Enhance i18n function to handle floats

[jmooring]

i18n/en.toml

[hour]
one = "{{ . }} hour"
other = "{{ . }} hours"

Test results with e4dc9a8:

ID	Statement	Actual	Expected	Status
1	T "hour" 1	1 hour	1 hour	Pass
2	T "hour" 1.5	1.5 hour	1.5 hours	Fail
3	T "hour" 2	2 hours	2 hours	Pass
4	T "hour" "1"	1 hour	1 hour	Pass
5	T "hour" "1.5"	1.5 hours	1.5 hours	Pass
6	T "hour" "2"	2 hours	2 hours	Pass

The existing float test passes because, in English, 22.5 and 22 both belong to the other cardinal category as defined by the CLDR plural rules.

I am unsure why test 5 passes while test 2 fails.

In Polish 100 and 100.0 should be handled differently.

i18n/pl.toml

[day]
one = "{{ . }} miesiąc"
few = "{{ . }} miesiące"
many = "{{ . }} miesięcy"
other = "{{ . }} miesiąca"

Test results with e4dc9a8:

ID	Statement	Actual	Expected	Status
7	T "day" 1	1 miesiąc	1 miesiąc	Pass
8	T "day" 2	2 miesiące	2 miesiące	Pass
9	T "day" 100	100 miesięcy	100 miesięcy	Pass
10	T "day" 100.0	100 miesięcy	100.0 miesiąca	Fail

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.