bug#48711: Crashes in lisp_string_width

[Juri Linkov]

>> BTW, now compilation highlights this line
>>
>>   Pure-hashed: 15274 strings, 3028 vectors, 41734 conses, 2570 bytecodes, 
>> 254 others
>>
>> using red error face.  Does this mean that something needs to be fixed?
>
> You mean, this is a result of the string-width changes?

I don't know, the red error face was in the compilation output only 1 day,
but it's weird that now the last compilation highlights the line

  Pure-hashed: 15274 strings, 3028 vectors, 41734 conses, 2570 bytecodes, 254 
others

with the blue face-lock-function-name-face correctly again.