Made loading gremlin.sh faster by not using maven
diff --git a/scripts/parse-classpath.py b/scripts/parse-classpath.py
new file mode 100755
index 0000000..2e60a1d
--- /dev/null
+++ b/scripts/parse-classpath.py
@@ -0,0 +1,34 @@
+#!/usr/bin/env python
+
+import sys
+from os import path
+from xml.dom import minidom
+
+# Location of the .classpath file relative to this script
+classpath_filename = "../.classpath"
+m2_repository = "~/.m2/repository"
+
+def parse_classpath_xml(classpath_file, abs_m2_repository):
+    xmldoc = minidom.parse(classpath_file)
+    classpathentries = xmldoc.getElementsByTagName('classpathentry')
+    classpath = ""
+    for entry in classpathentries:
+        kind = entry.attributes['kind'].value
+        if kind == "output" or kind == "var":
+            cp_entry = entry.attributes['path'].value + ":"
+            
+            classpath += cp_entry.replace("M2_REPO", abs_m2_repository)
+
+    print classpath[0:-1]
+
+if __name__ == "__main__":    
+    abs_m2_repository = path.expanduser("~/.m2/repository")
+
+    classpath_file = path.abspath(path.join(path.dirname(path.realpath(sys.argv[0])), classpath_filename))
+
+    try:
+        with open(classpath_file) as f:
+            parse_classpath_xml(f, abs_m2_repository)
+    except IOError:
+        print "Error reading classpath file from %s" % classpath_file
+        print "- Please check path is correct then run 'mvn eclipse:eclipse' to generate file"