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
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).
2004-09-13 Mark McLoughlin <mark@skynet.ie>
* configure.in: Version 2.12.0.
2004-08-25 Michael Haubenwallner <michael.haubenwallner@salomon.at>
* src/orb/orb-core/allocators.c (ORBit_freekids_via_TypeCode_T):
Step over align-holes at the end of structures (bug#151023)
* src/orb/orb-core/corba-any.c
(ORBit_marshal_value) (ORBit_demarshal_value)
(ORBit_copy_value_core) (ORBit_value_equivalent):
Ditto.
* test/everything/everything.idl
test/everything/client.c
test/everything/structServer.c:
Added testcase for bug#151023, align-holes at the end of structures.
* test/everything/constants.h:
Ditto.
Added constants for CHAR and OCTET.
2004-08-19 Michael Meeks <michael@ximian.com>
* Version 2.11.2
* src/orb/GIOP/giop-recv-buffer.c
(giop_recv_list_destroy_queue_entry): unref the
ent's connection without the message queue lock
held.
* test/everything/client.c: bin stderr printfs
during thread tests, to avoid unhelpful rescheduling.
* src/orb/orb-core/orbit-small.c
(ORBit_small_unlisten_for_broken_full): use the
peek method, no need to try re-connecting just to
disconnect the 'broken' signal.
* src/orb/orb-core/corba-object.c
(CORBA_Object_release_cb): after a long audit;
unlock / lock around the giop connection unref;
this avoids the final connection release
deadlocking vs. a pending call in the I/O
thread needing this lock.
(ORBit_object_peek_connection): impl.
2004-08-19 Magnus Therning <magnus@therning.org>
* src/idl-compiler/orbit-idl-backend.c: fix
bit-field assignments.
2004-08-10 Michael Meeks <michael@ximian.com>
* test/everything/basicServer.c
(BasicServer_testBoolString): impl.
2004-08-11 Kjartan Maraas <kmaraas@gnome.org>
* src/idl-compiler/orbit-idl-c-backend.c: (orbit_idl_output_c):
0 vs NULL.
* src/idl-compiler/orbit-idl-c-common.c: (cc_output_typecodes):i
Same here.
* src/idl-compiler/orbit-idl-c-utils.c:
(orbit_cbe_get_typespec_str): Same here too.
* src/orb/dynamic/dynany.c: (DynamicAny_DynUnion_member_name):
0 vs NULL and add some casts.
* src/orb/orb-core/allocators.c: (ORBit_freekids_via_TypeCode_T):
0 vs NULL fixes.
* src/orb/orb-core/corba-object.c: (ORBit_objref_find):
Same again.
* src/services/name/boot.c: (main): And again.
* test/echo-srv.c: (echo_srv_finish_object), (echo_srv_finish_poa):
And again.
2004-08-10 Mark McLoughlin <mark@skynet.ie>
* src/orb/orb-core/iop-profiles.c: (IOP_start_profiles),
(IOP_TAG_CODE_SETS_demarshal): don't initialize the sequence
using structure initialization, but rather zero the structure
contents. Fixes bug #147255, which admittedly is really a
gcc on powerpc bug, but this code is nicer anyway.
2004-08-08 Gustavo J. A. M. Carneiro <gustavo@users.sourceforge.net>
* src/orb/orb-core/corba-any.c (ORBit_sequence_remove):
Impl. method to remove a single element in an ORBit sequence.
* include/orbit/orb-core/orbit-small.h: Add ORBit_sequence_remove
prototype.
* test/everything/client.c (testSequenceHelpers): Test
ORBit_sequence_remove.
2004-07-30 Mark McLoughlin <mark@skynet.ie>
* test/test-corbaloc.c: "\%" isn't a valid escape sequence,
so just use "%". From d.binderman@virgin.net.
2004-07-07 Frank Rehberger <frehberg@mole.xtradyne.de>
* Revert to Version 2.11.x and Add Version ORBit2-2.11.1
(developer release only)
2004-07-07 Frank Rehberger <frehberg@gnome-de.org>
* Add Version ORBit2-2.12.0
2004-06-22 Chris Landrieu <perotin101@yahoo.com>
* src/orb/orb-core/allocators.c (ORBit_freekids_via_TypeCode_T):
add missing pre-align for struct/except types. Complicate the
union alignment rules.
2004-05-30 Christophe Saout <christophe@saout.de>
* src/orb/orb-core/corba-any.c:
(ORBit_marshal_value, ORBit_demarshal_value): Fix broken wstring
support. Marshal/demarshal the correct number of bytes on the
wire and negotiate the UTF-16 byte ordering using an optional
endianess marker. Also works now with JacORB.
2004-05-30 Christophe Saout <christophe@saout.de>
* src/services/imodule/orbit-imodule-utils.c
(ORBit_imodule_get_typecode): set c_align in nested array
typecodes that are not handled by ORBit_imodule_find_c_align.
2004-05-17 Michael Meeks <michael@ximian.com>
* ORBit-2.0.pc.in: prune pkgconfig cruft.
2004-05-14 Fernando Herrera <fherrera@onirica.com>
* src/idl-compiler/orbit-idl-c-backend.c: (out_for_pass):
clobber LC_ALL because cpp tranlates some strings in some
locales during dep files generation that makes compilation
fails. Fixes bug #142546.
2004-05-08 Frank Rehberger <frehberg@gnome-de.org>
* src/orb/orb-core/iop-profiles.c (IOP_start_profiles): orbit
generated profiles contain full codeset data structures
(IOP_component_free): freeing codeset structures
(CodeSetComponent_marshal): marshal codeset structures
(CodeSetComponent_demarshal): demarshal codeset structures
* src/orb/orb-core/iop-profiles.h (IOP_PROFILES_CODE_SET_*):
declare known code-sets, but UTF8 and UTF16 supported only.
(struct IOP_TAG_CODE_SETS_info): complete CodeSet data structure
using CONV_FRAME_CodeSetComponentInfo.
* src/orb/orb-core/corba-orb.c: variable orbit_use_corbaloc must
be of type gboolean.
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.
