This is in media-video/mjpegtools/mjpegtools-1.6.0-r7.ebuild # Portage currently chokes on the following nested conditional. # SRC_URI="mirror://sourceforge/mjpeg/${P}.tar.gz # quicktime? ( !alpha? ( # mirror://sourceforge/mjpeg/quicktime4linux-1.4-patched.tar.gz # ) )" if [ "$ARCH" = alpha ]; then SRC_URI="mirror://sourceforge/mjpeg/${P}.tar.gz" else SRC_URI="mirror://sourceforge/mjpeg/${P}.tar.gz quicktime? ( http://download.sourceforge.net/mjpeg/quicktime4linux-1.4-patched.tar.gz )" fi Reproducible: Always Steps to Reproduce: 1. Uncomment the SRC_URI with the nested conditional 2. emerge mjpegtools Actual Results: # emerge mjpegtools Calculating dependencies ...done! >>> emerge (1 of 1) media-video/mjpegtools-1.6.0-r7 to / >>> Downloading !alpha? --15:22:32-- http://!alpha/? => `/usr/distfiles/index.html' Resolving www-proxy.zko.dec.com... done. Connecting to www-proxy.zko.dec.com[16.47.32.60]:8080... connected. Proxy request sent, awaiting response... 502 Proxy Error ( Host was not found. For more information about this event, see ISA Server Help. ) 15:22:37 ERROR 502: Proxy Error ( Host was not found. For more information about this event, see ISA Server Help. ). !!! Couldn't download !alpha?. Aborting. Expected Results: It should have skipped downloading quicktime4linux since I'm on an alpha.
You cannot wrap ANYTHING in bash condidtionals that portage needs to use. Portage _CANNOT_ figure that out. SRC_URI="alpha? ( http://blah1.alpha : http://blah1.notalpha )"
Carpaski, I think you missed the point of this bug. The bash conditionals were just supposed to be a workaround (sorry if a bad one). The issue is that SRC_URI isn't working with nested conditionals, as far as I can tell. Here is what I'm trying to do: SRC_URI="mirror://sourceforge/mjpeg/${P}.tar.gz quicktime? ( !alpha? ( mirror://sourceforge/mjpeg/quicktime4linux-1.4-patched.tar.gz ) )" Portage barfs as described earlier in this bug report. I've also tried this format: SRC_URI="mirror://sourceforge/mjpeg/${P}.tar.gz alpha? ( : quicktime? ( mirror://sourceforge/mjpeg/quicktime4linux-1.4-patched.tar.gz ) )" In this case it doesn't unpack the second source on x86. How should I accomplish this correctly? Thanks, Aron
Oh. Completely ignored the comments. :-/ Ok. It'll be fixed for portage-2.1... not sure about before then. I'll look at it for 47-r4.
I worked around this for the moment by downloading the second source file on alpha even though it won't be built on that platform.
maybe it's just me, but why cant we consolidate the 2 functions used by DEPEND and SRC_URI into the same function ? that way when we add features to one, both will get it ? plus, the DEPEND one is more advancded ...
This is a feature really needed for maintaining all JDK's (blackdown,sun,ibm,....)
ill keep bugging portage team until this gets resolved =)
http://dev.gentoo.org/~drobbins/SRC_URI-fix.diff Bugs like this need to get closed *much* quicker. The functions are all there. They just need to get called. No new code was required (just a couple of different function calls). I'll commit this fix to cvs so it should show up in the next release.
This is still not fixed in 2.0.49-r18.
This bug was fixed in .50pre19 at least. But .49* is not fixed. carpaski: do we need to fix it in .49*?
supposed to be fixed in 2.0.50 which is stable now. If this bug is not fixed please reopen.
Sorry not fixed yet. bash-2.05b# emerge info Portage 2.0.50-r1 (default-x86-1.4, gcc-3.3.2, glibc-2.3.2-r9, 2.6.1-gentoo) ================================================================= System uname: 2.6.1-gentoo i686 Celeron (Coppermine) Gentoo Base System version 1.4.3.10 distcc 2.11.1 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled] Autoconf: sys-devel/autoconf-2.58 Automake: sys-devel/automake-1.7.7 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-march=pentium3 -O3" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.1/share/config /usr/kde/3/share/config /usr/share/config /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" CXXFLAGS="-O2 -mcpu=i686 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache cvs sandbox userpriv usersandbox" GENTOO_MIRRORS="ftp://usename:pass@mirror.internode.on.net/pub/gentoo http://usename:pass@premium.planetmirror.com/pub/gentoo http://gentoo.oregonstate.edu http://distro.ibiblio.org/pub/linux/distributions/gentoo" MAKEOPTS="-j2" PKGDIR="/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.au.gentoo.org/gentoo-portage" USE="3dfx X acpi alsa apache2 apm arts artswrappersuid avi berkdb bonobo cdr crypt cups debug doc encode esd ethereal fam fastcgi flash foomaticdb gd gdbm gif gnome gphoto2 gpm gtkhtml guile imagemagick imlib java jpeg kde ldap libg++ libgda libwww mad mbox mikmod mmx motif mozilla mpeg mysql nas ncurses oggvorbis openal oss pam pdflib perl php pic plotutils png postgres ppds psyco python qt quicktime readline samba sasl scanner sdl skey slang slp snmp socks5 sox spell sse ssl svga tcltk tetex tiff truetype usb vhost x86 xinerama xml xml2 xmms xv zeo zlib" ebuild uclibc-0.9.26.ebuild digest >>> Downloading x86? --11:54:06-- http://x86/? => `/usr/portage/distfiles/index.html' Resolving x86... failed: Host not found. !!! Couldn't download x86?. Aborting. #Ebuild uclibc-0.9.26.ebuild extract. inherit eutils crosscompile DESCRIPTION="Embedded C libraries" HOMEPAGE="http://www.uclibc.org/" SRC_URI="mirror://kernel/linux/libs/uclibc/uClibc-${PV}.tar.bz2 nls? ( x86? ( mirror://kernel/linux/libs/uclibc/uClibc-locale-030818.tgz ) )" #
Hi, It works for me. Can you attach the ebuild?
Created attachment 25527 [details] uclibc-0.9.26.ebuild Don't expect this ebuild do more than unpack. Still very much work in progress. bash-2.05b# ebuild uclibc-0.9.26.ebuild digest >>> Downloading x86? --22:34:05-- http://x86/? => `/usr/portage/distfiles/index.html' Resolving x86... failed: Host not found. !!! Couldn't download x86?. Aborting. bash-2.05b# pwd /usr/local/portage/dev-libs/uclibc bash-2.05b#
nls? ( x86? ( mirror://kernel/linux/libs/uclibc/uClibc-locale-030818.tgz ) )" this can't work just use nls or x86 but not both
OK. I found the problem. It occurs when FEATURES="cvs". I just fixed it in cvs. Thank you for reporting.
martin: the point of this bug is that it should work nakano seems to have fixed it for us
*** Bug 42144 has been marked as a duplicate of this bug. ***
*** Bug 42315 has been marked as a duplicate of this bug. ***
what about a portage-2.0.50-r2 release ;)
works fine now in -r2 -> close :)
there is still(or again) a bug in that, i have the following src_uri: x86? ( !static? ( http://snapshot.opera.com/unix/${OPERAVER}/intel-linux/en/${PN}-${OPERAVER2}.5-shared-qt.i386-en.tar.bz2 ) ) x86? ( static? ( http://snapshot.opera.com/unix/${OPERAVER}/intel-linux/en/${PN}-${OPERAVER2}.1-static-qt.i386-en.tar.bz2 ) ) ppc? ( !static? ( http://snapshot.opera.com/unix/${OPERAVER}/ppc-linux/en/${PN}-${OPERAVER2}.2-shared-qt.ppc-en.tar.bz2 ) ) ppc? ( static? ( http://snapshot.opera.com/unix/${OPERAVER}/ppc-linux/en/${PN}-${OPERAVER2}.1-static-qt.ppc-en.tar.bz2 ) ) sparc? ( !static? ( http://snapshot.opera.com/unix/${OPERAVER}/sparc-linux/en/${PN}-${OPERAVER2}.2-shared-qt.sparc-en.tar.bz2 ) ) sparc? ( static? ( http://snapshot.opera.com/unix/${OPERAVER}/sparc-linux/en/${PN}-${OPERAVER2}.1-static-qt.sparc-en.tar.bz2 ) ) portage downloads all files, but also wants to download a file called "!static"
It works here in 2.0.50-r6/portage-2.0.51_pre3 Do you have "cvs" in FEATURES? If so, portage would download all files to make a digest file. anyway, please show us 'emerge info'.
i have portage-2.0.50-r6 with cvs in FEATURES and tried to make an digest for opera-7.50_alpha4, after downloading all 6 tbz2's it does the following: >>> Downloading ftp://ftp.tu-clausthal.de/pub/linux/gentoo/distfiles/static? --02:37:35-- ftp://ftp.tu-clausthal.de/pub/linux/gentoo/distfiles/static? => `/usr/portage/distfiles/.listing' Resolving ftp.tu-clausthal.de... 139.174.2.36 Connecting to ftp.tu-clausthal.de[139.174.2.36]:21... connected. Logging in as anonymous ... Logged in! ==> SYST ... done. ==> PWD ... done. ==> TYPE I ... done. ==> CWD /pub/linux/gentoo/distfiles ... done. ==> PASV ... done. ==> LIST ... done. [ <=> ] 928,300 89.47K/s 02:37:46 (85.19 KB/s) - `/usr/portage/distfiles/.listing' saved [928300] Removed `/usr/portage/distfiles/.listing'. No matches on pattern `static?'. >>> Downloading ftp://ftp.tu-clausthal.de/pub/linux/gentoo/distfiles/!static? --02:37:46-- ftp://ftp.tu-clausthal.de/pub/linux/gentoo/distfiles/!static? => `/usr/portage/distfiles/.listing' Resolving ftp.tu-clausthal.de... 139.174.2.36 Connecting to ftp.tu-clausthal.de[139.174.2.36]:21... connected. Logging in as anonymous ... Logged in! ==> SYST ... done. ==> PWD ... done. ==> TYPE I ... done. ==> CWD /pub/linux/gentoo/distfiles ... done. ==> PASV ... done. ==> LIST ... done. [ <=> ] 928,300 89.76K/s 02:37:58 (85.36 KB/s) - `/usr/portage/distfiles/.listing' saved [928300] Removed `/usr/portage/distfiles/.listing'. No matches on pattern `!static?'. >>> Generating digest file... <<< !static? !!! Given file does not appear to be readable. Does it exist? !!! File: /usr/portage/distfiles/!static?
heino@heino opera $ emerge info Portage 2.0.50-r6 (default-x86-1.4, gcc-3.3.2, glibc-2.3.3_pre20040207-r0, 2.6.3) ================================================================= System uname: 2.6.3 i686 AMD Athlon(tm) XP 2600+ Gentoo Base System version 1.4.9 ccache version 2.3 [enabled] Autoconf: sys-devel/autoconf-2.58-r1 Automake: sys-devel/automake-1.8.3 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-march=athlon-xp -fomit-frame-pointer -Os -pipe" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.2/share/config /usr/kde/3/share/config /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/bind /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-march=athlon-xp -fomit-frame-pointer -Os -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache cvs digest sandbox strict" GENTOO_MIRRORS="ftp://ftp.tu-clausthal.de/pub/linux/gentoo/" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/home/heino/projects/gentoo/cvs/gentoo-x86" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="X aalib alsa apache2 apm arts avi berkdb cdr crypt cups dvd encode esd flash foomaticdb gd gdbm gif gnome gpm gtk gtk2 imap imlib jpeg kde ldap libg++ libwww lirc mad mikmod motif mozilla mpeg mysql ncurses nls nptl oggvorbis opengl pam pdflib perl png python qt quicktime readline ruby samba scanner sdl slang spell sse ssl svga tcltk tcpd tetex truetype unicode wmf x86 xml xml2 xmms xv zlib"
Sorry, I misunderstood your comment #22. I've just confirm the bug exists in 2.0.50-r6. But the bug has alrealy fixed in 2.0.51_pre*. I don't know when the version will be stable..
*** Bug 53679 has been marked as a duplicate of this bug. ***
Created attachment 33107 [details, diff] .50-r8 evaluate fix Seems this has popped up in .50-r8. The issue is in evaluate, basically for blah? [ blah2? [ uri1, uri2 ] ] if blah was set, then evaluate automatically skipped to the next arguement- basically skipping over the following arg regardless of what it may've been ( in this case, a list w/ the first element being another conditional). Included patch fixes the issue w/ evaluate not going recursive when it's supposed to, *and* fixes an oversight in evaluate, where it's not honoring the passed in allon (eg, if I passed in allon=1, it doesn't propogate when evaluate goes recursive). Bit wordy, but fixes the issue and a bug w/ allon support.
*** Bug 54102 has been marked as a duplicate of this bug. ***
*** Bug 54338 has been marked as a duplicate of this bug. ***
*** Bug 54705 has been marked as a duplicate of this bug. ***
*** Bug 55997 has been marked as a duplicate of this bug. ***
*** Bug 56246 has been marked as a duplicate of this bug. ***
*** Bug 57118 has been marked as a duplicate of this bug. ***
*** Bug 58442 has been marked as a duplicate of this bug. ***
*** Bug 58878 has been marked as a duplicate of this bug. ***
Nick, This bug alone warrants getting another .50 release out. I realize getting .51 out the door is a goal, but people keep hitting this. In bug #57261, all evaluate oddness (including this reocurring bug) is corrected. I realize you're busy, but this was identified broke in -r8, yet we're at -r9. The eclass IUSE issues were identified in r6, and are still broke. It would be nice to just schlop in the patches, and nail down these final issues w/ .50 so people stop running into them. Getting .51 stabled and marked ~arch is important, but I don't think .50 should be neglected, or that .51 should be pushed out early to correct .50 issues.
*** Bug 60595 has been marked as a duplicate of this bug. ***
This should be fixed in both .50 and .51
