file pkgsrc/mk/bsd.pkg.clean.mk.
* Reimplement the clean target so that we don't need to invoke a separate
make process just to do clean-depends.
* Reimplement clean-depends using a pipe to avoid command-line length
issues.
show-depends-dirs
show-all-depends-dirs
show-all-depends-dirs-excl
While we're here, reimplement them using the new pkg_paths and
depends-depth-first.awk scripts. As a side effect, this speeds up
the execution of these targets by around 5-15% depending on the
complexity of the dependency graph.
NOTE: This *does* change the behavior of these targets slightly because
the output is now newline-separated instead of space-separated,
and the order of the output lines doesn't match the original
targets because the dependency graph is now walked depth-first
instead of breadth first.
the code that performs the dependency graph traversal (in depth-first
fashion). This script has a hook that allows executing a shell command
line upon visiting a dependency's package directory in either prefix
or postfix order, and may be used to simplify the code in bsd.pkg.mk
that iterates over dependencies.
This awk script requires the target "show-depends-pkgpaths", which is
defined in a new Makefile pkgsrc/mk/bsd.utils.mk. This file should
accumulate "utility" targets that current exist in bsd.pkg.mk, i.e.,
"helper" targets for various actions.
for manipulating PLISTs. This module is not used by default pending
more widespread testing -- currently the variable _USE_PLIST_MODULE
must be defined in /etc/mk.conf to enable its use.
The main features of the new PLIST module are:
(1) Splits out the PLIST-handling code from bsd.pkg.mk into a
separate "plist" module.
(2) Splits out giant, multi-line awk scripts stored in make
variables into separate awk scripts that may be joined
together to post-process PLISTs. Each of these awk scripts
consolidates the processing for one set of files, e.g.,
man pages, info pages, etc., and is more easily commented
than a make variable.
(3) Splits out the print-PLIST code from the regular PLIST code
since they have no common pieces (print-plist.mk vs.
plist.mk).
(4) Completely re-implements the shared-library handling to be
more efficient. Along the way, this also fixes a problem
for Mac OS X users where the PLISTs incorrectly contained
absolute paths.
(5) Completely re-implements the info-file handling so that we
can migrate from INFO_FILES definitions to just adding
info/foo.info entries in the static PLISTs.
(6) Adds commented-out support for automatically compressed or
decompressed info page entries based on the value of MANZ.
These changes will be activated after texinfo.mk has been
replaced by something that is built using the more modern
primitives now available in pkgsrc.
(7) Move the file compression logic into a separate script
"doc-compress" that compresses or decompresses files while
minding symlinks. This script is now called by bsd.pkg.mk
to do the "autmoatic man page handling". In the future,
it will also handle the "automatic info page handling" and
possible others.
In general, the idea is to move stuff out of the Makefiles and into
separate files where we don't need to worry about quoting rules
and where each file can have a separate history of commits. This
simplifies the makefile logic (especially in terms of readability)
and also simplifies maintenance of the code.
that the interpreter of "#!"-style scripts exists. It is disabled by
default, and can be enabled by setting CHECK_INTERPRETER=yes. As for the
other check-* targets, CHECK_INTERPRETER_SKIP is a list of shell globs
that can be used to to skip certain files.
Ok'ed by jlam.
packages from branch to branch:
When a package is considered broken (for whatever reason) at branch
time, we'll mark it with BROKEN_IN=name-of-branch. At the next
branch, we can easily determine which ones have remained broken and
remove them.
BROKEN_IN is visible to users iff a build fails, when it appends a
warning message to the above effect.
With feedback from dillo. Reviewed by jlam.
BINPKG_SITES entry separately, construct an appropriate PKG_PATH
and call pkg_add only once. Patch from Chapman Flack in PR 30929.
Use PKGNAME_REQD when installing package dependencies.
Patch from Chapman Flack in PR 30954.
configure script itself so that the generated config.status script
does what we want (just exit if asked to "recheck"). This ensures
the timestamp for config.status is earlier than the timestamps for
the files that config.status generates (Makefile, config.h, etc.).
This fixes some problems where some packages end up "rebuilding" as
part of the install target, which makes the rebuilt files owned by
root and makes cleaning the work directory fail.
framework should be used. This is implemented by creating a small
file mk/install/pkginstall.mk that guards the implementation makefile
mk/install/bsd.pkginstall.mk. This guard file just checks whether
one of the pkginstall-related variables is non-empty, and if so, then
the implementation file is automatically included. This completely
deprecates USE_PKGINSTALL, which no longer has any affect in pkgsrc.
if NO_CHECKSUM is set to "yes". This allows to run "make makepatchsum"
in the top level pkgsrc directory without having to worry about lots of
additional distinfo files being created.
to force an early exit as soon as a "cd" command fails. Otherwise, "/"
would have been added as a dependency, leading to undefined behavior.
See PR 32202 for details.
Fixes the second item of PR 32202.
every top level target that is intended to be called by the user or by a
package different from the current package. It provides the same action
as the "main" targets like "fetch", "expand", "build" before in case
PKG_FAIL_REASON is set, that is it prints all PKG_FAIL_REASONs and
fails.
Fixes PR 32202.
Implementation notes:
- The target names have the "do-" prefix to not pollute the
bsd.pkg.check.mk namespace.
- The PKG_SKIP_REASON has no influence on the do-check-pkg-fail-reason
target, although both are handled with the same code.
(This also fixes problem where on some Linux systems, ldd
outputs linux-gate.so.1 entry without corresponding library.)
This was noticed on Linux.
This was discussed and okayed on the tech-pkg list in February,
2002. I have been using this on NetBSD and Linux since then. (Also
just tested on DragonFly.)
${WRKDIR}, and just always create it as part of a "make extract". This
should fix problems where if an old work directory is lying around that
doesn't already have .pkgdb, then a "make reinstall" won't break.
completely. The only difference to revision 1.1748 is that the
PATH=$$PATH:${SU_CMD_PATH_APPEND} argument is included in single quotes.
This will lead to problems when the PATH should contain single quotes
but works in all other cases. (Single quotes in the PATH hadn't worked
before either.) The tricky part regarding this code is that the PATH and
.CURDIR bypass the SU_CMD. They are evaluated before and inserted as
literals into the command that is executed by the SU_CMD. Preserving
these variables this way circumvents interpretation and modification
through the SU_CMD, but leads to ugly rules for quoting which are
currently not handled completely correctly.
ALLOW_VULNERABLE_PACKAGES settings that applies to all packages, there can
now be per-package lists of allowed vulnerability ids:
ALLOW_VULNERABILITIES.<pkgname>=<space separated list of vulnids>
To avoid duplication of code, audit-packages is now used to do these checks.
It can be skipped altogether by setting:
SKIP_AUDIT_PACKAGES=yes
variable before if it had multiple adjacent white-space characters in
it. Removed the dependency on awk(1), as a simple shell script can do
this job, too. Another change is that all kinds of weird pathnames can
be used for PREPEND_PATH, like /bin, "/bin/Program Files", "/ ",
/foo\ bar, and so on. This had not been possible before, too.
Related to PR 32079.
Ok'ed by jlam.
