New UDP tracker client for testing purposes

[josecelano]

You can use it with:

cargo run --bin udp_tracker_client 144.126.245.19:6969 9c38422213e30bff212b30c360d26f9a02136422

and the output should be something like:

AnnounceIpv4(
    AnnounceResponse {
        transaction_id: TransactionId(
            -888840697,
        ),
        announce_interval: AnnounceInterval(
            300,
        ),
        leechers: NumberOfPeers(
            0,
        ),
        seeders: NumberOfPeers(
            4,
        ),
        peers: [
            ResponsePeer {
                ip_address: xx.yy.zz.254,
                port: Port(
                    51516,
                ),
            },
            ResponsePeer {
                ip_address: xx.yy.zz.20,
                port: Port(
                    59448,
                ),
            },
            ResponsePeer {
                ip_address: xx.yy.zz.224,
                port: Port(
                    58587,
                ),
            },
        ],
    },
)

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (62fddba) 77.56% compared to head (f0710d3) 77.59%.

Files	Patch %	Lines
src/shared/bit_torrent/tracker/udp/client.rs	93.33%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #627      +/-   ##
===========================================
+ Coverage    77.56%   77.59%   +0.02%     
===========================================
  Files          125      125              
  Lines         8337     8349      +12     
===========================================
+ Hits          6467     6478      +11     
- Misses        1870     1871       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[josecelano]

This binary prints the Rust struct instead of a JSON object like in the http_tracker_client.

This is because Aquatic does not implement serialize for responses (only bencode format).

I will keep it this way for the time being until I start working on the E2E tests. I plan to use this binary for E2E testing but maybe I can only use the exit code. It depends on the asserts. I can modify it later if the json format fits better or just to keep both clients similar.

It would also be nice to use the domain instead of the IP.

[josecelano]

ACK afc4731

[josecelano]

I don't know why the code coverage report is failing. It's working on my local machine. I have to heck if I'm using the same Rust version.

[josecelano]

I don't know why the code coverage report is failing. It's working on my local machine. I have to heck if I'm using the same Rust version.

By the way @WarmBeer I do not know why we are generating the code coverage report in two workflows. We have a specific workflow for code coverage. SHould we remove that step from the Testing workflow?

[josecelano]

I don't know why the code coverage report is failing. It's working on my local machine. I have to heck if I'm using the same Rust version.

By the way @WarmBeer I do not know why we are generating the code coverage report in two workflows. We have a specific workflow for code coverage. SHould we remove that step from the Testing workflow?

Maybe it was because we wanted to see the report on the workflow execution log. But we can move the step to the coverage wirkflow.

[mickvandijke]

I don't know why the code coverage report is failing. It's working on my local machine. I have to heck if I'm using the same Rust version.

By the way @WarmBeer I do not know why we are generating the code coverage report in two workflows. We have a specific workflow for code coverage. SHould we remove that step from the Testing workflow?

I did not setup the current workflow, so I do not know the need for two "code coverage" steps.

But it is weird that the first one passed while the second one in the Testing workflow didn't 😅.

[josecelano]

I don't know why the code coverage report is failing. It's working on my local machine. I have to heck if I'm using the same Rust version.

By the way @WarmBeer I do not know why we are generating the code coverage report in two workflows. We have a specific workflow for code coverage. SHould we remove that step from the Testing workflow?

I did not setup the current workflow, so I do not know the need for two "code coverage" steps.

But it is weird that the first one passed while the second one in the Testing workflow didn't 😅.

OK, I've opened a new issue: #628

By the way @WarmBeer, I intended to mention @da2ce7 who was the last one refactoring the workflows :-)

[josecelano]

ACK f0710d3