Distinction between Assert::set_stdin() and with_stdin is confusing

[zetok]

No idea if it doesn't actually work, or if I'm doing something wrong and API is simply allowing for it, but either way it should be fixed.

assert_cmd = "0.6"

my test case:

#[test]
fn broek_dep() {
    Command::new("cat")
        .assert()
        .set_stdin(vec![0])
        .stdout("\0")
        .success();
}

Which results in:

thread 'broek_dep' panicked at 'Unexpected stdout
command=`"cat"`
stdin=``````
code=0
stdout=``````
stderr=``````
',

---

Also on a side note, output of assert_cmd when there's an error is rather confusing and not helpful when you're trying to deal with input/output that consists of raw, often unprintable, bytes. Debug output without additional ``` would work a lot better IMO.

[epage]

No idea if it doesn't actually work, or if I'm doing something wrong and API is simply allowing for it, but either way it should be fixed.

#[test]
fn broek_dep() {
    Command::new("cat")
        .assert()
        .set_stdin(vec![0])
        .stdout("\0")
        .success();
}

Are you converting over from assert_cli? That wouldn't run the program until the expression was completed. assert_cmd runs at the time of assert or earlier. I've started an issue for documenting migration, see #30

So that set_stdin call you are making doesn't do what you think. Instead of writing it out (see instead with_stdin), the content from with_stdin is passed to set_stdin.

I can see how this isn't letting people fall into the pit of success. Any ideas for better names or other means to help people use the right function and avoid the one more so meant for internal purposes (but is pub for anyone writing custom logic)

[zetok]

Are you converting over from assert_cli?

No, but I've found out about assert_cli first from a blog post, and then I've looked at docs of both crates to see what's different.

Any ideas for better names or other means to help people use the right function and avoid the one more so meant for internal purposes (but is pub for anyone writing custom logic)

I think that having something like a bold Warning in the documentation of the function explaining that this might not be what they want and recommending using with_stdin instead would do the trick.

[epage]

Forgot to reply to:

Also on a side note, output of assert_cmd when there's an error is rather confusing and not helpful when you're trying to deal with input/output that consists of raw, often unprintable, bytes. Debug output without additional ``` would work a lot better IMO.

Please open a distinct issue for this

[epage]

I think that having something like a bold Warning in the documentation of the function explaining that this might not be what they want and recommending using with_stdin instead would do the trick.

Documentation just patches over problems and only helps so far as it is read (e.g. the current documentation already pretty clearly states how the parameter to set_stdin works).

I'm considering things like

• Changing the name (adding "context")
• Making this functions pub(crate) until there is a need and handle it then
• Try to make clearer the idea that the problem has already run by the time Assert exists.