Date: Fri, 10 Aug 2001 00:12:49 -0400 (EDT) From: Joe Clarke <marcus@marcuscom.com> To: Jason Quigley <jasonq@mac.com> Cc: Pete Fritchman <petef@databits.net>, <ports@FreeBSD.ORG> Subject: Re: FreeBSD Port: netatalk-1.5p6_1 Message-ID: <20010810001029.P32016-100000@shumai.marcuscom.com> In-Reply-To: <78C7DFBB-8D38-11D5-AE67-000A277D821C@>
next in thread | previous in thread | raw e-mail | index | archive | help
I have committed patches to the Netatalk CVS repo that fixes the -43
errors in OS X. I think that's what you're showing me here. When 1.5p7
comes out, they will be there. There is going to be a big change with
Netatalk soon. They're changing from BSD license to GPL. While I'm not a
big fan of this, they claim it will bring a bunch of new developers to the
project.
If 1.5p7 doesn't come out soon, I'll snap some patches together for the
FreeBSD port. Thanks for the interest, Jason!
Joe Clarke
On Fri, 10 Aug 2001, Jason Quigley wrote:
> Hi Pete & Marcus!
>
> Doh!!! Must remember to purge out the old ports directory!
>
> Anyway, that compiles nicely. I have also included some patches
> pertaining to OS X (still get the horrible -43 error :-( ) which you may
> wish to include in the package. I just split these out of the unified
> diff in CVS. Can the ports build system accept unified diffs?
>
>
>
> patch-aa:
>
> *** etc/afpd/directory.c.orig Wed Feb 28 10:53:24 2001
> --- etc/afpd/directory.c Wed Feb 28 10:53:24 2001
> @@ -943,13 +943,13 @@
> break;
>
> case DIRPBIT_UID :
> - aint = st->st_uid;
> + aint = htonl(st->st_uid);
> memcpy( data, &aint, sizeof( aint ));
> data += sizeof( aint );
> break;
>
> case DIRPBIT_GID :
> - aint = st->st_gid;
> + aint = htonl(st->st_gid);
> memcpy( data, &aint, sizeof( aint ));
> data += sizeof( aint );
> break;
> @@ -1188,7 +1188,7 @@
> memcpy( &aint, buf, sizeof(aint));
> buf += sizeof( aint );
> if ( (curdir->d_did == DIRDID_ROOT) &&
> - (setdeskowner( aint, -1 ) < 0)) {
> + (setdeskowner( ntohl(aint), -1 ) < 0)) {
> switch ( errno ) {
> case EPERM :
> case EACCES :
> @@ -1208,7 +1208,7 @@
> break;
> }
> }
> - if ( setdirowner( aint, -1, vol_noadouble(vol) ) < 0 ) {
> + if ( setdirowner( ntohl(aint), -1, vol_noadouble(vol) ) <
> 0 ) {
> switch ( errno ) {
> case EPERM :
> case EACCES :
> @@ -1225,12 +1225,11 @@
> }
> }
> break;
> -
> case DIRPBIT_GID :
> memcpy( &aint, buf, sizeof( aint ));
> buf += sizeof( aint );
> if (curdir->d_did == DIRDID_ROOT)
> - setdeskowner( -1, aint );
> + setdeskowner( -1, ntohl(aint) );
>
> #if 0 /* don't error if we can't set the desktop owner. */
> switch ( errno ) {
> @@ -1253,7 +1252,7 @@
> }
> #endif
>
> - if ( setdirowner( -1, aint, vol_noadouble(vol) ) < 0 ) {
> + if ( setdirowner( -1, ntohl(aint), vol_noadouble(vol) ) <
> 0 ) {
> switch ( errno ) {
> case EPERM :
> case EACCES :
>
>
>
>
> patch-ab:
>
> *** etc/afpd/messages.c.orig Sun Apr 15 21:14:11 2001
> --- etc/afpd/messages.c Sun Apr 1 11:11:38 2001
> @@ -115,9 +115,6 @@
> return AFPERR_BITMAP;
> }
>
> - if(strlen(message)==0)
> - return AFP_OK;
> -
> /* output format:
> * message type: 2 bytes
> * bitmap: 2 bytes
>
>
>
> patch-ac:
>
> *** etc/uams/uams_dhx_pam.c.orig Sun Apr 15 19:15:08 2001
> --- etc/uams/uams_dhx_pam.c Tue Feb 27 11:07:43 2001
> @@ -19,6 +19,7 @@
> #ifdef HAVE_UNISTD_H
> #include <unistd.h>
> #endif /* HAVE_UNISTD_H */
> +#include <errno.h>
>
> #include <security/pam_appl.h>
>
> @@ -78,6 +79,8 @@
>
> #define COPY_STRING(s) (s) ? strdup(s) : NULL
>
> + errno = 0;
> +
> if (num_msg < 1) {
> /* Log Entry */
> syslog(LOG_INFO, "uams_dhx_pam.c :PAM DHX Conversation
> Err -- %m");
> @@ -454,7 +457,7 @@
> memset(rbuf, 0, PASSWDLEN); /* zero out the password */
> *uam_pwd = dhxpwd;
> /* Log Entry */
> - syslog(LOG_INFO, "uams_dhx_pam.c :PAM: PAM Auth OK!: %s --
> %m", AFP_OK);
> + syslog(LOG_INFO, "uams_dhx_pam.c :PAM: PAM Auth OK!");
> /* Log Entry */
> return AFP_OK;
>
>
>
> Cheers,
> Jason.
>
>
> On Thursday, August 9, 2001, at 04:26 PM, Pete Fritchman wrote:
>
> > ++ 09/08/01 12:05 +0200 - Jason Quigley:
> > | When I try to make netatalk, I get the prompts/failures detailed
> > below.
> > | patch-aa seems like it wants to patch the Makefile, which doesn't yet
> > | exist as configure hasn't been run. What needs to be done to make the
> > | port work cleanly?
> > |
> > | Thanks,
> > | Jason.
> > |
> >
> > Well, you have the latest Makefile, but however you updated your ports
> > tree didn't remove files/patch-aa (and whatever else it might have
> > needed to). For the current patches, see
> > http://www.freebsd.org/cgi/cvsweb.cgi/ports/net/netatalk/files/.
> >
> > | [vulcan:ports/net/netatalk] jason# make
> > | ===> Extracting for netatalk-1.5p6_1
> >
> > [ this is the current version/revision ]
> >
> > | >> Patch patch-aa failed to apply cleanly.
> >
> > [ patch-aa no longer exists though ]
> >
> >
> > -pete
> >
> > --
> > Pete Fritchman <petef@databits.net>
> > Databits Network Services, Inc. <http://databits.net>;
> > finger petef@databits.net for PGP key
>
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-ports" in the body of the message
>
>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010810001029.P32016-100000>