fix(abi): do not error on the nothing-to-repair path
A slot with verdicts recorded but nothing unsafe died on its own success
message:
Error in sprintf("%s/%s: %d objects recorded, none unsafe. Nothing to repair.", ...)
RPostgres returns count(*) as integer64, which %d rejects. The pipeline step
then failed for a slot that was in the best possible state, which is the wrong
signal entirely.
This commit is contained in:
parent
e3b7212eaf
commit
fcbc4e8e92
1 changed files with 4 additions and 2 deletions
|
|
@ -141,11 +141,13 @@ if (nrow(rows) == 0L) {
|
||||||
call. = FALSE
|
call. = FALSE
|
||||||
)
|
)
|
||||||
}
|
}
|
||||||
|
# `count(*)` arrives as a double (or integer64), and `%d` rejects both, so the
|
||||||
|
# nothing-to-do path used to die on its own success message.
|
||||||
message(sprintf(
|
message(sprintf(
|
||||||
"%s/%s: %d objects recorded, none unsafe. Nothing to repair.",
|
"%s/%s: %s objects recorded, none unsafe. Nothing to repair.",
|
||||||
arch,
|
arch,
|
||||||
codename,
|
codename,
|
||||||
total
|
format(as.numeric(total), scientific = FALSE)
|
||||||
))
|
))
|
||||||
quit(status = 0)
|
quit(status = 0)
|
||||||
}
|
}
|
||||||
|
|
|
||||||
Loading…
Reference in a new issue