- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] A tiny fix in the Polish translation.
by Jan Stępień 19 Feb '09

19 Feb '09

From: Jan Stępień <jstepien(a)users.sourceforge.net> The previous translation of 'targets' meant literally 'it targets' and it sounded awkwardly. The current version is a plural of a 'target'. Signed-off-by: Jan Stępień <jstepien(a)users.sourceforge.net> --- po/pl.po | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/po/pl.po b/po/pl.po index eaafb3b..cf027ac 100644 --- a/po/pl.po +++ b/po/pl.po @@ -900,7 +900,7 @@ msgstr "Żadnych" #, c-format msgid "Targets (%d):" -msgstr "Celuje (%d):" +msgstr "Cele (%d):" #, c-format msgid "Total Download Size: %.2f MB\n" -- 1.6.1.3

2 1

[pacman-dev] about pacdiff
by Xavier 17 Feb '09

17 Feb '09

I like a lot pacdiff for dealing with pacnew files. I always notice that it miss the grub config because it lies in /boot/ while pacdiff only checks in /etc/ I was curious so I decided to have a look at how many backup config files are not in /etc/ :) $ sh backup.sh > backup-list $ cat backup-list | cut -d'/' -f1 | sort -u boot conf.d home opt srv sysctl.conf usr var Note that we have three crazy files there : empty, conf.d and sysctl.conf. Here are the three problematic files/packages : conf.d/ntp-client.conf / /var/abs/extra/ntp/PKGBUILD /etc/foomatic/filter.conf / /var/abs/testing/foomatic-filters/PKGBUILD sysctl.conf / /var/abs/community/network/ufw/PKGBUILD I should probably report bugs for these, though it was not my original intention. So after this manual filtering, here is the list : boot home opt srv usr var That would considerably slow down pacdiff though so probably not worth it. It is disappointing :)

4 5

[pacman-dev] Fwd: [PATCH] repo-add: fix eval and quote issues
by Dan McGee 16 Feb '09

16 Feb '09

Yes, I think I sent this to myself on accident. Looks like it is time for another cup of coffee. ---------- Forwarded message ---------- From: Dan McGee <dan(a)archlinux.org> Date: Sun, Feb 8, 2009 at 12:30 PM Subject: Re: [PATCH] repo-add: fix eval and quote issues To: Dan McGee <dan(a)archlinux.org> On Sun, Feb 8, 2009 at 12:22 PM, Dan McGee <dan(a)archlinux.org> wrote: > eval was ugly and dirty, and bit us here. Instead, use a safer form of > variable declaration to ensure quotes don't foil us in pkgdesc or any other > fields. > > This fixes FS#10837. > > Signed-off-by: Dan McGee <dan(a)archlinux.org> > --- We apparently had never handled quotes in descriptions correctly, so this patch should fix these issues. I'm going to push this patch to maint if there are no objections, although I expect our next release will come from master. > scripts/repo-add.sh.in | 25 +++++++++++++------------ > 1 files changed, 13 insertions(+), 12 deletions(-) > > diff --git a/scripts/repo-add.sh.in b/scripts/repo-add.sh.in > index be0859e..93fdd52 100644 > --- a/scripts/repo-add.sh.in > +++ b/scripts/repo-add.sh.in > @@ -152,18 +152,19 @@ db_write_entry() > > # read info from the zipped package > local line I did make one small fix here- added var and val to the local declaration. > - for line in $(bsdtar -xOf "$pkgfile" .PKGINFO | \ > - grep -v "^#" | sed 's|$\w*$\s*=\s*$.*$|\1="\2"|'); do > - eval "$line" > - case "$line" in > - group=*) _groups="$_groups$group\n" ;; > - depend=*) _depends="$_depends$depend\n" ;; > - backup=*) _backups="$_backups$backup\n" ;; > - license=*) _licenses="$_licenses$license\n" ;; > - replaces=*) _replaces="$_replaces$replaces\n" ;; > - provides=*) _provides="$_provides$provides\n" ;; > - conflict=*) _conflicts="$_conflicts$conflict\n" ;; > - optdepend=*) _optdepends="$_optdepends$optdepend\n" ;; > + for line in $(bsdtar -xOf "$pkgfile" .PKGINFO | grep -v '^#'); do > + var="$(echo $line | sed 's|$\w*$\s*=\s*$.*$|\1|')" > + val="$(echo $line | sed 's|$\w*$\s*=\s*$.*$|\2|')" > + declare $var="$val" > + case "$var" in > + group) _groups="$_groups$group\n" ;; > + depend) _depends="$_depends$depend\n" ;; > + backup) _backups="$_backups$backup\n" ;; > + license) _licenses="$_licenses$license\n" ;; > + replaces) _replaces="$_replaces$replaces\n" ;; > + provides) _provides="$_provides$provides\n" ;; > + conflict) _conflicts="$_conflicts$conflict\n" ;; > + optdepend) _optdepends="$_optdepends$optdepend\n" ;; > esac > done > > -- > 1.6.1.2

7 12

[pacman-dev] [PATCH] Mac OS X Full path of sed
by Kevin Barry 16 Feb '09

16 Feb '09

> Please resend both of these patches as separate emails- git tools are > meant to have one patch per email, whether using format-patch or am. Sure thing. I'm new to git. Change configure.ac to use the full path of stat when on darwin/mac. This is needed for situations when a user installs the GNU/coreutils and places it in their path before /usr/bin, but the SIZECMD is already configured for Darwin's version of stat. Signed-off-by: Kevin Barry <barryk gmail com> --- configure.ac | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/configure.ac b/configure.ac index 2e7f433..216cf76 100644 --- a/configure.ac +++ b/configure.ac @@ -190,7 +190,7 @@ case "${host_os}" in ;; darwin*) host_os_darwin=yes - SIZECMD="stat -f %z" + SIZECMD="/usr/bin/stat -f %z" ;; esac -- 1.6.1.2

3 2

[pacman-dev] [PATCH] repo-add: bash implementation of realpath
by Yun Zheng Hu 16 Feb '09

16 Feb '09

Replaced readlink -f / realpath with a more portable bash implementation. Signed-off-by: Yun Zheng Hu <yunzheng.hu(a)gmail.com> --- scripts/repo-add.sh.in | 37 ++++++++++++++++++++++++------------- 1 files changed, 24 insertions(+), 13 deletions(-) diff --git a/scripts/repo-add.sh.in b/scripts/repo-add.sh.in index c89e2a5..f266719 100644 --- a/scripts/repo-add.sh.in +++ b/scripts/repo-add.sh.in @@ -107,7 +107,7 @@ write_list_entry() { db_write_delta() { # blank out all variables and set deltafile to absolute path - local deltafile=$($realpath "$1") + local deltafile=$(realpath "$1") local filename=$(basename "$deltafile") local deltavars pkgname fromver tover arch csize md5sum @@ -138,7 +138,7 @@ db_write_delta() db_write_entry() { # blank out all variables and set pkgfile to an absolute path - local pkgfile=$($realpath "$1") + local pkgfile=$(realpath "$1") local pkgname pkgver pkgdesc url builddate packager csize size \ group depend backup license replaces provides conflict force \ _groups _depends _backups _licenses _replaces _provides _conflicts \ @@ -267,6 +267,27 @@ db_remove_entry() { popd 2>&1 >/dev/null } # end db_remove_entry +# bash implementation of realpath / readlink -f +# arg1 - filename +realpath() +{ + fname=${1%/} # strips trailing '/' + while [ -L "$fname" ]; do + oldfname="$fname" + fname="$(command ls -l $fname)" + fname="${fname#*\> }" + if [ "$fname" = . ] ; then + fname="$(dirname $oldfname)" + elif echo $fname | grep -vq '^/' - ; then + fname="$(dirname $oldfname)/$fname" + fi + done + pushd $(dirname $fname) > /dev/null + fname=$(pwd -P)/$(basename $fname) + popd > /dev/null + echo $fname +} + # PROGRAM START # determine whether we have gettext; make it a no-op if we do not @@ -294,16 +315,6 @@ if [ $# -lt 2 ]; then exit 1 fi -# check for and store the name of a realpath-like program -if [ $(type -t realpath) ]; then - realpath='realpath' -elif [ $(type -t readlink) ]; then - realpath='readlink -f' -else - error "$(gettext "Either realpath or readlink are required by repo-add.")" - exit 1 # $E_MISSING_PROGRAM -fi - # main routine gstmpdir=$(mktemp -d /tmp/repo-tools.XXXXXXXXXX) || (\ error "$(gettext "Cannot create temp directory for database building.")"; \ @@ -326,7 +337,7 @@ for arg in "$@"; do QUIET=1 elif [ -z "$REPO_DB_FILE" ]; then # store absolute path to repo DB - REPO_DB_FILE=$($realpath "$arg") + REPO_DB_FILE=$(realpath "$arg") if [ -f "$REPO_DB_FILE" ]; then if ! test_repo_db_file $cmd; then error "$(gettext "Repository file '%s' is not a proper pacman database.")" "$REPO_DB_FILE" -- 1.6.1.2

5 6

[pacman-dev] [PATCH] BSD/Mac OS X sed inplace backups with mktemp
by Kevin Barry 16 Feb '09

16 Feb '09

This patch changes the way that sed inplace is run in makepkg when running in versionpkg mode. Previous, sed would run with no backup file, called in the GNU way. It seems there is no cross platform method of calling sed inplace with no backup (BSD/sed uses sed -i '', GNU/sed uses sed -i'' or simply sed -i). Instead, this patch creates a backup file (from an mktemp extension) and removes it after the successful completetion of the sed operation. This means the user should not see the backup file, unless something went wrong (Full partition for example) in which case they don't lose the previous PKGBUILD as they would in previous versions. Further information about GNU/sed versus BSD/sed: On Mac OS X (Man page is from FreeBSD, so I believe it would be the same there): sed -i -e 's/foo/bar/' filename will create a backup file named filename-e sed -i -e 's/foo/bar/' -e 's/asd/dsa/' filename will give an error message because of the second -e sed -i 's/foo/bar/' filename will give an error message sed -i '' 's/foo/bar/' filename will work without creating a backup, but fails on GNU/sed sed -i'' 's/foo/bar/' filename sed -i 's/foo/bar/' filename Work only on GNU/sed Signed-off-by: Kevin Barry <barryk gmail com> --- scripts/makepkg.sh.in | 9 +++++++-- 1 files changed, 7 insertions(+), 2 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index e10c345..5af4add 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -1199,8 +1199,13 @@ devel_update() { if [ -n "$newpkgver" ]; then if [ "$newpkgver" != "$pkgver" ]; then if [ -f "./$BUILDSCRIPT" ]; then - sed -i "s/^pkgver=[^ ]*/pkgver=$newpkgver/" "./$BUILDSCRIPT" - sed -i "s/^pkgrel=[^ ]*/pkgrel=1/" "./$BUILDSCRIPT" + # Create temporary filename for sed backup. + # Needed for BSD/GNU cross-platform compatibility + TEMP_NAME=$(mktemp -u "./${BUILDSCRIPT}.XXXXXX") + TEMP_NAME=${TEMP_NAME#"./${BUILDSCRIPT}."} + sed -i".${TEMP_NAME}" -e "s/^pkgver=[^ ]*/pkgver=$newpkgver/" \ + -e "s/^pkgrel=[^ ]*/pkgrel=1/" "./$BUILDSCRIPT" && + rm -f "./${BUILDSCRIPT}.${TEMP_NAME}" source "$BUILDSCRIPT" fi fi -- 1.6.1.2

1 0

[pacman-dev] [PATCH] BSD/Mac OS X sed inplace backups
by Kevin Barry 16 Feb '09

16 Feb '09

> What does BSD sed currently do? I am assuming it leaves a backup file(s) but can you clarify the numbers and names of these? On Mac OS X (Man page is from FreeBSD, so I believe it would be the same there): sed -i -e 's/foo/bar/' filename will create a backup file named filename-e sed -i -e 's/foo/bar' -e '/asd/dsa/' filename will give an error message because sed -i 's/foo/bar' filename will give an error message sed -i '' 's/foo/bar/' filename will work without creating a backup file, but fails on GNU/sed It's really annoying that BSD sed doesn't like sed -i'' and GNU/sed doesn't like sed -i '' This patch changes the way that sed inplace is run in makepkg when running in versionpkg mode. Previous, sed would run with no backup file, called in the GNU way. It seems there is no cross platform method of calling sed inplace with no backup (BSD/sed uses sed -i '', GNU/sed uses sed -i'' or simply sed -i). Instead, this patch creates a backup file (.pkgsave) and removes it after the succesful completion of the sed operation. This means the user should not see the backup file, unless something went run (full hard drive?) in which case they don't lose the PKGBUILD as they would have before. Signed-off-by: Kevin Barry <barryk gmail com> --- scripts/makepkg.sh.in | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index e10c345..8ddf88f 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -1199,8 +1199,9 @@ devel_update() { if [ -n "$newpkgver" ]; then if [ "$newpkgver" != "$pkgver" ]; then if [ -f "./$BUILDSCRIPT" ]; then - sed -i "s/^pkgver=[^ ]*/pkgver=$newpkgver/" "./$BUILDSCRIPT" - sed -i "s/^pkgrel=[^ ]*/pkgrel=1/" "./$BUILDSCRIPT" + sed -i.pkgsave -e "s/^pkgver=[^ ]*/pkgver=$newpkgver/" \ + -e "s/^pkgrel=[^ ]*/pkgrel=1/" "./$BUILDSCRIPT" && + rm -f "./${BUILDSCRIPT}.pkgsave" source "$BUILDSCRIPT" fi fi -- 1.6.1.2

3 2

[pacman-dev] [PATCH] Mac OS X [1] Full path of stat (SIZECMD) and [2] sed -i
by Kevin Barry 15 Feb '09

15 Feb '09

Change configure.ac to use the full path of stat when on darwin/mac. This is needed for situations when a user installs the GNU/coreutils and places it in their path before /usr/bin, but the SIZECMD is already configured for Darwin's version of stat. Signed-off-by: Kevin Barry <barryk gmail com> --- configure.ac | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/configure.ac b/configure.ac index 2e7f433..216cf76 100644 --- a/configure.ac +++ b/configure.ac @@ -190,7 +190,7 @@ case "${host_os}" in ;; darwin*) host_os_darwin=yes - SIZECMD="stat -f %z" + SIZECMD="/usr/bin/stat -f %z" ;; esac -- 1.6.1.2 Patch Number Two This patch changes the way that sed inplace is run in makepkg when running in versionpkg mode. Previous, sed would run with no backup file, called in the GNU way. It seems there is no cross platform method of calling sed inplace with no backup (BSD/sed uses sed -i '', GNU/sed uses sed -i'' or simply sed -i). Instead, this patch creates a backup file (.pkgsave) and removes it after the succesful completion of the sed operation. This means the user should not see the backup file, unless something went run (full hard drive?) in which case they don't lose the PKGBUILD as they would have before. Signed-off-by: Kevin Barry <barryk gmail com> --- scripts/makepkg.sh.in | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index e10c345..8ddf88f 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -1199,8 +1199,9 @@ devel_update() { if [ -n "$newpkgver" ]; then if [ "$newpkgver" != "$pkgver" ]; then if [ -f "./$BUILDSCRIPT" ]; then - sed -i "s/^pkgver=[^ ]*/pkgver=$newpkgver/" "./$BUILDSCRIPT" - sed -i "s/^pkgrel=[^ ]*/pkgrel=1/" "./$BUILDSCRIPT" + sed -i.pkgsave -e "s/^pkgver=[^ ]*/pkgver=$newpkgver/" \ + -e "s/^pkgrel=[^ ]*/pkgrel=1/" "./$BUILDSCRIPT" && + rm -f "./${BUILDSCRIPT}.pkgsave" source "$BUILDSCRIPT" fi fi -- 1.6.1.2

3 2

[pacman-dev] Non-reentrancy of libalpm, why?
by Patrick Walton 15 Feb '09

15 Feb '09

Hi, I noticed that libalpm isn't reentrant/thread-safe, even though it defines pmhandle_t to store all global information instead of using multiple global variables. This is unfortunate for my use case. I'm thinking about writing a cross-compiling makepkg-like utility, which would need to access two package handles simultaneously at once: that of the host and that of the target. Is there a reason why libalpm isn't reentrant? And how frozen is the libalpm API supposed to be? Thanks, Patrick

2 1

[pacman-dev] PLEASE take my patches. PLEASE.
by Bryan Ischo 15 Feb '09

15 Feb '09

I hate to be reduced to begging, but after weeks of no one taking my patches, this is what I have to do. I jumped through every hoop I was asked to jump through to get my patches accepted (and to be honest, there were ALOT of hoops). My most recent patches apply to the source code without any conflicts and address all outstanding issues raised on this list. Dan McGee said that if I addressed those issues, he'd take my patches. Well I did, weeks ago, and yet, my patches are not in the tree. So this is one last attempt on my part to try to get a pacman dev to apply my patches to the master git tree. It should be trivially easy - should take all of a few seconds. Thanks, Bryan

3 3