support assertions about String, PathBuf, etc

[sourcefrog]

assert_that(&stdout).starts_with(&"Created new archive");

doesn't seem to work if stdout is a String: I need to explicitly call .as_str()

Am I doing this wrong? Or if not, would you be open to a PR that adds a specialization for String, and similarly for PathBuf?

[cfrancia]

Adding in String and PathBuf specialisations sounds like a great idea. If you do a PR for it, I'd be more than happy to review and merge in.

[cfrancia]

@sourcefrog -- are you still planning on looking into this? If not, I'm happy to make the required changes.

[sourcefrog]

I probably won't get to it right away, so go ahead

…

On Tue, Nov 29, 2016, 11:41 PM cfrancia ***@***.***> wrote: @sourcefrog <https://github.com/sourcefrog> -- are you still planning on looking into this? If not, I'm happy to make the required changes. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#5 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAVI8x0z-IFUHrj8WSyVJsBJt3uNmFeeks5rDSi6gaJpZM4KqvE5> .

[sourcefrog]

Could you also handle String?

[cfrancia]

@sourcefrog -- I thought I'd resolved that one as well via: f203392. That wasn't what you were expecting with this issue?

[sourcefrog]

You did, I just missed it.

I was hoping this would work too but maybe it's a different problem

    assert_that(&stderr).is_equal_to("");

error[E0277]: the trait bound `&str: std::borrow::Borrow<std::string::String>` is not satisfied
  --> tests/blackbox.rs:48:26
   |
48 |     assert_that(&stderr).is_equal_to("");
   |                          ^^^^^^^^^^^ the trait `std::borrow::Borrow<std::string::String>` is not implemented for `&str`

[cfrancia]

Ah okay, I see it. Yeah, is_equal_to(..) expects to receive the same type as the subject (so String in your case).

All of the functions under the StrAssertions trait expect &str (although I use the Borrow trait to simplify things), which is why contains(..) etc work as expected.

Would it be useful to have an is_empty() (or similarly named) assertion against String/str for your particular case, or you'd be wanting to match on an exact string?

[sourcefrog]

On Sun, Dec 11, 2016 at 11:39 PM, cfrancia ***@***.***> wrote: Would it be useful to have an is_empty() (or similarly named) assertion against String/str for your particular case, or you'd be wanting to match on an exact string?

I think both would be useful. A specific `is_empty` would be parallel to the method on `String` itself and other Spectral matchers. And, comparisons of `Strings` to literal strings seem like a common thing to do.

[cfrancia]

An is_empty assertion will be pretty simple to add in, so I'll do that soon.

Equality between different types in theory is simple (PartialEq should handle those cases), except that I may have complicated things by my usage of the borrow trait, which seems to restrict the is_equal_to method to exact type matches.

I'll need to think more about a solution to that problem, otherwise I could end up solving one problem and creating another (or unnecessarily breaking compatibility).