[pacman-dev] [PATCH 1/3] bacman: allow for parallel packaging

Gordian Edenhofer gordian.edenhofer at gmail.com
Sun Aug 14 19:15:17 UTC 2016


On Sun, 2016-08-14 at 14:23 -0400, Dave Reisner wrote:
> On Sun, Aug 14, 2016 at 08:10:52PM +0200, Gordian Edenhofer wrote:
> > 
> > * move the actual assembly process into its own function
> > * allow for packaging multiple packages with one command
> > * handle SIGHUP SIGINT SIGTERM and remove working dirs accordingly
> > * add some comments
> 
> Can't this be done already by running multiple instances of bacman?
> In
> other words, what's gained by adding this complexity to bacman
> itself?

Yes, sure this basically achieves the same purpose. Though I would
argue the added complexity is negligible since it merely puts the
assembly process in its own function, nothing more. The git-diff may
suggest something different but this is mainly due to the changed
indentation.

> Bonus, if you're on a system that has bash 4.3, you can do better
> than
> the parallelize function you wrote and avoiding a tight loop and the
> CPU
> it'll eat up:
> 
>   pbacman() {
>     local max_jobs=$1; shift
> 
>     for arg; do
>       while $(( $(jobs -p | wc -l) > max_jobs )); do
>         wait -n
>       done
> 
>       bacman "$arg" &
>     done
>     wait
>   }
> 
> We can't use this in bacman without bumping the bash version
> requirement
> which probably isn't something the pacman maintainers are willing to
> do.

I am aware this probably isn't an elegant way to do things but it is
simple and gets the job done. The CPU usage can be drastically
decreased by adding a short sleep in between the checks. But I guess
there is still room for improvement.

> > 
> > 
> > Signed-off-by: Gordian Edenhofer <gordian.edenhofer at gmail.com>
> > ---
> >  contrib/bacman.sh.in | 460 ++++++++++++++++++++++++++-------------
> > ------------
> >  1 file changed, 238 insertions(+), 222 deletions(-)
> >  mode change 100644 => 100755 contrib/bacman.sh.in
> > 
> > diff --git a/contrib/bacman.sh.in b/contrib/bacman.sh.in
> > old mode 100644
> > new mode 100755
> > index a611c1a..4cd78e4
> > --- a/contrib/bacman.sh.in
> > +++ b/contrib/bacman.sh.in
> > @@ -33,9 +33,18 @@ ARGS=("$@")
> >  
> >  m4_include(../scripts/library/output_format.sh)
> >  
> > -#
> > -# User Friendliness
> > -#
> > +# Lazy recursive clean up of temporary dirs
> > +work_dir_root="${TMPDIR:-/tmp}/bacman"
> > +clean_up() {
> > +	rm -r "${work_dir_root}".*
> > +	echo
> > +	exit
> > +}
> > +
> > +# Trap termination signals
> > +trap clean_up SIGHUP SIGINT SIGTERM
> > +
> > +# Print usage information
> >  usage() {
> >  	echo "${myname} (pacman) v${myver}"
> >  	echo
> > @@ -46,12 +55,14 @@ usage() {
> >  	echo "Example: ${myname} linux-headers"
> >  }
> >  
> > +# Print version information
> >  version() {
> >  	printf "%s %s\n" "$myname" "$myver"
> >  	echo 'Copyright (C) 2008 locci
> > <carlocci_at_gmail_dot_com>'
> >  	echo 'Copyright (C) 2008-2016 Pacman Development Team <pac
> > man-dev at archlinux.org>'
> >  }
> >  
> > +# Check for specified arguments
> >  while [[ ! -z $1 ]]; do
> >  	if [[ $1 == "--nocolor" ]]; then
> >  		USE_COLOR='n'
> > @@ -64,13 +75,16 @@ while [[ ! -z $1 ]]; do
> >  	fi
> >  done
> >  
> > +# Configure colored output
> >  m4_include(../scripts/library/term_colors.sh)
> >  
> > -if (( $# != 1 )); then
> > +# Break if no argument was given
> > +if (( $# < 1 )); then
> >  	usage
> >  	exit 1
> >  fi
> >  
> > +# Print usage or version if requested
> >  if [[ $1 = -@(h|-help) ]]; then
> >  	usage
> >  	exit 0
> > @@ -79,9 +93,7 @@ elif [[ $1 = -@(V|-version) ]]; then
> >  	exit 0
> >  fi
> >  
> > -#
> > -# Fakeroot support
> > -#
> > +# Run in fakeroot if EUID is not root
> >  if (( EUID )); then
> >  	if [[ -f /usr/bin/fakeroot ]]; then
> >  		msg "Entering fakeroot environment"
> > @@ -94,264 +106,268 @@ if (( EUID )); then
> >  	fi
> >  fi
> >  
> > -#
> > -# Setting environmental variables
> > -#
> > +# Source environmental variables and specify fallbacks
> >  if [[ ! -r @sysconfdir@/pacman.conf ]]; then
> >  	error "unable to read @sysconfdir@/pacman.conf"
> >  	exit 1
> >  fi
> > -
> >  eval $(awk '/DBPath/ {print $1$2$3}' @sysconfdir@/pacman.conf)
> >  pac_db="${DBPath:- at localstatedir@/lib/pacman/}/local"
> > -
> >  if [[ ! -r @sysconfdir@/makepkg.conf ]]; then
> >  	error "unable to read @sysconfdir@/makepkg.conf"
> >  	exit 1
> >  fi
> > -
> >  source "@sysconfdir@/makepkg.conf"
> >  if [[ -r ~/.makepkg.conf ]]; then
> >  	source ~/.makepkg.conf
> >  fi
> > -
> >  pkg_dest="${PKGDEST:-$PWD}"
> >  pkg_pkger=${PACKAGER:-'Unknown Packager'}
> >  
> > -pkg_name="$1"
> > -pkg_dir=("$pac_db/$pkg_name"-+([^-])-+([^-]))
> > -pkg_namver=("${pkg_dir[@]##*/}")
> > -
> > -#
> > -# Checks everything is in place
> > -#
> > +# Check for an existing database
> >  if [[ ! -d $pac_db ]]; then
> >  	error "pacman database directory ${pac_db} not found"
> >  	exit 1
> >  fi
> >  
> > -if (( ${#pkg_dir[@]} != 1 )); then
> > -	error "%d entries for package %s found in pacman database"
> > \
> > -		${#pkg_dir[@]} "${pkg_name}"
> > -	msg2 "%s" "${pkg_dir[@]}"
> > -	exit 1
> > -fi
> > -
> > -if [[ ! -d $pkg_dir ]]; then
> > -	error "package %s is found in pacman database,"
> > "${pkg_name}"
> > -	plain "       but '%s' is not a directory" "${pkg_dir}"
> > -	exit 1
> > -fi
> > -
> > -#
> > -# Begin
> > -#
> > -msg "Package: ${pkg_namver}"
> > -work_dir=$(mktemp -d "${TMPDIR:-/tmp}/bacman.XXXXXXXXXX")
> > -cd "$work_dir" || exit 1
> > -
> > -#
> > -# File copying
> > -#
> > -msg2 "Copying package files..."
> > -
> > -while read i; do
> > -	if [[ -z $i ]]; then
> > -		continue
> > +# Assemble a single package: $1 = pkgname
> > +fakebuild() {
> > +	pkg_name="$1"
> > +	pkg_dir=("$pac_db/$pkg_name"-+([^-])-+([^-]))
> > +	pkg_namver=("${pkg_dir[@]##*/}")
> > +
> > +	# Checks database for specified package
> > +	if (( ${#pkg_dir[@]} != 1 )); then
> > +		error "%d entries for package %s found in pacman
> > database" \
> > +			${#pkg_dir[@]} "${pkg_name}"
> > +		msg2 "%s" "${pkg_dir[@]}"
> > +		exit 1
> >  	fi
> > -
> > -	if [[ $i == %+([A-Z])% ]]; then
> > -		current=$i
> > -		continue
> > +	if [[ ! -d $pkg_dir ]]; then
> > +		error "package %s is found in pacman database,"
> > "${pkg_name}"
> > +		plain "       but '%s' is not a directory"
> > "${pkg_dir}"
> > +		exit 1
> >  	fi
> >  
> > -	case "$current" in
> > -		%FILES%)
> > -			local_file="/$i"
> > -			package_file="$work_dir/$i"
> > +	# Create working directory
> > +	msg "Package: ${pkg_namver}"
> > +	work_dir=$(mktemp -d "${work_dir_root}.XXXXXXXXXX")
> > +	cd "$work_dir" || exit 1
> >  
> > -			if [[ ! -e $local_file ]]; then
> > -				warning "package file $local_file
> > is missing"
> > -				continue
> > -			fi
> > -			;;
> > -
> > -		%BACKUP%)
> > -			# Get the MD5 checksum.
> > -			original_md5="${i##*$'\t'}"
> > -			# Strip the md5sum after the tab.
> > -			i="${i%$'\t'*}"
> > -			local_file="/$i.pacnew"
> > -			package_file="$work_dir/$i"
> > -
> > -			# Include unmodified .pacnew files.
> > -			local_md5="$(md5sum "$local_file" | cut
> > -d' ' -f1)"
> > -			if [[ $INCLUDE_PACNEW == 'n' ]] \
> > -			|| [[ ! -e $local_file ]] \
> > -			|| [[ $local_md5 != $original_md5 ]]; then
> > -				# Warn about modified files.
> > -				local_md5="$(md5sum "/$i" | cut
> > -d' ' -f1)"
> > -				if [[ $local_md5 != $original_md5
> > ]]; then
> > -					warning "package file /$i
> > has been modified"
> > -				fi
> > -				# Let the normal file be included
> > in the %FILES% list.
> > -				continue
> > -			fi
> > -			;;
> > +	# Assemble list of files which belong to the package and
> > tar them
> > +	msg2 "Copying package files..."
> > +	while read i; do
> > +		if [[ -z $i ]]; then
> > +			continue
> > +		fi
> >  
> > -		*)
> > +		if [[ $i == %+([A-Z])% ]]; then
> > +			current=$i
> >  			continue
> > -			;;
> > -	esac
> > +		fi
> >  
> > -	ret=0
> > -	bsdtar -cnf - -s'/.pacnew$//' "$local_file" 2> /dev/null |
> > bsdtar -xpf - 2> /dev/null
> > +		case "$current" in
> > +			%FILES%)
> > +				local_file="/$i"
> > +				package_file="$work_dir/$i"
> > +
> > +				if [[ ! -e $local_file ]]; then
> > +					warning "package file
> > $local_file is missing"
> > +					continue
> > +				fi
> > +				;;
> > +
> > +			%BACKUP%)
> > +				# Get the MD5 checksum.
> > +				original_md5="${i##*$'\t'}"
> > +				# Strip the md5sum after the tab.
> > +				i="${i%$'\t'*}"
> > +				local_file="/$i.pacnew"
> > +				package_file="$work_dir/$i"
> > +
> > +				# Include unmodified .pacnew
> > files.
> > +				local_md5="$(md5sum "$local_file"
> > | cut -d' ' -f1)"
> > +				if [[ $INCLUDE_PACNEW == 'n' ]] \
> > +				|| [[ ! -e $local_file ]] \
> > +				|| [[ $local_md5 != $original_md5
> > ]]; then
> > +					# Warn about modified
> > files.
> > +					local_md5="$(md5sum "/$i"
> > | cut -d' ' -f1)"
> > +					if [[ $local_md5 !=
> > $original_md5 ]]; then
> > +						warning "package
> > file /$i has been modified"
> > +					fi
> > +					# Let the normal file be
> > included in the %FILES% list.
> > +					continue
> > +				fi
> > +				;;
> >  
> > -	# Workaround to bsdtar not reporting a missing file as an
> > error
> > -	if ! [[ -e $package_file || -L $package_file ]]; then
> > -		error "unable to add $local_file to the package"
> > -		plain "       If your user does not have
> > permission to read this file, then"
> > -		plain "       you will need to run $myname as
> > root."
> > +			*)
> > +				continue
> > +				;;
> > +		esac
> > +
> > +		# Tar files
> > +		ret=0
> > +		bsdtar -cnf - -s'/.pacnew$//' "$local_file" 2>
> > /dev/null | bsdtar -xpf - 2> /dev/null
> > +		# Workaround to bsdtar not reporting a missing
> > file as an error
> > +		if ! [[ -e $package_file || -L $package_file ]];
> > then
> > +			error "unable to add $local_file to the
> > package"
> > +			plain "       If your user does not have
> > permission to read this file, then"
> > +			plain "       you will need to run $myname
> > as root."
> > +			rm -rf "$work_dir"
> > +			exit 1
> > +		fi
> > +	done < "$pkg_dir"/files
> > +
> > +	ret=$?
> > +	if (( ret )); then
> >  		rm -rf "$work_dir"
> >  		exit 1
> >  	fi
> > -done < "$pkg_dir"/files
> >  
> > -ret=$?
> > -if (( ret )); then
> > -	rm -rf "$work_dir"
> > -	exit 1
> > -fi
> > +	# Calculate package size
> > +	pkg_size=$(du -sk | awk '{print $1 * 1024}')
> >  
> > -pkg_size=$(du -sk | awk '{print $1 * 1024}')
> > -
> > -#
> > -# .PKGINFO stuff
> > -# TODO adopt makepkg's write_pkginfo() into this or
> > scripts/library
> > -#
> > -msg2 "Generating .PKGINFO metadata..."
> > -echo "# Generated by $myname $myver"    > .PKGINFO
> > -if [[ $INFAKEROOT == "1" ]]; then
> > -	echo "# Using $(fakeroot -v)"    >> .PKGINFO
> > -fi
> > -echo "# $(LC_ALL=C date)"            >> .PKGINFO
> > -echo "#"                    >> .PKGINFO
> > -
> > -while read i; do
> > -	if [[ -z $i ]]; then
> > -		continue;
> > +	# Reconstruct .PKGINFO from database
> > +	# TODO adopt makepkg's write_pkginfo() into this or
> > scripts/library
> > +	msg2 "Generating .PKGINFO metadata..."
> > +	echo "# Generated by $myname $myver"    > .PKGINFO
> > +	if [[ $INFAKEROOT == "1" ]]; then
> > +		echo "# Using $(fakeroot -v)"    >> .PKGINFO
> >  	fi
> > -
> > -	if [[ $i == %+([A-Z])% ]]; then
> > -		current=$i
> > -		continue
> > +	echo "# $(LC_ALL=C date)"    >> .PKGINFO
> > +	echo "#"    >> .PKGINFO
> > +	while read i; do
> > +		if [[ -z $i ]]; then
> > +			continue;
> > +		fi
> > +		if [[ $i == %+([A-Z])% ]]; then
> > +			current=$i
> > +			continue
> > +		fi
> > +
> > +		case "$current" in
> > +			# desc
> > +			%NAME%)
> > +				echo "pkgname = $i"    >> .PKGINFO
> > +				;;
> > +			%VERSION%)
> > +				echo "pkgver = $i"    >> .PKGINFO
> > +				;;
> > +			%DESC%)
> > +				echo "pkgdesc = $i"    >> .PKGINFO
> > +				;;
> > +			%URL%)
> > +				echo "url = $i"    >> .PKGINFO
> > +				;;
> > +			%LICENSE%)
> > +				echo "license = $i"    >> .PKGINFO
> > +				;;
> > +			%ARCH%)
> > +				echo "arch = $i"    >> .PKGINFO
> > +				pkg_arch="$i"
> > +				;;
> > +			%BUILDDATE%)
> > +				echo "builddate = $(date -u
> > "+%s")"    >> .PKGINFO
> > +				;;
> > +			%PACKAGER%)
> > +				echo "packager =
> > $pkg_pkger"        >> .PKGINFO
> > +				;;
> > +			%SIZE%)
> > +				echo "size = $pkg_size"        >>
> > .PKGINFO
> > +				;;
> > +			%GROUPS%)
> > +				echo "group = $i"    >> .PKGINFO
> > +				;;
> > +			%REPLACES%)
> > +				echo "replaces = $i"    >>
> > .PKGINFO
> > +				;;
> > +			%DEPENDS%)
> > +				echo "depend = $i"   >> .PKGINFO
> > +				;;
> > +			%OPTDEPENDS%)
> > +				echo "optdepend = $i" >> .PKGINFO
> > +				;;
> > +			%CONFLICTS%)
> > +				echo "conflict = $i" >> .PKGINFO
> > +				;;
> > +			%PROVIDES%)
> > +				echo "provides = $i"  >> .PKGINFO
> > +				;;
> > +			%BACKUP%)
> > +				# Strip the md5sum after the tab
> > +				echo "backup = ${i%%$'\t'*}"   >>
> > .PKGINFO
> > +				;;
> > +		esac
> > +	done < <(cat "$pkg_dir"/{desc,files})
> > +
> > +	comp_files=".PKGINFO"
> > +
> > +	# Add instal file if present
> > +	if [[ -f $pkg_dir/install ]]; then
> > +		cp "$pkg_dir/install" "$work_dir/.INSTALL"
> > +		comp_files+=" .INSTALL"
> > +	fi
> > +	if [[ -f $pkg_dir/changelog ]]; then
> > +		cp "$pkg_dir/changelog" "$work_dir/.CHANGELOG"
> > +		comp_files+=" .CHANGELOG"
> >  	fi
> >  
> > -	case "$current" in
> > -		# desc
> > -		%NAME%)
> > -			echo "pkgname = $i"    >> .PKGINFO
> > -			;;
> > -		%VERSION%)
> > -			echo "pkgver = $i"    >> .PKGINFO
> > -			;;
> > -		%DESC%)
> > -			echo "pkgdesc = $i"    >> .PKGINFO
> > -			;;
> > -		%URL%)
> > -			echo "url = $i"    >> .PKGINFO
> > -			;;
> > -		%LICENSE%)
> > -			echo "license = $i"    >> .PKGINFO
> > -			;;
> > -		%ARCH%)
> > -			echo "arch = $i"    >> .PKGINFO
> > -			pkg_arch="$i"
> > -			;;
> > -		%BUILDDATE%)
> > -			echo "builddate = $(date -u "+%s")"    >>
> > .PKGINFO
> > -			;;
> > -		%PACKAGER%)
> > -			echo "packager = $pkg_pkger"        >>
> > .PKGINFO
> > -			;;
> > -		%SIZE%)
> > -			echo "size = $pkg_size"        >> .PKGINFO
> > -			;;
> > -		%GROUPS%)
> > -			echo "group = $i"    >> .PKGINFO
> > -			;;
> > -		%REPLACES%)
> > -			echo "replaces = $i"    >> .PKGINFO
> > -			;;
> > -		%DEPENDS%)
> > -			echo "depend = $i"   >> .PKGINFO
> > -			;;
> > -		%OPTDEPENDS%)
> > -			echo "optdepend = $i" >> .PKGINFO
> > -			;;
> > -		%CONFLICTS%)
> > -			echo "conflict = $i" >> .PKGINFO
> > -			;;
> > -		%PROVIDES%)
> > -			echo "provides = $i"  >> .PKGINFO
> > -			;;
> > -
> > -		# files
> > -		%BACKUP%)
> > -			# Strip the md5sum after the tab
> > -			echo "backup = ${i%%$'\t'*}"   >> .PKGINFO
> > -			;;
> > -	esac
> > -done < <(cat "$pkg_dir"/{desc,files})
> > -
> > -comp_files=".PKGINFO"
> > -
> > -if [[ -f $pkg_dir/install ]]; then
> > -	cp "$pkg_dir/install" "$work_dir/.INSTALL"
> > -	comp_files+=" .INSTALL"
> > -fi
> > -if [[ -f $pkg_dir/changelog ]]; then
> > -	cp "$pkg_dir/changelog" "$work_dir/.CHANGELOG"
> > -	comp_files+=" .CHANGELOG"
> > -fi
> > +	# Fixes owner:group and permissions for .PKGINFO,
> > .CHANGELOG, .INSTALL
> > +	chown root:root "$work_dir"/{.PKGINFO,.CHANGELOG,.INSTALL}
> > 2> /dev/null
> > +	chmod 644 "$work_dir"/{.PKGINFO,.CHANGELOG,.INSTALL} 2>
> > /dev/null
> >  
> > -#
> > -# Fixes owner:group and permissions for .PKGINFO, .CHANGELOG,
> > .INSTALL
> > -#
> > -chown root:root "$work_dir"/{.PKGINFO,.CHANGELOG,.INSTALL} 2>
> > /dev/null
> > -chmod 644 "$work_dir"/{.PKGINFO,.CHANGELOG,.INSTALL} 2> /dev/null
> > +	# Generate the package
> > +	msg2 "Generating the package..."
> >  
> > -#
> > -# Generate the package
> > -#
> > -msg2 "Generating the package..."
> > -
> > -pkg_file="$pkg_dest/$pkg_namver-$pkg_arch${PKGEXT}"
> > -ret=0
> > -
> > -# TODO: Maybe this can be set globally for robustness
> > -shopt -s -o pipefail
> > -bsdtar -cf - $comp_files * |
> > -case "$PKGEXT" in
> > -	*tar.gz)  gzip -c -f -n ;;
> > -	*tar.bz2) bzip2 -c -f ;;
> > -	*tar.xz)  xz -c -z - ;;
> > -	*tar.Z)   compress -c -f ;;
> > -	*tar)     cat ;;
> > -	*) warning "'%s' is not a valid archive extension." \
> > -	"$PKGEXT"; cat ;;
> > -esac > "${pkg_file}"; ret=$?
> > -
> > -if (( ret )); then
> > -	error "Unable to write package to $pkg_dest"
> > -	plain "       Maybe the disk is full or you do not have
> > write access"
> > +	pkg_file="$pkg_dest/$pkg_namver-$pkg_arch${PKGEXT}"
> > +	ret=0
> > +
> > +	# TODO: Maybe this can be set globally for robustness
> > +	shopt -s -o pipefail
> > +	bsdtar -cf - $comp_files * |
> > +	case "$PKGEXT" in
> > +		*tar.gz)  gzip -c -f -n ;;
> > +		*tar.bz2) bzip2 -c -f ;;
> > +		*tar.xz)  xz -c -z - ;;
> > +		*tar.Z)   compress -c -f ;;
> > +		*tar)     cat ;;
> > +		*) warning "'%s' is not a valid archive
> > extension." \
> > +		"$PKGEXT"; cat ;;
> > +	esac > "${pkg_file}"; ret=$?
> > +
> > +	# Move compressed package to destination
> > +	if (( ret )); then
> > +		error "Unable to write package to $pkg_dest"
> > +		plain "       Maybe the disk is full or you do not
> > have write access"
> > +		rm -rf "$work_dir"
> > +		exit 1
> > +	fi
> > +
> > +	# Clean up working directory
> >  	rm -rf "$work_dir"
> > -	exit 1
> > -fi
> > +	msg "Done."
> > +}
> >  
> > -rm -rf "$work_dir"
> > +# Run fakebuild in parralel with at maximum $MAX_JOBS jobs
> > +# By default only run one job
> > +MAX_JOBS=${MAX_JOBS:-1}
> > +parallelize() {
> > +	while [[ $# -gt 0 ]] ; do
> > +		job_count=($(jobs -p))
> > +		if [[ ${#job_count[@]} -lt $MAX_JOBS ]] ; then
> > +			fakebuild "$1" &
> > +			shift
> > +		fi
> > +	done
> > +	wait
> > +}
> >  
> > -msg "Done."
> > +# Initiate assembly function
> > +if [[ $MAX_JOBS -gt "1" ]]; then
> > +	parallelize "$@"
> > +else
> > +	for PKG in $@; do fakebuild $PKG; done
> > +fi
> >  
> >  exit 0
> >  
> > -- 
> > 2.9.2
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part
URL: <https://lists.archlinux.org/pipermail/pacman-dev/attachments/20160814/d4e7de25/attachment-0001.asc>


More information about the pacman-dev mailing list