apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Joe Orton <jor...@redhat.com>
Subject Re: Stas's proposed symbol renames
Date Thu, 09 Dec 2004 22:57:34 GMT
On Tue, Dec 07, 2004 at 07:39:08PM -0500, Stas Bekman wrote:
> Joe Orton wrote:
> >Well, since you really want feedback... ;) I often think this "naming
> >correctness" stuff goes too far, it just makes the API too verbose and
> >it can become unwieldy and ugly.  Do you genuinely get confused by the
> >existing names?
> 
> I do. But not confused. I find it hard to figure out what's a constant 
> means when reading code. Adding a non-ambiguous prefix helps resolve that 
> issue.
>
> Short constants work great when they are used in enum, which is used in 
> struct, so you always have a field name which is unambigous. When the 
> constant is hanging in the air alone, it's not easy to get what does it do.

But that's rare, it's so much more likely that e.g. the APR_FOPEN_*
constants sit inside or near an apr_file_open call, where they are
pretty unambiguous.  The only bad one is APR_READ vs APR_UREAD.  I'm not
trying to say it's black and white just that the trend worries me, if
the consensus is that these are a good thing then go for it.

> But that's not the point. I've posted the proposal back in May, it's been 
> dragging since then with 2 +1s and zero feedback in-between. Since there 
> is no general acceptance to this rename idea, I don't feel like it's the 
> right thing to do. But on the other hand none said no. So I'm lost. Now 
> I've committed one rename, and it's not clear if it's going to stay or 
> not. So if the majority says no, I just back it out and move on. If yes, 
> then I commit the other things and move on. But silence brings me nowhere.

I don't know, what's the consensus, who else cares about the naming?  If
I'm sat in my own little pedantic world on this then I don't want to get
in your way :)

joe

Mime
View raw message