Show the derived/implemented traits directly on the type when the mouse is hovering over the type name.

[iddm]

For example, when my mouse cursor is above the std::net::TcpListener in my code, I get this in the "tooltip":

std::net::tcp
pub struct TcpListener // size = 4, align = 4
A TCP socket server, listening for connections.

After creating a TcpListener by [bind]ing it to a socket address, it listens for incoming TCP connections. These can be accepted by calling [accept] or by iterating over the [Incoming](vscode-file://vscode-app/opt/visual-studio-code/resources/app/out/vs/code/electron-sandbox/workbench/workbench.html) iterator returned by [incoming](vscode-file://vscode-app/opt/visual-studio-code/resources/app/out/vs/code/electron-sandbox/workbench/workbench.html).

The socket will be closed when the value is dropped.

The Transmission Control Protocol is specified in [IETF RFC 793](vscode-file://vscode-app/opt/visual-studio-code/resources/app/out/vs/code/electron-sandbox/workbench/workbench.html).

Examples
use std::net::{TcpListener, TcpStream};

fn handle_client(stream: TcpStream) {
    // ...
}

fn main() -> std::io::Result<()> {
    let listener = TcpListener::bind("127.0.0.1:80")?;

    // accept connections and process them serially
    for stream in listener.incoming() {
        handle_client(stream?);
    }
    Ok(())
}
[10 implementations](vscode-file://vscode-app/opt/visual-studio-code/resources/app/out/vs/code/electron-sandbox/workbench/workbench.html)

This is all fine and dandy; however, it would also be really useful if all the trait implementations were shown for this type, close to the full path to the type in the first line. The size and alignment are both great, but having the traits implemented listed would be a more useful thing, in my opinion. Especially it would be helpful since many crates provide trait implementations during the code generation stage, so it is really hard to find out if a certain trait is implemented for the type or not, as what you see in the code is some macro magic with many different conditions, where you might not know which one is truth and so, leads to some trait being implemented for this type.

[HKalbasi]

Isn't that 10 implementations button enough for you? Including them always would be a little spammy, and I think has negative effect on hover performance.

[iddm]

The 10 implementations link navigates to actual implementations in the code; this is insufficient for various reasons. The first thing - as I mentioned, it would be much handier to know the trait names immediately, rather than scrolling all the way down to this link, looking at all these 10 implementations sequentially, parsing the code with your eyes on the way, instead of just listing the trait names right near the type name as I suggested. Another problem with this is that sometimes such 10 implementations links lead to the code generation code, which is certainly unclear. However, the rust analyzer parser must already know what traits a specific type has implementations for and what not; it would be nice if we had just used this knowledge and showed the trait names implemented for the type instead of only providing this button and the places those are allegedly implemented. An example may be the Vulkan ash crate, where most of the types use compile-time macros to generate certain trait implementations for the type, and these implementations are impossible to know even with this button, as it leads to the macro usage code.

Regarding the performance, I suggest just caching a list of strings of trait names for each type somewhere. We don't need the lifetimes (if we need them, we will read the docs for the trait); we just need the trait names. Should occupy much space, will save precious CPU time, and the performance won't suffer).

[zzhaolei]

I like the way intellij-rust implements. This is more explicit and doesn't interfere too much.