Print a message when no lints found

[IndrajeetPatil]

Closes #2640

library(lintr)
lint(text = "x <- 1L", linters = all_linters())
#> ℹ No lints found.

^{Created on 2024-08-04 with reprex v2.1.1}

[IndrajeetPatil]

Windows failures seem spurious:

> checking top-level files ... NOTE
  Non-standard file/directory found at top level:
    'install.ps1'

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 97.96%. Comparing base (f0d9407) to head (63b0017).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2643   +/-   ##
=======================================
  Coverage   97.96%   97.96%           
=======================================
  Files         126      126           
  Lines        5760     5761    +1     
=======================================
+ Hits         5643     5644    +1     
  Misses        117      117           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[MichaelChirico]

this is a classed signal:

class(tryCatch(cli::cli_inform("xxx"), condition=identity))
# [1] "rlang_message" "message"       "condition"

Is that appropriate? I might have expected just cat() here, I guess {cli} adds nice aesthetics -- are there any analogues for "plain text to terminal"? cli::cli_text(c(i = "No lints found.")) drops the i formatting.

(or perhaps a message is appropriate, WDYT?)

[AshesITR]

Message seems ok to me.

[IndrajeetPatil]

Yes, we could use cat(), but I also think there are benefits to having this as a classed signal over cat() (e.g. the messages will show up in logs).

If we don't want a classed signal, either cli_alert() or its sibling should work:

cli::cli_alert_success("No lints found")
#> ✔ No lints found

class(tryCatch(cli::cli_alert_success("No lints found"), condition=identity))
#> [1] "cli_message" "condition"

^{Created on 2024-08-05 with reprex v2.1.1}

So we can have our pretty aesthetics cake and eat it too.

[AshesITR]

LGTM. The Windows failure looks like it needs reporting as a bug to r-lib/actions.

[IndrajeetPatil]

LGTM. The Windows failure looks like it needs reporting as a bug to r-lib/actions.

Done: r-lib/actions#898

[MichaelChirico]

Still waver a bit on whether signalling a condition is really the right thing to do for code that's performing fine, in fact congratulatory.

But, I don't think it actually matters -- we are a "front end" package & I can't think of any scenario where the difference between cat()/message() would actually matter for our users. So, let's proceed for simplicity's sake.