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!
Related
Is it possible to write an extension which will do code analysis? I looked at the api and I didnt see how I can get all variables that are declared in the document/xref on specific variable and get it flow in the code. If someone know api in visual studio code which actually helps with it, it will be great.
Creating such an extension is certainly possible, however you have to do your own language processing. AFAIK there's no way to access the internal structures (syntax tree, symbol table etc.) of the existing JS/TS (or other language for that matter) extensions.
I did my best to overcome this problem but it was in vain. I am developing tests in Robot Framework (RF) in Visual Studio Code and I want it to either show "signatures" of keywords from standard libraries or to be able to go to their definitions. Go to definition and signature displaying when hovering with Ctrl key are behaving this way:
works for my keywords written in RF (hoorah!)
works for my keywords written in Python only if a keyword consists of one word (I think conversion between undescores and spaces is failing)
does not work for keywords from standard libraries even if the keyword is one-word (e.g. "Fail"), regardless whether the keyword comes from built-in libraries or other ones (e.g. SeleniumLibrary)
When failing, Robot Framework Intellisense Server gives me message "Keyword definition 'Blah Blah' not found from the workspace".
I am using plugins Robot Framework Intellisense FORK and/or Robot Framework Language Server. I tried to configure them carefully according their documentation (Details), but the best state I reached is described above.
Can you help me please? I do not need to use a specific plugin, I just need to have the signature or documentation (or implementation) of every keyword in my code to be one click far.
According to their documentation they support
Goto definition
For variables
For user keywords
I was able to partially solve this when I not only open the folder with source code, but create a workspace and add the folder containing standard libraries to it as well (Add Folder to Workspace). I do not know exactly which folder is the best to add, it seems ...\Python\Python37\Lib\site-packages\ to me.
I am totally new to creating extensions in VS Code, and all the official examples of extensions are written in Typescript/Javascript, which I have no experience with. Is it possible to create VS Code extensions in other languages, such as Python or C++?
If so, could anyone point me to any resources to get me started?
It is possible by creating a C++ module for Node.js, which can then be loaded like any other node module. Of course, some glue code written in JS or TS is necessary to register the extension and translate calls to/from vscode.
I've gone this way in my ANTLR4 extension, but gave up eventually, because of the troubles I had due to incompatible dependencies (you have to make sure the extension uses the exact same V8 version, which was used to build the underlying Node.js used by vscode, on all supported platforms).
This situation might have change, I don't know, but with that in the background I don't recommend it.
If you want to add support for a new language in vscode you can also write a separate language server, as is mentioned in the linked SO answer. For other type of work, I'm afraid, you have no alternative to use.
No, as #rioV8 said, since VSCode is an electron app and runs on Javascript.
Hi I am interested in understanding how eclipse autocomplete works. I want to understand how eclipse distinguishes between local and global variables in a piece of Java code. I would also like to understand how eclipse stores method signatures for an infinite number of classes and how it associates a method to a given class. And is it possible for one person to develop an autocomplete feature for a language like JavaScript.
There is already an AutoComplete feature for Javascript. You just need to let Eclipse install the appropriate extensions.
Eclipse maintains a model of your program, including the project and all the dependencies. It's big, but it's not infinite. When you hit the dot, it figures out based on the variable type what the target type can be, and then displays the relevant methods based on its internal model.
This is easy for Java because you can usually know the static type. Much harder in other languages.
The Eclipse plug-in developer's guide discusses how different things, including the internal model and auto completion works. There are extension points to implement yiur own.
Being really new to wx, I'm wondering if there is an IDE (especially for Linux) which would help me lay out a frame or dialog or whatever just to help me see what I'm doing. That means also creating the code for those changes.
I remember way back when using resource compilers for OS/2 and Windows that produced binaries that would then create the window, and was hoping for something similar (though obviously not binary if wx doesn't support that).
I use wxFormBuilder. It is written in wxWidgets, so it works on Linux quite well. It can generate C++ code or XRC files. Make sure you understand its philosophy, and use it like this:
generate C++ code for the GUI
don't edit the code wxFormBuilder generated, but create new files
in new files, derive new classes from the classes it generated
implement event handlers in you own class (wxFB creates virtual function for each event handler you wish to use)
I usually name the wxFormBuilder generated classes/files like, for example, MainFrameGUI, and one with implementation (derived one in which I write all my code) would be just MainFrame. This enables you to change the visual layout and regenerate C++ files from wxFB at any time without overwriting your code.
DialogBlocks works quite well for me, although sometimes you need to edit the code to fix errors manually. It has a property editor that seems advanced enough.
Just another options is wxGlade. It does not have the that much features as the others mentioned seem to have, but it works just good enough for me to not daring to switch.
I use Code::Blocks IDE from http://www.codeblocks.org which has
- built-in GUI editor
- Cross compilable, so you can use it under Linux, OSX and Windows.
But I still use wxFormBuilder with it instead of built-in wxSmith editor. But they are compatible with internal wxSmith.
For windows you've got "wx-devcpp" which is Blodsheed Dev C++ with some addons providing what you looking for
Here is project page
http://wxdsgn.sourceforge.net/