Hold back blocking packages instead of failing the whole upgrade

A repo package that replaces something an installed AUR package still depends
on aborted the entire transaction, and would have done so on every subsequent
run - one stale AUR package was enough to cut a machine off from all updates
indefinitely. Observed in the wild: percona-server-clients replaces
libperconaserverclient without providing it, while heidisql-qt6-bin hard-depends
on it, blocking 213 unrelated package updates.

pacman names the offending package in its dependency errors, so it is now
extracted and passed to --ignore for one retry: the other 213 packages go
through and the blocker is reported. The hold is per-run, never written to
IgnorePkg, so it disappears by itself once upstream catches up.

The single retry is also now a bounded recovery loop, because fixing one
problem regularly uncovers the next - a conflict resolved with --ask=20 can
surface a dependency error behind it. Each remedy is applied at most once.

The held-back set is shown in the menu and notified only when it changes, so a
blocker waiting on an upstream fix does not produce the same message daily.

Makefile

@@ -5,7 +5,10 @@
 # targets degrade to a no-op when msgfmt/scdoc are missing, so the tree stays
 # usable for development without the build dependencies installed.
 
-VERSION      := 1.0.0
+# Overridable so a packager can pass the version it is actually building
+# (`make VERSION=$pkgver`). The literal below is the fallback for builds
+# straight from a checkout, and is what a release tag has to carry.
+VERSION      ?= 1.0.5
 
 PREFIX       ?= /usr
 DESTDIR      ?=

po/cachy-auto-update.pot

@@ -134,7 +134,7 @@ msgstr ""
 msgid "Quit"
 msgstr ""
 
-msgid "Press Enter to continue..."
+msgid "Press any key to continue..."
 msgstr ""
 
 #. Commands
@@ -243,3 +243,14 @@ msgstr ""
 
 msgid "pacman's lock file looks left over from an interrupted update. Updates are paused until it is cleared."
 msgstr ""
+
+msgid "Some packages were held back"
+msgstr ""
+
+#, c-format
+msgid "%s could not be updated and was skipped. Everything else is up to date."
+msgstr ""
+
+#, c-format
+msgid "Held back: %s"
+msgstr ""

po/de.po

@@ -135,8 +135,8 @@ msgstr "Aktuelle Bedingungen anzeigen"
 msgid "Quit"
 msgstr "Beenden"
 
-msgid "Press Enter to continue..."
+msgid "Press any key to continue..."
-msgstr "Zum Fortfahren Enter drücken …"
+msgstr "Zum Fortfahren eine Taste drücken …"
 
 #. Commands
 msgid "Automatic updates are on."
@@ -244,3 +244,14 @@ msgstr "Paketdatenbank gesperrt"
 
 msgid "pacman's lock file looks left over from an interrupted update. Updates are paused until it is cleared."
 msgstr "Die Sperrdatei von pacman scheint von einem abgebrochenen Update übrig zu sein. Bis sie entfernt ist, pausieren die Updates."
+
+msgid "Some packages were held back"
+msgstr "Einige Pakete wurden zurückgehalten"
+
+#, c-format
+msgid "%s could not be updated and was skipped. Everything else is up to date."
+msgstr "%s konnte nicht aktualisiert werden und wurde übersprungen. Alles andere ist aktuell."
+
+#, c-format
+msgid "Held back: %s"
+msgstr "Zurückgehalten: %s"

src/cachy-auto-update

@@ -170,7 +170,9 @@ cau_do_status() {
 
 cau_do_run() {
 	cau_need_root
-	exec "$CAU_LIBEXECDIR/cachy-auto-update-run" "$@"
+	# Not exec'd: this is also called from the menu, which has to survive the
+	# run and redraw afterwards.
+	"$CAU_LIBEXECDIR/cachy-auto-update-run" "$@"
 }
 
 cau_do_log() {

src/cachy-auto-update-run

@@ -40,6 +40,13 @@ if [[ $EUID -ne 0 ]]; then
 	exit 2
 fi
 
+# Force a neutral locale here rather than relying on the unit's Environment=,
+# so a run started by hand behaves exactly like one started by the timer.
+# pacman failures are classified by matching its output, and on a German system
+# that output is German. Text aimed at a person does not come through here - a
+# notification is rendered in the recipient's own locale by cau_msg_in.
+export LC_ALL=C LANGUAGE=
+
 # Re-exec under an inhibitor so a suspend or shutdown cannot land in the middle
 # of a pacman transaction and leave the database half-written. Done before
 # anything else so the whole pass is covered, including the lock.
@@ -195,6 +202,22 @@ else
 	fi
 fi
 
+# Packages skipped so the rest of the upgrade could proceed. The notification
+# only fires when the set changes: a blocker waiting on an upstream fix would
+# otherwise produce the same message every single day.
+prev_held="$(cau_state_read held_back '')"
+if [[ -n $CAU_PACMAN_HELD ]]; then
+	cau_state_write held_back "$CAU_PACMAN_HELD"
+	cau_warn "Held back: $CAU_PACMAN_HELD"
+	if [[ $CAU_PACMAN_HELD != "$prev_held" && $CFG_NOTIFY_ERROR == yes ]]; then
+		cau_notify normal "Some packages were held back" \
+			"%s could not be updated and was skipped. Everything else is up to date." \
+			"$CAU_PACMAN_HELD"
+	fi
+else
+	cau_state_clear held_back
+fi
+
 if cau_pacman_reboot_needed; then
 	cau_state_write reboot_needed 1
 	cau_info "A kernel update needs a restart"

src/lib/common.sh

@@ -146,11 +146,18 @@ cau_run_logged() {
 # Terminal helpers for the CLI
 # ---------------------------------------------------------------------------
 
+# Set by cau_bad and cau_note. The menu redraws immediately after an action,
+# which would wipe the screen; this marks that something was printed that the
+# user still has to read, so only those cases wait for a keypress. A plain
+# success needs no acknowledgement - the status block at the top of the menu
+# already shows the new state.
+CAU_UI_NEEDS_ACK=''
+
 cau_say()  { printf '%s\n' "$*"; }
 cau_head() { printf '\n%s%s%s\n\n' "$CAU_C_BOLD$CAU_C_BLUE" "$*" "$CAU_C_RESET"; }
 cau_ok()   { printf '%s✔%s %s\n' "$CAU_C_GREEN" "$CAU_C_RESET" "$*"; }
-cau_bad()  { printf '%s✘%s %s\n' "$CAU_C_RED" "$CAU_C_RESET" "$*" >&2; }
+cau_bad()  { CAU_UI_NEEDS_ACK=1; printf '%s✘%s %s\n' "$CAU_C_RED" "$CAU_C_RESET" "$*" >&2; }
-cau_note() { printf '%s•%s %s\n' "$CAU_C_DIM" "$CAU_C_RESET" "$*"; }
+cau_note() { CAU_UI_NEEDS_ACK=1; printf '%s•%s %s\n' "$CAU_C_DIM" "$CAU_C_RESET" "$*"; }
 
 # ---------------------------------------------------------------------------
 # State files

src/lib/menu.sh

@@ -87,6 +87,13 @@ cau_ui_status() {
 			"$(cau_msg "A restart is recommended to finish a kernel update.")" \
 			"$CAU_C_RESET"
 	fi
+
+	local held
+	held="$(cau_state_read held_back '')"
+	if [[ -n $held ]]; then
+		printf '\n  %s%s%s\n' "$CAU_C_YELLOW" \
+			"$(cau_msg "Held back: %s" "$held")" "$CAU_C_RESET"
+	fi
 }
 
 # cau_ui_status_conditions
@@ -145,31 +152,49 @@ cau_ui_menu() {
 		printf '  [q] %s\n' "$(cau_msg "Quit")"
 		printf '\n  > '
 
-		read -r choice || { printf '\n'; return 0; }
+		# One keypress, no Enter. -s keeps the raw character out of the
+		# display so the echo below is the only thing printed, and a failing
+		# read means EOF (Ctrl-D, or a script piping input) - that quits.
+		read -rsn1 choice || { printf '\n'; return 0; }
+		printf '%s\n' "$choice"
 
 		case "$choice" in
+			# The two switches flip and return straight to the menu, where the
+			# status block shows the result. Only a warning or an error holds
+			# the screen (see CAU_UI_NEEDS_ACK).
 			1)
+				CAU_UI_NEEDS_ACK=''
 				if [[ $CFG_ENABLED == yes ]]; then cau_do_disable; else cau_do_enable; fi
-				cau_pause
+				if [[ -n $CAU_UI_NEEDS_ACK ]]; then cau_pause; fi
 				;;
 			2)
+				CAU_UI_NEEDS_ACK=''
 				if [[ $CFG_NOTIFICATIONS == yes ]]; then
 					cau_do_notifications off
 				else
 					cau_do_notifications on
 				fi
-				cau_pause
+				if [[ -n $CAU_UI_NEEDS_ACK ]]; then cau_pause; fi
 				;;
 			3) cau_do_run --force; cau_pause ;;
 			4) cau_do_log; cau_pause ;;
 			5) cau_ui_status_conditions; cau_pause ;;
-			q|Q|'') return 0 ;;
+			q|Q) return 0 ;;
+			$'\e')
+				# Arrow keys and friends arrive as ESC [ X. Swallow the rest so
+				# one keypress does not redraw the menu three times. Escape is
+				# deliberately not a quit key: that would make a stray arrow
+				# key close the menu.
+				read -rsn2 -t 0.05 _ 2>/dev/null || true
+				;;
+			# Anything else, Enter included, just redraws.
 			*) ;;
 		esac
 	done
 }
 
 cau_pause() {
-	printf '\n  %s' "$(cau_msg "Press Enter to continue...")"
+	printf '\n  %s' "$(cau_msg "Press any key to continue...")"
-	read -r _ || true
+	read -rsn1 _ || true
+	printf '\n'
 }

src/lib/pkg_pacman.sh

@@ -10,6 +10,9 @@
 CAU_PACMAN_COUNT=0
 CAU_PACMAN_PENDING=''
 
+# Packages that had to be skipped so the rest of the upgrade could go through.
+CAU_PACMAN_HELD=''
+
 # Base flags for every unattended pacman invocation.
 cau_pacman_flags() {
 	printf '%s\n' --noconfirm --color never --noprogressbar --disable-download-timeout
@@ -55,6 +58,8 @@ _cau_pacman_classify() {
 
 	if grep -qiE 'are in conflict|unresolvable package conflicts' "$log"; then
 		printf 'conflict\n'
+	elif grep -qiE 'could not satisfy dependencies|breaks dependency|unable to satisfy dependency' "$log"; then
+		printf 'dependency\n'
 	elif grep -qiE 'signature from .* is (unknown trust|marginal trust|invalid)|invalid or corrupted package \(PGP signature\)|key ".*" is unknown|keyring is not writable' "$log"; then
 		printf 'keyring\n'
 	elif grep -qiE 'exists in filesystem' "$log"; then
@@ -64,6 +69,25 @@ _cau_pacman_classify() {
 	fi
 }
 
+# _cau_pacman_blockers <logfile>
+# The packages standing in the way of an otherwise fine upgrade. pacman names
+# them in its dependency errors:
+#
+#   :: removing libperconaserverclient breaks dependency 'libperconaserverclient'
+#      required by heidisql-qt6-bin
+#   :: unable to satisfy dependency 'foo' required by bar
+#
+# In the first form the package being removed is the one to keep; in the second
+# it is the package that cannot be installed.
+_cau_pacman_blockers() {
+	local log="$1"
+
+	{
+		sed -nE "s/.*removing ([^ ]+) breaks dependency.*/\\1/p" "$log"
+		sed -nE "s/.*unable to satisfy dependency '[^']*' required by ([^ ]+).*/\\1/p" "$log"
+	} | grep -E '^[A-Za-z0-9@._+-]+$' | sort -u
+}
+
 # cau_pacman_update
 # Returns 0 on success (including "nothing to do"), 1 on a failure the user
 # needs to hear about. CAU_PACMAN_COUNT holds how many packages moved.
@@ -76,15 +100,31 @@ cau_pacman_update() {
 		return 0
 	fi
 
+	if [[ -n $CAU_PACMAN_PENDING ]]; then
 		CAU_PACMAN_COUNT="$(grep -c . <<< "$CAU_PACMAN_PENDING")"
 		[[ $CAU_PACMAN_COUNT =~ ^[0-9]+$ ]] || CAU_PACMAN_COUNT=0
 		cau_info "Updating $CAU_PACMAN_COUNT repository package(s)"
+	else
+		# checkupdates is unavailable, so the list is unknown and pacman is
+		# asked to work it out itself.
+		CAU_PACMAN_COUNT=0
+		cau_info "Running a full system upgrade (pending list unavailable)"
+	fi
 
 	mapfile -t flags < <(cau_pacman_flags)
 	log="$(mktemp)" || return 1
 
-	if pacman -Syu "${flags[@]}" > "$log" 2>&1; then
+	# Recovery loop rather than a single retry: fixing one problem regularly
+	# uncovers the next (a conflict resolved into a dependency error, say).
+	# Each remedy is applied at most once, so this always terminates.
+	local -a extra=() blockers=() tried=()
+	local attempt=0 b
+
+	while true; do
+		if pacman -Syu "${flags[@]}" "${extra[@]}" > "$log" 2>&1; then
 			cat "$log" >> "$CAU_RUNLOG" 2>/dev/null
+			grep -E '^(removing|replacing) ' "$log" 2>/dev/null \
+				| while read -r line; do cau_info "  $line"; done
 			rm -f "$log"
 			return 0
 		fi
@@ -93,6 +133,11 @@ cau_pacman_update() {
 		kind="$(_cau_pacman_classify "$log")"
 		cau_warn "pacman -Syu failed ($kind)"
 
+		if (( ++attempt > 3 )) || [[ " ${tried[*]} " == *" $kind "* ]]; then
+			break
+		fi
+		tried+=("$kind")
+
 		case "$kind" in
 			keyring)
 				# A stale keyring is the one failure that is always safe to fix
@@ -104,12 +149,6 @@ cau_pacman_update() {
 				if (( ${#keyrings[@]} )); then
 					cau_run_logged pacman -Sy --noconfirm --color never "${keyrings[@]}" || true
 				fi
-			if pacman -Syu "${flags[@]}" > "$log" 2>&1; then
-				cat "$log" >> "$CAU_RUNLOG" 2>/dev/null
-				rm -f "$log"
-				return 0
-			fi
-			cat "$log" >> "$CAU_RUNLOG" 2>/dev/null
 				;;
 
 			conflict)
@@ -119,18 +158,29 @@ cau_pacman_update() {
 				# question bitmask: 4 = CONFLICT_PKG, 16 = REMOVE_PKGS.
 				if [[ $CFG_RESOLVE_CONFLICTS != yes ]]; then
 					cau_error "Package conflict requires a decision (AutoResolveConflicts is off)"
-				rm -f "$log"
+					break
-				return 1
 				fi
 				cau_info "Resolving package conflicts automatically and retrying"
-			if pacman -Syu "${flags[@]}" --ask=20 > "$log" 2>&1; then
+				extra+=(--ask=20)
-				cat "$log" >> "$CAU_RUNLOG" 2>/dev/null
+				;;
-				grep -E '^(removing|replacing) ' "$log" 2>/dev/null \
+
-					| while read -r line; do cau_info "  $line"; done
+			dependency)
-				rm -f "$log"
+				# Something installed still depends on a package the repos want
-				return 0
+				# to drop or replace - almost always an AUR package that has not
+				# caught up yet. Nothing here can fix that, and it is not worth
+				# failing over: letting one stuck package block every other
+				# update indefinitely is far worse on an unattended machine.
+				# Hold the blockers back and upgrade everything else.
+				mapfile -t blockers < <(_cau_pacman_blockers "$log")
+				if (( ${#blockers[@]} == 0 )); then
+					cau_error "Dependency problem with no package to hold back"
+					break
 				fi
-			cat "$log" >> "$CAU_RUNLOG" 2>/dev/null
+				for b in "${blockers[@]}"; do
+					extra+=(--ignore "$b")
+				done
+				CAU_PACMAN_HELD="${blockers[*]}"
+				cau_warn "Holding back ${blockers[*]} and retrying without them"
 				;;
 
 			filesystem)
@@ -138,13 +188,18 @@ cau_pacman_update() {
 				# silently clobber something the user put there deliberately, so
 				# this one stays a human decision.
 				cau_error "Files on disk conflict with the update; manual review needed"
-			rm -f "$log"
+				break
-			return 1
+				;;
+
+			*)
+				break
 				;;
 		esac
+	done
 
 	rm -f "$log"
 	CAU_PACMAN_COUNT=0
+	CAU_PACMAN_HELD=''
 	return 1
 }