Used to find the filesize of all duplicates found with duplicate.sh.

1 files changed, 70 insertions, 0 deletions

diff --git a/bin/duplicate.sh b/bin/duplicate.sh
new file mode 100755
index 0000000..655aff2
--- /dev/null
+++ b/bin/duplicate.sh
@@ -0,0 +1,70 @@
+#!/bin/bash
+
+# Example usage, find all doubles in all sub directory
+# 	for i in *; do if [ -d $i ]; then double.sh $i; fi; done
+#
+# To find duplicates in specific directory
+# 	double.sh /path/to/dir /path/to/other/dir ./relativ/path
+#
+# With zero paramater, the script uses the working diroctory.
+
+DATABASE="/tmp/double.db"
+DIR="./found_doubles"
+COUNT=0
+
+if [ $# -eq 0 ]; then
+	LOOP=false
+else
+	LOOP=true
+fi
+
+echo ""
+
+while true; do 
+
+	if [ $LOOP = true ]; then
+		if [ ! -d "$1" ]; then
+			echo "Can't chdir to $1."
+			exit 1
+		else
+			echo "Changing directory to '$1'."
+			cd "$1"
+			shift
+		fi
+	else
+		echo "Working in directory '$(pwd)'."
+	fi
+
+	sqlite3 $DATABASE "CREATE TABLE files (hash TEXT UNIQUE)"
+	mkdir -p $DIR
+	TMP=0
+	for i in *; do
+        	if [ -f "$i" ]; then
+                	HASH=$(md5sum "$i" | awk '{ print $1 }')
+	                sqlite3 $DATABASE "INSERT INTO files VALUES ('$HASH')" 2>/dev/null
+			if [ ! $? -eq 0 ]; then
+				mv -v "$i" "$DIR"
+				((COUNT ++))
+				((TMP ++))
+			fi
+        	fi
+	done
+
+	if [ $TMP -eq 0 ]; then
+		rmdir $DIR
+	fi
+
+	if [ $LOOP = true ]; then
+		sqlite3 $DATABASE "DROP TABLE files"
+		echo ""
+	else
+		break
+	fi
+
+	if [ $# -eq 0 ]; then
+		break;
+	fi
+done
+
+echo "Found and moved $COUNT files to $DIR."
+rm -f $DATABASE