bug#70988: (read FUNCTION) uses Latin-1 [PATCH]

[Mattias Engdegård]

17 maj 2024 kl. 12.45 skrev Eli Zaretskii <eliz@gnu.org>:

>>>>> Is it an accident that the code does the same only _after_ the call to
>>>>> readbyte?
>>>> 
>>>> Yes, I have no reason to believe otherwise.
>>> 
>>> To me, it actually looks as done on purpose.
>> 
>> You could very well be right about that. What I meant is that the order 
>> doesn't matter at all.
> 
> Doesn't it affect what the readbyte call does?

No -- the `*multibyte = ...` assignment is just an extra return value, which 
indicates whether the returned values come from a unibyte or multibyte source. 
For any given source (READCHARFUN, in the terminology of lread.c), the 
characters will all be unibyte or multibyte, so this returned `multibyte` flag 
will typically only be used once by the caller and saved for future reference.

But you are right to question it because lread.c is a royal mess and many 
changes have not been made in a clean way. It is unclear whether it's worth 
returning the `multibyte` flag at all; it's only used in special cases.