I'm trying to fix some incorrect calls to ParallelFor() that are appearing in performance profiles. I can't seem to find it in the Blueprint scripts in the Unreal Engine 4 project I'm working with. The Unreal documentation is sparse, and only tells me how to use it in C++.
Any ideas? I'm really concerned that I can't do a plain-text search for functions like this inside the Blueprint scripts. The Unreal Engine dev forums didn't help. The existing search mechanism via the search boxes appears to be for variables.
Using grep in the project folder fails.
There are many factors to muti-threading speedup, number of cores, utilization, memory, scheduler...
The problem may be external to the code.
On to the question:
Many of the built-in blueprint functions are implemented natively in C++.
I would suggest looking at the profiler call stack(tree view); Follow it up to something named similar to a node. This technique may fail for cross thread dispatches.
You will have to download the source code and look there to find the calls.
The other way is to build a debug build(with symbols) of the game and attach a debugger to the process.
The call was built-in to the framework in my case.
Thanks for pointing me in the right direction.
You can easily search trough all your BPs from withing the editor with
CTRL+SHIFT+F or window -> developer tools -> find in blueprints.
You get also there from within a blueprint, CTRL+F enter what you'r looking for, and on the right side there is kind of a book symbol in the same line where you can search all blueprints.
Related
I'm looking to improve my own productivity by trying to figure out how to provide cmd+click functionality for variables declared at runtime. I work with a library that allows you to register objects to it at runtime to be accessed later — it would be great if I could cmd+click to definitions of these! How do you think I could achieve this? I'm open to any ideas. Thanks!
Edit:
Thoughts I have so far inspired by feedback from various Slack communities:
There is no such thing as 'runtime' when providing functionality in IDEs
We need to be able to teach the IDE about what definitions to expect. Whether this means writing an extension to use the VS Code API specifically for my use-case, or otherwise working within a system that generates files that the IDE already knows to look for, which allows it to connect the dots and provide the precious peek / goto definition functionality
Edit:
I'm starting work on a VS Code extension that reads a dotfile for configuration — I'll share what I find out as I go!
Nowadays I'm trying to adapt new popular features called LSP(language Server Protocol) into my emacs. In the middle of that I found that some of those language server like as ccls, cquery can support a feature "call hierarchy". But I wonder whether it can find call hierarchy in a huge project like as Linux kernel. Unless it has, then I'd like to know if there is an another way to figure out a call hierarchy in a project.
Could you give me some advice, please?
Thanks.
I am working on an academic research regarding some very long functions in the Linux kernel (link, link).
For that research, I would like to use some code flow visualization tool, that would be able to plot a graph in which each vertex is a decision point and each edge is a piece of code which runs in a consequent way.
Do you know of any good, open source project that can visualize C code?
Perhaps a tool like KCacheGrind would be of help. It generates call graphs based on actual calls and cannot pre-generate a call graph without actually running the program, which may not suit your needs, but then it again it may.
History flow's are very neat for changes/diff across multiple versions.
Codeplex has a project, Dependency Visualizer which does support C also.
Gprof2Dot can render oprofile, this would get you dynamic info also.
CodeViz also (static tool) would work.
If your using gcc, gcc-xml has an introspector plugin also todo this.
You appears to want to acquire a flowchart of C source code ("decisions", "code blocks").
Something like this C flowchart?
To do this correctly, esp. for Linux kernal code, I'd expect you to have to preprocess the code first to get rid of macros and conditionals. I would assume that GCC would construct such a graph internally and that you ought to be able to get your hands on that graph.
Doxygen does some amount of 'visualization',
but you need to work on the code a bit for it to be usable.
Another interesting thing to check would be lxr
Linux Cross Referencer is a software toolset for indexing and presenting source code repositories. LXR was initially targeted at the Linux source code, but has proved usable for a wide range of software projects. lxr.linux.no is currently running an experimental fork of the LXR software.
I can recommend Sourcetrail. Can work with a compile_commands.json. Not sure if it's still maintained, though. But it's foss and you can fork it!
I'm evaluating the possibility of developing an Eclipse plugin to modify the source code of some Java files.
The Eclipse plugin should:
add one menu option or context menu option to launch the modification process.
add a key binding
only alter the UI in that way when an editor has been open on a Java file.
the modification process would not open a dialog, or maybe, a very simple one.
the modification process would traverse the AST of the Java file and would modify it.
Considering that we have no experience with Eclipse plugins and we need spend time in reading docs, how much time do you estimate in developing that plugin?
Thanks in advance.
It's really not that difficult at all... I had students in my design patterns class doing it for an assignment (adding/removing javabean getters and setters)
See http://help.eclipse.org/ganymede/topic/org.eclipse.jdt.doc.isv/guide/jdt_api_manip.htm
[EDIT: added the following article reference]
And a great article on it at http://www.eclipse.org/articles/article.php?file=Article-JavaCodeManipulation_AST/index.html (from 2006 -- there may be a few API changes since)
Yes, writing plugins takes a little getting used to, but so does any API.
And you can modify the AST -- see the page I reference above.
(I should note that the above link is from the eclipse help, which can also be accessed via Help->Help Contents inside Eclipse -- there's a lot of good info in there, but it's just a starting point)
You'll probably spend quite some time cursing the complexity of the eclipse plugin system. There are some example plugin development projects that can be very helpful if they cover the area you're working in.
I'd say you're looking at 2-4 days of work, spent mainly getting familiar with the platform - someone with a lot of experience writing eclipse plugins would probably take no more than an hour.
However, your step 5 could be tricky. I don't know how easy it is to access and change the Java AST; my experience is based on developing an editor plugin for an exotic file format rather than Java code.
Well, the four first points are easy to achieve, even by monkey coders that look at the eclipse PDE documentation shipped with Eclipse. These can be achieve in 1 day of work, maybe 2.
The hardest point is really the fifth one and the kind of modification you expect to do. Acting directly on the editor content is simple, accessing the editor internal AST and modifying it is really a bigger challenge and I doubt that it could be achieve in less than a week by unexperimented people (it can take longer, depending of what kind of modification you want to apply).
Having programmed through emacs and vi for years and years at this point, I have heard that using an IDE is a very good way of becoming more efficient.
To that end, I have decided to try using Eclipse for a lot of coding and seeing how I get on.
Are there any suggestions for easing the transition over to an IDE. Obviously, some will think none of this is worth the bother, but I think with Eclipse allowing emacs-style key bindings and having code completion and in-built debugging, I reckon it is well worth trying to move over to a more feature-rich environment for the bulk of my development worth.
So what suggestions do you have for easing the transition?
Eclipse is the best IDE I've used, even considering its quite large footprint and sluggishness on slow computers (like my work machine... Pentium III!).
Rather than trying to 'ease the transition', I think it's better to jump right in and let yourself be overwhelmed by the bells and whistles and truly useful refactorings etc.
Here are some of the most useful things I would consciously use as soon as possible:
ctrl-shift-t finds and opens a class via incremental search on the name
ctrl-shift-o automatically generates import statements (and deletes redundant ones)
F3 on an identifier to jump to its definition, and alt-left/right like in web browsers to go back/forward in navigation history
The "Quick fix" tool, which has a large amount of context-sensitive refactorings and such. Some examples:
String messageXml = in.read();
Message response = messageParser.parse(messageXml);
return response;
If you put the text cursor on the argument to parse(...) and press ctrl+1, Eclipse will suggest "Inline local variable". If you do that, then repeat with the cursor over the return variable 'response', the end result will be:
return messageParser.parse(in.read());
There are many, many little rules like this which the quick fix tool will suggest and apply to help refactor your code (including the exact opposite, "extract to local variable/field/constant", which can be invaluable).
You can write code that calls a method you haven't written yet - going to the line which now displays an error and using quick fix will offer to create a method matching the parameters inferred from your usage. Similarly so for variables.
All these small refactorings and shortcuts save a lot of time and are much more quickly picked up than you'd expect. Whenever you're about to rearrange code, experiment with quick fix to see if it suggests something useful.
There's also a nice bag of tricks directly available in the menus, like generating getters/setters, extracting interfaces and the like. Jump in and try everything out!
One thing that helped me transition from Emacs to other IDEs was the idea that IDEs are terrible editors. I scoffed at that person but I now see their point.
An editor, like Emacs or Vim, can really focus on being a good editor first and foremost.
An IDE, like Visual Studio or Eclipse, really focuses on being a good project management tool with a built in way to modify files.
I find that keeping the above in mind (and keeping Emacs handy) helps me to not get frustrated when the IDE du jour is not meeting my needs.
If you've been using emacs/vi for years (although you listed both, so it seems like you may not be adapted fully to one of them), using said editor will probably be faster for you than an IDE. The level of mind-meld a competant emacs/vi user can achieve with a customized setup and years of muscle memory is astounding.
Some free ones:
XCode on the Mac
Eclipse
Lazarus (Open Source clone of Delphi)
Visual Studio Express
Editions
Try making a couple of test applications just to get your feet wet. At first, it will probably feel more cumbersome. The benefits of IDEs don't come until you begin having a good understanding of them and their various capabilities. Once you know where everything is and start to understand the key commands, life gets easier, MUCH easier.
I think you'll find IDE's invaluable once you get into them. The code complete and navigation features, integrated running/debugging, and all the other little benefits really add up.
Some suggestions for starting out and easing transition:
- start by going through a tutorial or demonstration included with the IDE documentation to get familar with where things are in the GUI.
- look at different kinds of sample projects (usually included with the IDE or as a separate download) for different types of areas you may be coding (web applications, desktop applications, etc) to see how they are laid out and structured in the IDE.
- once comfortable, create your own project from existing code that you know well, ideally not something overly complex, and get it all compiling/working.
- explore the power! Debug your code, use refactorings, etc. The right click menu is your friend until you learn the keyboard shortcuts just to see all the things you can do. Right click different areas of your code to see what is possible and learn (or re-map) the keyboard shortcuts.
Read the doc...
And see what shortcuts/keybindings equivalents are with your familiar ones. Learn the new ones...
Old question, but let me suggest that in some circumstances, something like Notepad++ might be appropriate for the OP's situation which may be encountered by others. Especially if you are looking for something lightweight, Notepad++ can be part of a developer's arsenal of tools. Eclipse, Visual Studio and others are resource hogs with all their automagic going on and if you are looking to whip out something pretty quick with a whole bunch of keyboard shortcuts and the like or if you are interested in viewing someone else's source, this can be quite useful. Oh yeah, and it is free too.