On Wed, 12 Sep 2001, Steve Langasek wrote:
> Rather than a single jumbo patch, perhaps it would help to first add autoconf
> support and a sample usage of getpwnam_r()? After that, adding proper
> reentrant support to all of the modules should be fairly parallelizable.
In fact, let's go ahead and get started on this.
I've just committed changes to the autoconf code which add HAVE_GETPWNAM_R and
HAVE_GETGRNAM_R defines if those functions are available. More can be added
for the other NSS functions as we get farther along.
Is the following ok for a first implementation, or do you have more ambitious
plans? I believe you mentioned providing a wrapper for systems that don't
have getpwnam_r(), but I'm personally quite content with this.
Regards,
Steve Langasek
postmodern programmer
struct passwd *pwd = NULL;
#if HAVE_GETPWNAM_R
char buf[512];
struct passwd pwd_buf;
pwd = &pwd_buf;
#endif
...
#if HAVE_GETPWNAM_R
getpwnam_r(name, pwd, buf, sizeof(buf), &pwd);
#else
pwd = getpwnam(name);
#endif
