[HOWTO] Hassle-free emerge -uD world (update-world v1.8)

[mahler]

I don't see the point of an ebuild. I see what you are saying, but you have to remember, update-world is just a simple script. There's not much point in maintaining an ebuild for installing 1 file.

Ebuilds have one major advantage, for even the smallest scripts: automatic updates.
Without an ebuild people keep having to remember to check for updates,
and keep track of which version they have actually got.

[haubi]

I don't see the point of an ebuild. I see what you are saying, but you have to remember, update-world is just a simple script. There's not much point in maintaining an ebuild for installing 1 file.

That's true from ebuild-maintainer's POV.
Thing is is that this file has versions, and sometimes receives upgrades which I don't want to miss.
Currently, I manually have to find the link and see if there is a new version. This is what the portage-tree is for IMO...

[MrVahn]

I agree. This is such a great script that it could even be included in portage someday. Having to check back the forums for updates of a script you use quite so often is a hassle. A portage ebuild will be fine for us.

[mgorny]

I modified the script a little, adding support for 'genlop -p'. That is, if genlop is available on the system (checked with 'which'), then 'update-world -p' passes ebuild list to it, and outputs estimated build time.

http://mgorny.pastebin.ca/460914

[vithar]

This command creates a list called 'emergelist' in ~/.update-world/.

I suggest to replace ~/.update-world with $XDG_CONFIG_HOME/update-world according to XDG Base Directory Specification.

[DancesWithWords]

New Version 1.8 (2/16/07)
I've added a lot of new features to the newest version of update-world. It seems pretty stable (to me). Let me know if there are any problems.
Changelog v1.8:
-Cleaned up and modularized code. It should be a lot more readable now.
-Script now sources /etc/make.conf. Several portage variables are used by this script now.
-Added ability to 'emerge -p <package> | update-world. (realize that this is really only meant for updating existing packages...if you're installing a new one, you probably can't just skip a package failure, as it's likely a required dependency)
-Added interface with portage 2.1's elog capability. You can type 'update-world --log' to view build messages after updating (mesages stored in ~/update-world/<date>/notice/. Check out http://www.gentoo.org/doc/en/handbook/h ... doc_chap4. if you haven't set up elog yet.
-Added interface with revdep-rebuild. You can run it alone (update-world --revdep) to prepare a list of packages to install from revdep rebuild, or you can use it in conjunction with --install (update-world --revdep --install) to automatically run revdep-rebuild when the update is complete.

snip...

Started using this a week ago and boy am I happy to have found this script. It will make looking after my Gentoo boxes easier. I've three here at home and 2 at work. Thanks for this it is appreciated.

--
bruce

[merlijn]

VERY nice script indeed, just a small suggestion:

Code: Select all

prepare ()
{
if [[ -e "$emergelist" ]]
   then echo -en "The file 'emergelist' already exists and will be overwritten.\nContinue anyway? (Y/n) "
   read continue
   if [[ "$continue" = 'y' || "$continue" = 'yes' || "$continue" = 'Y' || "$continue" = 'Yes' || "$continue" = 'YES' || "$continue" = '' ]]
      then :
      else echo exiting.
      exit 0
   fi
   else :
fi

can this condition be changed to "if !(emergelist doesnt exist || emergelist file is empty)" ?

In most cases a world update works hassle free for me, but if there are no packages to be fixed the emergefile is still there with nothing in it, thus giving this error on a new `update-world -p`.

Thank you for your time and effort on this!

[Gerben]

In version 1.8 of the script update-world --log only works after a cd into $homedir (~/.update-world)

If the command is issued in another directory it will present the directories in your current directories as options instead of the date directories in $homedir

current code is
...
# update-world --log
if [[ -n `echo "$arguments" | sed -n 's/-l/&/p'` || -n `echo "$arguments" | sed -n 's/--log/&/p'` ]]
then echo "Choose a date: "
select directory in `ls -Al | grep ^d | sed 's/.*\ //'`
do
...

The reason is the absence of $homedir in the ls command of:
select directory in `ls -Al | grep ^d | sed 's/.*\//'`

Changing this into:
select directory in `ls -Al $homedir| grep ^d | sed 's/.*\//'`

will make update-world --log work from any directory

[wrc1944]

Tried this script, and got this- it bails out. The new one in GWN bails out too, and doesn't work, at least on my system.

Code: Select all

gentoo wrc # update-world --prepare
Preparing list of packages...

These are the packages that would be merged, in order:

Calculating world dependencies... done!
[ebuild   R   ] x11-proto/evieext-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/trapproto-3.4.3  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/xf86dgaproto-2.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] sys-apps/sed-4.1.5  USE="nls -static" USERLAND="(GNU%*)" 0 kB [?=>0]
[ebuild   R   ] sys-apps/pciutils-2.2.6-r1  USE="zlib* -network-cron" 0 kB [0]
[ebuild  N    ] dev-util/boost-build-1.34.1  0 kB [0]
[ebuild   R   ] x11-misc/xorg-cf-files-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXprintUtil-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/xcursorgen-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/bigreqsproto-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/resourceproto-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/xf86rushproto-1.1.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/xcmiscproto-1.1.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/xf86bigfontproto-1.1.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/scrnsaverproto-1.1.0  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] app-text/ghostscript-gpl-8.57  USE="X cups gtk -bindist -cjk -djvu -jpeg2k (-emacs%)" 0 kB [?=>0]
[ebuild   R   ] sys-apps/dbus-1.0.2-r2  USE="X -debug -doc (-selinux)" KERNEL="(linux%*) (-FreeBSD)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/xdriinfo-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libFS-1.0.0  USE="ipv6 -debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/xmodmap-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/setxkbmap-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/liboldX-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-misc/gccmakedep-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] dev-python/qscintilla-1.7.1  USE="-debug% -doc" 0 kB [?=>0]
[ebuild   R   ] media-libs/libdvdread-0.9.7  USE="(-static%)" 0 kB [?=>0]
[ebuild  N    ] dev-libs/boost-1.34.1  USE="-debug -doc -icu -pyste -tools" 0 kB [0]
[ebuild     U ] sys-apps/hal-0.5.9.1-r1 [0.5.9-r1] USE="crypt -acpi -debug -dell -disk-partition -doc -pcmcia (-selinux)" 0 kB [?=>0]
[ebuild   R   ] x11-misc/imake-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXprintAppUtil-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-themes/xcursor-themes-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXevie-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXTrap-1.0.0  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-base/xorg-server-1.3.0.0  USE="dri ipv6 nptl sdl xorg -3dfx -debug -dmx -kdrive -minimal -xprint" INPUT_DEVICES="evdev keyboard mouse -acecad -aiptek -calcomp -citron -digitaledge -dmc -dynapro -elo2300 -elographics -fpit -hyperpen -jamstudio -joystick -magellan -microtouch -mutouch -palmax -penmount -spaceorb -summa -synaptics -tek4957 -ur98 -vmmouse -void -wacom" VIDEO_CARDS="fbdev radeon vesa vga -apm -ark -chips -cirrus -cyrix -dummy -epson -fglrx% -glint -i128 -i740 -i810 (-impact) -imstt -mach64 -mga -neomagic (-newport) -nsc -nv -nvidia -r128 -rendition -s3 -s3virge -savage -siliconmotion -sis -sisusb (-sunbw2) (-suncg14) (-suncg3) (-suncg6) (-sunffb) (-sunleo) (-suntcx) -tdfx -tga -trident -tseng -v4l -via -vmware -voodoo" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXxf86dga-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] gnome-base/gnome-menus-2.18.2  USE="python -debug" KERNEL="(linux%*)" 0 kB [?=>0]
[ebuild     U ] media-sound/ardour-2.0.3 [0.99.3] USE="nls sse (-altivec) -debug -fftw%" 0 kB [?=>0]

Total: 36 packages (2 upgrades, 2 new, 32 reinstalls), Size of downloads: 0 kB
Portage tree and overlays:
 [0] /usr/portage
 [?] indicates that the source repository could not be determined

These are the packages that will be installed.  If you want to alter this list,
just edit the file 'emergelist' in /root/.update-world. When finished,
run 'update-world --install'

gentoo wrc #
gentoo wrc # update-world --install
Calculating dependencies... done!
>>> Verifying ebuild Manifests...
>>> starting parallel fetching

>>> Emerging (1 of 36) dev-util/boost-build-1.34.1 to /
 * boost_1_34_1.tar.bz2 RMD160 ;-) ...                                                                              [ ok ]
 * boost_1_34_1.tar.bz2 SHA1 ;-) ...                                                                                [ ok ]
 * boost_1_34_1.tar.bz2 SHA256 ;-) ...                                                                              [ ok ]
 * boost_1_34_1.tar.bz2 size ;-) ...                                                                                [ ok ]
 * checking ebuild checksums ;-) ...                                                                                [ ok ]
 * checking auxfile checksums ;-) ...                                                                               [ ok ]
 * checking miscfile checksums ;-) ...                                                                              [ ok ]
 * checking boost_1_34_1.tar.bz2 ;-) ...                                                                            [ ok ]
>>> Unpacking source...
>>> Unpacking boost_1_34_1.tar.bz2 to /var/tmp/portage/dev-util/boost-build-1.34.1/work
>>> Source unpacked.
>>> Compiling source in /var/tmp/portage/dev-util/boost-build-1.34.1/work/boost_1_34_1/tools ...
###
### Using 'cc' toolset.
###
rm -rf bootstrap
mkdir bootstrap
i686-pc-linux-gnu-gcc -o bootstrap/jam0 -Os -march=athlon-xp -frename-registers -fweb -pipe -fomit-frame-pointer -ftree-vectorize -funit-at-a-time -freorder-blocks -fno-ident -freorder-blocks-and-partition -fmerge-all-constants -combine -fno-strict-aliasing command.c compile.c debug.c execunix.c expand.c fileunix.c glob.c hash.c hdrmacro.c headers.c jam.c jambase.c jamgram.c lists.c make.c make1.c newstr.c option.c parse.c pathunix.c pathvms.c regexp.c rules.c scan.c search.c subst.c timestamp.c variable.c modules.c strings.c filesys.c builtins.c pwd.c class.c native.c w32_getreg.c modules/set.c modules/path.c modules/regex.c modules/property-set.c modules/sequence.c modules/order.c execnt.c filent.c
make1.c: In function 'alloc_state':
make1.c:131: error: incompatible implicit declaration of function 'profile_memory'
debug.c:64: error: previous definition of 'profile_memory' was here
make1.c: In function 'call_timing_rule':
make1.c:706: error: incompatible implicit declaration of function 'evaluate_rule'
compile.c:910: error: previous definition of 'evaluate_rule' was here
newstr.c: In function 'newstr':
newstr.c:130: error: incompatible implicit declaration of function 'profile_memory'
debug.c:64: error: previous definition of 'profile_memory' was here
builtins.c: In function 'append_if_exists':
builtins.c:679: error: incompatible implicit declaration of function 'timestamp'
timestamp.c:73: error: previous definition of 'timestamp' was here
native.c: In function 'declare_native_rule':
native.c:27: error: incompatible implicit declaration of function 'lol_build'
builtins.c:1574: error: previous definition of 'lol_build' was here
modules/set.c: At top level:
modules/set.c:35: warning: conflicting types for 'init_set'
builtins.c:342: warning: previous implicit declaration of 'init_set' was here
modules/path.c:26: warning: conflicting types for 'init_path'
builtins.c:343: warning: previous implicit declaration of 'init_path' was here
modules/regex.c:91: warning: conflicting types for 'init_regex'
builtins.c:344: warning: previous implicit declaration of 'init_regex' was here
modules/property-set.c:103: warning: conflicting types for 'init_property_set'
builtins.c:345: warning: previous implicit declaration of 'init_property_set' was here
modules/sequence.c:35: warning: conflicting types for 'init_sequence'
builtins.c:346: warning: previous implicit declaration of 'init_sequence' was here
modules/order.c:137: warning: conflicting types for 'init_order'
builtins.c:347: warning: previous implicit declaration of 'init_order' was here
 *
 * ERROR: dev-util/boost-build-1.34.1 failed.
 * Call stack:
 *   ebuild.sh, line 1648:   Called dyn_compile
 *   ebuild.sh, line 988:   Called qa_call 'src_compile'
 *   ebuild.sh, line 44:   Called src_compile
 *   boost-build-1.34.1.ebuild, line 57:   Called die
 *
 * building bjam failed
 * If you need support, post the topmost build error, and the call stack if relevant.
 * A complete build log is located at '/var/log/portage/dev-util:boost-build-1.34.1:20070802-231443.log'.
 *

*** Hit Control-C to exit, or just wait to continue with emerge.

*** Continuing with emerge.


*** dev-util/boost-build-1.34.1 compile failed, skipping.
*** Continuing with emerge world
Calculating dependencies... done!
>>> Verifying ebuild Manifests...
>>> starting parallel fetching

>>> Emerging (1 of 36) dev-util/boost-build-1.34.1 to /
 * boost_1_34_1.tar.bz2 RMD160 ;-) ...                                                                              [ ok ]
 * boost_1_34_1.tar.bz2 SHA1 ;-) ...                                                                                [ ok ]
 * boost_1_34_1.tar.bz2 SHA256 ;-) ...                                                                              [ ok ]
 * boost_1_34_1.tar.bz2 size ;-) ...                                                                                [ ok ]
 * checking ebuild checksums ;-) ...                                                                                [ ok ]
 * checking auxfile checksums ;-) ...                                                                               [ ok ]
 * checking miscfile checksums ;-) ...                                                                              [ ok ]
 * checking boost_1_34_1.tar.bz2 ;-) ...                                                                            [ ok ]
>>> Unpacking source...
>>> Unpacking boost_1_34_1.tar.bz2 to /var/tmp/portage/dev-util/boost-build-1.34.1/work
>>> Source unpacked.
>>> Compiling source in /var/tmp/portage/dev-util/boost-build-1.34.1/work/boost_1_34_1/tools ...
###
### Using 'cc' toolset.
###
rm -rf bootstrap
mkdir bootstrap
i686-pc-linux-gnu-gcc -o bootstrap/jam0 -Os -march=athlon-xp -frename-registers -fweb -pipe -fomit-frame-pointer -ftree-vectorize -funit-at-a-time -freorder-blocks -fno-ident -freorder-blocks-and-partition -fmerge-all-constants -combine -fno-strict-aliasing command.c compile.c debug.c execunix.c expand.c fileunix.c glob.c hash.c hdrmacro.c headers.c jam.c jambase.c jamgram.c lists.c make.c make1.c newstr.c option.c parse.c pathunix.c pathvms.c regexp.c rules.c scan.c search.c subst.c timestamp.c variable.c modules.c strings.c filesys.c builtins.c pwd.c class.c native.c w32_getreg.c modules/set.c modules/path.c modules/regex.c modules/property-set.c modules/sequence.c modules/order.c execnt.c filent.c
make1.c: In function 'alloc_state':
make1.c:131: error: incompatible implicit declaration of function 'profile_memory'
debug.c:64: error: previous definition of 'profile_memory' was here
make1.c: In function 'call_timing_rule':
make1.c:706: error: incompatible implicit declaration of function 'evaluate_rule'
compile.c:910: error: previous definition of 'evaluate_rule' was here
newstr.c: In function 'newstr':
newstr.c:130: error: incompatible implicit declaration of function 'profile_memory'
debug.c:64: error: previous definition of 'profile_memory' was here
builtins.c: In function 'append_if_exists':
builtins.c:679: error: incompatible implicit declaration of function 'timestamp'
timestamp.c:73: error: previous definition of 'timestamp' was here
native.c: In function 'declare_native_rule':
native.c:27: error: incompatible implicit declaration of function 'lol_build'
builtins.c:1574: error: previous definition of 'lol_build' was here
modules/set.c: At top level:
modules/set.c:35: warning: conflicting types for 'init_set'
builtins.c:342: warning: previous implicit declaration of 'init_set' was here
modules/path.c:26: warning: conflicting types for 'init_path'
builtins.c:343: warning: previous implicit declaration of 'init_path' was here
modules/regex.c:91: warning: conflicting types for 'init_regex'
builtins.c:344: warning: previous implicit declaration of 'init_regex' was here
modules/property-set.c:103: warning: conflicting types for 'init_property_set'
builtins.c:345: warning: previous implicit declaration of 'init_property_set' was here
modules/sequence.c:35: warning: conflicting types for 'init_sequence'
builtins.c:346: warning: previous implicit declaration of 'init_sequence' was here
modules/order.c:137: warning: conflicting types for 'init_order'
builtins.c:347: warning: previous implicit declaration of 'init_order' was here
 *
 * ERROR: dev-util/boost-build-1.34.1 failed.
 * Call stack:
 *   ebuild.sh, line 1648:   Called dyn_compile
 *   ebuild.sh, line 988:   Called qa_call 'src_compile'
 *   ebuild.sh, line 44:   Called src_compile
 *   boost-build-1.34.1.ebuild, line 57:   Called die
 *
 * building bjam failed
 * If you need support, post the topmost build error, and the call stack if relevant.
 * A complete build log is located at '/var/log/portage/dev-util:boost-build-1.34.1:20070802-231551.log'.
 *

*** Hit Control-C to exit, or just wait to continue with emerge.

*** Continuing with emerge.
The failed package dev-util/boost-build-1.34.1 could not be merged and must be successfully installed before continuing.
gentoo wrc #

COMMENT:
Apparently, it just bails out after the second try at boost, and didn't continue the -uD world, as advertised. I then tried the other script in the latest Gentoo Weekly News (see below), and it too bails out with Dependency Errors, and "installed when not in list" problems. What am I missing here, or don't any of these type scripts actually work in all cases? Do I need to edit the script or somehow pass an option for a special case?

Code: Select all

gentoo wrc # updatescript.sh
Emerging -uDN world then running revdep-rebuild.
You will be asked for confirmation; CTRL-C to abort at any time.
Resolving...

These are the packages that would be merged, in order:

Calculating world dependencies  ..... ..... ..... done!
[ebuild   R   ] sys-apps/sed-4.1.5  USE="nls -static" USERLAND="(GNU%*)" 0 kB [?=>0]
[ebuild   R   ] sys-apps/pciutils-2.2.6-r1  USE="zlib* -network-cron" 0 kB [0]
[ebuild  N    ] dev-util/boost-build-1.34.1  0 kB [0]
[ebuild   R   ] x11-misc/xorg-cf-files-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXprintUtil-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/xcursorgen-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/bigreqsproto-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/resourceproto-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/xf86rushproto-1.1.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/xcmiscproto-1.1.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/xf86bigfontproto-1.1.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-proto/scrnsaverproto-1.1.0  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] app-text/ghostscript-gpl-8.57  USE="X cups gtk -bindist -cjk -djvu -jpeg2k (-emacs%)" 0 kB [?=>0]
[ebuild   R   ] sys-apps/dbus-1.0.2-r2  USE="X -debug -doc (-selinux)" KERNEL="(linux%*) (-FreeBSD)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/xdriinfo-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libFS-1.0.0  USE="ipv6 -debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXevie-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXTrap-1.0.0  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/xmodmap-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXxf86dga-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-apps/setxkbmap-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/liboldX-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-misc/gccmakedep-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] dev-python/qscintilla-1.7.1  USE="-debug% -doc" 0 kB [?=>0]
[ebuild   R   ] media-libs/libdvdread-0.9.7  USE="(-static%)" 0 kB [?=>0]
[ebuild  N    ] dev-libs/boost-1.34.1  USE="-debug -doc -icu -pyste -tools" 0 kB [0]
[ebuild     U ] sys-apps/hal-0.5.9.1-r1 [0.5.9-r1] USE="crypt -acpi -debug -dell -disk-partition -doc -pcmcia (-selinux)" 0 kB [?=>0]
[ebuild   R   ] x11-misc/imake-1.0.2  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-libs/libXprintAppUtil-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-themes/xcursor-themes-1.0.1  USE="-debug (-X%*)" 0 kB [?=>0]
[ebuild   R   ] x11-base/xorg-server-1.3.0.0  USE="dri ipv6 nptl sdl xorg -3dfx -debug -dmx -kdrive -minimal -xprint" INPUT_DEVICES="evdev keyboard mouse -acecad -aiptek -calcomp -citron -digitaledge -dmc -dynapro -elo2300 -elographics -fpit -hyperpen -jamstudio -joystick -magellan -microtouch -mutouch -palmax -penmount -spaceorb -summa -synaptics -tek4957 -ur98 -vmmouse -void -wacom" VIDEO_CARDS="fbdev radeon vesa vga -apm -ark -chips -cirrus -cyrix -dummy -epson -fglrx% -glint -i128 -i740 -i810 (-impact) -imstt -mach64 -mga -neomagic (-newport) -nsc -nv -nvidia -r128 -rendition -s3 -s3virge -savage -siliconmotion -sis -sisusb (-sunbw2) (-suncg14) (-suncg3) (-suncg6) (-sunffb) (-sunleo) (-suntcx) -tdfx -tga -trident -tseng -v4l -via -vmware -voodoo" 0 kB [?=>0]
[ebuild   R   ] gnome-base/gnome-menus-2.18.2  USE="python -debug" KERNEL="(linux%*)" 0 kB [?=>0]
[ebuild     U ] media-sound/ardour-2.0.3 [0.99.3] USE="nls sse (-altivec) -debug -fftw%" 0 kB [?=>0]

Total: 33 packages (2 upgrades, 2 new, 29 reinstalls), Size of downloads: 0 kB
Portage tree and overlays:
 [0] /usr/portage
 [?] indicates that the source repository could not be determined
* 33 in total.
About to cancel (Y to proceed)?

** Rebuilding sys-apps/sed-4.1.5 - 1 of 33.
Patching:...done
Completed compile

* IMPORTANT: 1 config files in '/usr/kde/3.5/share/config' need updating.

** sys-apps/sed-4.1.5 installed successfully.
Packages installed so far: 0 of 0.      33 remaining.
!! Unable to resolve sys-apps/pciutils-2.2.6-r1.

** Installing dev-util/boost-build-1.34.1 - 1 of 33.
* ERROR: dev-util/boost-build-1.34.1 failed.
* Call stack:
*   ebuild.sh, line 1648:   Called dyn_compile
*   ebuild.sh, line 988:   Called qa_call 'src_compile'
*   ebuild.sh, line 44:   Called src_compile
*   boost-build-1.34.1.ebuild, line 57:   Called die
* building bjam failed
* If you need support, post the topmost build error, and the call stack if relevant.
* A complete build log is located at '/var/log/portage/dev-util:boost-build-1.34.1:20070802-232855.log'.
Completed compile


** dev-util/boost-build-1.34.1 failed to install.
Packages installed so far: 0 of 1.      32 remaining.
** Failed so far: dev-util/boost-build-1.34.1.
!! Unable to resolve x11-misc/xorg-cf-files-1.0.2.

** Rebuilding x11-libs/libXprintUtil-1.0.1 - 2 of 33.
Checking dependencies..done
* Elibtoolize: libXprintUtil-1.0.1
Patching:..done
Completed compile

* IMPORTANT: 1 config files in '/usr/kde/3.5/share/config' need updating.

** x11-libs/libXprintUtil-1.0.1 installed successfully.
Packages installed so far: 0 of 1.      32 remaining.
** Failed so far: dev-util/boost-build-1.34.1.

** Rebuilding x11-apps/xcursorgen-1.0.1 - 2 of 33.
Checking dependencies..done
** updatescript.sh: Dependency Error - x11-apps/xcursorgen-1.0.1 requires waiting which is not in the install list!

sys-apps/sed-4.1.5 installed when not in list
x11-libs/libXprintUtil-1.0.1 installed when not in list
!! ** Config file updates pending **.
gentoo wrc #

[bluecog6]

@wrc1944 - this may be a completely dumb thing to ask (I'm a noob), but couldn't you edit the file that contains the list of packages to be emerged and remove boost-build?

I am very new to this -- I can't wait to try this script though -- I have to have 630+ packages complied to be world updated -- that's a lot, but I have a weekend roadtrip coming up... we'll see what happens

[thermal]

I`ve added support for using genlop to estimate the build time when preparing the emergelist. Here`s the patch:

Code: Select all

--- update-world        2007-08-30 11:54:25.000000000 +0200
+++ update-world-new    2007-08-30 11:54:53.000000000 +0200
@@ -133,6 +133,11 @@
 emerge $parameter world > $emergelist
 errorcheck
 emerge $parameter world
+if [ -x "$(which genlop 2> /dev/null)" ]; then
+       genlop -p < $emergelist | grep "Estimated update time"
+else
+       echo "Emerge genlop to calculate approximate update time"
+fi
 echo -e "\nThese are the packages that will be installed.  If you want to alter this list,\njust edit the file 'emergelist' in $homedir. When finished,\nrun 'update-world --install'"
 cat $emergelist | sed -n 's/^\[ebuild[[:upper:][:lower:] ]*\]\ \([[:lower:][:upper:]0-9._+\/-]*\).*/\1/p' | sed 's/.*/=&/' | sort -d > $emergetemp
 cp $emergetemp $emergelist

[sLumpia]

It was discussed in another thread about combining this script with emerging packages in the correct order, a la this thread by Guenther Brunthaler (thanks for a job well done, Guenther )

has you combine it? it would be great
thanks for this great script

[steveL]

Apparently, it just bails out after the second try at boost, and didn't continue the -uD world, as advertised. I then tried the other script in the latest Gentoo Weekly News (see below), and it too bails out with Dependency Errors, and "installed when not in list" problems. What am I missing here, or don't any of these type scripts actually work in all cases? Do I need to edit the script or somehow pass an option for a special case?

Er, that looks like a bug in update (interesting one too) but you really should have reported it in the [topic=546828]update thread[/topic]. We're on version 0.1.4_beta2a atm. We also hang in #friendly-coders on IRC for bugfixing, if anyone wants to help out or needs BASH help and has been mangled by #bash ;)

[cbart]

Code: Select all

modules/property-set.c:103: warning: conflicting types for 'init_property_set'
builtins.c:345: warning: previous implicit declaration of 'init_property_set' was here
modules/sequence.c:35: warning: conflicting types for 'init_sequence'
builtins.c:346: warning: previous implicit declaration of 'init_sequence' was here
modules/order.c:137: warning: conflicting types for 'init_order'
builtins.c:347: warning: previous implicit declaration of 'init_order' was here
 *
 * ERROR: dev-util/boost-build-1.34.1 failed.
 * Call stack:
 *   ebuild.sh, line 1648:   Called dyn_compile
 *   ebuild.sh, line 988:   Called qa_call 'src_compile'
 *   ebuild.sh, line 44:   Called src_compile
 *   boost-build-1.34.1.ebuild, line 57:   Called die
 *
 * building bjam failed
 * If you need support, post the topmost build error, and the call stack if relevant.
 * A complete build log is located at '/var/log/portage/dev-util:boost-build-1.34.1:20070802-231443.log'.
 *

I had this problem too. In my case it was a bad CFLAG, -combine; if it is it simply add a file:

Code: Select all

# nano -w /etc/portage/env/dev-utils/boost-build
CFLAGS="(your cflags here without the -combine CFLAG)"
CXXFLAGS="${CFLAGS} (your additional cxxflags here)"

Hope I helped

[wrc1944]

Thanks cbart,
I did have -combine (it was a default in the Conrad install I used on this box. Come to think of it, I think I recall reading that Conrad did have second thoughts about this and remove remove this flag. I'll give your solution a try, or remove -combine completely from make.conf..

[dioon]

Is this script still available,cannot seem to find a site to download from.

[ryker]

Is this script still available,cannot seem to find a site to download from.

Both sites seem to be working fine for me.

http://www.mindlesstechie.net/gentoo/update-world/
http://dixieflatline.homelinux.org/files/gentoo/

[trapdoor]

Thanks for your work. Just tried it, and it's a nice solution. Simple and functional!

[.arris]

The pipe feature does not seem to work for me:

Code: Select all

gentoo ~ # emerge -p nano | update-world
No arguments supplied.  If you're piping from the emerge command (emerge -p <package> | update-world), this is fine.  Otherwise, cancel and type 'update-world --help' for usage information.
Error: Invalid Command.
Run 'update-world --help' for usage.

I'm planning to pull in all packages listed in the world file of a different machine by running a command like this:

Code: Select all

emerge -p $(<world) | update-world

Of course this won't work wothout the piping feature of update-world
Anyone have an idea how to fix this?

[Kasumi_Ninja]

update-world fails with dev-haskell/filepath

Code: Select all

*** Hit Control-C to exit, or just wait to continue with emerge.

*** Continuing with emerge.
The failed package dev-haskell/filepath-1.0 could not be merged and must be successfully installed before continuing.

[muhsinzubeir]

thanks for the script....save me lots of hours...

[pjv]

I couldn't succeed in doing something like 'emerge -p <package> | update-world'. By the way, in the first post, in the green code text you forgot '-p'.

I was trying:

Code: Select all

emerge -p coreutils | update-world.sh

Your script says:

Code: Select all

No arguments supplied.  If you're piping from the emerge command (emerge -p <package> | update-world), this is fine.  Otherwise, cancel and type 'update-world --help' for usage information.
Error: Invalid Command.
Run 'update-world --help' for usage.

The output in between would be:

Code: Select all

These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild     U ] sys-devel/automake-1.10.1 [1.10] 
[ebuild  N    ] app-arch/lzma-utils-4.32.6  USE="-nocxx" 
[ebuild     U ] sys-apps/coreutils-6.10-r2 [6.9-r1] USE="-vanilla%" 

The emergelist is empty and it didn't even ask me if I wanted to overwrite (well it could have been that it asked me once, I said no, and it still wiped but didnt put in the new stuff).

Last of all, while I appreciate your work, I'm a bit worried you didn't enhance the already existing pye script, or more importantly that this kind of functionality still is missing from standard portage (allthough paludis apparently has it).

[kirk427]

this is what I changed to make update-world work with piped commands:

comment out line 412 (until [[ `echo "$piped...)

then change the loop to look like this:

Code: Select all

while read line
        do
        piped="${piped}
$line"
        done

something in the original until was causing update-world to fail. I didn't have time to figure out which part.

[minor_prophets]

I've been using this script now for well over a year now on 3 machines and just wanted to thank you again for such a great script.