`timeit` hardcodes plural "secs"

[StanFromIreland]

Bug report

Bug description:

cpython/Lib/timeit.py

Lines 342 to 347 in 8687b9d

	s = "" if number == 1 else "s"
	print(
	f"{number} loop{s} "
	f"{theme.punctuation}-> "
	f"{theme.timing}{time_taken:.{precision}g} secs{reset}"
	)

This should instead account for the (unlikely) situation that time_taken == 1 sec, like we do for loops.

CPython versions tested on:

CPython main branch

Operating systems tested on:

No response

Linked PRs

• gh-148254: Use singular "sec" in timeit verbose output #148290
• [3.14] gh-148254: Use singular "sec" in timeit verbose output (GH-148290) #148303
• [3.13] gh-148254: Use singular "sec" in timeit verbose output (GH-148290) #148304

[hugovk]

We can always say "sec", we use singular for the other units:

❯ ./python.exe -m timeit --verbose "import time, random; time.sleep(random.random() * 5)"
1 loop -> 3.01 secs

raw times: 110 msec, 4.57 sec, 2.07 sec, 1.07 sec, 4.21 sec

1 loop, best of 5: 110 msec per loop

:0: UserWarning: The test results are likely unreliable. The worst time (4.57 sec) was more than four times slower than the best time (110 msec).