fix(examples): fix compilation errors in examples

[bioinformatist]

Hi Greg,

The examples hacknews and hacknews_axum, who can be used within both csr and ssr mode, raise error during trunk serve with latest cargo:

2025-03-25T09:12:13.675539Z  INFO 🚀 Starting trunk 0.21.9
2025-03-25T09:12:14.274171Z  INFO 📦 starting build
warning: output filename collision.
The bin target `hackernews` in package `hackernews v0.1.0 (/home/ysun/leptos/examples/hackernews)` has the same output filename as the lib target `hackernews` in package `hackernews v0.1.0 (/home/ysun/leptos/examples/hackernews)`.
Colliding filename is: /home/ysun/leptos/examples/hackernews/target/wasm32-unknown-unknown/debug/hackernews.wasm
The targets should have unique names.
Consider changing their names to be unique or compiling them separately.
This may become a hard error in the future; see <https://github.com/rust-lang/cargo/issues/6313>.
warning: output filename collision.
The bin target `hackernews` in package `hackernews v0.1.0 (/home/ysun/leptos/examples/hackernews)` has the same output filename as the lib target `hackernews` in package `hackernews v0.1.0 (/home/ysun/leptos/examples/hackernews)`.
Colliding filename is: /home/ysun/leptos/examples/hackernews/target/wasm32-unknown-unknown/debug/hackernews.wasm.dwp
The targets should have unique names.
Consider changing their names to be unique or compiling them separately.
This may become a hard error in the future; see <https://github.com/rust-lang/cargo/issues/6313>.

(Compilation details omitted...)

Caused by:
    0: HTML build pipeline failed (1 errors), showing first
    1: error from asset pipeline
    2: running cargo build
    3: found more than one target artifact: ["hackernews", "hackernews"]:
        * consider adding `<link data-trunk rel="rust" data-bin={bin} />` to the index.html to build only the specified binary
        * or adding `<link data-trunk rel="rust" data-target-name={artifact} />` to select the specific artifact by name

My output of cargo --version:

cargo 1.85.1 (d73d2caf9 2024-12-31)

While the error also occurs with 1.81.0-nightly(a2b58c3da 2024-07-16).

Though it's raised by trunk, people believe it's a cargo problem: trunk-rs/trunk#740.

It seems cargo used to produce duplicate artifacts but the checking rules may be changed at some time: rust-lang/cargo#6313 (comment)

To fix it, I explicitly defined the binary artifact name and specified the name in index.html. I'm not sure if it's the best way though 😅

Also, the simple_logger was removed from the example counter_isomorphic, which indeed was not needed before, but somehow cause another compilation error:

[gbj]

Just double-checking: You say it's an error, but when I try it (on 1.87.0-nightly 2025-03-05), it's a warning, and the log output you post has it as a warning, right? At present it works completely as intended, but tells you it might break in the future?

To be honest I'd be surprised if the change to convert it from a warning to a hard error ever landed.

[bioinformatist]

Sorry it's my fault, I haven't clarified what happened.

My cargo version is 1.85.1 (d73d2caf9 2024-12-31) and the trunk version is 0.21.9.

The warnings are thrown by cargo:

While the error is raised by trunk:

All these abnormalities occurs within csr mode by trunk serve.

We have main() function in the source code, so there will be a binary named as hackernews after compiling. Meanwhile, our ssr mode need cdylib as one of the crate types, which may produce another hackernews artifact. I know it's not binary but it has an ambiguous name. Actually, the csr mode only need the binary one, so I renamed it with hackernews_bin and specify this name in index.html for trunk. That's it.

Sorry for wasting your time before.

[gbj]

Oh, weird. trunk does not have an error; if I go to hackernews and try trunk serve --open it works just fine.

I guess if trunk is actually erroring for some users then it makes sense to make the change.

Could you revert the change adding console_log/log to the Axum one however, and just remove that initialization line in main. log is not actually used anywhere in the example.

[bioinformatist]

Done 😄

[bioinformatist]

Currently the PR fails to pass the CIs. I can reproduce it locally by cargo leptos build:

It seems like a cargo-leptos problem, which is similar to: leptos-rs/cargo-leptos#358

Maybe we can wait leptos-rs/cargo-leptos#422 to solve this problem? If there's anything I can do for it, plz feel easy to let me know!

[benwis]

This seems to be an issue with rust's latest nightly, which as mentioned in the mentioned commits. There is an artifact collision warning but that shouldn't prevent compilation. I'll close this as a separate issue, but if you feel the warning is an issue, let's track that elsewhere