This changes the buildlink3.mk files to use an include guard for the
recursive include. The use of BUILDLINK_DEPTH, BUILDLINK_DEPENDS,
BUILDLINK_PACKAGES and BUILDLINK_ORDER is handled by a single new
variable BUILDLINK_TREE. Each buildlink3.mk file adds a pair of
enter/exit marker, which can be used to reconstruct the tree and
to determine first level includes. Avoiding := for large variables
(BUILDLINK_ORDER) speeds up parse time as += has linear complexity.
The include guard reduces system time by avoiding reading files over and
over again. For complex packages this reduces both %user and %sys time to
half of the former time.
and add a new helper target and script, "show-buildlink3", that outputs
a listing of the buildlink3.mk files included as well as the depth at
which they are included.
For example, "make show-buildlink3" in fonts/Xft2 displays:
zlib
fontconfig
iconv
zlib
freetype2
expat
freetype2
Xrender
renderproto
same as the gzip compressed one used previously (and the one from the
MASTER_SITE which is 2 bytes longer!?!).
Thanks to rui for pointing out where to get it the "old" gzip tarball.
RECOMMENDED is removed. It becomes ABI_DEPENDS.
BUILDLINK_RECOMMENDED.foo becomes BUILDLINK_ABI_DEPENDS.foo.
BUILDLINK_DEPENDS.foo becomes BUILDLINK_API_DEPENDS.foo.
BUILDLINK_DEPENDS does not change.
IGNORE_RECOMMENDED (which defaulted to "no") becomes USE_ABI_DEPENDS
which defaults to "yes".
Added to obsolete.mk checking for IGNORE_RECOMMENDED.
I did not manually go through and fix any aesthetic tab/spacing issues.
I have tested the above patch on DragonFly building and packaging
subversion and pkglint and their many dependencies.
I have also tested USE_ABI_DEPENDS=no on my NetBSD workstation (where I
have used IGNORE_RECOMMENDED for a long time). I have been an active user
of IGNORE_RECOMMENDED since it was available.
As suggested, I removed the documentation sentences suggesting bumping for
"security" issues.
As discussed on tech-pkg.
I will commit to revbump, pkglint, pkg_install, createbuildlink separately.
Note that if you use wip, it will fail! I will commit to pkgsrc-wip
later (within day).
* Fixed bug in ExpOutputProc that caused misbehavior during partial writes.
* Noted that expect_background ignores timeouts. Added to documentation.
* Patch for "mkpasswd -vo".
* Removed beta designation.
* Daniel A. Steffen <steffen@ics.mq.edu.au> provided patch for
MacOS to avoid panic-redefinition.
Changes 5.41.0:
* Simon Taylor <simon@unisolve.com.xau> provided fix for
interact -o which was completely broken by 5.40.1.
Changes 5.40.1:
* Added scroll support to official tkterm. Copied all fixes
from/to term_expect to/from tkterm.
* Kiran Madabhushi <maskiran@hotmail.xcom> encountered interact
diagnostics incorrectly pointing to expect_background. Also,
found multiple -o flags behaving unexpectedly. Added diag.
* Kristoffer Eriksson <ske@pkmab.xse> noted typo in SIMPLE code
in exp_inter.c. However, this is extremely unlikely to affect
any machines.
* Reinhard Max <max@suse.xcom> noted that "make test" failed when
run in the background. The log testcase was testing the
send_tty command. Added code in both Expect and in the test
to handle this.
in the process. (More information on tech-pkg.)
Bump PKGREVISION and BUILDLINK_DEPENDS of all packages using libtool and
installing .la files.
Bump PKGREVISION (only) of all packages depending directly on the above
via a buildlink3 include.
All library names listed by *.la files no longer need to be listed
in the PLIST, e.g., instead of:
lib/libfoo.a
lib/libfoo.la
lib/libfoo.so
lib/libfoo.so.0
lib/libfoo.so.0.1
one simply needs:
lib/libfoo.la
and bsd.pkg.mk will automatically ensure that the additional library
names are listed in the installed package +CONTENTS file.
Also make LIBTOOLIZE_PLIST default to "yes".
by moving the inclusion of buildlink3.mk files outside of the protected
region. This bug would be seen by users that have set PREFER_PKGSRC
or PREFER_NATIVE to non-default values.
BUILDLINK_PACKAGES should be ordered so that for any package in the
list, that package doesn't depend on any packages to the left of it
in the list. This ordering property is used to check for builtin
packages in the correct order. The problem was that including a
buildlink3.mk file for <pkg> correctly ensured that <pkg> was removed
from BUILDLINK_PACKAGES and appended to the end. However, since the
inclusion of any other buildlink3.mk files within that buildlink3.mk
was in a region that was protected against multiple inclusion, those
dependencies weren't also moved to the end of BUILDLINK_PACKAGES.
Fixes PR 20055 from Robert Elz. Updates tk-expect to version 5.38.
Also remove the pre-configure step from tk-expect that was removed from
tcl-expect a year ago.
Also use the standard test target rather than an explict one.
7/18/02 5.38.0 At request of Hugh Sasse <hgs@dmu.ac.uk> added md5 hash of gz
to homepage.
Dave Schooler <dave@stashtea.com> reported that send -s wasn't
handling certains chars correctly. Turned out to be those
that had multibyte UTF8 reps. send -s was just pumping out
hunks of bytes without regard to UTF boundaries and evidentally
Tcl's I/O engine thought that it should translate a partial
UTF8 character into, uh, something else.
Curt Shroeder <c.schroeder@computer.org> fixed bug in rftp - a
a filename looked enough like a 3-digit diagnostic that the
script got confused.
4/16/02 5.37.2 Multixterm couldn't find man page all the time.
4/16/02 5.37.1 Made multixterm handle user-supplied args.
4/15/02 5.37.0 Added multixterm to example directory.
4/8/02 5.36.1 Backed out CONST qualifiers. Too much trouble with older
versions of Tcl. I'll let someone else worry about them.
4/8/02 5.36.0 Made first cut at multixterm, a replacement for crlogin.
Fixed bug in background handler. If an action waited on the
same spawn id, esPtr would become invalidated.
Ryan Schmidt <rschmidt@mac.com> noted configure didn't
recognize MacOS X. Downloaded new config.guess.
Andreas Kupries <andreask@activestate.com> provided CONST
patches to accomodate Tcl changes per TIP 27.
2/25/02 5.35.0 Joe Eggleston <joe@arbor.net> noted bug in full_buffer test.
The test hadn't been I18'd properly and was testing chars
instead of bytes. Also fixed diagnostics so it printed when
it was testing full buffer even if there wasn't one.
2/7/02 5.34.1 Bruce Hartweg <brhartweg@bigfoot.com> noted that direct spawn
ids were not being tested so something like "expect -i exp9999"
would dump core. Evidentally a bug from the 5.31 transition.
12/20/01 5.34.0 Don Porter <don.porter@nist.gov> provided package-related
fixes for test suite.
Brian Theado <brian.theado@usa.net> noted that interact's -re
support broke when offsets kicked in. Turned out that the
regexp engine supports them during execution but the results
are delivered RELATIVE to the offset. (I suspect this was done
due to expediency.)
Changes from 5.32.1 (the last pkgsrc version):
10/1/01 5.33.0 <mark@doradosoftware.com> found that expect's diagnostics
didn't include the "no" after testing for a full buffer.
Hemang Lavana <hlavana@cisco.com> noted that "debug" (Dbg_On)
calls didn't always force the debugger into step mode.
Martin Kammerhofer <dada@sbox.tugraz.at> noted that the man
page neglected to document interpreter -eof.
Chris Clare <clarec@nortelnetworks.com> provided fix for
multiple decl in C lib.
Sheng Wang <wangs@sh.bel.alcatel.be> found interact's
can-match code had broken. It was missing the special hook
that Henry had added just for this purpose. How strange.
Dieter Fiebelkorn <dieter@fiebelkorn.net> requested addition
to config.guess for Power*Macintosh:Darwin for MacOSX.
Aside - to download latest config.guess:
cvs -d :pserver:anoncvs@subversions.gnu.org:/cvs checkout \
config
Added pipeline example to unbuffer man page.
8/4/00 5.32.2 Allen J. Newton <anewton@alturia.fleet.org> provided code for
generating passwords with special characters in mkpasswd.
Brent Welch <welch@ajubasolutions.com> changed the fix1line
install script so that "autoexpect" and other scripts that
get installed into the platform-independent bin directory
generically invoke "expect" from the users PATH instead
of hardwiring the platform-specific expect pathname.
foo-* to foo-[0-9]*. This is to cause the dependencies to match only the
packages whose base package name is "foo", and not those named "foo-bar".
A concrete example is p5-Net-* matching p5-Net-DNS as well as p5-Net. Also
change dependency examples in Packages.txt to reflect this.
out of date - it was based on a.out OBJECT_FMT, and added entries in the
generated PLISTs to reflect the symlinks that ELF packages uses. It also
tried to be clever, and removed and recreated any symbolic links that were
created, which has resulted in some fun, especially with packages which
use dlopen(3) to load modules. Some recent changes to our ld.so to bring
it more into line with other Operating Systems also exposed some cracks.
+ Modify bsd.pkg.mk and its shared object handling, so that PLISTs now contain
the ELF symlinks.
+ Don't mess about with file system entries when handling shared objects in
bsd.pkg.mk, since it's likely that libtool and the BSD *.mk processing will
have got it right, and have a much better idea than we do.
+ Modify PLISTs to contain "ELF symlinks"
+ On a.out platforms, delete any "ELF symlinks" from the generated PLISTs
+ On ELF platforms, no extra processing needs to be done in bsd.pkg.mk
+ Modify print-PLIST target in bsd.pkg.mk to add dummy symlink entries on
a.out platforms
+ Update the documentation in Packages.txt
With many thanks to Thomas Klausner for keeping me honest with this.
