Merge pull request #276 from kmike/fix-strings-tutorial

fix deprecated imports in "Unicode and passing strings" tutorial

docs/src/tutorial/strings.rst

@@ -139,13 +139,13 @@ exception, e.g. due to insufficient memory.  If you need to
 :c:func:`free()` the string after the conversion, you should wrap
 the assignment in a try-finally construct::
 
-    cimport stdlib
+    from libc.stdlib cimport free
     cdef bytes py_string
     cdef char* c_string = c_call_creating_a_new_c_string()
     try:
         py_string = c_string
     finally:
-        stdlib.free(c_string)
+        free(c_string)
 
 To convert the byte string back into a C :c:type:`char*`, use the
 opposite assignment::
@@ -376,8 +376,7 @@ conversions in general) in dedicated functions, as this needs to be
 done in exactly the same way whenever receiving text from C.  This
 could look as follows::
 
-    cimport python_unicode
-    cimport stdlib
+    from libc.stdlib cimport free
 
     cdef unicode tounicode(char* s):
         return s.decode('UTF-8', 'strict')
@@ -391,7 +390,7 @@ could look as follows::
         try:
             return s[:length].decode('UTF-8', 'strict')
         finally:
-            stdlib.free(s)
+            free(s)
 
 Most likely, you will prefer shorter function names in your code based
 on the kind of string being handled.  Different types of content often