Bugzilla Bug 69439

While trying to emerge tetex-2.0.4, I get

./mf ./tests/online
This is METAFONT, Version 2.7182 (Web2C 7.4.5)
(./tests/online.mfXlib: connection to ":0.0" refused by server
Xlib: No protocol specified

Error: Can't open display: :0.0
make[2]: *** [mf-check] Error 1
make[2]: Leaving directory `/var/tmp/portage/tetex-2.0.2-r4/work/tetex-src-2.0.2/texk/web2c'
make[1]: *** [check] Error 1
make[1]: Leaving directory `/var/tmp/portage/tetex-2.0.2-r4/work/tetex-src-2.0.2/texk'
make: *** [check] Error 2

!!! ERROR: app-text/tetex-2.0.2-r4 failed.
!!! Function src_test, Line 520, Exitcode 0
!!! Make check failed. See above for details.
!!! If you need support, post the topmost build error, NOT this status mess

Reproducible: Always
Steps to Reproduce:
1.
2.
3.




# emerge info
Portage 2.0.51-r2 (default-x86-1.4, gcc-3.3.4, glibc-2.3.4.20041021-r0,
2.6.8-20040814 i686)
=================================================================
System uname: 2.6.8-20040814 i686 AMD Athlon(tm) Processor
Gentoo Base System version 1.6.4
Autoconf: sys-devel/autoconf-2.59-r5
Automake: sys-devel/automake-1.8.5-r1
Binutils: sys-devel/binutils-2.15.92.0.2-r1
Headers:  sys-kernel/linux26-headers-2.6.8.1-r1
Libtools: sys-devel/libtool-1.5.2-r5
ACCEPT_KEYWORDS="x86 ~x86"
AUTOCLEAN="yes"
CFLAGS="-march=athlon -O2 -pipe -fstack-protector"
CHOST="i686-pc-linux-gnu"
COMPILER=""
CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config
/usr/kde/3.3/env /usr/kde/3.3/share/config /usr/kde/3.3/shutdown
/usr/kde/3/share/config /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 -O2 -pipe -fstack-protector"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs candy ccache distlocks maketest sandbox userpriv usersandbox"
GENTOO_MIRRORS="ftp://linux.rz.ruhr-uni-bochum.de/gentoo-mirror/
http://ftp.easynet.nl/mirror/gentoo/
http://www.mirror.ac.uk/sites/www.ibiblio.org/gentoo/ http://gentoo.inode.at/
http://linux.rz.ruhr-uni-bochum.de/download/gentoo-mirror/"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage/local /usr/local/portage/uncleowen
/usr/local/portage/kde-split"
SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage"
USE="3dnow X aalib alsa apache2 apm avi bitmap-fonts cdr crypt cups directfb dvd
dvdr encode esd f77 foomaticdb gd gif gnome gpm gtk gtk2 guile imlib ipv6 java
joystick jpeg libg++ libwww mad mikmod mmx mmx2 mpeg ncurses nls nvidia
oggvorbis opengl oss pam pdflib perl png python qt quicktime readline sdl slang
spell sse sse2 ssl svga tcltk tcpd tetex truetype x86 xml2 xmms xv zlib"

Use xhost or something to allow root to access your X.

Yeah, sure, I know that, but: Isn't it a bug in itself that a package tries to
access X (or any other ressource it doesn't own) during any phase of the
emerge?

Hum, but this is a "test" phase to make sure that the package 
build is completed, and if the test of the package necessarily 
require X, accessing to X should be given. It might be a good 
idea to check if root has access to X, though. If you write a
patch to achieve this, I'm also happy to include it.

Created an attachment (id=43314) [details]
tetex-2.0.2-r4.ebuild

Proposed ebuild that uses the virtualx eclass.

Works for me

Yeah, this is ideal patch if this works.... However,
it adds X to DEPEND, while tetex could be compiled
without it. As we cannot inherit eclasses conditionally
at the moment (Portage's limitation to make portage
internal cache correctly), I cannot take this patch. 
I'm really sorry for it.

This is rather old but the problem persists even with tetex-3.
The status is as follows:
- Conditional inherits will never be supported by portage, but maybe some of
the stuff going on in GLEP 33 will help us instead.
- Another solution would be to have something like TESTDEPEND (see bug 69021).
- Too many people wont like X being pulled in just to satisfy maketest, so
thats not an option.

The virtualx eclass only depends on x11 if the USE flag X is set:

DEPEND="X? ( || ( x11-base/xorg-server virtual/x11 ) )"
IUSE="X"

So what's the problem?

You are absolutely right :-)
I have added it to CVS.

Thanks!

What about having a check inside the test code. If the possibly required Xvfb
is not available the test might just be skipped with a warning. Currently this
forces one to either disable X11 client support, or to install Xvfb and
everything that pulls in. Not really useful for a headless system.

If you provide the test I'll be happy to include it.
Text-markup are currently under-staffed, so fixing test targets are currently
not prioritized very high, especially since the current state doesn't break
installs.

But either way I think you got a valid point, so please open a new bug so we
will remember to fix it when we got the time (I don't think it falls under the
category of this bug).