Ingorable Rust snippets

[dragostis]

Is there a way to add ignorable Rust snippets like in rustdoc?

[azerupi]

There is a PR open #404 to add a no_run attribute on code blocks. Does that cover your use case?

[dragostis]

no_run probably tries to compile the code, so it probably won't work. I would only need a code snippet that highlights Rust code.

[budziq]

Hi @dragostis

The code snippets are not compiled unless you:

1. use the play button - then it is compiled remotely by the https://play.rust-lang.org/
2. call mdbook test - then it is compiled by rustdoc
3. use some third party test runner like rust-skeptic which integrates with cargo test

In all cases all you need is one of two annotations for code blocks:

• no_run - code gets compiled but tests are not run
• ignore - code does not get compiled at all

in case 1. you'll need PR #404 to be merged for the play button to respect no_run and ignore but mdbook test and rust-skeptic already understand these annotations. So I guess that your use case is already covered.

[dragostis]

Oh, I see. That makes complete sense. However, is there a way to add crate dependencies?

[budziq]

It depends on which of the three cases are you thinking.

1. play button - nothing. You'll get only the crates that are installed on https://play.rust-lang.org/. PR Dynamic play button based on no_run property and used crates availability  #404 disables play button if the used crates are not available there.
2. mdbook test - you'll have to use -L /path/to/compiled/deps introduced since https://github.com/azerupi/mdBook/pull/340/ which requires that your project somehow depends on the extern crates (ie. these are listed in Cargo.toml like in case of 3.)
3. third party runners - well its depends on which one you use. For instance rust-skeptic requires that you add dependent crates to the Cargo.toml

I hope that solves your problem.

[dragostis]

Thank you for the detailed answer. That makes sense. I guess this should probably be covered in book.

[budziq]

That makes sense. I guess this should probably be covered in book.

@dragostis Indeed it should! We are planning to release a new version (which is mostly blocked on minor regressions to be fixed and documentation update). Unfortunately the progress with docs update is slow as we do not have as much time on our hands as we would like ;) If you are willing to contribute some time to provide a PR that would update the docs somewhat It would be more than welcome :)