update clone of file iterator to store state

examples/file_iter.py

@@ -15,9 +15,13 @@
 for id, value in iter:
     print id, value
 
-iter2 = iter.copy()
-iter3 = iter.copy()
+print '----'
+iter1 = tk.get_file_iterator('file_iter_test.txt', type='text')
+for id, value in iter1:
+    print id, value
+    break
+iter2 = iter1.copy()
+for id, value in iter1:
+    print id, value
 for id, value in iter2:
     print id, value
-for id, value in iter3:
-    print id, value

rltk/file_iterator.py

@@ -1,6 +1,8 @@
 import json
 import csv
 import hashlib
+import itertools
+
 from jsonpath_rw import parse
 
 
@@ -35,20 +37,31 @@ def __init__(self, file_path, type='text', **kwargs):
             self._id_prefix = hashlib.md5(file_path).hexdigest()[:6]
 
     def __copy__(self):
-        return FileIterator(self._file_path, self._type, **self._kwargs)
+        """
+            Clone the iterator include states
+        """
+        # https://docs.python.org/2/library/itertools.html#itertools.tee
+        # tee is not that helpful here, and it will also occupy a lot of memory
+        # self._file_handler, new_iter = itertools.tee(self._file_handler)
+
+        new_iter = FileIterator(self._file_path, self._type, **self._kwargs)
+        for _ in new_iter:
+            if new_iter._count == self._count:
+                break
+        return new_iter
 
     def copy(self):
         return self.__copy__()
 
     def next(self):
         """
         Returns:
-            misc, dict: object id, value
+            str, dict: object id, value
         """
         try:
             oid, value = None, None
             if self._type == 'json_line':
-                line = self._file_handler.next()
+                line = next(self._file_handler)
                 line = json.loads(line)
 
                 matches = self._id_path_parser.find(line)
@@ -79,14 +92,15 @@ def next(self):
             return oid, value
 
         except StopIteration as e:
-            self._file_handler.close()
+            # self._file_handler.close()
             raise e
 
     def __iter__(self):
         return self
 
     def __del__(self):
         try:
-            self._file_handler.close()
+            pass
+            # self._file_handler.close()
         except:
             pass