[airflow] branch v1-10-test updated: [AIRFLOW-4472] Use json.dumps/loads for templating lineage data (#5253)

[di...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

dimberman pushed a commit to branch v1-10-test
in repository https://gitbox.apache.org/repos/asf/airflow.git


The following commit(s) were added to refs/heads/v1-10-test by this push:
     new 132ea62  [AIRFLOW-4472] Use json.dumps/loads for templating lineage data (#5253)
132ea62 is described below

commit 132ea62137624ce5fafdd7e1d809fa5a115cf0c8
Author: bolkedebruin <bo...@users.noreply.github.com>
AuthorDate: Tue May 7 22:42:28 2019 +0200

    [AIRFLOW-4472] Use json.dumps/loads for templating lineage data (#5253)
    
    jinja2 cannot use dict/lists as templates hence converting
    it to json solves this while keeping complexity down.
    
    (cherry picked from commit a6daeb544e815fe350a96d24ae3bb14aee4079a7)
---
 airflow/lineage/datasets.py    | 11 +++++++++--
 airflow/models/baseoperator.py |  7 ++++---
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/airflow/lineage/datasets.py b/airflow/lineage/datasets.py
index 2602770..3d61e5d 100644
--- a/airflow/lineage/datasets.py
+++ b/airflow/lineage/datasets.py
@@ -16,6 +16,7 @@
 # KIND, either express or implied.  See the License for the
 # specific language governing permissions and limitations
 # under the License.
+import json
 import six
 
 from typing import List
@@ -62,7 +63,11 @@ class DataSet(object):
         if attr in self.attributes:
             if self.context:
                 env = Environment()
-                return env.from_string(self._data.get(attr)).render(**self.context)
+                # dump to json here in order to be able to manage dicts and lists
+                rendered = env.from_string(
+                    json.dumps(self._data.get(attr))
+                ).render(**self.context)
+                return json.loads(rendered)
 
             return self._data.get(attr)
 
@@ -82,7 +87,9 @@ class DataSet(object):
         env = Environment()
         if self.context:
             for key, value in six.iteritems(attributes):
-                attributes[key] = env.from_string(value).render(**self.context)
+                attributes[key] = json.loads(
+                    env.from_string(json.dumps(value)).render(**self.context)
+                )
 
         d = {
             "typeName": self.type_name,
diff --git a/airflow/models/baseoperator.py b/airflow/models/baseoperator.py
index b61c980..470c14f 100644
--- a/airflow/models/baseoperator.py
+++ b/airflow/models/baseoperator.py
@@ -24,10 +24,11 @@ import functools
 import logging
 import sys
 import warnings
+from datetime import timedelta, datetime
+from typing import Callable, Dict, Iterable, List, Optional, Set
 
 from abc import ABCMeta, abstractmethod
 from datetime import datetime, timedelta
-from typing import Any, Callable, ClassVar, Dict, FrozenSet, Iterable, List, Optional, Set, Type, Union
 
 
 import attr
@@ -428,8 +429,8 @@ class BaseOperator(LoggingMixin):
         self._log = logging.getLogger("airflow.task.operators")
 
         # lineage
-        self.inlets = []  # type: Iterable[DataSet]
-        self.outlets = []  # type: Iterable[DataSet]
+        self.inlets = []   # type: List[DataSet]
+        self.outlets = []  # type: List[DataSet]
         self.lineage_data = None
 
         self._inlets = {