Note that the GUI especially hasn't been getting a lot of love, and is
probably useful only as a proof-of-concept for now.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>

As suggested by Neil Brown, remove some more autogenerated files and fix
the Makefile so that a

	make
	make install

will suffice to generate everything necessary.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>

Signed-off-by: J. Bruce Fields <bfields@redhat.com>

Signed-off-by: J. Bruce Fields <bfields@redhat.com>

Signed-off-by: Malahal Naineni <malahal@us.ibm.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>

The sizeof operator, when applied to a parameter declared to have array,
yields the size of the adjusted (pointer) type, even if the parameter
declaration specifies a length.

Signed-off-by: Malahal Naineni <malahal@us.ibm.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>

After upgrading from 0.3.2 to 0.3.3 to fix some segfaulting issues we
had, 0.3.3 now rejects ACEs that have empty mask fields.  This is
because the function parse_alloc_fields checks whether the three strings
are zero length.  Having a zero length "type" or "who" definitely
doesn't make sense, but having an empty mask is entirely possible.  It's
allowed in the old versions and the ZFS back-end in our setup allows
them too.  Even nfs4_getfacl prints them out, producing the frustrating
situation where nfs4_getfacl can produce ACLs that you can't apply with
nfs4_setfacl.

This patch modifies the function to not check if the mask is an empty
string.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>

Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Improve parse_alloc_fields()'s readability by switching from strchr() to
strsep() and make sure we have the correct number of fields in the ACE.

Signed-off-by: Johann Dahm <johann.dahm@gmail.com>
Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: Johann Dahm <johann.dahm@gmail.com>
Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Also renames a loop variable for better readability.

Signed-off-by: Johann Dahm <johann.dahm@gmail.com>
Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: Johann Dahm <johann.dahm@gmail.com>
Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: Johann Dahm <johann.dahm@gmail.com>
Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: Johann Dahm <johann.dahm@gmail.com>
Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: Johann Dahm <johann.dahm@gmail.com>
Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Properly define maximum sizes for principal names, ACE sizes, and ACL size.
Previously, both the size of a given ACE and the number of ACEs allowed in an
ACL were artificially constrained and caused a variety of bugs.

Special thanks to Brian De Wolf for help both on and off the linux-nfs
mailing list in reporting and debugging these issues.

Signed-off-by: David M. Richter <richterd@citi.umich.edu>

Signed-off-by: David M. Richter <richterd@citi.umich.edu>