Teach portsnap how to ignore unwanted parts of the ports tree. A line

of the form "REFUSE foo" in portsnap.conf will result in parts of the
tree matching "^foo" being (a) not extracted by "portsnap extract", (b)
not updated by "portsnap update", and (c) not having any patches or new
ports downloaded by "portsnap fetch" or "portsnap cron". The example
shown in portsnap.conf demonstrates ignoring all the language categories.

As mentioned in portsnap.conf.5, the use of an imcomplete ports tree is
not officially supported; but this is something which many users have
requested, so I'm adding it anyway.

PR:		bin/85619 (but not the patch provided therein)
MFC after:	1 month

3 files changed, 108 insertions, 8 deletions

diff --git a/etc/portsnap.conf b/etc/portsnap.conf
index 665061e..3497edb 100644
--- a/etc/portsnap.conf
+++ b/etc/portsnap.conf
@@ -16,3 +16,9 @@ SERVERNAME=portsnap.FreeBSD.org
 # a PGP-signed email from <security-officer@FreeBSD.org> telling you to
 # change it and explaining why.
 KEYPRINT=9b5feee6d69f170e3dd0a2c8e469ddbd64f13f978f2f3aede40c98633216c330
+
+# Example of ignoring parts of the ports tree.  If you know that you
+# absolutely will not need certain parts of the tree, this will save
+# some bandwidth and disk space.  See the manual page for more details.
+# REFUSE arabic chinese french german hebrew hungarian japanese
+# REFUSE korean polish portuguese russian ukrainian vietnamese
diff --git a/share/man/man5/portsnap.conf.5 b/share/man/man5/portsnap.conf.5
index eeae1f1..ba632166 100644
--- a/share/man/man5/portsnap.conf.5
+++ b/share/man/man5/portsnap.conf.5
@@ -87,14 +87,48 @@ is used.
 .Pp
 If more than one line of any of the above forms is included in
 .Nm
-then only the last one will take effect.  Any lines not of the above
-forms will be ignored.
+then only the last one will take effect.
+.Pp
+Finally, a line of the form
+.Dl REFUSE foo bar
+will instruct
+.Xr portsnap 8
+to ignore parts of the ports tree with paths starting with
+.Ar foo 
+or
+.Ar bar ,
+which are interpreted as extended regular expressions by
+.Xr egrep 1 .
+This will result in those parts of the tree not being updated
+in the compressed snapshot when the
+.Cm fetch
+and
+.Cm cron
+commands are used and not being extracted when the
+.Cm extract
+command is used (unless a specific
+.Ar path
+is passed to
+.Xr portsnap 8 ) ,
+and if those parts of the ports tree are present they
+will not be updated when the
+.Cm update
+command is used.
+Unlike the other options, the parameters in REFUSE lines
+accumulate and all such lines are considered.
+.Bf Em
+Note that operating with an incomplete ports tree is not
+supported and may cause unexpected results.
+.Ef
+.Pp
+Any lines not of the above forms will be ignored.
 .Sh FILES
 .Bl -tag -width "/etc/portsnap.conf"
 .It /etc/portsnap.conf
 Default location of the portsnap configuration file.
 .El
 .Sh SEE ALSO
+.Xr egrep 1
 .Xr fetch 1
 .Xr portsnap 8
 .Xr sha256 8
diff --git a/usr.sbin/portsnap/portsnap/portsnap.sh b/usr.sbin/portsnap/portsnap/portsnap.sh
index 38098e7..de263f0 100644
--- a/usr.sbin/portsnap/portsnap/portsnap.sh
+++ b/usr.sbin/portsnap/portsnap/portsnap.sh
@@ -81,6 +81,7 @@ init_params() {
 	DDSTATS=""
 	INDEXONLY=""
 	SERVERNAME=""
+	REFUSE=""
 }
 
 # Parse the command line
@@ -173,6 +174,8 @@ default_conffile() {
 # Read {KEYPRINT, SERVERNAME, WORKDIR, PORTSDIR} from the configuration
 # file if they haven't already been set.  If the configuration
 # file doesn't exist, do nothing.
+# Also read REFUSE (which cannot be set via the command line) if it is
+# present in the configuration file.
 parse_conffile() {
 	if [ -r "${CONFFILE}" ]; then
 		for X in KEYPRINT WORKDIR PORTSDIR SERVERNAME; do
@@ -182,6 +185,13 @@ parse_conffile() {
 				    cut -f 2- -d '=' | tail -1`
 			fi
 		done
+
+		if grep -qE "^REFUSE[[:space:]]" ${CONFFILE}; then
+			REFUSE="^(`
+				grep -E "^REFUSE[[:space:]]" "${CONFFILE}" |
+				    cut -c 7- | xargs echo | tr ' ' '|'
+				`)"
+		fi
 	fi
 }
 
@@ -668,6 +678,21 @@ fetch_update() {
 	    cut -f 2 -d '|'`.gz > INDEX.new
 	fetch_index_sanity || return 1
 
+# If we have decided to refuse certain updates, construct a hybrid index which
+# is equal to the old index for parts of the tree which we don't want to
+# update, and equal to the new index for parts of the tree which gets updates.
+# This means that we should always have a "complete snapshot" of the ports
+# tree -- with the caveat that it isn't actually a snapshot.
+	if [ ! -z "${REFUSE}" ]; then
+		echo "Refusing to download updates for ${REFUSE}"	\
+		    >${QUIETREDIR}
+
+		grep -Ev "${REFUSE}" INDEX.new > INDEX.tmp
+		grep -E "${REFUSE}" INDEX |
+		    sort -m -k 1,1 -t '|' - INDEX.tmp > INDEX.new
+		rm -f INDEX.tmp
+	fi
+
 # Generate a list of wanted ports patches
 	join -t '|' -o 1.2,2.2 INDEX INDEX.new |
 	    fetch_make_patchlist > patchlist
@@ -761,14 +786,34 @@ extract_indices() {
 	echo "done."
 }
 
-# Create .portsnap.INDEX
+# Create .portsnap.INDEX; if we are REFUSEing to touch certain directories,
+# merge the values from any exiting .portsnap.INDEX file.
 extract_metadata() {
-	sort ${WORKDIR}/INDEX > ${PORTSDIR}/.portsnap.INDEX
+	if [ -z "${REFUSE}" ]; then
+		sort ${WORKDIR}/INDEX > ${PORTDIR}/.portsnap.INDEX
+	elif [ -f ${PORTSDIR}/.portsnap.INDEX ]; then
+		grep -E "${REFUSE}" ${PORTSDIR}/.portsnap.INDEX	\
+		    > ${PORTSDIR}/.portsnap.INDEX.tmp
+		grep -vE "${REFUSE}" ${WORKDIR}/INDEX | sort |
+		    sort -m - ${PORTSDIR}/.portsnap.INDEX.tmp	\
+		    > ${PORTSDIR}/.portsnap.INDEX
+		rm -f ${PORTSDIR}/.portsnap.INDEX.tmp
+	else
+		grep -vE "${REFUSE}" ${WORKDIR}/INDEX | sort \
+		    > ${PORTSDIR}/.portsnap.INDEX
+	fi
 }
 
 # Do the actual work involved in "extract"
 extract_run() {
-	if ! grep "^${EXTRACTPATH}" ${WORKDIR}/INDEX | while read LINE; do
+	if !
+		if ! [ -z "${EXTRACTPATH}" ]; then
+			grep "^${EXTRACTPATH}" ${WORKDIR}/INDEX
+		elif ! [ -z "${REFUSE}" ]; then
+			grep -vE "${REFUSE}" ${WORKDIR}/INDEX
+		else
+			cat ${WORKDIR}/INDEX
+		fi | while read LINE; do
 		FILE=`echo ${LINE} | cut -f 1 -d '|'`
 		HASH=`echo ${LINE} | cut -f 2 -d '|'`
 		echo ${PORTSDIR}/${FILE}
@@ -813,14 +858,29 @@ update_run() {
 		return 0
 	fi
 
+# If we are REFUSEing to touch certain directories, don't remove files
+# from those directories (even if they are out of date)
 	echo -n "Removing old files and directories... "
-	sort ${WORKDIR}/INDEX | comm -23 ${PORTSDIR}/.portsnap.INDEX - |
-	    cut -f 1 -d '|' | lam -s "${PORTSDIR}/" - | xargs rm -rf
+	if ! [ -z "${REFUSE}" ]; then 
+		sort ${WORKDIR}/INDEX |
+		    comm -23 ${PORTSDIR}/.portsnap.INDEX - | cut -f 1 -d '|' |
+		    grep -vE "${REFUSE}" |
+		    lam -s "${PORTSDIR}/" - | xargs rm -rf
+	else
+		sort ${WORKDIR}/INDEX |
+		    comm -23 ${PORTSDIR}/.portsnap.INDEX - | cut -f 1 -d '|' |
+		    lam -s "${PORTSDIR}/" - | xargs rm -rf
+	fi
 	echo "done."
 
 # Install new files
 	echo "Extracting new files:"
-	if ! sort ${WORKDIR}/INDEX |
+	if !
+		if ! [ -z "${REFUSE}" ]; then
+			grep -vE "${REFUSE}" ${WORKDIR}/INDEX | sort
+		else
+			sort ${WORKDIR}/INDEX
+		fi |
 	    comm -13 ${PORTSDIR}/.portsnap.INDEX - |
 	    while read LINE; do
 		FILE=`echo ${LINE} | cut -f 1 -d '|'`