[Top][All Lists]

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

Re: Is it time to remove INTERNAL_FIELD?

From: Eli Zaretskii
Subject: Re: Is it time to remove INTERNAL_FIELD?
Date: Thu, 23 Apr 2015 16:53:38 +0300

> From: Oleh Krehel <address@hidden>
> Date: Thu, 23 Apr 2015 15:30:26 +0200
> Cc: address@hidden
> we should remove the macros that don't do anything.

What this macro does is allow you to use field names like 'foo', when
the field is really called 'foo_'.

I think it's okay to remove INTERNAL_FIELD, but I think we should keep
the trailing underscore appended in BVAR and KVAR.  That's how all
this started: the fields were renamed to have a trailing underscore so
that code that used foo->bar instead of BVAR (foo, bar) would be
immediately flagged by the compiler.

> As for accidental access, I'm sure these rare errors will be caught by
> the code review / test suite.

We don't want to rely on code reviews, since they are very informal
and their coverage is too low to be efficient.

Based on bitter past experience with similar errors that lay low for
months, sometimes for years, I'd rather not give up those underscores
in BVAR and KVAR, which means the struct fields should retain them.

reply via email to

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