[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Nmh-workers] A € for your thoughts - should we fix UTF-8 subject ou

From: Tethys
Subject: Re: [Nmh-workers] A € for your thoughts - should we fix UTF-8 subject output in scan for 1.5?
Date: Mon, 21 May 2012 09:09:12 +0100

Ken Hornstein writes:

>Well, I guess it depends on your perspective.  Is this a portability
>problem?  Linux (the one I tested) returns 0 for isspace() &&
>iscntrl() > 127.  If that's true for you, it works fine.

That's not strictly true, either. Both isspace() and iscntrl() return
zero for me for everything above 127. But although the subject line of
your message is correctly decoded by scan, it's broken when decoded by
show (and by exmh, which I'm guessing is just shelling out to show).
This is on Linux with 1.4.

Should this be fixed? Absolutely. But I'm not seeing a massive benefit 
in rushing it through into 1.5 this late in the release process, though. 
It's been broken forever. Waiting until 1.6 won't hurt.

Incidentally, I'd say OS X is very buggy if it's interpreting bytes
above 127 as whitespace in a UTF-8 locale, given that no single byte
in that range is valid UTF-8. There's an argument for iscntrl(), but
even that's a bit tenuous. I'd probably just say that all of the 
isxxxx() functions should return false for those values.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]