Re: 23.0.60; Assuming errstring unibyte is incorrect and leads to Emacs crashes

[Eli Zaretskii]

> From: Kenichi Handa <address@hidden>
> CC: address@hidden, address@hidden, address@hidden
> Date: Thu, 28 Aug 2008 09:40:01 +0900
> 
> > Can we explain in a comment what is special about a slash as the
> > second character of the error message?
> 
> We should, but I don't know the reason.  By grepping
> ChangeLog files, I found this item:
> 
> 1991-02-21  Richard Stallman  (address@hidden)
> 
>       * fileio.c (report_file_error): Don't downcase "I/O".
> 
> Is this the reason?

Yes, that's what I remembered as well.  But if the reason is "I/O",
why not test for that explicitly?  Do certain locales translate "I/O"
into a different string, which still has a slash as its second
character?

(I'm having a strange feeling that we already discussed this at some
point in the past.  Perhaps searching the archives will bring some
useful hits.)