[Top][All Lists]

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

Re: Use of @encode(char) not portable.

From: Richard Frith-Macdonald
Subject: Re: Use of @encode(char) not portable.
Date: Tue, 8 May 2001 09:02:47 +0100

On Monday, May 7, 2001, at 04:29 AM, Adam Fedor wrote:

I've encountered a bug in the coding functions of the base library
(perhaps others). @encode(char) is used in several places, but this is
not portable. @encode(char) returns an unsigned char (_C_UCHR) on some
machines and signed (_C_CHR) on others. It's easy to change all
occurances to @encode(unsigned char), but I'm not sure how many people
this will mess up. I'm not sure if it's worth the trouble though to
create a new encoding version in every class it's needed.

The other choice is to make the decoder a tad more forgiving...

I think perhaps we should replace all occurrences of @encode(char) with
@encode(signed char) ... since all the other types are consistent in that
omitting the signed/unsigned qualifier implies signed.

Making the decoders more tolerant also sounds like a good idea ... perhaps
we should do both?

reply via email to

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