Re: [PATCH v12 04/10] utf8: teach same_encoding() alternative UTF encoding names

>> On 16 Mar 2018, at 00:25, Eric Sunshine <sunshine@sunshineco.com> wrote:
>>>        if (is_encoding_utf8(src) && is_encoding_utf8(dst))
>>>                return 1;
>>> +       if (same_utf_encoding(src, dst))
>>> +               return 1;
>>>        return !strcasecmp(src, dst);
>>> }
>>
>> This seems odd. I would have expected the newly-added generalized
>> conditional to replace the original UTF-8-specific conditional, not
>> supplement it. That is, shouldn't the entire function body be:
>>
>>    if (same_utf_encoding(src, dst))
>>        return 1;
>>    return !strcasecmp(src, dst);
>
> No, because is_encoding_utf8() returns "true" (=1) if the encoding
> is NULL. That is not the case for UTF-16 et al. The caller of
> same_encoding() might expect that behavior.
> I could have moved the "UTF-8" == NULL assumption into
> same_utf_encoding() but that did not feel right.
> Does this make sense?