Fix leaking sockets after v2 list operation

This fixes the same problem with leaking file descriptors after the port
to requests I16a7b02f2b10e506e91719712cf34ef0aea1afc0 does, but for the
v2 api client.

The paginate function in the list method has been refactored from a
recursive generator to a non-recursive generator, which avoids issues
with garbage collection holding the socket open.

Change-Id: Iaa7a421e8c1acafb7b23bb3f55e50b9d2a9d030a
Closes-Bug: #1428797

glanceclient/v2/images.py

@@ -54,40 +54,40 @@ class Controller(object):
         page_size = kwargs.get('page_size') or DEFAULT_PAGE_SIZE
 
         def paginate(url, page_size, limit=None):
+            next_url = url
 
-            if limit and page_size > limit:
-                # NOTE(flaper87): Avoid requesting 2000 images when limit is 1
-                url = url.replace("limit=%s" % page_size,
-                                  "limit=%s" % limit)
+            while True:
+                if limit and page_size > limit:
+                    # NOTE(flaper87): Avoid requesting 2000 images when limit
+                    # is 1
+                    next_url = next_url.replace("limit=%s" % page_size,
+                                                "limit=%s" % limit)
 
-            resp, body = self.http_client.get(url)
-            for image in body['images']:
-                # NOTE(bcwaldon): remove 'self' for now until we have
-                # an elegant way to pass it into the model constructor
-                # without conflict.
-                image.pop('self', None)
-                yield self.model(**image)
-                # NOTE(zhiyan): In order to resolve the performance issue
-                # of JSON schema validation for image listing case, we
-                # don't validate each image entry but do it only on first
-                # image entry for each page.
-                self.model.validate = empty_fun
+                resp, body = self.http_client.get(next_url)
+                for image in body['images']:
+                    # NOTE(bcwaldon): remove 'self' for now until we have
+                    # an elegant way to pass it into the model constructor
+                    # without conflict.
+                    image.pop('self', None)
+                    yield self.model(**image)
+                    # NOTE(zhiyan): In order to resolve the performance issue
+                    # of JSON schema validation for image listing case, we
+                    # don't validate each image entry but do it only on first
+                    # image entry for each page.
+                    self.model.validate = empty_fun
 
-                if limit:
-                    limit -= 1
-                    if limit <= 0:
-                        raise StopIteration
+                    if limit:
+                        limit -= 1
+                        if limit <= 0:
+                            raise StopIteration
 
-            # NOTE(zhiyan); Reset validation function.
-            self.model.validate = ori_validate_fun
+                # NOTE(zhiyan); Reset validation function.
+                self.model.validate = ori_validate_fun
 
-            try:
-                next_url = body['next']
-            except KeyError:
-                return
-            else:
-                for image in paginate(next_url, page_size, limit):
-                    yield image
+                try:
+                    next_url = body['next']
+                except KeyError:
+                    return
 
         filters = kwargs.get('filters', {})
         # NOTE(flaper87): We paginate in the client, hence we use