in this man page: depoted packages are kept in their depot directories, and
pkg_view(1) manages instances of those packages in views. Also copy the
warning in the pkg_delete(1) man page instead of the one from the
pkg_add(1) man page since the former correctly refers to package control
files already installed on the system.
* Teaching pkg_add(1) to be smarter about how to add a depoted packge.
* "pkg_delete foo" only deletes foo if it's a unique package.
* Teach pkg_admin(1) two new commands "add" and "delete" for working
with pkgdb.byfile.db.
* pkg_view(1) uses the new pkg_admin(1) commands to keep the
pkgdb.byfile.db files consistent in each view.
unique package. This makes deleting packages less surprising in a pkgviews
world where you can have multiple versions of packages installed at the
same time in the package depot.
Also make "pkg_delete -f" work depoted packages if they have been added to
views by also running "pkg_delete -f" for the package instances in the
views.
add pkg ...
For each listed package, write the absolute pathnames of the
files listed in its +CONTENTS file together with the package
they belong to into the package database. This should be used
only by pkg_view(1).
delete pkg ...
For each listed package, remove all file entries in the
package database that belong to the package. This should be
used only by pkg_view(1).
These commands will be used by pkg_view(1) to keep the pkgdb.byfile.db
files for each view consistent with what's stored in each view.
Since we stole these names from the debug code, change the names of the
debug commands to "addkey" and "delkey".
but the size of the data field in the pkgdb.byfile.db file is the length of
the string plus 1 for the NUL. Fix this check so that removing all of the
files belonging to pkgname from pkgdb.byfile.db works again.
if our other MODE SENSEs fail. Use this code for the optical device case,
at least for now. (We could query the optical media type and do a table
lookup for the geometry, but why bother? Actually, why bother with geometry
at all, but I digress...)
empty for others. x86_64 (and s390, which we do not have) have .eh_frame
as a read-only section, so __EH_FRAME_BEGIN__ and the register/deregister
functions must take a const foo *.
This also fixes an ICE in gcc 3.3.1 (see gcc PR #9552)
wiz