Commit graph

4000 commits

Author SHA1 Message Date
jlam
1ec8131688 Set PKG_REFCOUNT_DBDIR in the environment for the +INSTALL script, just
like pkg_add(1) does in pkg_install-20050204.
2005-02-05 01:17:59 +00:00
jmmv
ee2419f704 Add tl locale, initially used by glib 2.6.2. 2005-02-04 11:20:39 +00:00
jlam
2ab02f1ddd Bump PKGTOOLS_REQD to 20050204. This version of pkg_install passes
the correct information to the +INSTALL and +DEINSTALL scripts to fix
problems with binary packages incorrectly locating the reference counts
database, e.g. /var/db.refcount.
2005-02-04 09:19:41 +00:00
jlam
9896c47ccf Create a new INSTALL script action "UNPACK" that unpacks all of the
sub-scripts.
2005-02-03 23:29:08 +00:00
jlam
a6f8c45bfc Next round of install script cleanup -- we now create +FILES,
+RCD_SCRIPTS, and +PERMS subscripts to handle copying config files
and rc.d scripts, and setting special permissions on files.  The +FILES
and +RCD_SCRIPTS are basically identical except for different embedded
packets of data, and they feature reference-counting for the files in
case multiple packages share the same config file.

Garbage-collect unused functions and definitions in the install scripts
now that the subscripts are self-contained.
2005-02-02 10:33:01 +00:00
abs
6a63f7037d allow postgresql80 (from wip), similar to java-vm.mk and jdk14 2005-02-01 11:07:31 +00:00
jlam
a767f84684 In the usage message, note that the CHECK-* actions can take an optional
metadatadir argument.
2005-01-31 21:41:06 +00:00
jlam
db02fdf8d8 Note in the usage that CHECK-* take an optional metadatadir argument. 2005-01-31 20:32:58 +00:00
jlam
7ebb4d0225 Add missing "fi". 2005-01-31 19:53:31 +00:00
jlam
d621d74579 Fix automatically adding users and groups after the {GROUP,USER}ADD
definitions were removed from mk/install/headers.  The checks for
whether {GROUP,USER}ADD are defined are now moved to the usergroup
script.
2005-01-31 18:41:08 +00:00
jlam
7bb70d34b2 Add RCS Id. 2005-01-31 17:42:20 +00:00
jmmv
da5adc365e Convert the Java related packages to use the alternatives system. To do
this, we only need some simple logic in java-env.mk to automatically
generate an ALTERNATIVES file for us.

There are two exceptions, though: fastjar and jikes (pointed out by tv@),
which do not use this file; they have to be handled manually.

Bump PKGREVISION for all affected packages.

While doing this, remove the java-wrapper package, obsoleted by the new
functionality provided by pkg_alternatives.
2005-01-30 23:18:37 +00:00
jlam
642fa3e0e6 Don't make pkgsrc depend on dc(1) to be present in the system. Since
we already use awk so pervasively in pkgsrc, simply use awk in place
of dc for simple computations.
2005-01-28 21:05:59 +00:00
jlam
0a8bdc04af Avoid hardcoding the refcount database into the INSTALL scripts. We
now simply make ${PKG_REFCOUNT_DBDIR} always be ${PKG_DBDIR}.refcount
so that it always follows the location of ${PKG_DBDIR}.  This preserves
the ability for PKG_DBDIR (and PKG_REFCOUNT_DBDIR) to be different on
different machines despite using the same binary packages.
2005-01-28 19:06:07 +00:00
jlam
8f3180c914 Move the default directory for the reference counts database outside
of ${PKG_DBDIR} to avoid problems with the package tools thinking that
it is a bad package.  The default directory is ${PKG_DBDIR} with
".refcount" appended to the name.  This may be set explicitly through
PKG_REFCOUNT_DBDIR in /etc/mk.conf (bootstrap users may want to do
this, although the default value should do the right thing).
2005-01-28 18:32:40 +00:00
jlam
39695d15ef Remove entries for GROUPADD and USERADD from header since they're all
handled within the usergroup script.
2005-01-28 18:05:34 +00:00
jlam
fc16f1f48c Pretty-print the users and groups that may need to be removed during
CHECK-REMOVE.  The users and groups are now printed on the same line
with whitespace separating the entries (for easy copy-and-pasting info
commands), and wrapping to successive lines if we overflow the length
of the current line.
2005-01-28 18:00:28 +00:00
jlam
b474d00694 If PKG_SYSCONFSUBDIR is non-empty, then add ${PKG_SYSCONFDIR} to the
list of +DIRS-maintained directories.  This allows for a package to
specify where its config files should go and creates the directory
automatically, even if it doesn't have any example config files to
put in place.
2005-01-28 08:17:53 +00:00
jlam
0d547bd690 Note the end of the install-dirs unpack template within the INSTALL
script.
2005-01-28 07:39:13 +00:00
jlam
f9d620373a Continue with install script cleanup -- we now create a +USERGROUP
script that is unpacked by the +INSTALL script at PRE-INSTALL time
before any other actions take place, and invoke +USERGROUP to create
any necessary users and groups.  Remove the now-unused code for
PKG_USERS and PKG_GROUPS from the install and deinstall templates.

We also reference count the users and groups and store the reference
counts in ${PKG_DBDIR}/.refcount/{users,groups}.  This allows multiple
packages to register that they use same users and groups, and allows
+USERGROUP to be invoked at any time to repair an installed package.

Also fix the install and deinstall templates to invoke the CHECK-*
actions with ${PKG_METADATA_DIR} so that the correct PKGNAME can be
derived.  This fixes the weird messages asking the user to create
directories for ".pkgdb" during a "make install".
2005-01-28 07:37:55 +00:00
jlam
5aa13ad4ee Use reference counts to properly account for the creation and removal
of directories needed for the proper functioning of each package.

The +INSTALL script unpacks a +DIRS script that adds and removes
directories.  The +DIRS script entirely encapsulates the directory
creation and removal, and completely replaces the code in the
mk/install/install and mk/install/deinstall templates that handled
{MAKE,OWN}_DIRS and {MAKE,OWN}_DIRS_PERMS.

The +DIRS script is meant to be executed from within the package
meta-data directory, e.g.  /var/db/pkg/<pkgname>.  It's usage is:

	./+DIRS ADD|REMOVE|CHECK-ADD|CHECK-REMOVE

The ADD and REMOVE actions cause the necessary directories to be added
or removed from the system.  The CHECK-ADD and CHECK-REMOVE actions
print out informative messages prompting the user to either create or
remove some necessary directories.

The behaviour of "ADD" is such that if the directory already exists
on the system and is not already ref-counted, then that directory is
marked as "pre-existing".  On "REMOVE", pre-existing directories are
left untouched on the filesystem.

At any time, the root user can sanity-check the directories needed by
packages by invoking all of the +DIRS scripts with the "CHECK-ADD"
action.  If there are missing directories, then invoking all of the
+DIRS scripts with the "ADD" action will ensure that any missing
directories are created.

The reference counts database is stored in ${PKG_DBDIR}/.refcount.
The reference counts related to directories managed by the +DIRS script
are stored in ${PKG_DBDIR}/.refcount/dirs.  If the directory reference
counts database is removed, then invoking all of the +DIRS scripts
with the "ADD" action will reconstruct the database; however, directories
may be marked as being pre-existing, so they won't be removed at
package de-installation (although a message will be displayed informing
the user that those directories can be removed).
2005-01-28 06:30:58 +00:00
jlam
ce33cd0c70 Move the creation of ${PKG_DB_TMPDIR} to when we create ${WRKDIR}.
This ensures that it's always created with the same user/group/mode
as ${WRKDIR}, so if a non-root user thinks he has permissions to remove
${WRKDIR}, then that expectation will remain true for ${PKG_DB_TMPDIR}.
2005-01-27 18:32:20 +00:00
sketch
b6aec517bc Pass the correct programs to the mkdatabase script. 2005-01-27 17:55:42 +00:00
tv
1d26911a04 Fix handling of PREPEND_PATH in dependency-recursed builds. 2005-01-27 11:21:46 +00:00
tv
974ce33409 Cleanup: wrapper-defs.mk is no longer optional; remove its conditional.
(...and if it were optional, it should have been an .sinclude anyway.)

Sanity: If mk/platform/${OPSYS}.mk is missing, don't assume NetBSD is it.
pkgsrc now depends on a valid platform file for an OS, so require it.
(Still includes NetBSD.mk, but sets PKG_FAIL_REASON.)
2005-01-27 04:16:47 +00:00
tv
bcce274c8b Move the PREPEND_PATH for LOCALBASE/bin (and optionally X11BASE/bin) to a
spot that will come before compiler.mk (in bsd.prefs.mk).  Previously,
LOCALBASE/bin was appearing earlier in the path than work/.<compiler>/bin,
which could cause the Wrong Thing to happen.
2005-01-27 04:05:08 +00:00
jlam
c858a36842 Also substitute for PWD_CMD so we can use it in the INSTALL scripts. 2005-01-26 23:15:03 +00:00
jlam
b09a6d5308 Create a new variable "INSTALL_UNPACK_TMPL" that points to a list of
shell script fragments that unpack sub-scripts before the any of the
other PRE-INSTALL stages are run.
2005-01-26 23:14:12 +00:00
jlam
a2a889c638 Fix so that we preserve /usr/lib/foo.so on the command line. 2005-01-26 05:03:11 +00:00
cube
07b9c2fded Set argmatch=yes when we transform /some/where/libfoo.so into
-L/some/where -lfoo so that the arguments we push on the stack are actually
used in logic.

Should solve the issue build bash with option 'static' enabled reported on
netbsd-users.

OK'd by jlam@.
2005-01-25 23:27:34 +00:00
tv
e2d4e995a8 Relax need to put GNU_CONFIGURE before inclusion of bsd.prefs.mk:
Convert _OPSYS_MAX_CMDLEN to a plain _OPSYS_MAX_CMDLEN_CMD variable, which
is not evaluated by a shell until CONFIGURE_ENV is expanded (and only then
if USE_LIBTOOL+GNU_CONFIGURE are both set).
2005-01-25 18:15:58 +00:00
jmmv
fb5121604f Explicitly use -g to avoid relying on default behavior. 2005-01-25 16:48:53 +00:00
jmmv
e1b9c0a3c2 Redo previous per suggestion from reed@. Do not process the alternatives
until the pkg_alternatives package has been manually installed.  This way
we drop the dependency on that package.
2005-01-25 16:29:16 +00:00
jmmv
b071961bea Make the usage of alternatives optional through the PKG_ALTERNATIVES variable,
which is similar to PKG_CONFIG.  This does not change how the binary packages
are built, only whether the pkg_alternatives command should be run at package
installation/deinstallation time or not.  The variable can take either YES or
NO as values and defaults to YES.
2005-01-25 15:37:25 +00:00
jmmv
f70b9a4716 Add alternatives.mk, a file used to manipulate the alternatives system.
A package can optionally provide an ALTERNATIVES file which contains pairs
of wrapper/alternative, one per line.  The file is then used by the install
and deinstall scripts to register the alternatives and to create the
appropriate wrappers.

Make bsd.pkg.mk include this new file.  This happens unconditionally to
keep all the alternatives logic in an independent file.  Otherwise, some
of the stuff could have to be moved to bsd.pkg.mk, breaking consistency.
2005-01-25 13:13:35 +00:00
grant
857b601b7b only try to run the build if "cd /usr/pkgsrc" is successful. 2005-01-25 08:47:55 +00:00
tv
48c1dc3c38 Add python22, python24 to base addresses list. 2005-01-24 21:47:00 +00:00
tv
443bb330bc Disallow -rdynamic (was being attempted by shells/scsh). 2005-01-24 21:21:45 +00:00
tv
e1c33bd2a4 Merge a little visual-only cleanup from the tv-derecurse branch:
Collect many modifiable hard-to-find variables scattered throughout
bsd.pkg.mk into a common section near the top of the file.  (Not necessarily
"complete", but helps to reduce HEAD-branch divergence.)
2005-01-24 19:57:42 +00:00
tv
3c440a0121 Merge down deferred PREPEND_PATH handling from the tv-derecurse branch.
Also merge in deferred CLASSPATH handling, which can now be moved back
to java-vm.mk for cleanliness.
2005-01-24 19:32:33 +00:00
tv
6a1da4db28 Make CHECK_FILES_SKIP_CMD into a single grep, using a ${var:@foo@bar@} loop. 2005-01-24 18:44:38 +00:00
tv
4284bdc494 Merge down deferred EVAL_PREFIX handling from tv-derecurse branch.
Also move its definition later in bsd.pkg.mk, so that it actually works
for compiler/*.mk (which is why those files originally had to provide
"defaults" for their variables -- they never did get evaluated).
2005-01-24 18:20:57 +00:00
tv
dace9b647e Merge down janitorial change from tv-derecurse branch:
Move definition of PKGSRC_COMPILER to defaults/mk.conf, to make it more
readily seen by the user.
2005-01-24 18:07:56 +00:00
tv
db5bb1ad1d Merge down janitorial change from tv-derecurse branch:
No need to do .som_done dance; just show the message as part of the
pre-extract phase.  If that means showing it more than once if the extract
is interrupted and restarted, no harm done.
2005-01-24 18:04:24 +00:00
jschauma
0c0134061b Per default, only create md5 and sha1 checksums.
(These are available on all platforms via digest(1).)
Set commands for other checksums on a per OPSYS basis.

Set MKSUSM=yes back as the default.
2005-01-24 15:00:48 +00:00
tv
51a7a67d08 Not all OS's have all the checksum tools available to make MKSUMS=yes work.
Default it to "no" instead.
2005-01-24 13:21:13 +00:00
jschauma
d004e03093 Add MKSUMS and SIGN_AS (the former defaulting to yes, the latter commented
out).
2005-01-24 13:08:19 +00:00
xtraeme
2494cf89ca Fix breakage in previous, installed pkg db files should be installed into
${PKG_DB_TMPDIR} not ${_PKG_DB_TMPDIR}, patch from Geoff Wing.
2005-01-24 09:31:06 +00:00
jschauma
bb18a86029 Add bits to allow bulk-builders to generate checksums for the binary
packages they upload (by setting the optional variable MKSUMS=yes) and,
also optionally, PGP signing them (by setting SIGN_AS=username@NetBSD.org,
for example).
2005-01-24 03:41:34 +00:00
jlam
1e0e784ec4 Move the creation of ${PKG_DB_TMPDIR} earlier so that it's created as
the build user instead of the root user.  This should allow "make
clean" as a non-root user to work again.
2005-01-23 22:23:40 +00:00
jlam
af900af65a Rearrange the register-pkg and post-install-fake-pkg code. All of
the meta-data files are generated and stored in ${PKG_DB_TMPDIR} by
the *-install-fake-pkg targets, and then all of the contents of that
directory are simply copied into ${PKG_DBDIR}/${PKGNAME} by register-pkg.

This is intended to make (DE)INSTALL scripts be more powerful.
Currently, they have no way to keep state on their own.  Now, they
can keep state in the current working directory.  When invoked by
pkg_add(1), the current working directory is ${PKG_DBDIR}/${PKGNAME},
so the state files are already stored in the correct place.  When
invoked by bsd.pkg.mk, the current working directory is ${PKG_DB_TMPDIR},
and any state files generated by the (DE)INSTALL scripts are copied
into the correct place by the register-pkg target.
2005-01-23 21:57:38 +00:00
jlam
daf62c2000 Rework the way that packages are registered during a "make install". We
split the fake-pkg target into three smaller pieces:

pre-install-fake-pkg - Create files in ${WRKSRC}/.pkgdb meant to be
	installed into ${PKG_DBDIR}/${PKGNAME} that don't depend on
	the installed files.

post-install-fake-pkg - Create files in ${WRKSRC}/.pkgdb meant to be
	installed into ${PKG_DBDIR}/${PKGNAME} that may depend on the
	installed files.

register-pkg - Copy the files in ${WRKSRC}/.pkgdb and create a few
	additional meta-files in ${PKG_DBDIR}/${PKGNAME}.

The idea is that ${WRKSRC}/.pkgdb contains as much of the state as
possible that will be copied into ${PKG_DBDIR}/${PKGNAME}.

The INSTALL and DEINSTALL scripts generated by bsd.pkg.install.mk are
now invoked with ${WRKSRC}/.pkgdb as the working directory, so the
scripts may use the working directory to store temporary files,
regardless of whether they're invoked from bsd.pkg.mk or by pkg_add.
2005-01-23 20:45:22 +00:00
grant
13ed8b06aa message whitespace fix 2005-01-22 09:46:55 +00:00
tv
49fe383c7d Re-add direct symlink logic, optional under GCC_USE_SYMLINKS=yes. Saves
one fork/exec step on platforms where gcc is compiled with hardcoded gcc-lib
pathnames.  This applies to most platforms, but the default currently
remains not to use symlinks so as to avoid unexpected gotchas for users.
2005-01-21 21:59:14 +00:00
tv
23c895bb83 The pipe triggers a subshell even though () is not used. So use "exit" and
|| on the outside of the command to determine whether the packages found
in the loop are up to date.
2005-01-21 13:24:13 +00:00
tv
737cb046a6 Fix missing {} around variable name, noted by wiz.
Change one instance of $(...) to `...` per policy.
2005-01-21 13:11:46 +00:00
tv
6e2055d909 Add NOLINT=1 for BSD-.mk based packages. 2005-01-21 13:07:12 +00:00
tv
e24f55470c Add ruby16/ruby18 to base addresses list. 2005-01-21 06:14:23 +00:00
tv
6a8e8e01db Interix doesn't have a "flex" program per se in the base system. Its "lex"
is flex, but libfl.a and FlexLexer.h are missing.  So use pkgsrc flex
on an as-needed basis.
2005-01-21 02:23:34 +00:00
tv
d5938364f3 Defer the check of GNU_CONFIGURE until later; should fix build of net/ORBit
(and perhaps other packages) which set GNU_CONFIGURE only after bsd.prefs.mk.
2005-01-21 02:09:10 +00:00
jlam
e4efb0343f Also match NetBSD security releases, e.g. 2.0.1, when deciding whether
using dlopen() implies linking against libpthread.  Idea for fix from
PR pkg/29022 and probably also fixes PR pkg/28800.
2005-01-20 15:22:39 +00:00
tv
913dfa54a2 Add LINTPKGSRC_CACHE support to reuse .lintpkgsrc.db created by pre-build.
"cd" to $USR_PKGSRC automagically so this can be called from any dir.
2005-01-20 02:02:50 +00:00
tv
76a832fa68 Fix error in previous; add explicit version check first, then the optionally
excluded timestamp check.
2005-01-19 22:12:27 +00:00
tv
5b254ab9a7 * Add USE_BULK_TIMESTAMPS (default "yes") to control whether dependency
binpkg timestamps, and ${PKGDIR} file timestamps, will be used to
  determine if a rebuild is necessary.  If changed to "no", then rebuilds
  will only happen when PKGNAME actually changes.

* Add LINTPKGSRC_DB (and companion USE_LINTPKGSRC_CACHE, default "no") to
  make pre-build generate a reusable database with "lintpkgsrc -E"; helps
  with nfs-mounted pkgsrc.  A future change will allow use of this in the
  upload script.

* Add entry for PRUNEPACKAGES in build.conf-example (this was missing).
2005-01-19 20:58:00 +00:00
markd
2c56afb401 Add no_NY locale used by kdbg and kbiff. Also hsb and nds locales used
by kde.
2005-01-19 13:19:23 +00:00
tv
291f7ac5d5 Force manpages to be man-only (no catpages) when using BSD .mk files in
a package build.  This may change once groff is fixed, but for now, there's
no guarantee of a nroff available on the host system.
2005-01-18 22:36:37 +00:00
tv
54ce782081 Add <bsd.lib.mk> to base addresses list. 2005-01-18 19:37:08 +00:00
tv
198bb2cf3c Add python23 to the base-address list. 2005-01-18 18:22:56 +00:00
jlam
5977984da2 Split out the argument pre-processing logic into its own script
arg-pp-main, and allow for wrapper-specific hooks to extend the
pre-processing through _WRAP_ARG_PP.<wrappee>.  Move the Darwin
GCC-related preprocessing into arg-pp-darwin-gcc, and create a
arg-pp-mipspro-cc that understands -LANG:<feature> and -LIST:<feature>
so that they're not confused with the usual -Ldir options.
2005-01-18 17:25:13 +00:00
tv
81e40c318f Turn distfile pruning back off by suggestion of wiz@. It's only binpkg
pruning that needs to be on by default.
2005-01-18 15:04:34 +00:00
tv
9a80a01fb8 Turn on all pruning by default. This avoids problems with @blddep matching
in the current incarnation of pkg_install.
2005-01-18 14:11:11 +00:00
jlam
1cdb69ea24 Teach the wrapper scripts about HP-UX *.sl shared libraries. They
work the same way as ELF *.so shared libraries from pkgsrc's point of
view -- just the extension is different.
2005-01-18 05:24:47 +00:00
jmmv
cd4de569dd Add a blank line between multiple "broken due to a broken dependency"
messages for better readability.
2005-01-17 08:52:50 +00:00
jlam
2374c335bd Rework code to more easily allow listing other platform that, like NetBSD,
use GCC and can accept -pthread.  Add DragonFly(BSD) to this list (from
PR pkg/28964).
2005-01-17 08:29:30 +00:00
jmmv
af92a2b644 - Move all CHECK_FILES_SKIP definitions to a central place in bsd.pkg.mk.
- Move the previous block of code down in the file so that all used variables
  are defined (specially PKG_SYSCONFDIR).  Fixes a problem noticed by
  kristerw@'s bulk build in comms/minicom.
- Ignore diff's return code, which aborts make in NetBSD 1.6.2.  Also noticed
  by kristerw@'s bulk build.
- Use full paths to do the checks, instead of relative to ${PREFIX}.  Less
  ambiguity.  Matches should be turned into regular expressions that anchor
  to a whole line (tried that, but found some problems).
- Turn CHECK_FILES to NO by default.  As said in the previous point, there
  are still some problems that have to be fixed and minor improvements to
  be done.  And I have no time to fix this ATM.  Yes, this definitely needs
  more testing.  I'm sorry for all the noise.  (But hey!  you should set and
  try this feature locally! ;-)
2005-01-15 10:50:17 +00:00
adrianp
204ac90729 Fix endif typo 2005-01-15 08:59:28 +00:00
tv
d59d4df63a Work around an undef-in-string warning caused by unexpanded $NetBSD$
id tags.  Noticed by kristerw.
2005-01-15 01:53:00 +00:00
jmmv
a12ae77f04 Also skip ${INFO_DIR}/dir in the check-files functionality, as this file
is not controlled by any package (so it won't be part of any PLIST).
2005-01-14 18:41:00 +00:00
jmmv
99909219cc Automatically add any of the {CONF,SUPPORT}_FILES and {MAKE,OWN}_DIRS stuff
to CHECK_FILES_SKIP to avoid some false positives.  These directories are
created in the pre-install stage so are included in the generated file list.
The files are also added to silence some problems that may arise during
"make replace".  Found by wiz@ in the gtk2 package.
2005-01-14 18:02:38 +00:00
tv
782903ea16 "uptodate-pkgtools" is actually a dead target these days. Use "fetch" in
pkgtools/pkglint to determine if pkg_install needs an update.
2005-01-14 15:32:34 +00:00
jmmv
8cf5329682 Add the check-files target: when enabled, this verifies that installed
packages do not install more files than expected (nor delete existing
files), aside other sanity checks in PKG_SYSCONFDIR and VARBASE.

This behavior is only enabled if PKG_DEVELOPER is set and CHECK_FILES
is YES (the default).  Should let us catch problems in other systems,
as some packages install different files depending on the OS they are
being built (which is different to see).

Furthermore, since the sanity checks done in PKG_SYSCONFDIR and VARBASE
are quite agressive, only enable them when CHECK_FILES_STRICT is YES
(defaults to NO).  Developers should enable this feature to detect errors,
but this can't be a default yet.  Otherwise, lots of packages could be
marked broken in bulk builds (they really are, according to "cleanliness"
rules, but most of them are non-trivial to fix).
2005-01-14 14:36:31 +00:00
tv
7983e9bf35 Only do the config.* overrides iff GNU_CONFIGURE. (defined() check needed
to make older bmakes happy.)
2005-01-14 14:08:06 +00:00
jmmv
e05f3f488a For those packages where the maintainer field can't be determined by
grepping, run make to get MAINTAINER's value.  I've tested this with
a script that tries to "simulate" this one without problems, but I
can't test it "in place".
2005-01-14 13:08:28 +00:00
jlam
bff1eecc97 Modify linux-pam and solaris-pam builtin.mk files to be more general
in their tests for built-in versions of the PAM implementations.  The
MacOS X case now collapses nicely into the linux-pam case.  Allow
pam.buildlink3.mk to use solaris-pam as an accepted PAM implementation.
2005-01-14 07:54:20 +00:00
jlam
643b26af82 We need to check that IS_BUILTIN.<pam> is "yes", not just that it's empty. 2005-01-14 07:37:39 +00:00
jlam
151f14b269 Add a builtin.mk file to detect Solaris' PAM, used by pam.buildlink3.mk. 2005-01-14 06:29:55 +00:00
jlam
1b5734f517 Create a pam.buildlink3.mk file that is used by PAM-using packages.
It includes the correct buildlink3.mk file from either Linux-PAM
(security/PAM) or OpenPAM (security/openpam) and eventually will
support solaris-pam.  pam.buildlink3.mk will:

	* set PAMBASE to the base directory of the PAM files;
	* set PAM_TYPE to the PAM implementation used.

There are two variables that can be used to tweak the selection of
the PAM implementation:

PAM_DEFAULT is a user-settable variable whose value is the default
	PAM implementation to use.

PAM_ACCEPTED is a package-settable list of PAM implementations
	that may be used by the package.

Modify most packages that include PAM/buildlink3.mk to include
pam.buildlink3.mk instead.
2005-01-14 05:15:39 +00:00
tv
089aec77ed Add $NetBSD$. 2005-01-14 00:10:50 +00:00
tv
f8b3f96fb6 Add override for uncommon "config.rpath" GNU config snippet. Use platform
file to indicate that override logic should be turned on.  (AFAICT, only
Interix is afflicted at the moment.)
2005-01-14 00:10:01 +00:00
tv
609eec1da7 Put back the proper variable construct for LIBTOOL_REQD that was borken in
previous.  (_OPSYS_LIBTOOL_REQD is only defined on some platforms, so the
:U clause provides the next best choice if that's not defined.)
2005-01-13 20:19:57 +00:00
tv
e4cc383421 * Clean up syntax ambiguities, e.g. (my $foo = $bar) =~ ....
* Add optional verbosity to show most commands as they are being executed.

* "use strict" and "use warnings".

* Slurp in the varables from build.conf and bmake in one shot (amazing
  startup speed boost with nfs pkgsrc); put them in a hash to make
  "use strict" much happier with the namespace.

* Fix a bunch of undef-dereference errors evidenced by "use warnings".

* Exclude PKG_DBDIR from leftovers list if it is inside LOCALBASE.

* Convert some <a name="..."/> constructs to <a name="..."></a> to make
  non-XHTML-compliant browsers happier.
2005-01-13 19:29:28 +00:00
seb
680703690d Add TNF copyright. 2005-01-13 09:49:06 +00:00
grant
7240d7d5bd add my copyright/license. 2005-01-12 22:58:29 +00:00
jlam
1fc74ae68e Correct my name to what I legally use. 2005-01-12 22:56:41 +00:00
tv
7ac5f2db44 Fix checking of up to date binary packages. The problem with DEPENDS
containing constructs of the form

    {perl>=5.8.3nb1,perl-thread>=5.8.3nb1}

is that a Bourne shell "for" interprets this as two separate items in the
list, nuking the {}s.  The above will never succeed and thus always cause
Perl-dependent packages to be rebuilt unconditionally.

Just adding more \s isn't enough -- the parsing of the depends list has to
be moved out of the "for" clause.  So, echo them one per line and use "read"
to pull them in without allowing the shell to peek at those characters.
2005-01-12 22:51:54 +00:00
recht
e03832caa6 set _PATCH_CAN_BACKUP=no to avoid problems with DragonFly's native patch.
Patch by Todd Willey in PR 28947
2005-01-12 21:11:41 +00:00
jlam
d6dfee8415 Use ${AWK} instead of a bare "awk". 2005-01-12 18:42:42 +00:00
jlam
ffd5ad15de Rearrange how LANGUAGES.<compiler> is set so that we can more flexibly
detect when only parts of a compiler toolchain are present, e.g. the
SunPro C compiler is installed, but not the Fortran-77 compiler.
2005-01-12 18:37:52 +00:00
jlam
c39637ec68 Put a TNF copyright on files that I authored. 2005-01-12 18:36:11 +00:00
jlam
d9168be817 Whitespace. 2005-01-12 17:40:48 +00:00