<div dir="ltr"><div class="gmail_default" style="font-family:monospace">> That was what i referred to as "Map-solvable" problems,<br>> but if this way of doing it is a thing, then I agree with<br>> the point.<br><br>Doing it the Map way would not only be slower, but it would force you to include more concepts than you need. Your identifier is your index. Lists/arrays have index. Bringing a map in only complicates things.<br><br></div><div class="gmail_default" style="font-family:monospace">As for the rest of your response, I think I understand the trouble here. You put it best when you said that you are introducing this feature with the intent of improving the developer experience, specifically through making code more concise.</div><div class="gmail_default" style="font-family:monospace"><br></div><div class="gmail_default" style="font-family:monospace">I can understand that goal, and it is a good goal to strive for, but conciseness is one of the last things Java aims to provide as a language or through its standard library. I won't tell you not to do it, but I will tell you that there will be significant push back from folks here, and for good reason -- conciseness is not the priority here at Java. At Java, the lack of conciseness is not a weakness. Sure, unneeded verbosity can be a weakness, but the solution I suggested above is basically the tempo that Java strives for -- compose simple concepts so that you can create what you need.<br></div><div class="gmail_default" style="font-family:monospace"><br></div><div class="gmail_default" style="font-family:monospace">All of that is to say, you have made your point clear. You goal is conciseness and ease of access. Understanding that, it's clear that my solution is not concise enough to meet that goal, so I won't push it further.<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Apr 19, 2024 at 7:26 PM ІП-24 Олександр Ротань <<a href="mailto:rotan.olexandr@gmail.com">rotan.olexandr@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">"
<span style="font-family:monospace">1. Fetching data from same-sized lists -- This is a trick from Data-Oriented Design (not to be confused with Data-Oriented Programming) where you filter down to the indexes that you want, and use the indexes to fetch related values from the respective lists. So, I would filter down to Player1, Player4, and Player6, then I would use their indexes to do lookups in PlayerItemList, PlayerSkinList, etc. This tactic is especially useful when working with structs (or value classes when Java gets them)."</span><br style="font-family:monospace"><div><br></div><div>That was what i referred to as "Map-solvable" problems, but if this way of doing it is a thing, then I agree with the point.</div><div><br></div><div>"<span style="font-family:monospace">By this logic, many of the methods in the Stream library should instead be in the Collection library. There are several emails on the lambda mailing list that address why this wasn't done. I imagine that Rémi himself could point them out to you."</span></div><div><br></div><div>It may be just my opinion, but I don't see anything harmful in some shortcuts that could be optimized, especially if they still converge to a single point under the hood. I would love to read mails about maintainers' opinions on why some features of Streams couldnt be also a part of Collection library if it would enhance developer experience, so, if it wouldn't be hard for you, could you attach some links? Also I guess a good compromise would be extension methods, I am currently exploring possibilities of their implementation.</div><div><br></div><div><span style="font-family:monospace">"Well hold on. I am not at all saying that it is redundant. I am saying that the functionality that you desire can be easily recreated by using a more far-reaching feature, the .withIndex() idea that Rémi suggested. I am just trying to say that your idea probably should not be done on the List interface."</span><br></div><div><br></div><div>I am not saying that .withIndex() is a bad idea in any way. I can remember several times when I was desperately looking for something like python enumerate(), but all I found were some IntStream-based approaches. What I am saying is, while there is definitely room for both, withIndex does not solve the initial problem that motivated me to propose such changes: redundant (emphasis on this) verbosity. The main reason I decided to become an openjdk contributor is because I love to create tools that make development easier and more pleasant, and I feel more sophisticated, although a more specific API would deal much better with this task, and as this is my personal priority, I would stand my ground here.</div><div><br></div><div>"<span style="font-family:monospace">Regardless, Valhalla is not here yet. So at least for now, I will concede your point about performance."</span></div><div><br></div><div>Valhalla, at least for now, is nowhere near to a release even as a preview, would be great if it will get to Java releases until the next LTS. I think that even, even for a 2-4 years, this point is valid, then it at least should be taken into consideration.</div><div><br></div><div>PS: As you might notice, I am a developer experience guy rather than a flexible guy. My opinion is a result of some personal beliefs, but also is an aggregation of opinions of numbers of other programmers that I introduced to Java, especially switchers. Many of then feel those "wordiness" of Java and it repulses potential developers, especially when there is C# just by the corner. I think that Java has really strong sides, but I would love to cover its "weaknesses" as well, especially if it doesn't come with any maintenance burdens.</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">сб, 20 апр. 2024 г. в 02:00, David Alayachew <<a href="mailto:davidalayachew@gmail.com" target="_blank">davidalayachew@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div class="gmail_default" style="font-family:monospace">No confusion. You spoke clearly enough to be understood the first time.<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Apr 19, 2024 at 6:28 PM ІП-24 Олександр Ротань <<a href="mailto:rotan.olexandr@gmail.com" target="_blank">rotan.olexandr@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Regarding the first point in my last message. "Them" is referring to the developer, and "evaluate" means to be processed by the developer. Sorry for the potential confusion</div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">сб, 20 апр. 2024 г. в 00:52, ІП-24 Олександр Ротань <<a href="mailto:rotan.olexandr@gmail.com" target="_blank">rotan.olexandr@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">I think that, while finding the index of a matching element is not a very common task, especially among commercial developers, when discussing indexed streams, this would be the most common use case. To be honest, I don't even see any other use cases right now, besides maybe some processing based on divisibility of index (odd/even as the most common one). I have a pretty strong background in competitive programming, more than 5 years, and can remember anything more complex than that that could not be replaced with Map usage.<div><br></div><div>Regarding clarity, I fully agree with you. I am not a fan of those unreadable abbreviations in C/C++ or even in python. However, while it's hard to argue that your approach is more flexible, it has two downsides:</div><div><br></div><div>1. Although indexed streams are flexible, the flexibility might come at the cost of readability. Syntax like list.FindIndex is pretty straightforward and speaks for itself, while indexed-stream-based approach requires developer to explicitly evaluate what that chain of method does, which would distract them from understanding of general program logic flow, while usually being no more then utility. The reason for existence of findIndex is the same as for indexOf, so saying one is redundant means saying other one is redundant too.</div><div><br></div><div>2. Streams are performance-costy, thats the point that is hard to argue with. While list implementations could provide specified implementations of findIndex based on their internal structure, streams work only with generic interfaces. Moreover, even default implementation that uses ListIterator would be times faster and less resource consuming then generic stream-based approach. While I understand that performance is not as valuable as flexibility and clarity, it is still crucial, at least because of it being the main factor server maintenance cost after all.</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">сб, 20 апр. 2024 г. в 00:32, David Alayachew <<a href="mailto:davidalayachew@gmail.com" target="_blank">davidalayachew@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div class="gmail_default" style="font-family:monospace">It is not the goal of Java to be verbose, but it is also not the goal of Java to be concise either. It is the goal of Java to be clear, correct, and readable, in order to maximize maintainability, correctness, and flexibility. When you focus on accomplishing that goal first, you will find that conciseness takes care of itself.<br></div><div class="gmail_default" style="font-family:monospace"><br></div><div class="gmail_default" style="font-family:monospace">I understand that my approach is wordier than yours. Please also consider that my suggestion does exactly one task -- stream an element with its respective index. And yet, the number of ways that that can be used covers hundreds of use cases while still being incredibly easy to maintain and optimize. Far more than what you suggest. My feature easily composes with others, with anything that can take a Stream<T> in fact.<br></div><div class="gmail_default" style="font-family:monospace"><br></div><div class="gmail_default" style="font-family:monospace">The bar to adding a feature to Java, even to its libraries, is high. In fact, there is even a breaking point where a language/library can just be too dense to work with. Too many methods, too much complexity.<br></div><div class="gmail_default" style="font-family:monospace"><br></div><div class="gmail_default" style="font-family:monospace">Java makes it a priority to limit adding features or functionality wherever possible, only relenting when the benefit significantly outweighs the cost. Adding a method like yours that only serves a tiny slice of use cases just does not meet the goal of what Java is aiming for -- to get the biggest bang for buck when adding a feature to language/library.</div><div class="gmail_default" style="font-family:monospace"><br></div><div class="gmail_default" style="font-family:monospace">I think that your pain point is significant and worth addressing. But I think it should be done in a way that services a much larger group of users and use cases. I think my idea does that.</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Apr 19, 2024 at 5:14 PM ІП-24 Олександр Ротань <<a href="mailto:rotan.olexandr@gmail.com" target="_blank">rotan.olexandr@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">streamWithIndex() is indeed an interesting approach, however, working with pairs of values is not always comfortable as this may make code too wordy, which may be fixed using things like properties and tuples (which I am currently working on right now btw), but still code like list.streamWithIndex().filter(Predicate).findFirst().orElse(-1) is still long, thats why i think there is room for both in Java. <div><br></div><div>Also maybe instead of returning traditional -1 it would be more suitable to use OptionalInt instead, but I would like to hear community opinion about that.</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">сб, 20 апр. 2024 г. в 00:02, David Alayachew <<a href="mailto:davidalayachew@gmail.com" target="_blank">davidalayachew@gmail.com</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><span style="font-family:monospace"><br>No Rémi, I don't think your idea is the right approach. You are working on the wrong level of abstraction.<br><br>Many users ask requests like this all the time, and what you are suggesting would be even more error-prone than the equivalent for loop or <span class="gmail_default" style="font-family:monospace">the </span>IntStream suggestion that the user above requested. Not to mention that getting it to parallelize would be a task many users are likely to mess up -- either in correctness or in performance.<br><br>I think you would get far more mileage from adding 2 methods on the list interface streamWithIndex() and parallelStreamWithIndex() that would return a Stream<WithIndex<T>>, as opposed to just Stream<T>.<br><br></span><div><span style="font-family:monospace">That way, users are not writing a custom Gatherer each time they want to work with the index. They just have the index be a field in the object. They can work with it the same way they would any other object field.</span></div><div><span style="font-family:monospace"><br></span></div><div style="font-family:monospace" class="gmail_default">Furthermore, doing it this way makes the correct answer obvious. If I need to do something with an index, stream with the index.</div><div style="font-family:monospace" class="gmail_default"><br></div><div style="font-family:monospace" class="gmail_default">On top of that, it significantly enhances readability by making it clear to the reader that, whatever this stream is doing will require use of the index, so watch out for that.<br></div><br><div><br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Apr 19, 2024, 1:47 PM Remi Forax <<a href="mailto:forax@univ-mlv.fr" target="_blank">forax@univ-mlv.fr</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><div id="m_-1185381824599906472m_-2806104675124158725m_-6921175329925079697m_-5860725732801130010m_1734352192913307536m_3960311137366901599m_107230642713004337m_6383933055071446295m_-4157093933715390893m_9119089240644209637zimbraEditorContainer" style="font-family:arial,helvetica,sans-serif;font-size:12pt;color:rgb(0,0,0)"><div>Hello,</div><div>for me, it seems what you want is Collector on Stream which is able to short-circuit,<br></div><div>so you can write<br></div><div> list.stream().collect(Collectors.findFirst(s -> s.contains("o")))<br></div><div>and in reverse<br></div><div> list.reversed().stream().collect(Collectors.findFirst(s -> s.contains("o")))<br></div><div> <br></div><div>Using a Stream here is more general and will work with other collections like a LinkedHashSet for example.<br></div><div>Sadly, there is no way to define a short-circuiting collector :(<br></div><div><br></div><div>You can have a short-circuiting Gatherer like this<br></div><div> <T> Gatherer<T, ?, Integer> findIndex(Predicate<? super T> predicate) {<br> return Gatherer.ofSequential(<br> () -> new Object() { int index; },<br> Integrtor.ofGreedy((state, element, downstream) -> {<br> var index = state.index++;<br> if (predicate.test(element)) {<br> return downstream.push(index);<br> }<br> return true;<br> }));<br> }<br></div><div><br></div><div>and use it like this:<br></div><div> list.stream().gather(findIndex(s -> s.contains("o"))).findFirst().orElse(-1);<br></div><div><br></div><div>But it's more verbose.<br></div><div><br></div><div>I wonder if at the same time that the Gatherer API is introduced, the Collector API should be enhanced to support short-circuiting collectors ?<br></div><div><br></div><div>regards,<br></div><div>Rémi<br></div><div><br></div><div><br></div><hr id="m_-1185381824599906472m_-2806104675124158725m_-6921175329925079697m_-5860725732801130010m_1734352192913307536m_3960311137366901599m_107230642713004337m_6383933055071446295m_-4157093933715390893m_9119089240644209637zwchr"><div><blockquote style="border-left:2px solid rgb(16,16,255);margin-left:5px;padding-left:5px;color:rgb(0,0,0);font-weight:normal;font-style:normal;text-decoration:none;font-family:Helvetica,Arial,sans-serif;font-size:12pt"><b>From: </b>"ІП-24 Олександр Ротань" <<a href="mailto:rotan.olexandr@gmail.com" rel="noreferrer" target="_blank">rotan.olexandr@gmail.com</a>><br><b>To: </b>"core-libs-dev" <<a href="mailto:core-libs-dev@openjdk.org" rel="noreferrer" target="_blank">core-libs-dev@openjdk.org</a>><br><b>Sent: </b>Friday, April 19, 2024 5:59:39 PM<br><b>Subject: </b>Addition of Predicate-based findIndex and findLastIndex methods to java.util.List<br></blockquote></div><div><blockquote style="border-left:2px solid rgb(16,16,255);margin-left:5px;padding-left:5px;color:rgb(0,0,0);font-weight:normal;font-style:normal;text-decoration:none;font-family:Helvetica,Arial,sans-serif;font-size:12pt"><div dir="ltr">Subject<br>Addition of Predicate-based findIndex and findLastIndex methods to java.util.List<br><br>Motivation<br>The motivation behind this proposal is to enhance the functionality of the List interface by providing a more flexible way to find the index of an element. Currently, the indexOf and lastIndexOf methods only accept an object as a parameter. This limits the flexibility of these methods as they can only find the index of exact object matches.<br><br>Here I want to propose methods that would accept a Predicate as a parameter, allowing users to define a condition that the desired element must meet. This would provide a more flexible and powerful way to find the index of an element in a list.<br><br>The changes I am proposing are implemented in this PR: <a href="https://github.com/openjdk/jdk/pull/18639" rel="noreferrer" target="_blank">https://github.com/openjdk/jdk/pull/18639</a>. Here is a brief overview of the changes made in this pull request:<br><br>Added the findIndex (Predicate<? super E> filter) method to the List interface.<br>Added the findLastIndex (Predicate<? super E> filter) method to the List interface.<br>Implemented these methods in all non-abstract classes that implement the List interface, as well as List itself (default impl).<br>The changes have been thoroughly tested to ensure they work as expected and do not introduce any regressions. The test cases cover a variety of scenarios to ensure the robustness of the implementation.<br><br>For example, consider the following test case:<br><br>List<String> list = new ArrayList<>();<br>list.add("Object one");<br>list.add("NotObject two");<br>list.add("NotObject three");<br><br>int index1 = list.findIndex(s -> s.contains("ct t"));<br>System.out.println(index1); // Expected output: 1<br>int index2 = list. findLastIndex(s -> s.startsWith("NotObject"));<br>System.out.println(index2); // Expected output: 2<br>Currently, to achieve the same result, we would have to use a more verbose approach:<br><br>int index1 = IntStream.range(0, list.size())<br> .filter(i -> list.get(i).contains("ct t"))<br> .findFirst()<br> .orElse(-1);<br>System.out.println(index1); // Output: 1<br>int index2 = IntStream.range(0, list.size())<br> .filter(i -> list.get(i).startsWith("NotObject"))<br> .reduce((first, second) -> second)<br> .orElse(-1);<br>System.out.println(index2); // Output: 2<br><div>Or other approaches that require additional instructions and, therefore, can`t be used in all scopes (like passing argument to method).</div><div><br>I believe these additions would greatly enhance the functionality and flexibility of the List interface, making it more powerful and user-friendly. I look forward to your feedback and am open to making any necessary changes based on your suggestions.<br><br>The main reason I am publishing this proposal in the mailing system is to gather feedback from the Java developers community, especially about possible caveats related to backward compatibility of your projects. Would appreciate every opinion!</div><br><div>Best regards</div></div><br></blockquote></div></div></div></blockquote></div>
</blockquote></div>
</blockquote></div>
</blockquote></div>
</blockquote></div>
</blockquote></div>
</blockquote></div>
</blockquote></div>