Github user vanzin commented on a diff in the pull request:
https://github.com/apache/spark/pull/11796#discussion_r58104272
--- Diff: dev/run-tests.py ---
@@ -371,9 +371,10 @@ def build_apache_spark(build_tool, hadoop_version):
build_spark_sbt(hadoop_version)
-def detect_binary_inop_with_mima():
+def detect_binary_inop_with_mima(hadoop_version):
+ build_profiles = get_hadoop_profiles(hadoop_version) +
modules.root.build_profile_flags
set_title_and_block("Detecting binary incompatibilities with MiMa",
"BLOCK_MIMA")
- run_cmd([os.path.join(SPARK_HOME, "dev", "mima")])
+ run_cmd([os.path.join(SPARK_HOME, "dev", "mima")] + build_profiles)
--- End diff --
I added these arguments so that when the mima script would end up
re-running the packaging step, it would not copy new files with different
versions into the assembly `jars` directory. That was causing test failures
earlier if you look at the comment history.
Later I added a flag to disable copying dependencies altogether, and used
it in the mima invocation, making this not entirely necessary. But in the end,
I think this is still more correct than the previous code - and, as you say,
has the nice effect of making sure binary compatibility doesn't change because
of different dependency versions.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
