Commit Graph

1161 Commits

Author SHA1 Message Date
espie
303bd978e6 pattern I use a lot: start right after a given dir.
Easy to do.
2010-05-01 10:27:56 +00:00
ajacoutot
056a331311 Override MAKE in CONFIGURE_ENV when USE_GMAKE is set to Yes. Some ports
seem to need this more and more and it makes sense anyway.

tested in a bulk by jasper@, ok espie@
2010-04-24 09:54:42 +00:00
espie
3f48860b4c fix and recommit, my bad. 2010-04-20 21:00:35 +00:00
landry
9a3be8dc0f Revert to r1.991, this temporarly broke the portstree..
agreed by kili@ and sthen@
2010-04-20 19:58:09 +00:00
espie
09ed1dcb61 use _MAKE and _MAKESYS, simpler to read 2010-04-20 10:15:09 +00:00
espie
8565ca1737 provide a _MAKE internal that does the cd ${.CURDIR} && exec,
use it to avoid the obj/ problem in /usr/ports.
provide a _MAKESYS internal for bsd.port.mk with SYSTRACE too
2010-04-20 10:03:36 +00:00
espie
5b287b6cbf make sure SUBDIR doesn't escape into individual ports directory. 2010-04-20 10:02:39 +00:00
espie
8beda10cb7 fix logic, so that BROKEN-arch shows in the correct places
when MULTI_PACKAGES is set.
2010-04-17 10:16:11 +00:00
espie
51b217c20a forgot vax 2010-04-17 09:49:52 +00:00
espie
6d10bb1b49 add a list of ALL_ARCHS so that I can dump BROKEN-<arch> in dump-vars. 2010-04-12 13:08:20 +00:00
espie
39f6cafc40 fix buglet 2010-04-05 14:03:26 +00:00
espie
03931ebeb5 USE_GROFF framework for the mandoc switch 2010-04-05 14:02:50 +00:00
espie
c52b023249 don't know how this broke, but make it work again (and more sturdy against
multiple commas)

noticed by aja
2010-04-05 13:59:03 +00:00
espie
abbf59d6e3 typo, noticed by Andreas Kähäri <ak@ebi.ac.uk> 2010-03-22 20:19:12 +00:00
espie
a07ce53b70 use :Q to avoid any problem 2010-03-21 17:12:10 +00:00
espie
0a363398c3 ouch, pass TERM only if it's defined. Add TERMCAP while there. And do the same
thing to http_proxy/ftp_proxy, please...
2010-03-21 17:00:46 +00:00
ajacoutot
c4c7ece3ba Add a new APM_ARCHS variable.
comments from espie@ naddy@ jasper@
"looks good" jasper@
2010-03-21 11:34:38 +00:00
espie
cd50be5feb move to newer style -D for pkg_add. 2010-03-20 19:14:49 +00:00
espie
ae1f947e6d pass TERM to pkg_add so that Term::Cap will be able to use more capabilities 2010-03-20 19:14:06 +00:00
espie
feec1c7f7a introduce BROKEN-arch 2010-03-20 19:11:51 +00:00
espie
228149426e support for more stats in dpb, basically harmless 2010-03-05 07:49:29 +00:00
espie
a090480935 TRY_BROKEN, idea from FreeBSD (by popular demand, jasper + phessler) 2010-02-26 19:20:24 +00:00
espie
7a81866fd7 warn about FULLPKGNAME without FULLPKGPATH, as it is a bit difficult to spot
in complicated ports
2010-02-26 19:05:23 +00:00
espie
9e0b3d6c97 don't iterate on IGNORE, this breaks java's jrl message 2010-02-13 21:11:35 +00:00
espie
f7f5b4ff0a some important improvements to bsd.port.mk
- simplify IGNORE handling, always define it, even when NO_IGNORE is set,
and only use it for the targets it's meant to influence.
- make it possible to multiply-IGNORE a port, so stuff that's BROKEN and
NOT_FOR_ARCH will show up as both.
- DESCRIBE_TARGETs can die, since we no longer need special IGNORE dance
for them.
- add an IGNORE_IS_FATAL tweak, so that ignored ports can actually error out.
- reorg dump-vars to choose what to show, including new IGNORE, and COMES_WITH.
- name an explicit prepare step, which does check build dependencies and
install them (useful for build timing purposes)
- move the do-fetch test out, so that we can add pre-extract and do-extract
there eventually.
2010-02-12 12:00:19 +00:00
jasper
923a684afc - mention mips64el where applicable
ok espie@ ajacoutot@
2010-02-05 13:06:03 +00:00
espie
2a6cec5724 duplicate RUN_DEPENDS, it should be indexed only. 2010-01-28 10:09:06 +00:00
steven
191ff969ee allow USE_LIBTOOL to be set to 'gnu'; 'gnu' and 'yes' do the same for now.
ok espie
2009-10-14 13:01:03 +00:00
landry
fb2acb23e7 Change WRKOBJDIR (again...) to default to ${PORTSDIR}/pobj instead of obj,
it confuses make mirror-maker (among others..)
req'd by espie@, agreed by several@ in budapest.
2009-10-13 14:39:23 +00:00
simon
456bb1cca4 Finally add CONFIGURE_STYLE=modinst. This adds p5-Module-Install to
BUILD_DEPENDS, and --skipdeps to CONFIGURE_ARGS.  Additionally it nukes
existing ${WRKSRC}/inc/Module/*Install* files in pre-configure to force
regeneration of bundled and often outdated Module::Install versions.

feedback/testing by msf@, jim@ and landry@
2009-08-12 22:36:48 +00:00
espie
5552fa34d2 maintainer usability: some targets are usually not invoked recursively,
but they can (sometimes) be useful: add DANGEROUS knob to prevent running
them accidentally, but allow people to run makesum/update-plist/update-patches
recursively.
2009-07-26 12:14:05 +00:00
espie
ef63e25ff6 small mirror-maker tweaks:
- there's no need for :: on pure dependencies, : works just fine for
accumulating (and :: is not 100% standard on all systems)
- append directly to _FETCH_MAKEFILE_NAMES. This avoids an untidy blank space
2009-07-15 23:44:36 +00:00
espie
3ea520053a when using peek-ftp, I often start with make show=DISTFILES, so save myself
the trouble.
2009-07-13 12:21:44 +00:00
simon
e47a3e8294 let Module::Build tests behave like ExtUtils::MakeMaker ones
ok landry@, sthen@
2009-06-18 13:41:28 +00:00
landry
27251dbdc9 systrace isn't too happy when WRKOBJDIR contains a symlink, which can
happen often (/usr/ports NFS mounted & /usr/ports/obj a symlink to a
local dir, /usr/ports as a symlink to /home/wherever..)
Use readlink -fn only if USE_SYSTRACE is set so less things are run when
bsd.port.mk is included. Should make everyone happy.

Issue reported by and original fix ajacoutot@, sounds reasonable to espie@
2009-06-17 13:42:49 +00:00
sthen
df3fbaad7d set RECURSIVE_FETCH_LIST to No by default. ok sturm, ajacoutot, espie. 2009-06-12 17:26:51 +00:00
landry
861925b221 Provide saner defaults for WRKOBJDIR, using ${PORTSDIR}/obj. No more w-*
dirs forgotten all around. One can still unset it in /etc/mk.conf to get
back to the older behaviour, or still override it with /usr/obj/ports or
/usr/wobj or whatever. At least, now it can be a separate partition.
Most of us were already overriding this value, OpenBSD is all about sane
& simple default settings.
${PORTSDIR}/obj suggestion from jakemsr@.
Discussed at length with several, 'Works for me' krw@ ok todd@ wcmaier@.
2009-06-12 13:39:30 +00:00
espie
3cb8b96f11 don't export brutally everything from bsd.own.mk into each port's Makefile,
but select some variables instead.

checked by naddy@ on a bulk-build.

(apart from making stuff less easy to break, it also means Makefiles stop
inheriting a WIDE set of .PHONY targets, which is a good thing)
2009-06-09 17:46:58 +00:00
simon
f9808159d2 allow ports to set REGRESS_IS_INTERACTIVE=X11 in order to get rid of
redundant makefile parts found at many places in the tree

DISPLAY and XAUTHORITY variables/checks to be nuked soon from many ports

ok ajacoutot@, sthen@, jasper@
2009-05-16 22:18:50 +00:00
martynas
fb518f37e5 make "doesn't seem to exist on this system" and "Size matches for"
implicit;  since they filled the whole screen making important stuff
unnoticable.  ok espie@, sthen@
2009-05-05 20:58:38 +00:00
espie
3286c9fba0 remove trailing dot 2009-04-24 08:53:33 +00:00
espie
7112ac8117 add support for MATCHDIR: will only recurse into dirs that match the
pattern...
2009-04-11 14:56:31 +00:00
espie
b466c39b7a minor tweaks:
- totally forbid overriding do-fetch
- slightly better diagnostics in a few cases
2009-04-11 14:55:31 +00:00
ajacoutot
21e51b9ca9 - add old behaviour back where MODGNU_AUTOMAKE_DEPENDS is always set even
when not using CONFIGURE_STYLE=automake
this fixes ports depending on a specific version of automake while e.g
using the gnu CONFIGURE_STYLE

ok espie@
2009-03-14 10:46:56 +00:00
espie
5bcba4cec0 fix automake dependencies to read like automake->=1.4,<1.5 instead
of automake-1.4.*
2009-03-09 23:07:33 +00:00
espie
db7f3a147c say exactly WHAT was skipped, instead of giving the pattern 2009-01-24 11:32:43 +00:00
ajacoutot
126913162c - fix some unquoted shell expansions
this fixes an old bug where package creation would fail if one has some
particular file under PKGPATH because it would give pkg_create the wrong
args

from Wim Lewis on ports@, thank you!
"looks right" sthen@, "sounds reasonable" landry@, ok jasper@ espie@
2009-01-05 12:43:49 +00:00
espie
65e7dd9971 somewhat better error indications after some user feedback ;) 2008-11-18 11:45:41 +00:00
ajacoutot
3b893bbd83 - revert; USE_GCC* does not belong to the gcc* MODULES usage and
the man page has always been wrong about it

requested by and ok steven@
2008-10-28 13:29:05 +00:00
ajacoutot
c78568ec50 - reverse logic so that gcc3/4 is actually used when USE_GCC* is set
- set USE_GCC* to Yes by default when using a gcc* MODULE which makes
much more sense

discussed with and ok robert@
2008-10-28 12:26:39 +00:00