pre-MDEV-30157 & pre-MDEV-28669: fixes before the main corrections

This commit adds even more correct handling of parameters
with paths when they contain leading or trailing spaces and/or
slashes. Also it fixes problems that occur when the user specified
explicit paths to additional directories, but these paths match
the specified path of the data directory - in this case, additional
subdirectories should be treated (in relation to the data directory)
in the same way as if these paths were not specified or as if they
are implicitly specified as "." or "./". But prior to this fix,
existing code treated any values as if they were completely
separate directories, whether or not they actually point to the
same location to which datadir points to - and this sometimes
resulted in incorrect file transfers.

This fix does not contain separate tests, as tests will be
part of the main commit(s). This fix has been made as a separate
commit to facilitate review for major substantive fixes related
to MDEV-30157 and MDEV-28669.

3 files changed, 118 insertions, 51 deletions

diff --git a/scripts/wsrep_sst_common.sh b/scripts/wsrep_sst_common.sh
index 6a94cb0f706..407a4c9dd24 100644
--- a/scripts/wsrep_sst_common.sh
+++ b/scripts/wsrep_sst_common.sh
@@ -47,18 +47,51 @@ trim_string()
 
 trim_dir()
 {
-    local t=$(trim_string "$1")
-    if [ "$t" != '/' ]; then
-        if [ "${t%/}" != "$t" ]; then
-            t=$(trim_string "${t%/}")
+    if [ -n "$BASH_VERSION" ]; then
+        local pattern="![:space:]${2:-}"
+        local x="${1#*[$pattern]}"
+        local z=${#1}
+        x=${#x}
+        if [ $x -ne $z ]; then
+            local y="${1%[$pattern/]*}"
+            y=${#y}
+            x=$(( z-x-1 ))
+            y=$(( y-x+1 ))
+            x="${1:$x:$y}"
+            [ -z "$x" ] && x='.'
+            printf '%s' "$x"
+        else
+            printf ''
         fi
     else
-        t='.'
+        local pattern="[:space:]${2:-}"
+        local x=$(echo "$1" | sed -E "s/^[$pattern]+|[$pattern/]+\$//g")
+        if [ -n "$x" ]; then
+            echo "$x"
+        elif "${1#*/}" != "$1"; then
+            echo '.'
+        else
+            echo ''
+        fi
     fi
+}
+
+trim_right()
+{
     if [ -n "$BASH_VERSION" ]; then
-        printf '%s' "$t"
+        local pattern="[![:space:]${2:-}]"
+        local z=${#1}
+        local y="${1%$pattern*}"
+        y=${#y}
+        if [ $y -ne $z ]; then
+            y=$(( y+1 ))
+            printf '%s' "${1:0:$y}"
+        else
+            printf ''
+        fi
     else
-        echo "$t"
+        local pattern="[[:space:]${2:-}]"
+        echo "$1" | sed -E "s/$pattern+\$//g"
     fi
 }
 
@@ -111,7 +144,7 @@ INNOEXTRA=""
 while [ $# -gt 0 ]; do
 case "$1" in
     '--address')
-        WSREP_SST_OPT_ADDR="$2"
+        WSREP_SST_OPT_ADDR=$(trim_string "$2")
         #
         # Break address string into host:port/path parts
         #
@@ -119,20 +152,22 @@ case "$1" in
         \[*)
             # IPv6
             # Remove the starting and ending square brackets, if present:
-            addr_no_bracket="${WSREP_SST_OPT_ADDR#\[}"
+            addr="${WSREP_SST_OPT_ADDR#\[}"
+            addr=$(trim_right "${addr%%\]*}")
             # Some utilities and subsequent code require an address
             # without square brackets:
-            readonly WSREP_SST_OPT_HOST_UNESCAPED="${addr_no_bracket%%\]*}"
+            readonly WSREP_SST_OPT_HOST_UNESCAPED="$addr"
             # Square brackets are needed in most cases:
-            readonly WSREP_SST_OPT_HOST="[$WSREP_SST_OPT_HOST_UNESCAPED]"
+            readonly WSREP_SST_OPT_HOST="[$addr]"
             # Mark this address as IPv6:
             readonly WSREP_SST_OPT_HOST_IPv6=1
             # Let's remove the leading part that contains the host address:
             remain="${WSREP_SST_OPT_ADDR#*\]}"
             ;;
         *)
-            readonly WSREP_SST_OPT_HOST="${WSREP_SST_OPT_ADDR%%[:/]*}"
-            readonly WSREP_SST_OPT_HOST_UNESCAPED="$WSREP_SST_OPT_HOST"
+            addr=$(trim_right "${WSREP_SST_OPT_ADDR%%[:/]*}")
+            readonly WSREP_SST_OPT_HOST="$addr"
+            readonly WSREP_SST_OPT_HOST_UNESCAPED="$addr"
             readonly WSREP_SST_OPT_HOST_IPv6=0
             # Let's remove the leading part that contains the host address:
             remain="${WSREP_SST_OPT_ADDR#*[:/]}"
@@ -154,17 +189,18 @@ case "$1" in
         else
             readonly WSREP_SST_OPT_PATH=""
         fi
+        WSREP_SST_OPT_ADDR_PORT=$(trim_right "$WSREP_SST_OPT_ADDR_PORT")
         # Remove the module name part from the string, which ends with "/":
         remain="${WSREP_SST_OPT_PATH#*/}"
         # This operation removes the tail after the very first occurrence
         # of the "/" character, inclusively:
-        readonly WSREP_SST_OPT_MODULE="${WSREP_SST_OPT_PATH%%/*}"
+        readonly WSREP_SST_OPT_MODULE=$(trim_right "${WSREP_SST_OPT_PATH%%/*}")
         # If there is one more "/" in the string, then everything before
         # it will be the LSN, otherwise the LSN is empty:
         if [ "$remain" != "$WSREP_SST_OPT_PATH" ]; then
             # Extract the part that matches the LSN by removing all
             # characters starting from the very first "/":
-            readonly WSREP_SST_OPT_LSN="${remain%%/*}"
+            readonly WSREP_SST_OPT_LSN=$(trim_right "${remain%%/*}")
             # Exctract everything after the first occurrence of
             # the "/" character in the string:
             source="$remain"
@@ -176,7 +212,7 @@ case "$1" in
                 # Let's extract the version number by removing the tail
                 # after the very first occurence of the "/" character
                 # (inclusively):
-                readonly WSREP_SST_OPT_SST_VER="${remain%%/*}"
+                readonly WSREP_SST_OPT_SST_VER=$(trim_right "${remain%%/*}")
             else
                 readonly WSREP_SST_OPT_SST_VER=""
             fi
@@ -218,41 +254,46 @@ case "$1" in
         shift
         ;;
     '--defaults-file')
-        readonly WSREP_SST_OPT_DEFAULT="$1=$2"
-        readonly WSREP_SST_OPT_DEFAULTS="$1='$2'"
+        file=$(trim_string "$2")
+        readonly WSREP_SST_OPT_DEFAULT="$1=$file"
+        readonly WSREP_SST_OPT_DEFAULTS="$1='$file'"
         shift
         ;;
     '--defaults-extra-file')
-        readonly WSREP_SST_OPT_EXTRA_DEFAULT="$1=$2"
-        readonly WSREP_SST_OPT_EXTRA_DEFAULTS="$1='$2'"
+        file=$(trim_string "$2")
+        readonly WSREP_SST_OPT_EXTRA_DEFAULT="$1=$file"
+        readonly WSREP_SST_OPT_EXTRA_DEFAULTS="$1='$file'"
         shift
         ;;
     '--defaults-group-suffix')
-        readonly WSREP_SST_OPT_SUFFIX_DEFAULT="$1=$2"
-        readonly WSREP_SST_OPT_SUFFIX_VALUE="$2"
+        suffix=$(trim_string "$2")
+        readonly WSREP_SST_OPT_SUFFIX_DEFAULT="$1=$suffix"
+        readonly WSREP_SST_OPT_SUFFIX_VALUE="$suffix"
         shift
         ;;
     '--host')
-        case "$2" in
+        addr=$(trim_string "$2")
+        case "$addr" in
         \[*)
             # IPv6
             # Remove the starting and ending square brackets, if present:
-            addr_no_bracket="${2#\[}"
+            addr="${addr#\[}"
+            addr=$(trim_right "${addr%%\]*}")
             # Some utilities and subsequent code require an address
             # without square brackets:
-            readonly WSREP_SST_OPT_HOST_UNESCAPED="${addr_no_bracket%%\]*}"
+            readonly WSREP_SST_OPT_HOST_UNESCAPED="$addr"
             # Square brackets are needed in most cases:
-            readonly WSREP_SST_OPT_HOST="[${WSREP_SST_OPT_HOST_UNESCAPED}]"
+            readonly WSREP_SST_OPT_HOST="[$addr]"
             # Mark this address as IPv6:
             readonly WSREP_SST_OPT_HOST_IPv6=1
             ;;
         *)
-            readonly WSREP_SST_OPT_HOST="$2"
-            readonly WSREP_SST_OPT_HOST_UNESCAPED="$2"
+            readonly WSREP_SST_OPT_HOST="$addr"
+            readonly WSREP_SST_OPT_HOST_UNESCAPED="$addr"
             readonly WSREP_SST_OPT_HOST_IPv6=0
             ;;
         esac
-        WSREP_SST_OPT_ADDR="$WSREP_SST_OPT_HOST"
+        WSREP_SST_OPT_ADDR="$addr"
         shift
         ;;
     '--local-port')
@@ -272,11 +313,11 @@ case "$1" in
         shift
         ;;
     '--role')
-        readonly WSREP_SST_OPT_ROLE="$2"
+        readonly WSREP_SST_OPT_ROLE=$(trim_string "$2")
         shift
         ;;
     '--socket')
-        readonly WSREP_SST_OPT_SOCKET="$2"
+        readonly WSREP_SST_OPT_SOCKET=$(trim_string "$2")
         shift
         ;;
     '--user')
@@ -284,23 +325,23 @@ case "$1" in
         shift
         ;;
     '--gtid')
-        readonly WSREP_SST_OPT_GTID="$2"
+        readonly WSREP_SST_OPT_GTID=$(trim_string "$2")
         shift
         ;;
     '--binlog'|'--log-bin')
-        readonly WSREP_SST_OPT_BINLOG="$2"
+        readonly WSREP_SST_OPT_BINLOG=$(trim_string "$2")
         shift
         ;;
     '--binlog-index'|'--log-bin-index')
-        WSREP_SST_OPT_BINLOG_INDEX="$2"
+        WSREP_SST_OPT_BINLOG_INDEX=$(trim_string "$2")
         shift
         ;;
     '--log-basename')
-        readonly WSREP_SST_OPT_LOG_BASENAME="$2"
+        readonly WSREP_SST_OPT_LOG_BASENAME=$(trim_string "$2")
         shift
         ;;
     '--gtid-domain-id')
-        readonly WSREP_SST_OPT_GTID_DOMAIN_ID="$2"
+        readonly WSREP_SST_OPT_GTID_DOMAIN_ID=$(trim_string "$2")
         shift
         ;;
     '--mysqld-args')
diff --git a/scripts/wsrep_sst_mariabackup.sh b/scripts/wsrep_sst_mariabackup.sh
index 63ef8be8690..7465b2a3b08 100644
--- a/scripts/wsrep_sst_mariabackup.sh
+++ b/scripts/wsrep_sst_mariabackup.sh
@@ -439,9 +439,10 @@ get_footprint()
         -regex '.*undo[0-9]+$\|.*\.ibd$\|.*\.MYI$\|.*\.MYD$\|.*ibdata1$' \
         -type f -print0 | du --files0-from=- --block-size=1 -c -s | \
         awk 'END { print $1 }')
-
     local payload_undo=0
-    if [ -n "$ib_undo_dir" -a -d "$ib_undo_dir" ]; then
+    if [ -n "$ib_undo_dir" -a "$ib_undo_dir" != '.' -a \
+         "$ib_undo_dir" != "$DATA_DIR" -a -d "$ib_undo_dir" ]
+    then
         cd "$ib_undo_dir"
         payload_undo=$(find . -regex '.*undo[0-9]+$' -type f -print0 | \
             du --files0-from=- --block-size=1 -c -s | awk 'END { print $1 }')
@@ -451,7 +452,7 @@ get_footprint()
     wsrep_log_info \
         "SST footprint estimate: data: $payload_data, undo: $payload_undo"
 
-    payload=$(( payload_data + payload_undo ))
+    payload=$(( payload_data+payload_undo ))
 
     if [ "$compress" != 'none' ]; then
         # QuickLZ has around 50% compression ratio
@@ -1220,13 +1221,16 @@ else # joiner
         INNODB_DATA_HOME_DIR=$(trim_dir "$INNODB_DATA_HOME_DIR")
     fi
 
-    if [ -n "$INNODB_DATA_HOME_DIR" -a "$INNODB_DATA_HOME_DIR" != '.' ]; then
+    if [ -n "$INNODB_DATA_HOME_DIR" -a "$INNODB_DATA_HOME_DIR" != '.' -a \
+         "$INNODB_DATA_HOME_DIR" != "$DATA_DIR" ]
+    then
         # handle both relative and absolute paths:
         cd "$DATA"
         [ ! -d "$INNODB_DATA_HOME_DIR" ] && mkdir -p "$INNODB_DATA_HOME_DIR"
         cd "$INNODB_DATA_HOME_DIR"
         ib_home_dir="$(pwd)"
         cd "$OLD_PWD"
+        [ "$ib_home_dir" = "$DATA_DIR" ] && ib_home_dir=""
     fi
 
     # if no command line argument and INNODB_LOG_GROUP_HOME is not set,
@@ -1236,13 +1240,16 @@ else # joiner
         INNODB_LOG_GROUP_HOME=$(trim_dir "$INNODB_LOG_GROUP_HOME")
     fi
 
-    if [ -n "$INNODB_LOG_GROUP_HOME" -a "$INNODB_LOG_GROUP_HOME" != '.' ]; then
+    if [ -n "$INNODB_LOG_GROUP_HOME" -a "$INNODB_LOG_GROUP_HOME" != '.' -a \
+         "$INNODB_LOG_GROUP_HOME" != "$DATA_DIR" ]
+    then
         # handle both relative and absolute paths:
         cd "$DATA"
         [ ! -d "$INNODB_LOG_GROUP_HOME" ] && mkdir -p "$INNODB_LOG_GROUP_HOME"
         cd "$INNODB_LOG_GROUP_HOME"
         ib_log_dir="$(pwd)"
         cd "$OLD_PWD"
+        [ "$ib_log_dir" = "$DATA_DIR" ] && ib_log_dir=""
     fi
 
     # if no command line argument and INNODB_UNDO_DIR is not set,
@@ -1252,13 +1259,16 @@ else # joiner
         INNODB_UNDO_DIR=$(trim_dir "$INNODB_UNDO_DIR")
     fi
 
-    if [ -n "$INNODB_UNDO_DIR" -a "$INNODB_UNDO_DIR" != '.' ]; then
+    if [ -n "$INNODB_UNDO_DIR" -a "$INNODB_UNDO_DIR" != '.' -a \
+         "$INNODB_UNDO_DIR" != "$DATA_DIR" ]
+    then
         # handle both relative and absolute paths:
         cd "$DATA"
         [ ! -d "$INNODB_UNDO_DIR" ] && mkdir -p "$INNODB_UNDO_DIR"
         cd "$INNODB_UNDO_DIR"
         ib_undo_dir="$(pwd)"
         cd "$OLD_PWD"
+        [ "$ib_undo_dir" = "$DATA_DIR" ] && ib_undo_dir=""
     fi
 
     if [ -n "$backup_threads" ]; then
@@ -1500,11 +1510,15 @@ else # joiner
                 binlogs=$(ls -d -1 "$binlog_base".[0-9]* 2>/dev/null || :)
             fi
             cd "$DATA_DIR"
-            if [ -n "$binlog_dir" -a "$binlog_dir" != '.' ]; then
+            if [ -n "$binlog_dir" -a "$binlog_dir" != '.' -a \
+                 "$binlog_dir" != "$DATA_DIR" ]
+            then
                 [ ! -d "$binlog_dir" ] && mkdir -p "$binlog_dir"
             fi
             index_dir=$(dirname "$binlog_index");
-            if [ -n "$index_dir" -a "$index_dir" != '.' ]; then
+            if [ -n "$index_dir" -a "$index_dir" != '.' -a \
+                 "$index_dir" != "$DATA_DIR" ]
+            then
                 [ ! -d "$index_dir" ] && mkdir -p "$index_dir"
             fi
             if [ -n "$binlogs" ]; then
diff --git a/scripts/wsrep_sst_rsync.sh b/scripts/wsrep_sst_rsync.sh
index ddf41559c29..ee3e848025b 100644
--- a/scripts/wsrep_sst_rsync.sh
+++ b/scripts/wsrep_sst_rsync.sh
@@ -185,7 +185,9 @@ if [ -z "$INNODB_LOG_GROUP_HOME" ]; then
     INNODB_LOG_GROUP_HOME=$(trim_dir "$INNODB_LOG_GROUP_HOME")
 fi
 
-if [ -n "$INNODB_LOG_GROUP_HOME" -a "$INNODB_LOG_GROUP_HOME" != '.' ]; then
+if [ -n "$INNODB_LOG_GROUP_HOME" -a "$INNODB_LOG_GROUP_HOME" != '.' -a \
+     "$INNODB_LOG_GROUP_HOME" != "$DATA_DIR" ]
+then
     # handle both relative and absolute paths:
     cd "$DATA"
     [ ! -d "$INNODB_LOG_GROUP_HOME" ] && mkdir -p "$INNODB_LOG_GROUP_HOME"
@@ -201,7 +203,9 @@ if [ -z "$INNODB_DATA_HOME_DIR" ]; then
     INNODB_DATA_HOME_DIR=$(trim_dir "$INNODB_DATA_HOME_DIR")
 fi
 
-if [ -n "$INNODB_DATA_HOME_DIR" -a "$INNODB_DATA_HOME_DIR" != '.' ]; then
+if [ -n "$INNODB_DATA_HOME_DIR" -a "$INNODB_DATA_HOME_DIR" != '.' -a \
+     "$INNODB_DATA_HOME_DIR" != "$DATA_DIR" ]
+then
     # handle both relative and absolute paths:
     cd "$DATA"
     [ ! -d "$INNODB_DATA_HOME_DIR" ] && mkdir -p "$INNODB_DATA_HOME_DIR"
@@ -217,7 +221,9 @@ if [ -z "$INNODB_UNDO_DIR" ]; then
     INNODB_UNDO_DIR=$(trim_dir "$INNODB_UNDO_DIR")
 fi
 
-if [ -n "$INNODB_UNDO_DIR" -a "$INNODB_UNDO_DIR" != '.' ]; then
+if [ -n "$INNODB_UNDO_DIR" -a "$INNODB_UNDO_DIR" != '.' -a \
+     "$INNODB_UNDO_DIR" != "$DATA_DIR" ]
+then
     # handle both relative and absolute paths:
     cd "$DATA"
     [ ! -d "$INNODB_UNDO_DIR" ] && mkdir -p "$INNODB_UNDO_DIR"
@@ -504,7 +510,9 @@ EOF
                             if [ "$first" = '-' -o "$first" = '@' ]; then
                                 bin_base="./$bin_base"
                             fi
-                            if [ -n "$bin_dir" -a "$bin_dir" != '.' ]; then
+                            if [ -n "$bin_dir" -a "$bin_dir" != '.' -a \
+                                 "$bin_dir" != "$DATA_DIR" ]
+                            then
                                 tar $tar_options "$BINLOG_TAR_FILE" \
                                     -C "$bin_dir" "$bin_base" >&2
                             else
@@ -872,7 +880,7 @@ EOF
         binlog_cd=0
         # Change the directory to binlog base (if possible):
         if [ -n "$binlog_dir" -a "$binlog_dir" != '.' -a \
-             -d "$binlog_dir" ]
+             "$binlog_dir" != "$DATA_DIR" -a -d "$binlog_dir" ]
         then
             binlog_cd=1
             cd "$binlog_dir"
@@ -891,11 +899,15 @@ EOF
                tmpfile=$(TMPDIR="$tmpdir"; mktemp)
             fi
             index_dir=$(dirname "$binlog_index");
-            if [ -n "$index_dir" -a "$index_dir" != '.' ]; then
+            if [ -n "$index_dir" -a "$index_dir" != '.' -a \
+                 "$index_dir" != "$DATA_DIR" ]
+            then
                 [ ! -d "$index_dir" ] && mkdir -p "$index_dir"
             fi
             binlog_cd=0
-            if [ -n "$binlog_dir" -a "$binlog_dir" != '.' ]; then
+            if [ -n "$binlog_dir" -a "$binlog_dir" != '.' -a \
+                 "$binlog_dir" != "$DATA_DIR" ]
+            then
                 [ ! -d "$binlog_dir" ] && mkdir -p "$binlog_dir"
                 binlog_cd=1
                 cd "$binlog_dir"