httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dean Gaudet <>
Subject Re: The new %pp/pA/pI formats
Date Mon, 20 Apr 1998 09:47:42 GMT
Try %A and %I and then watch gcc -Wall complain.  That's why I posted the
patch for review a few weeks back.  I explained the logic of the %pA, %pI,
%pp choices then 'cause Martin wondered about it too.

I don't really care if the patch goes in or not.  I actually think that
gcc's warnings are a little too non-configurable.  I looked at the gcc
code and it shouldn't be TOO hard to make it extensible... but I just
don't have enough hours in the day :(

Note that alternate choices for printing inet addresses are, uh, lacking. 
For example, inet_ntoa() is not threadsafe.  I suppose something could be
pstrdup()d out of a pool... but that chews memory.  What you really want
is something along the lines of C++ streams which are easily extended with
type checking. 


On Sun, 19 Apr 1998, Jim Jagielski wrote:

> Is that safe? I'm guessing that there might be modules that have been
> assuming that the %p stuff works just like ANSI... Now they need to
> worry about %pp... The thing is that, until recently, we just used
> snprintf() (and our version by default) so modules may run into
> troubles... Also, don't modules like mod_perl and PHP simply use
> our snprintf()?
> Couldn't we just have %A and %I ?
> I'm not looking to create any arguments, but just whether this change
> really makes sense...
> -- 
> ===========================================================================
>    Jim Jagielski   |||   |||
>             "That's no ordinary rabbit... that's the most foul,
>             cruel and bad-tempered rodent you ever laid eyes on"

View raw message