[FEATURE] (util): implement csbucomparecsfiles.sh

util/README

@@ -21,7 +21,6 @@ csbulistcsfiles.sh
 csbucomparecsfiles.sh
   Check downstream data archives for checksum files. Report missing checksum
   files and checksum files differing from the upstream file.
-  --- not yet implemented ---
 
 csbucscheck.sh
   Recurse into a read-only subdirectory tree and verify checksums.

util/csbucomparecsfiles.sh

@@ -11,18 +11,26 @@
 # 
 # ============================================================================
 #
-VERSION="06/06/2016   V1.0"
+VERSION="13/06/2016   V1.0"
 
 # name of csback checksum files
 CHECKSUMFILE="checksumfile.cs"
 
+# current working directory
+HERE=$(pwd)
+
+# base name of temporary directory used for intermediate lists and for report
+# files
+TMPBASE=$HERE/report
+
+# ============================================================================
 # usage function
 function usage {
 cat << HERE
 This is <csbucomparecsfiles.sh>
 version: $VERSION
 
-usage: csbucomparecsfiles.sh [-v] [-d] dir [dir ...]
+usage: csbucomparecsfiles.sh [-v] [-d] [-r dir] basedir downdir
 or:    csbucomparecsfiles.sh -h
 
 Check downstream data archives for checksum files. Report missing checksum
@@ -30,6 +38,8 @@ files and checksum files differing from the upstream file.
 HERE
 }
 
+# ----------------------------------------------------------------------------
+
 # usage function
 function longusage {
 cat << HERE
@@ -37,31 +47,39 @@ cat << HERE
   -h        print detailed usage information
   -v        be verbose
   -d        print debug output
+  -r dir    base directory for report
+            if not provided, a subdirectory in the current directory will
+            be created; the default is
+            $TMPBASE
 
-  dir       one or more directories to be checked recursively
+  basedir   base directory of data collection
+  downdir   base directory of downstream data collection
 
   The name of checksumfiles to be searched for is: $CHECKSUMFILE
 HERE
 }
 
+# ============================================================================
+
 verbose=0
 debug=0
 
 # fetch commandline arguments
-while getopts 'hvd' Option;
+while getopts 'hvdr:' Option;
 do
   case ${Option} in
     h) hflag=1
       usage; longusage; exit 0;;
     v) verbose=1;;
     d) debug=1;;
+    r) TMPBASE="$OPTARG";;
     *) echo -e "ERROR: Unimplemented option chosen.\n"
       usage; exit 2;;
   esac
 done
 shift $((${OPTIND} - 1))
 
-MINPARAMS=1
+MINPARAMS=2
 # check number of commandline arguments
 if [ $# -lt ${MINPARAMS} ]
 then
@@ -69,5 +87,90 @@ then
   exit 2
 fi
 
+TMPBASEDIR=$(dirname $TMPBASE)
+if test ! \( -d $(dirname $TMPBASE) -a -w $(dirname $TMPBASE) \)
+then
+  echo report output base directory $TMPBASEDIR is not a directory
+  echo or is not writable
+  exit 2
+fi
+
+TMPDIR=$TMPBASE/report_$(basename $0 .sh)_$(basename $DATADIR)_$$
+mkdir -p $TMPDIR
+
+if test ! \( -d $TMPDIR -a -w $TMPDIR \)
+then
+  echo report output directory $TMPDIR is not a directory
+  echo or is not writable
+  exit 2
+fi
+
+DATADIR=$1
+DOWNDIR=$2
+
+if test ! \( -d $DATADIR -a -r $DATADIR \)
+then
+  echo data directory $DATADIR does not exits or is not readable
+  exit 2
+fi
+
+# ============================================================================
+# action
+# ======
+REPORT=$TMPDIR/report
+MISSING=$TMPDIR/missing
+DEVIANT=$TMPDIR/deviant
+touch $REPORT $MISSING $DEVIANT
+
+# report
+tee -a $REPORT <<HERE
+
+-----------------------------------------------------------------------
+$0
+version: $VERSION 
+started: $(date)
+
+check downstream data directory
+$DOWNDIR
+against data directory tree at
+$DATADIR
+-----------------------------------------------------------------------
+HERE
+
+cd $DATADIR
+
+nc=0
+for FILE in $(find . -name $CHECKSUMFILE | sort)
+do
+  echo check $FILE | tee -a $REPORT
+  DOWNFILE=$DOWNDIR/$FILE
+  if test -r $DOWNFILE
+  then
+    if /usr/bin/cmp -s $FILE $DOWNFILE
+    then
+      echo verified: $DOWNFILE | tee -a $REPORT
+    else
+      echo $DOWNFILE >> $DEVIANT
+      echo differs: $DOWNFILE | tee -a $REPORT
+      LOGFILE=$TMPDIR/$(echo $FILE | tr '/' '_')
+      diff $FILE $DOWNFILE > $LOGFILE
+    fi
+  else
+    echo $DOWNFILE >> $MISSING
+    echo missing: $DOWNFILE | tee -a $REPORT
+  fi
+  nc=$((nc+1))    
+done
+
+# ----------------------------------------------------------------------------
+
+tee -a $REPORT <<HERE
+
+$nc files were checked
+$(cat $MISSING | wc -l) are missing in downstream data archive
+$(cat $DEVIANT | wc -l) files in downstream archive do not match
+a report is provided in
+$TMPDIR
+HERE
 
 # ----- END OF csbucomparecsfiles.sh -----