Re: cvs commit: ports/editors/emacs-devel Makefile distinfo ports/editors/emacs-devel/files patch-Makefile.in patch-configure patch-configure.in patch-doc-emacs-Makefile.in patch-doc-lispintro-Makefil
Ashish SHUKLA writes:
> ashish 2011-04-28 15:08:25 UTC
>
> FreeBSD ports repository
>
> Modified files:
> editors/emacs-devel Makefile distinfo
> editors/emacs-devel/files patch-Makefile.in
> patch-doc-emacs-Makefile.in
> patch-doc-lispintro-Makefile.in
> patch-doc-lispref-Makefile.in
> patch-doc-misc-Makefile.in
> Added files:
> editors/emacs-devel/files patch-configure.in
> Removed files:
> editors/emacs-devel/files patch-configure patch-src_config.in
> Log:
> - Update to bzr revision 104000.
- Any reason to use autogen.sh instead of relying on AUTOTOOLS?
- Why patch-src_config.in was lost and not converted to patch-configure.in?
- No love for IMAGEMAGICK knob since
http://docs.freebsd.org/cgi/mid.cgi?86y6ahzpjp.fsf
%%
Index: editors/emacs-devel/Makefile
===
RCS file: /a/.cvsup/ports/editors/emacs-devel/Makefile,v
retrieving revision 1.46
diff -u -p -r1.46 Makefile
--- editors/emacs-devel/Makefile28 Apr 2011 15:08:25 - 1.46
+++ editors/emacs-devel/Makefile30 Apr 2011 12:43:07 -
@@ -26,8 +26,10 @@ CONFLICTS= emacs-19.* emacs-21.* emacs-2
EMACS_VER= 24.0.50
EMACS_REV= 104000
-GNU_CONFIGURE= yes
+USE_AUTOTOOLS= autoconf autoheader aclocal automake
+ACLOCAL_ARGS= -Im4
+AUTOMAKE_ARGS= --add-missing
USE_GMAKE= yes
USE_XZ=yes
CPPFLAGS+= -I${LOCALBASE}/include
@@ -35,7 +37,6 @@ LDFLAGS+= -L${LOCALBASE}/lib
CONFIGURE_ENV+=CPPFLAGS="${CPPFLAGS}" LDFLAGS="${LDFLAGS}"
CONFIGURE_ARGS=--localstatedir=/var
WRKSRC=${WRKDIR}/${PORTNAME}-${PORTVERSION}
-USE_AUTOTOOLS= autoconf:env automake:env aclocal:env
EMACS_DIRS=${DATADIR}/${EMACS_VER}/leim ${DATADIR}/${EMACS_VER}/lisp
${DATADIR}/${EMACS_VER}/src \
${PREFIX}/libexec/${PORTNAME}
@@ -256,9 +257,6 @@ add-plist-data:
.endfor
@${ECHO} "@unexec rmdir %D/${DATADIR_REL}/${EMACS_VER} 2>/dev/null ||
true" >>${TMPPLIST}
-pre-configure:
- @cd ${WRKSRC} && ./autogen.sh
-
.if defined(WITH_SOURCES)
PLIST_SUB+=SOURCES=""
Index: editors/emacs-devel/files/patch-configure.in
===
RCS file: /a/.cvsup/ports/editors/emacs-devel/files/patch-configure.in,v
retrieving revision 1.6
diff -u -p -r1.6 patch-configure.in
--- editors/emacs-devel/files/patch-configure.in28 Apr 2011 15:08:25
- 1.6
+++ editors/emacs-devel/files/patch-configure.in30 Apr 2011 12:49:52
-
@@ -44,3 +44,12 @@ $FreeBSD: ports/editors/emacs-devel/file
case "$opsys" in
dnl defined __FreeBSD__ || defined __NetBSD__ || defined __linux__
gnu-linux|freebsd|netbsd)
+@@ -3547,6 +3547,8 @@ AH_BOTTOM([
+
+ #ifdef HAVE_ALLOCA_H
+ # include
++#elif defined __FreeBSD__
++#include
+ #elif defined __GNUC__
+ # define alloca __builtin_alloca
+ #elif defined _AIX
%%
___
cvs-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/cvs-all
To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/editors/emacs-devel Makefile distinfo ports/editors/emacs-devel/files patch-Makefile.in patch-configure patch-configure.in patch-doc-emacs-Makefile.in patch-doc-lispintro-Makefil
ash...@freebsd.org (Ashish SHUKLA) writes: > Pan Tsu writes: >> - Any reason to use autogen.sh instead of relying on AUTOTOOLS? > > Because autogen.sh works fine, and is provided by the upstream to configure > the sources. So, I prefer to use upstream, unless it behaves badly. It makes sure autotools use and dependencies are tracked properly by the ports framework. For example, autoreconf (from autogen.sh) invokes autoheader which you did not specify in AUTOTOOLS. >> - Why patch-src_config.in was lost and not converted to patch-configure.in? > > As I see, in src/config.in (excerpt is pasted below) which get generated > already includes "stdlib.h", so I don't see any need of the re-including it, > unless you think it causes some issue. Including stdlib.h makes sure both alloca(3) and _STDLIB_H_ are defined at that point and no prior or future explicit inclusion of stdlib.h redefines it. The warning described ports/149049 is still present $ make gcc46 -DHAVE_CONFIG_H -I. -I../src -I/work/emacs-24.0.50.104000/src -I/usr/local/include -O2 -pipe -fno-strict-aliasing -MT mktime.o -MD -MP -MF .deps/mktime.Tpo -c -o mktime.o mktime.c In file included from mktime.c:25:0: ../src/config.h:1355:0: warning: "alloca" redefined [enabled by default] /usr/local/lib/gcc46/gcc/x86_64-portbld-freebsd9.0/4.6.1/include-fixed/stdlib.h:250:0: note: this is the location of the previous definition mv -f .deps/mktime.Tpo .deps/mktime.P and stems from... > > --8<---cut here---start->8--- > /* Define if the compiler supports function prototypes. It may do so but >not define __STDC__ (e.g. DEC C by default) or may define it as zero. */ > #undef PROTOTYPES > > #include > #include > > #ifdef HAVE_ALLOCA_H > # include > #elif defined __FreeBSD__ > #include > #elif defined __GNUC__ > # define alloca __builtin_alloca this definition for gcc/clang/pcc/etc. > #elif defined _AIX > # define alloca __alloca > #else > # include > # ifdef __cplusplus > extern "C" > # endif > void *alloca (size_t); > #endif > --8<---cut here---end--->8--- As HAVE_STDLIB_H is gone since r101711 HAVE_ALLOCA_H can be completely ignored on FreeBSD. %% Index: editors/emacs-devel/files/patch-configure.in === RCS file: /a/.cvsup/ports/editors/emacs-devel/files/patch-configure.in,v retrieving revision 1.6 diff -u -p -r1.6 patch-configure.in --- editors/emacs-devel/files/patch-configure.in28 Apr 2011 15:08:25 - 1.6 +++ editors/emacs-devel/files/patch-configure.in1 May 2011 02:50:27 - @@ -44,3 +44,20 @@ $FreeBSD: ports/editors/emacs-devel/file case "$opsys" in dnl defined __FreeBSD__ || defined __NetBSD__ || defined __linux__ gnu-linux|freebsd|netbsd) +@@ -3550,6 +3550,7 @@ + #include + #include + ++#ifndef __FreeBSD__ + #ifdef HAVE_ALLOCA_H + # include + #elif defined __GNUC__ +@@ -3563,6 +3564,7 @@ + # endif + void *alloca (size_t); + #endif ++#endif + + #ifndef HAVE_STRCHR + #define strchr(a, b) index (a, b) + %% ___ cvs-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/cvs-all To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/x11/nvidia-settings Makefile distinfo
Alexander Nedotsukov writes: > bland 2011-05-22 09:57:36 UTC > > FreeBSD ports repository > > Modified files: > x11/nvidia-settings Makefile distinfo > Log: > Update to 270.41.19 > > Revision ChangesPath > 1.34 +3 -2 ports/x11/nvidia-settings/distinfo > --- ports/x11/nvidia-settings/distinfo2011/03/19 12:37:56 1.33 > +++ ports/x11/nvidia-settings/distinfo2011/05/22 09:57:36 1.34 > @@ -1,2 +1,3 @@ > -SHA256 (nvidia-settings-270.18.tar.bz2) = > 11a775763f3484e18513bfbf7d911e45028c5e8e1dcd7a8d0bfe7fc04d904fc9 > -SIZE (nvidia-settings-270.18.tar.bz2) = 1542267 > +MD5 (nvidia-settings-270.41.19.tar.bz2) = 521dcb2f1261c303a915e20ad528b44b > +SHA256 (nvidia-settings-270.41.19.tar.bz2) = > 1fef9f80d0d637b6bd18efd91053269f740a396dc065318b3eba37dc44be3aee > +SIZE (nvidia-settings-270.41.19.tar.bz2) = 1542314 MD5 have crept into distinfo. It was previously mass-removed by miwi@. ___ cvs-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/cvs-all To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/emulators Makefile ports/emulators/wine-gecko Makefile distinfo pkg-descr pkg-plist
Gerald Pfeifer writes:
> gerald 2011-05-15 15:29:30 UTC
>
> FreeBSD ports repository
>
> Modified files:
> emulatorsMakefile
> Added files:
> emulators/wine-gecko Makefile distinfo pkg-descr pkg-plist
> Log:
> Add the new wine-gecko port, that complements the existing wine port.
>
> Wine implements its own version of Internet Explorer which is based on
> a custom version of Mozilla's Gecko Layout Engine. This engine is down-
> loaded automatically the first time a web page is loaded unless, as done
> by this port, the engine is pre-installed on the user's system.
>
> PR: 156001
> Submitted by: David Naylor
Does wine find gecko with DATADIR != PREFIX/share/wine ? Please, either
document it in the pkg-message that .msi file should be copied/symlinked
from PREFIX/wine-gecko/wine/gecko/*.msi or redefine DATADIR properly.
%%
Index: emulators/wine-gecko/Makefile
===
RCS file: /a/.cvsup/ports/emulators/wine-gecko/Makefile,v
retrieving revision 1.1
diff -u -p -r1.1 Makefile
--- emulators/wine-gecko/Makefile 15 May 2011 15:29:30 - 1.1
+++ emulators/wine-gecko/Makefile 23 May 2011 20:27:04 -
@@ -18,12 +18,15 @@ COMMENT=Gecko Layout Engine for Wine (H
NO_BUILD= yes
PLIST_SUB+=GECKO_VERSION="${DISTVERSION}"
+DATADIR= ${PREFIX}/share/wine
+PLIST_FILES= %%DATADIR%%/gecko/wine_gecko-%%GECKO_VERSION%%-x86.msi
+PLIST_DIRSTRY= %%DATADIR%%/gecko %%DATADIR%%
LICENSE= MPL GPLv2 GPLv3 LGPL21 LGPL3
LICENSE_COMB= multi
do-install:
- ${MKDIR} ${DATADIR}/wine/gecko
- ${INSTALL_DATA} ${DISTDIR}/wine_gecko-${DISTVERSION}-x86.msi
${DATADIR}/wine/gecko
+ ${MKDIR} ${DATADIR}/gecko
+ ${INSTALL_DATA} ${DISTDIR}/wine_gecko-${DISTVERSION}-x86.msi
${DATADIR}/gecko
.include
Index: emulators/wine-gecko/pkg-plist
===
RCS file: emulators/wine-gecko/pkg-plist
diff -N emulators/wine-gecko/pkg-plist
--- emulators/wine-gecko/pkg-plist 15 May 2011 15:29:30 - 1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,4 +0,0 @@
-%%DATADIR%%/wine/gecko/wine_gecko-%%GECKO_VERSION%%-x86.msi
-@dirrm %%DATADIR%%/wine/gecko
-@dirrmtry %%DATADIR%%/wine
-@dirrmtry %%DATADIR%%
%%
___
cvs-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/cvs-all
To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/emulators Makefile ports/emulators/wine-gecko Makefile distinfo pkg-descr pkg-plist
Gerald Pfeifer writes:
> On Tue, 24 May 2011, Alexey Dokuchaev wrote:
>> I believe DATADIR is also mishandled in main wine port: that is, extra
>> "wine" word crept in (paths look like /usr/local/share/wine/wine/fonts/...
>> for example).
>
> I've been wondering about that, too, but then it's been working for
> years and
>
> CONFIGURE_ARGS+=--datadir=${DATADIR} --verbose ...
>
> looks pretty innocent, doesn't it? Or is the concept of FreeBSD's
> ${DATADIR} so different from GNU's --datadir?
Can you prune it at the next update? It's bogus unless GNU_CONFIGURE's
--datadir was *then* not equal ${prefix}/share by default. For a new
user coming from pkgsrc et al PREFIX/share/wine/wine can be confusing.
Looking at bsd.port.mk *a user* changing DATADIR or any other dir in
hier(7)-like directory tree under PREFIX was never supported by most
ports, only PREFIX itself.
___
cvs-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/cvs-all
To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/sysutils/tmux Makefile ports/sysutils/tmux/files extra-patch-tty-keys.c patch-tty-keys.c
Wen Heping writes:
[...]
> Log:
> - Add an option to use patch-tty-key.c [2]
[...]
> Revision ChangesPath
> 1.1 +11 -0 ports/sysutils/tmux/files/extra-patch-tty-keys.c (new)
> 1.2 +0 -11 ports/sysutils/tmux/files/patch-tty-keys.c (dead)
Wouldn't providing alternate backspace character be better?
%%
Index: sysutils/tmux/Makefile
===
RCS file: /a/.cvsup/ports/sysutils/tmux/Makefile,v
retrieving revision 1.28
diff -u -p -r1.28 Makefile
--- sysutils/tmux/Makefile 16 Jun 2011 08:42:28 - 1.28
+++ sysutils/tmux/Makefile 17 Jun 2011 13:52:15 -
@@ -26,7 +26,7 @@ PORTEXAMPLES= *
OPTIONS= KQUEUE "Build without define HAVE_BROKEN_KQUEUE " On \
LIBEVENT2 "Use libevent version 2" Off \
LIBEVENT_STATIC "Build with static libevent" Off \
- BACKSPACE "Build with tty/keys patch" Off
+ BACKSPACE "Use BS for termios(4) ERASE (default is DEL)"
Off
# Now I set tmux build without #define HAVE_BROKEN_KQUEUE as default and an
option,
# If it still hang the system or other run error, try as upstream suggest:
@@ -73,7 +73,7 @@ EXTRA_PATCHES=${PATCHDIR}/extra-patch-f
.endif
.if defined(WITH_BACKSPACE)
-EXTRA_PATCHES= ${PATCHDIR}/extra-patch-tty-keys.c
+EXTRA_PATCHES= ${PATCHDIR}/extra-patch-bspace
.endif
post-patch:
Index: sysutils/tmux/files/extra-patch-bspace
===
RCS file: sysutils/tmux/files/extra-patch-bspace
diff -N sysutils/tmux/files/extra-patch-bspace
--- /dev/null 1 Jan 1970 00:00:00 -
+++ sysutils/tmux/files/extra-patch-bspace 17 Jun 2011 13:46:00 -
@@ -0,0 +1,26 @@
+Index: input-keys.c
+===
+--- input-keys.c (revision 2525)
input-keys.c (working copy)
+@@ -41,7 +41,7 @@
+
+ const struct input_key_ent input_keys[] = {
+ /* Backspace key. */
+- { KEYC_BSPACE, "\177", 0 },
++ { KEYC_BSPACE, "\010", 0 },
+
+ /* Function keys. */
+ { KEYC_F1, "\033OP", 0 },
+Index: window.c
+===
+--- window.c (revision 2525)
window.c (working copy)
+@@ -634,7 +634,7 @@
+ fatal("tcgetattr failed");
+ if (tio != NULL)
+ memcpy(tio2.c_cc, tio->c_cc, sizeof tio2.c_cc);
+- tio2.c_cc[VERASE] = '\177';
++ tio2.c_cc[VERASE] = '\010';
+ if (tcsetattr(STDIN_FILENO, TCSANOW, &tio2) != 0)
+ fatal("tcgetattr failed");
+
Index: sysutils/tmux/files/extra-patch-tty-keys.c
===
RCS file: sysutils/tmux/files/extra-patch-tty-keys.c
diff -N sysutils/tmux/files/extra-patch-tty-keys.c
--- sysutils/tmux/files/extra-patch-tty-keys.c 25 May 2011 00:29:33 -
1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,11 +0,0 @@
tty-keys.c.orig2010-06-04 08:45:52.0 +0800
-+++ tty-keys.c 2010-06-04 08:46:45.0 +0800
-@@ -456,8 +456,6 @@
-* used. termios should have a better idea.
-*/
- bspace = tty->tio.c_cc[VERASE];
-- if (bspace != _POSIX_VDISABLE && key == bspace)
-- key = KEYC_BSPACE;
- goto handle_key;
- }
-
%%
___
cvs-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/cvs-all
To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/games/wesnoth-devel Makefile
Pav Lucistnik writes:
> pav 2011-07-01 20:59:46 UTC
>
> FreeBSD ports repository
>
> Modified files:
> games/wesnoth-devel Makefile
> Log:
> - Mark BROKEN: incomplete plist
>
> Reported by:pointyhat
Can you try the following diff?
%%
Index: games/wesnoth-devel/Makefile
===
RCS file: /a/.cvsup/ports/games/wesnoth-devel/Makefile,v
retrieving revision 1.98
diff -u -p -r1.98 Makefile
--- games/wesnoth-devel/Makefile1 Jul 2011 20:59:46 - 1.98
+++ games/wesnoth-devel/Makefile1 Jul 2011 22:48:43 -
@@ -15,10 +15,8 @@ PKGNAMESUFFIX= -devel
MAINTAINER=phi...@freebsd.org
COMMENT= A fantasy turn-based strategy game
+EXTRACT_DEPENDS=${WRKSRC}/fonts/Junicode-Regular.ttf:${PORTSDIR}/x11-fonts/junicode:extract
LIB_DEPENDS= boost_thread.4:${PORTSDIR}/devel/boost-libs
-RUN_DEPENDS= ${NONEXISTENT}:${PORTSDIR}/x11-fonts/junicode:extract
-
-BROKEN=incomplete plist
LICENSE= GPLv2 GPLv3
LICENSE_COMB= dual
@@ -164,10 +162,12 @@ CMAKE_ARGS+= -DDOCDIR=none
IGNORE=game data is required, undefine NOPORTDATA
.endif
-post-install:
+post-extract: .SILENT
${INSTALL_DATA} $$(${MAKE} -V WRKSRC \
-C ${PORTSDIR}/x11-fonts/junicode)/fonts/Junicode-Regular.ttf \
- ${DATADIR}/fonts
+ ${WRKSRC}/fonts
+
+post-install:
-update-desktop-database ${PREFIX}/share/applications
.include
%%
___
cvs-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/cvs-all
To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/lang/sbcl Makefile
Stanislav Sedov writes: > stas2011-07-11 21:25:37 UTC > > FreeBSD ports repository > > Modified files: > lang/sbclMakefile > Log: > - Comment out THREADS option for now, sbcl fails to pass tests with threads > on. Why not explicitly mark it BROKEN rather than silently build without THREADS during BATCH mode? ___ cvs-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/cvs-all To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
Re: cvs commit: ports/Mk bsd.licenses.mk bsd.port.mk
Thomas Abthorpe writes: > tabthorpe2011-03-31 03:37:14 UTC > > FreeBSD ports repository > > Modified files: > Mk bsd.licenses.mk bsd.port.mk > Log: > - Disable license checking with DISABLE_LICENSES [1] > - Remove temp files from /tmp when build is interrupted [2] > > PR: ports/149837 [1], ports/155890 [2] ENOTIME? At least reopen ports/155890, I've sent an immediate followup back in 31 Mar. ___ cvs-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/cvs-all To unsubscribe, send any mail to "cvs-all-unsubscr...@freebsd.org"
