Fix typo in misc tests

[joelim-work]

If the unit test fails, it prints out the input twice instead of the expected value.

---

BTW I was having a look at the runeSliceWidthRange function because @raslop was working on this in #1029. Unfortunately the current implementation in the master branch does appear to be slightly buggy if you consider the following test cases:

{[]rune{'世', '界', '世', '界'}, 3, 3, []rune{}},
{[]rune{'世', '界', '世', '界'}, 4, 7, []rune{'世'}},
{[]rune{'世', '界', '世', '界'}, 4, 8, []rune{'世', '界'}},

In particular the current implementation panics when the beginning and end indexes are the same and happen to be in the middle of a rune, and also the case where the end occurs after the start of the last rune isn't handled properly. The following is a more robust implementation that does handle the above cases in addition to the existing cases:

func runeSliceWidthRange(rs []rune, beg, end int) []rune {
        if beg == end {
                return []rune{}
        }

        curr := 0
        b := 0
        foundb := false
        for i, r := range rs {
                w := runewidth.RuneWidth(r)
                if curr >= beg && !foundb {
                        b = i
                        foundb = true
                }
                if curr == end || curr+w > end {
                        return rs[b:i]
                }
                curr += w
        }

        return rs[b:]
}

That being said, runeSliceWidthRange is currently called in only a few places when truncating long filenames, and for these cases the filename input is sufficiently long so these edge cases don't matter, so I can understand if you don't want to make any changes to the existing code.

Anyway the suggestion about runeSliceWidthRange is a separate thing, and can be addressed later if at all. At the very least the change for correcting the test case error reporting should be merged.

[raslop]

The typo-fix commit here is exactly the same that I stumbled upon earlier today and also put into #1029.
The suggestion here is a better version of my try to fix runeSliceWidthRange() in #1029 (which works fine there btw.). I would like to see this suggestion merged too. Once this is done, I will then try to get #1029 (with only the feature commit left) merged.

[ilyagr]

The fix looks good, thanks!

[joelim-work]

The suggestion here is a better version of my try to fix runeSliceWidthRange() in #1029 (which works fine there btw.). I would like to see this suggestion merged too.

I have submitted this change, see #1273