Update book

[steveklabnik]

This PR does two things:

1. update the book to include Improve redirects book#1088
2. update to mdbook 0.1

Both of these things are big changes, so I want to land them now, well before the next branch, so we can kick the tires.

---

Locally, I'm seeing some weirdness around the reference and this:

Putting this PR up so others can try and build and see if it reproduces for them.

[rust-highfive]

r? @alexcrichton

(rust_highfive has picked a reviewer for you, use r? to override)

[alexcrichton]

in terms of r+, r=me whenever this is good to go! I imagine though @rust-lang/docs may wish to help out with the investigation

[QuietMisdreavus]

I can confirm the reference rendering strangely. There's an open PR on the reference that might help with this; should we merge that first and update it here?

[Michael-F-Bryan]

You'll probably want to make sure each document's book.toml is updated too. We changed the configuration file format around a bit, but there's a legacy shim which will try to detect when you're using the old format and use that, emitting a warning with instructions on how to update.

Let me know if there's anything we can do from mdbook's end!

[steveklabnik]

This will fail until rust-lang/mdBook#585 is in an mdbook release.

[steveklabnik]

this should be good to go now 🎊

[alexcrichton]

Looks like Travis may be failing?

[steveklabnik]

Hm. very odd.

[00:42:55] An error occured:
[00:42:55] Rendering failed

not super helpful...

[Michael-F-Bryan]

not super helpful...

Lol. We're using error-chain and chain_err() to give an error more context. If you print out the causes then it'll tell you what went wrong, we currently have a helper in mdbook for exactly this:

/// Prints a "backtrace" of some `Error`.
pub fn log_backtrace(e: &Error) {
    error!("Error: {}", e);

    for cause in e.iter().skip(1) {
        error!("\tCaused By: {}", cause);
    }
}

[steveklabnik]

@Michael-F-Bryan yeah, sorry, I typed that in haste; this behavior is totally reasonable!

I've pushed a commit that uses this, let's see what it says (the build succeeds for me locally)

[steveklabnik]

i was able to reproduce locally, and this is what I have:

Error: Rendering failed
        Caused By: Unable to copy across additional CSS and JS
        Caused By: Unable to copy theme/reference.css to C:\Users\steve\src\rs\build\x86_64-pc-windows-msvc\doc\reference\theme/reference.css
        Caused By: The system cannot find the path specified. (os error 3)

this make sense as to why it's only the reference that fails.

note that C:\Users\steve\src\rs\build\x86_64-pc-windows-msvc\doc\reference\theme/ does exist, this makes me think it's theme/reference.css that doesn't. but that does too: src\doc\reference\theme\reference.css is there.

i wonder if this is an mdbook bug, not respecting the path when passed in -d? that is, the error is just printing theme/reference.css, without the prefix and that makes me suspicious.

[steveklabnik]

Yes, this is the answer, I believe. The reference has theme/reference.css in its config, which is then un-modified by mdbook, leading to it not respecting -d. working on an mdbook patch now

[steveklabnik]

rust-lang/mdBook#598

[steveklabnik]

confirmed the fix locally, but pushed a commit up here to check on the builders too

we'll need an mdbook release in order to ship this

[steveklabnik]

well, uh looks like that's another bug: https://travis-ci.org/rust-lang/rust/builds/334450214#L2554

[steveklabnik]

okay! I'm going to wait until the release ships, and then rebase this to be more reasonable.

[steveklabnik]

Why is it invalid? In that we don't accept MPL based code? I was pretty sure we had some already... and it would be kinda weird to reject Mozilla's license, heh.

[kennytm]

@steveklabnik it would be invalid if these packages were used by the standard library since it should be strictly MIT/Apache2. As these are only used for generating the book, you can add exceptions to tidy by editing https://github.com/rust-lang/rust/blob/master/src/tools/tidy/src/deps.rs

[steveklabnik]

new commit pushed, that look correct? :)

[kennytm]

@bors r=alexcrichton

[bors]

📌 Commit abb162c has been approved by alexcrichton

[bors]

☔ The latest upstream changes (presumably #47991) made this pull request unmergeable. Please resolve the merge conflicts.

[steveklabnik]

@bors r=alexcrichton

[bors]

📌 Commit 983cc00 has been approved by alexcrichton