Fix build after PR #454

[SkiFire13]

PR #454 broke the build by adding a doctest that doesn't compile, in particular it gave the following error:

failures:

---- src/error.rs - error::ErrorVariant<R>::message (line 445) stdout ----
error[E0282]: type annotations needed for `pest::error::ErrorVariant<R>`
 --> src/error.rs:447:15
  |
5 | let variant = ErrorVariant::CustomError {
  |     -------   ^^^^^^^^^^^^^^^^^^^^^^^^^ cannot infer type for type parameter `R` declared on the enum `ErrorVariant`
  |     |
  |     consider giving `variant` the explicit type `pest::error::ErrorVariant<R>`, where the type parameter `R` is specified

error: aborting due to previous error

For more information about this error, try `rustc --explain E0282`.
Couldn't compile the test.

failures:
    src/error.rs - error::ErrorVariant<R>::message (line 445)

test result: FAILED. 59 passed; 1 failed; 2 ignored; 0 measured; 0 filtered out

This error made PRs #470's and #473's bors build fail. After this PR is merged bors should be re-run for them.

[CAD97]

I always forget whether I have permission to

bors: r+

[SkiFire13]

At least now it builds on stable, but it still fails on older versions.

The main problem is the bump of the sha-1 dependency's version from 0.8 to 0.9 in pest-meta made with PR #465, which even through it claims to support rust 1.21 has a transitive dependency on generic-array v.0.14 which requires a minimum rust version of 1.36 (which failed the "minimal versions" build) and a dependency on cpuid-bool v0.1.2 which uses the macro Kleene operator, a feature stabilized in rust 1.32, and fixed-size atomics, stabilized in 1.34 (which instead triggered on the "style" build). The options here are to either revert #465 or to increase the versions used by CI to at least 1.36 but I feel like this is a decision the maintainers should make.

Looks like pest/src/error.rs also needed a cargo fmt, that will be easy to fix.

As for why "coverage" paniced, that's a complete mistery to me.

[bors]

Canceled.

[CAD97]

This clearly improves the state of master, so I'm going to go ahead and merge it as-is currently. I think coverage has been broken for a while, but it's allowed to break because of weird issues like this (that were more prevalent when we initially set it up). I think I'm also going to set MSRV/"minimal versions" to allowed fail for the time being, so we can get master green again.

Unfortunately, I don't have time to spend actually working on projects right now, just time to review + merge what other people do.

[SkiFire13]

Unfortunately, I don't have time to spend actually working on projects right now, just time to review + merge what other people do.

If that's about PR #470 and #473, they were already reviewed by @dragostis, it's just the build that failed for this unrelated problem.

[CAD97]

Yeah, I've got time to review things and will re-r those once the build is fixed, it's just new development (such as actually making CI more reliable) which I don't have time for at the moment.