Date: Fri, 28 Aug 2015 10:33:05 +0000 From: bugzilla-noreply@freebsd.org To: freebsd-ports-bugs@FreeBSD.org Subject: [Bug 202404] [MAINTAINER] multimedia/mplayer2: updates to Makefile Message-ID: <bug-202404-13-Rqe4XvPW1m@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-202404-13@https.bugs.freebsd.org/bugzilla/> References: <bug-202404-13@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D202404 --- Comment #37 from Jan Beich <jbeich@FreeBSD.org> --- Comment on attachment 160430 --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=3D160430 v9 Let me try to enumerate what haven't landed yet. I wonder if a single blob commit is OK. - Expose A52, ENCA, DVDREAD, DVDNAV, RTCPU as options - Hide WIN32 option on non-i386 architectures - Drop no longer supported OSVERSION range - Drop DEBUG and PACKAGE_BUILDING hacks - Convert to option helpers - Sort options > + --disable-liba52 \ I thought you've converted it into A52 option since v6. > -ENCA_LIB_DEPENDS=3D libenca.so:${PORTSDIR}/converters/enca > -ENCA_CONFIGURE_OFF=3D --disable-enca Why remove since v8? mplayer2 directly links against libenca with ASS=3Don due to libass package having ENCA=3Don by default. Hidden dependencies make build unpredictable and may cause issues e.g., - libenca ABI changed -> bump PORTREVISION of consumers - libass is rebuilt with ENCA=3Doff then libenca can be pruned by |pkg autoremove| Maybe also make ENCA=3Don by default to keep the feature intact for users of freebsd.org packages. >+OPTIONS_DEFINE=3D A52 ASS CACA DEBUG DV DVDREAD DVDNAV GIF IPV6 JACK LADS= PA \ >+ LIBBLURAY LIBCDIO LIRC MAD OPENGL PORTAUDIO PULSEAUDIO \ >+ REALPLAYER RTC RTCPU SDL SMB SPEEX THEORA V4L VDPAU X11 XINERAMA The option is guarded in configure, so you need to provide it only for specific architectures. And it may need to be enabled by default to keep the feature for packages users intact. if test "$_runtime_cpudetection" =3D yes && ! x86 && ! ppc; then die "Runtime CPU detection only works for x86, x86-64 and PPC!" fi See below WIN32 example but ignore !PACKAGE_BUILDING guard. > .if !defined(PACKAGE_BUILDING) >+OPTIONS_DEFINE_i386+=3D WIN32 >+OPTIONS_DEFAULT_i386+=3D WIN32 > .endif > -WIN32_DESC=3D Win32 codec pack support > +WIN32_DESC=3D Enable win32 codec Why?=20 "Enable" is a noise word and can be removed from local descriptions for other options as well (as part of options cleanup). "win32" is not a codec but a "set" or "pack" of them. Original description used "codec set" but "codec pack" is more popular term on Windows. WIN32_DESC=3D Enable win32 codec set on the IA32 arch > +LIBCDIO_DESC=3D Enable libcdio support [...] > +LIBCDIO_LIB_DEPENDS=3D libcdio_paranoia.so:${PORTSDIR}/sysutils/libcdio-= paranoia > +LIBCDIO_CONFIGURE_OFF=3D --disable-libcdio I've renamed LIBCDIO to CDIO in ports r395434. This was supposed to be part of your v6 as CDIO_DESC is already defined in Mk/bsd.options.desc.mk. Please, rebase carefully. >-.if defined(WITH_KERN_HZ) >-DEFAULT_KERN_HZ=3D${WITH_KERN_HZ} >-.else >-DEFAULT_KERN_HZ=3D1024 >-.endif WITH_KERN_HZ still referenced later in Makefile and would break build if the assignment not restored or the reference not removed. And per comment 24 the conditional will work just fine without bsd.port.options.mk line. mplayer.c:4082:34: error: expected expression unsigned long irqp =3D ; /* 512 seemed OK. 128 is jerky. */ ^ Note, if you're removing a feature provide rationale to be used in commit message. It doesn't need to be more than one sentence long but should explain "why" and not "what" was done. (In reply to Carlos J Puga Medina from comment #33) > OK, I need to sanitize this before remove .include <bsd.port.options.mk> = line. > > .if ${PORT_OPTIONS:MRTC} Use target option helpers. They're documented in /usr/ports/CHANGES from 20150701 or https://www.freebsd.org/doc/en/books/porters-handbook/makefile-options.html= #options-targets-on --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-202404-13-Rqe4XvPW1m>