Merge "SLO: Make etag and size_bytes fully optional"

doc/source/api/large_objects.rst

@@ -50,29 +50,31 @@ Static large objects
 To create a static large object, divide your content into pieces and
 create (upload) a segment object to contain each piece.
 
-You must record the ``ETag`` response header that the **PUT** operation
-returns. Alternatively, you can calculate the MD5 checksum of the
-segment prior to uploading and include this in the ``ETag`` request
-header. This ensures that the upload cannot corrupt your data.
-
-List the name of each segment object along with its size and MD5
-checksum in order.
-
 Create a manifest object. Include the ``multipart-manifest=put``
 query string at the end of the manifest object name to indicate that
 this is a manifest object.
 
 The body of the **PUT** request on the manifest object comprises a json
-list, where each element contains the following attributes:
+list, where each element is an object representing a segment. These objects
+may contain the following attributes:
 
--  ``path``. The container and object name in the format:
+-  ``path`` (required). The container and object name in the format:
    ``{container-name}/{object-name}``
 
--  ``etag``. The MD5 checksum of the content of the segment object. This
-   value must match the ``ETag`` of that object.
+-  ``etag`` (optional). If provided, this value must match the ``ETag``
+   of the segment object. This was included in the response headers when
+   the segment was created. Generally, this will be the MD5 sum of the
+   segment.
 
--  ``size_bytes``. The size of the segment object. This value must match
-   the ``Content-Length`` of that object.
+-  ``size_bytes`` (optional). The size of the segment object. If provided,
+   this value must match the ``Content-Length`` of that object.
+
+-  ``range`` (optional). The subset of the referenced object that should
+   be used for segment data. This behaves similar to the ``Range`` header.
+   If omitted, the entire object will be used.
+
+Providing the optional ``etag`` and ``size_bytes`` attributes for each
+segment ensures that the upload cannot corrupt your data.
 
 **Example Static large object manifest list**
 

swift/common/middleware/slo.py

@@ -39,10 +39,10 @@ Key         Description
 =========== ========================================================
 path        the path to the segment object (not including account)
             /container/object_name
-etag        the ETag given back when the segment object was PUT,
-            or null
-size_bytes  the size of the complete segment object in
-            bytes, or null
+etag        (optional) the ETag given back when the segment object
+            was PUT
+size_bytes  (optional) the size of the complete segment object in
+            bytes
 range       (optional) the (inclusive) range within the object to
             use as a segment. If omitted, the entire object is used.
 =========== ========================================================
@@ -67,8 +67,8 @@ head every segment passed in to verify:
  5. if the user provided a range, it is a singular, syntactically correct range
     that is satisfiable given the size of the object.
 
-Note that the etag and size_bytes keys are still required; this acts as a guard
-against user errors such as typos. If any of the objects fail to verify (not
+Note that the etag and size_bytes keys are optional; if ommitted, the
+verification is not performed. If any of the objects fail to verify (not
 found, size/etag mismatch, below minimum size, invalid range) then the user
 will receive a 4xx error response. If everything does match, the user will
 receive a 2xx response and the SLO object is ready for downloading.
@@ -106,12 +106,10 @@ If a user uploads this manifest:
 
   .. code::
 
-     [{"path": "/con/obj_seg_1", "etag": null, "size_bytes": 2097152,
-       "range": "0-1048576"},
-      {"path": "/con/obj_seg_2", "etag": null, "size_bytes": 2097152,
+     [{"path": "/con/obj_seg_1", "size_bytes": 2097152, "range": "0-1048576"},
+      {"path": "/con/obj_seg_2", "size_bytes": 2097152,
        "range": "512-1550000"},
-      {"path": "/con/obj_seg_1", "etag": null, "size_bytes": 2097152,
-       "range": "-2048"}]
+      {"path": "/con/obj_seg_1", "size_bytes": 2097152, "range": "-2048"}]
 
 The segment will consist of the first 1048576 bytes of /con/obj_seg_1,
 followed by bytes 513 through 1550000 (inclusive) of /con/obj_seg_2, and
@@ -230,8 +228,8 @@ DEFAULT_MAX_MANIFEST_SEGMENTS = 1000
 DEFAULT_MAX_MANIFEST_SIZE = 1024 * 1024 * 2  # 2 MiB
 
 
-REQUIRED_SLO_KEYS = set(['path', 'etag', 'size_bytes'])
-OPTIONAL_SLO_KEYS = set(['range'])
+REQUIRED_SLO_KEYS = set(['path'])
+OPTIONAL_SLO_KEYS = set(['range', 'etag', 'size_bytes'])
 ALLOWED_SLO_KEYS = REQUIRED_SLO_KEYS | OPTIONAL_SLO_KEYS
 
 SYSMETA_SLO_ETAG = get_sys_meta_prefix('object') + 'slo-etag'
@@ -301,10 +299,10 @@ def parse_and_validate_input(req_body, req_path):
         if not isinstance(seg_dict['path'], six.string_types):
             errors.append("Index %d: \"path\" must be a string" % seg_index)
             continue
-        if not (seg_dict['etag'] is None or
+        if not (seg_dict.get('etag') is None or
                 isinstance(seg_dict['etag'], six.string_types)):
-            errors.append(
-                "Index %d: \"etag\" must be a string or null" % seg_index)
+            errors.append('Index %d: "etag" must be a string or null '
+                          '(if provided)' % seg_index)
             continue
 
         if '/' not in seg_dict['path'].strip('/'):
@@ -313,7 +311,7 @@ def parse_and_validate_input(req_body, req_path):
                 "the form /container/object." % seg_index)
             continue
 
-        seg_size = seg_dict['size_bytes']
+        seg_size = seg_dict.get('size_bytes')
         if seg_size is not None:
             try:
                 seg_size = int(seg_size)
@@ -932,10 +930,10 @@ class StaticLargeObject(object):
                 problem_segments.append(
                     [quote(obj_name),
                      'Too small; each segment must be at least 1 byte.'])
-            if seg_dict['size_bytes'] is not None and \
+            if seg_dict.get('size_bytes') is not None and \
                     seg_dict['size_bytes'] != head_seg_resp.content_length:
                 problem_segments.append([quote(obj_name), 'Size Mismatch'])
-            if seg_dict['etag'] is not None and \
+            if seg_dict.get('etag') is not None and \
                     seg_dict['etag'] != head_seg_resp.etag:
                 problem_segments.append([quote(obj_name), 'Etag Mismatch'])
             if head_seg_resp.last_modified:

test/functional/test_slo.py

@@ -473,9 +473,36 @@ class TestSlo(Base):
 
     def test_slo_missing_etag(self):
         file_item = self.env.container.file("manifest-a-missing-etag")
+        file_item.write(
+            json.dumps([{
+                'size_bytes': 1024 * 1024,
+                'path': '/%s/%s' % (self.env.container.name, 'seg_a')}]),
+            parms={'multipart-manifest': 'put'})
+        self.assert_status(201)
+
+    def test_slo_missing_size(self):
+        file_item = self.env.container.file("manifest-a-missing-size")
+        file_item.write(
+            json.dumps([{
+                'etag': hashlib.md5('a' * 1024 * 1024).hexdigest(),
+                'path': '/%s/%s' % (self.env.container.name, 'seg_a')}]),
+            parms={'multipart-manifest': 'put'})
+        self.assert_status(201)
+
+    def test_slo_path_only(self):
+        file_item = self.env.container.file("manifest-a-path-only")
+        file_item.write(
+            json.dumps([{
+                'path': '/%s/%s' % (self.env.container.name, 'seg_a')}]),
+            parms={'multipart-manifest': 'put'})
+        self.assert_status(201)
+
+    def test_slo_typo_etag(self):
+        file_item = self.env.container.file("manifest-a-typo-etag")
         try:
             file_item.write(
                 json.dumps([{
+                    'teag': hashlib.md5('a' * 1024 * 1024).hexdigest(),
                     'size_bytes': 1024 * 1024,
                     'path': '/%s/%s' % (self.env.container.name, 'seg_a')}]),
                 parms={'multipart-manifest': 'put'})
@@ -484,12 +511,13 @@ class TestSlo(Base):
         else:
             self.fail("Expected ResponseError but didn't get it")
 
-    def test_slo_missing_size(self):
-        file_item = self.env.container.file("manifest-a-missing-size")
+    def test_slo_typo_size(self):
+        file_item = self.env.container.file("manifest-a-typo-size")
         try:
             file_item.write(
                 json.dumps([{
                     'etag': hashlib.md5('a' * 1024 * 1024).hexdigest(),
+                    'siz_bytes': 1024 * 1024,
                     'path': '/%s/%s' % (self.env.container.name, 'seg_a')}]),
                 parms={'multipart-manifest': 'put'})
         except ResponseError as err:

test/unit/common/middleware/test_slo.py

@@ -168,6 +168,18 @@ class TestSloMiddleware(SloTestCase):
                   'size_bytes': 100,
                   'foo': 'bar', 'baz': 'quux'}])))
 
+        # This also catches typos
+        self.assertEqual(
+            'Index 0: extraneous keys "egat"\n',
+            self._put_bogus_slo(json.dumps(
+                [{'path': '/cont/object', 'egat': 'etagoftheobjectsegment',
+                  'size_bytes': 100}])))
+        self.assertEqual(
+            'Index 0: extraneous keys "siez_bytes"\n',
+            self._put_bogus_slo(json.dumps(
+                [{'path': '/cont/object', 'etag': 'etagoftheobjectsegment',
+                  'siez_bytes': 100}])))
+
     def test_bogus_input_ranges(self):
         self.assertEqual(
             "Index 0: invalid range\n",
@@ -568,9 +580,11 @@ class TestSloPutManifest(SloTestCase):
         ], sorted(errors))
 
     def test_handle_multipart_put_skip_size_check(self):
-        good_data = json.dumps(
-            [{'path': '/checktest/a_1', 'etag': 'a', 'size_bytes': None},
-             {'path': '/checktest/b_2', 'etag': 'b', 'size_bytes': None}])
+        good_data = json.dumps([
+            # Explicit None will skip it
+            {'path': '/checktest/a_1', 'etag': 'a', 'size_bytes': None},
+            # ...as will omitting it entirely
+            {'path': '/checktest/b_2', 'etag': 'b'}])
         req = Request.blank(
             '/v1/AUTH_test/checktest/man_3?multipart-manifest=put',
             environ={'REQUEST_METHOD': 'PUT'}, body=good_data)
@@ -618,9 +632,11 @@ class TestSloPutManifest(SloTestCase):
         self.assertIn('Etag Mismatch', cm.exception.body)
 
     def test_handle_multipart_put_skip_etag_check(self):
-        good_data = json.dumps(
-            [{'path': '/checktest/a_1', 'etag': None, 'size_bytes': 1},
-             {'path': '/checktest/b_2', 'etag': None, 'size_bytes': 2}])
+        good_data = json.dumps([
+            # Explicit None will skip it
+            {'path': '/checktest/a_1', 'etag': None, 'size_bytes': 1},
+            # ...as will omitting it entirely
+            {'path': '/checktest/b_2', 'size_bytes': 2}])
         req = Request.blank(
             '/v1/AUTH_test/checktest/man_3?multipart-manifest=put',
             environ={'REQUEST_METHOD': 'PUT'}, body=good_data)
@@ -686,6 +702,7 @@ class TestSloPutManifest(SloTestCase):
             '/v1/AUTH_test/checktest/man_3?multipart-manifest=put',
             environ={'REQUEST_METHOD': 'PUT'}, body=good_data)
         status, headers, body = self.call_slo(req)
+        self.assertEqual(('201 Created', ''), (status, body))
         expected_etag = '"%s"' % md5hex('ab:1-1;b:0-0;aetagoftheobjectsegment:'
                                         '10-40;')
         self.assertEqual(expected_etag, dict(headers)['Etag'])