Create all_equal_linter() #2885
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #2885 +/- ##
=======================================
Coverage 99.28% 99.28%
=======================================
Files 128 129 +1
Lines 7232 7270 +38
=======================================
+ Hits 7180 7218 +38
Misses 52 52 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Bisaloo
force-pushed
the
all_equal_linter
branch
from
a7111d8 to
a5e900b
Compare
| #' preceded by the negation operator `!`, are thus likely to generate unexpected | ||
| #' errors if the compared objects have differences. | ||
| #' An alternative is to use `identical()` to compare vector of strings or when | ||
| #' exact equality is expected. |
There was a problem hiding this comment.
BTW stopifnot() has some built-in logic for handling all.equal() checks, c.f.
stopifnot(all.equal(1, 1.001))
# Error: 1 and 1.001 are not equal:
# Mean relative difference: 0.001
I might actually lint this code while we're at it:
stopifnot(isTRUE(all.equal(x, y)))Since it gives a worse error:
stopifnot(isTRUE(all.equal(1, 1.001)))
# Error: isTRUE(all.equal(1, 1.001)) is not TRUESome hits:
We can do that as a follow-up if you'd like; in this PR at least I'd want to have some mention of stopifnot().
There was a problem hiding this comment.
We might as well test isFALSE(all.equal()) here too, it gets some usage:
https://github.com/search?q=lang%3AR+%2FisFALSE%5C%28all%5C.equal%5C%28%2F&type=code
There was a problem hiding this comment.
Yes, isFALSE(all.equal()) is definitely wrong since it'll always return FALSE. I wonder if that'd fit better under a istrue_linter() that check cases where authors wanted !isTRUE() rather than isFALSE().
There was a problem hiding this comment.
Similarly for stopifnot(), great catch but I wonder if it should live here or in stopifnot_all_linter() 🤔
There was a problem hiding this comment.
isFALSE(all.equal(...)) is pretty rare, I guess it doesn't warrant its own linter. Let's put it here.
I also think the stopifnot() bit belongs here, since it's about incorrect usage of all.equal(), whereas stopifnot_all_linter() is about incorrect usage of stopifnot(). WDYT @AshesITR / @IndrajeetPatil ?
There was a problem hiding this comment.
I view stopifnot_all_linter() as "stopifnot(check) has actually more bells and whistles than if (check) stop()". You can pass it all.equal() without isTRUE() the same way that you can pass it a logical vector without all(),
So it seems like a good fit there.
There was a problem hiding this comment.
One other thing to consider is that conjunct_test_linter() already has some logic related to stopifnot():
lintr/R/conjunct_test_linter.R
Line 110 in dc8f149
If we put the new all.equal() lint inside stopifnot_all_linter(), I'd want to rename it to just stopifnot_linter(), which would also create a tension with conjunct_test_linter() about where the check on stopifnot(A && B) should live. Fun :)
Not something that needs to be addressed in this PR: #2905
| @@ -51,6 +51,7 @@ | |||
|
|
|||
| ### New linters | |||
|
|
|||
| * `all_equal_linter()` warns about incorrect use of `all.equal()` in `if` clauses or preceded by `!` (#2885, @Bisaloo). | |||
There was a problem hiding this comment.
revisit the wording here
| Linter(linter_level = "expression", function(source_expression) { | ||
| all_equal_calls <- source_expression$xml_find_function_calls("all.equal") | ||
|
|
||
| dangerous_unwrapped_all_equal <- xml_find_all( |
There was a problem hiding this comment.
the logic is slightly off -- we get lints here in r-devel:
https://github.com/r-devel/r-svn/blob/6d2efbc65c5ae1c15ce4f1e25e2fcf5ddc3752d2/src/library/base/R/all.equal.R#L225-L226
https://github.com/r-devel/r-svn/blob/6d2efbc65c5ae1c15ce4f1e25e2fcf5ddc3752d2/src/library/base/R/eigen.R#L38-L41
I guess we're not being careful about which expr under IF is being linted:
if (A) all.equal(x, y)
|
Found some true positives :) |
Fix #2610