to iterators and none of them will create any garbage whatsoever.
In C#, just about everything is an IEnumerable. Since LINQ syntax, foreach loops, and the System.Linq namespace are all designed to work with IEnumerable, you’ve got lots of tools to use. Unfortunately, the core of IEnumerable is the GetEnumerator function which usually creates garbage and eventually causes memory fragmentation and GC framerate spikes. Do we simply stop using all of these nice tools? Normally the answer is “yes”, but today’s article shows you another way.
Last week I presented a problem: how do you iterate over multiple lists of multiple types in the order of some common field? For example, how would you iterate over a list of Player and a list of Enemy by both of their Health fields? In that article I showed two solutions to iterate over two lists in this way. What I didn’t show were any solutions to handle more than two lists. What if you needed to also iterate over a list of NPC? Today’s article discusses how to tackle this problem and ends up with a handy utility class that you can use for your own types no matter how many lists you have. Read on to see how!
We know that we should reduce the garbage our code produces to lighten the load on Unity’s garbage collector. The trouble is that many of the ways we’re creating garbage are hidden from us. One such way to inadvertently create a lot of garbage is to use a foreach loop… at least that’s what we’ve been told. Do foreach loops really create garbage for all types of arrays, lists, dictionaries, and the rest of the collections? Do they create garbage for every loop or just the first one? Today’s article investigates to put these questions to rest. Are you safe using foreach loops or should you re-write everything to use for. Read on to find out!
The first version of SafeList tried to address a common problem: inserting and removing elements into a List<T> while you loop over it. It had a lot of problems though and ended up being pretty much useless. Today’s article presents SafeList 2.0, a radically-improved version that really solves the problem so you can actually use it as a drop-in replacement for List<T>. Read on for the details, the source code, and even the unit tests that prove it handles all the nasty corner cases for you!
foreach loops are really convenient, but are for loops faster? It’s a simple question, but one that has really wide implications in almost any codebase. Today’s article tests them out to see which is faster for looping over arrays and Lists. Read on to see which is quicker!
Contrary to what you may have learned in a data structures class, linked lists are virtually always slower than just using arrays. The same goes for array wrapper classes like List. Today’s article discusses why this is the case and tests it out with a C# Unity app to make sure that the real world validates the theory.
Last week’s article compared the performance of arrays with List<T> and found List lacking. This week we’ll optimize both List and array to maximize performance regardless of which you choose to use.