The conflict checks compare the patterns first against the package
names without version (as reported by "pkg query "%n"), then - if
there was no match - agsinst the full package names including the
version (as reported by "pkg query "%n-%v").
Approved by: portmgr (blanket)
* Upstream changed build tool from GNU Autotools to CMake
* Set TEST_TARGET
* Add pre-test-TESTS-off target to display message that TESTS option
needs to be enabled to execute `make test`
* Switch to use pkg-messge instead of pkg-deinstall to display message
for those who removes ClamAV permanently from you system
* Stop checking in rc scripts if /var/run/clamav exists
* Pet portclippy
* Reformat Makefile with portfmt
Release Notes: https://blog.clamav.net/2021/09/clamav-01040-released.html
PR: 258565
Differential Revision: https://reviews.freebsd.org/D32354
* Add Long Term Support release of ClamAV
* Add CONFLICTS_INSTALL to security/clamav
* Add entry to UPDATING that explains
- Upstream changed their end-of-life policy and LTS release is
introduced
- How to switch from regular release (security/clamav) to LTS
release (security/clamav-lts)
Note about setting PORTEPOCH
In general new port should not set PORTEPOCH. But in this case it
should be set with following reason.
1. Recently upstream changed their end-of-life policy as following.
- Regular feature release (= 0.xyz.0 release) is released more
frequently, and life time of each feature release is shortened to
about 4 months.
- To compensate for the short lifetime of regular feature release,
Long Term Support feature release is introduced and its life time
is about 3 years.
2. First LTS starts with version 0.103.3 and it is same as current
version of security/clamav.
3. Because of short lifetime of regular feature release, it is
probable that not a few users of security/clamav want to switch to
security/clamav-lts after the latter is connected (and the former
is updated to 0.104.0).
3. For such users the entry is added to UPDATING about how to switch
from security/clamav to security/clamav-lts. And binary package
user is suggested to execute
`pkg set -o security/clamav:security/clamav-lts`. This command
changes the origin of already installed clamav packages. So user
can switch to LTS version without reinstalling.
4. But if PORTEPOCH isn't set in security/clamav-lts, binary package
user who executed above command will notice that version of
installed package (0.103.3,1) succeeds to current version of
security/clamav-lts (0.103.3). The situation doesn't change after
new patch release (0.103.4 for example) is released and it causes
the problem that installed package isn't properly upgraded with
`pkg upgrade`.
5. So PORTEPOCH is set in security/clamav-lts to prevent binary
package user from suffering from such problem.
- Remove MSPACK option and always use archivers/libmspack in
accordance with policy about bundled library described in 6.2 of
Porter's Handbook.
- Remove obsolete and no-op options from CONFIGURE_ARGS.
PR: 243914
Submitted by: Yasuhiro KIMURA <yasu@utahime.org> (maintainer)
Relnotes: https://blog.clamav.net/2020/02/clamav-01022-security-patch-released.html
Security: CVE-2020-3123
as a stand-in for "are we running on gcc".
For people already testing powerpc on clang, it is possible that they
already have both compilers in base. Thus, the assumption that "gcc is
in base" (e.g. libstdc++.so exists) always means "force use of GCC" is
already broken. It will be for everyone on -CURRENT once the switch is
made.
While here, standardize on compiler:c++11-lang instead of -lib (they are
equivalent these days), pet portlint, and do some other cleanup.
Approved by: portmgr (tier-2 blanket)
previous patch added the configure argument to find archivers/libmspack
installed, but neglected to set USES+= pkgconfig to actually have the binary
use that version. With it now properly finding libmspack.so, revert to
not trying to install libclammspack* if the MSPACK option is set.
PR: 233658
It adds a few more fixes and improvements.
- Solves the bug, by adding a MSPACK option (enabled by default) and a build dependency on pkg-config.
- Makes massive use of options helpers and similar tools to improve Makefile's readability. In the rewriting I have not maintained the dependency of option LDAP on option MILTER and the two options are now indipendent: if this is bad, it as to be fixed on my patch. I avoided to do it directly because I am unsure it is needed.
- Uses now libpcre2 consistently (before the port could link to libpcre2 by mistake).
- Removes useless arguments about zlib in CONFIGURE_ARGS.
- Reorders variables consistently with the standard order.
- Cleans header.
- Updates pkg-plist and sort if alphabetically.
- Bump PORTREVISION, needed because of some fixes.
PR: 228468
Submitted by: phascolarctos@protonmail.ch
Reported by: gondim@bsdinfo.com.br
