# XXX Accumulate commit message during each operation, so that a final
# commit operation will have a sensible message with all of the
# previous operations.
# XXX Fail if uid is 0, or perhaps != CVS meta files owner, to catch
# the case of invoking these targets as root.
and to support the "inet6" option instead.
Remaining usage of USE_INET6 was solely for the benefit of the scripts
that generate the README.html files. Replace:
BUILD_DEFS+= USE_INET6
with
BUILD_DEFS+= IPV6_READY
and teach the README-generation tools to look for that instead.
This nukes USE_INET6 from pkgsrc proper. We leave a tiny bit of code
to continue to support USE_INET6 for pkgsrc-wip until it has been nuked
from there as well.
automatically supply missing "basic" headers and libraries from an
older system, e.g. IRIX 5.x or Interix or AIX, etc.
Example usage:
USE_FEATURES+= snprintf glob regex
For now, we just pull in libnbcompat to supply the missing bits.
PATCHFILES nor DISTFILES, make fetch would complain about an unbuildable
dependency. Fix this by explicitly checking that _ALLFILES is not empty.
Reported by Sean Boudreau.
target by pkgsrc. This new target computes checksums only up until
the extract phase is complete.
The "checksum" target is now a target that can always be run by the
user at any time, regardless of the presence of the work directory.
These changes were proposed in PR pkg/36603 by Robert Elz.
README.html generation are broken. It will not find vulnerabilities in any
packages that have complex names in the pkg-vulnerabilties file.
e.g. php{4,5}-perl and sun-{jdk,jre}15
Post pkg_install-20070714 this is now fixed and only currently known
vulnerabilities are shown in the generated README.html files for packages.
You will need to update to pkg_install>=20070714 to get this fix.
provided in the distinfo file for those distfiles, rather than checking
specifically for the ones listed in _DIGEST_ALGORITHMS. If a distinfo
file actually provides checksums for other algorithms, e.g. TIGER,
then go ahead and verify them too.
run now at any time up until the "extract" phase completes, and the
"checksum" target no longer keeps a state (cookie) file.
From the checksum.mk file comments:
We do not actually wish to create a cookie file for the completion
of the checksum phase because we want to be able to detect if
any of the files have changed right up until the distfiles are
extracted. Therefore, we use the presence of the cookie file
from the "extract" phase to determine whether we need to continue
to verify checksums.
This addresses PR pkg/36262 and PR pkg/36603.
of an emulated operating system. Instead of proliferating things like
SUSE_VERSION_REQD, NETBSD_VERSION_REQD, SOLARIS_VERSION_REQD, etc., a
package can say:
EMUL_REQD= suse>=9.1 netbsd>=2.0 solaris>=10
all in one, succinct line.
a separate emulator-opsys.mk file.
The emulator-opsys.mk file defines EMUL_DISTRO and the various *EMUL*DIR*
variables, as well as any opsys-specific variables.
Include this file within compat_netbsd/Makefile.common so that the
*EXEC_FMT variables (defined by the compat*/emulator.mk files) are
defined. This fixes the build of compat* packages.
XXX emulator-opsys.mk will go away in the near future as we do more
XXX appropriate information hiding.
matches the native operating system. Use it in place of checking
whether EMUL_DISTRO matches "native-*" as EMUL_DISTRO is no longer
defined after bsd.prefs.mk is included.
This should fix PR pkg/36823 by Robert Elz.
