Check all targets for package-level tasks #5563
Conversation
|
We check the whole package by design. There's usually little benefit in checking only some things (as check is fast), and it's convenient if everything is checked. Does blanket |
|
No, that's why I've created this PR. Simple to check: make a compiler error in the tests, run the "-p" check command, there will be no errors. |
|
Oh, sorry, not sure about examples & benches for sure, but tests errors are not checked by the current command. Currently, the only way to verify that the tests compile is to actually run them, there's no check runnable RA provides for them. |
|
Ah, so it seems like we should pass `--all-targets` then! That was the
original design, didn't realise that it doesn't work :D
…On Wed, 29 Jul 2020 at 10:45, Kirill Bulatov ***@***.***> wrote:
Oh, sorry, not sure about examples & benches for sure, but tests errors
are not checked by the current command.
Currently, the only way to verify that the tests compile is to actually
run them, there's no check runnable RA provides for them.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#5563 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AANB3MYOG7VSARSZZAC7MX3R57ORJANCNFSM4PLAP75A>
.
|
SomeoneToIgnore
force-pushed
the
check-tests
branch
from
dbb97a5
to
85d71f4
Compare
|
I've added |
SomeoneToIgnore
changed the title
|
bors r+ |
|
Could that be that the functionality was removed due to the duplicate output?
That is indeed more verbose than needed now. |
When invoking "Select Runnable" with the caret on a runnable with a specific target (test, bench, binary), append the corresponding argument for the
cargo check -pmodule runnable.