#!/bin/sh
# count blank lines, white spaces and comments
# 17.8.98 Werner Heuser <r2d2c3po@zedat.fu-berlin.de>
#
# ToDo
# - look for tabs too
# - find  " ; ",  " { ",  " } " etc.
# - find the same items in *.gz
# - is there a corresponding approach for binary files?

rm -f tmp.txt

SUM_BLANK=0
SUM_COMMENT=0
SUM_BLANKS=0
SUM_X=0

for FILE in `find 1 2 -type f -print`
do

file $FILE | egrep "text" >/dev/null
if [ $? -eq 0 ]; then
 echo "FILE $FILE text"
 echo "FILE $FILE text ****************************************" >>tmp.txt

# find blank lines
 INC_B=`egrep "^$" $FILE | wc -l`
 egrep "^$" $FILE >>tmp.txt
 echo "INC_B $INC_B"

 SUM_BLANK=`bc<<EOF
 $SUM_BLANK + $INC_B
EOF`

# find comments at the beginning of a line
  INC_C=`egrep "^#" $FILE | egrep -v "^#!" | wc -c`
  egrep "^#" $FILE | egrep -v "^#!" >>tmp.txt
 echo "INC_C $INC_C"

 SUM_COMMENT=`bc<<EOF
 $SUM_COMMENT + $INC_C
EOF`

# find more than one blank in a line, which includes not a comment
  INC_X=`egrep -v "*[^$&\"/a-z]#" $FILE | egrep "  "| tr -d [:graph:] | wc -c`
  egrep -v "*[^$&\"/a-z]#" $FILE | egrep "  " >>tmp.txt
 echo "INC_X $INC_X"

 SUM_X=`bc<<EOF
 $SUM_X + $INC_X
EOF`

# find comments not at the beginning of a line, (" #") still not counted !!
  INC_M=`egrep "^[^#].*[^$&\"/a-z]#" $FILE | cut -f 2- -d"#" | wc -c`
  egrep "^[^#].*[^$&\"/a-z]#" $FILE  >>tmp.txt
 echo "INC_M $INC_M"

 SUM_BLANKS=`bc<<EOF
 $SUM_BLANKS + $INC_M
EOF`

else
 echo "FILE $FILE no text"
fi

done

SUM=`bc<<EOF
$SUM_BLANK + $SUM_COMMENT + $SUM_BLANKS + $SUM_X
EOF`

cat <<EOF
Blank Lines  $SUM_BLANK
Only Comments  $SUM_COMMENT
More then one Blank $SUM_BLANKS
Chars after Comments $SUM_X

SUM  $SUM
EOF