Uh oh!
There was an error while loading. Please reload this page.
[SPARK-12271][SQL] Improve error message when Dataset.as[ ] has incompatible schemas. - #10260
[SPARK-12271][SQL] Improve error message when Dataset.as[ ] has incompatible schemas.#10260nongli wants to merge 3 commits into
Conversation
SparkQA
commented
Dec 11, 2015
Test build #47560 has finished for PR 10260 at commit
|
There was a problem hiding this comment.
the simpleString of attribute includes expr id, which I think is noisy to users, should we generate the string manually or use StructType.fromAttributes to turn schema into StructType and call StrcutType.simpleString?
holdenk
commented
Dec 14, 2015
Could you maybe update the PR description to include the description from the JIRA? |
SparkQA
commented
Dec 14, 2015
Test build #47680 has finished for PR 10260 at commit
|
There was a problem hiding this comment.
Is there a reason to not just put this in CheckAnalysis and make it more general (i.e. not just Project)? I realize its only used here now, but if we ever do use it elsewhere, its invalid to remain after analysis.
marmbrus
commented
Dec 15, 2015
This looks great now! LGTM pending tests / conflict resolution |
Fixed a type erasure issue with wrap option.
SparkQA
commented
Dec 15, 2015
Test build #47747 has finished for PR 10260 at commit
|
SparkQA
commented
Dec 15, 2015
Test build #47754 has finished for PR 10260 at commit
|
cloud-fan
commented
Dec 16, 2015
LGTM |
marmbrus
commented
Dec 16, 2015
Thanks, merging to master. |
I think it was a mistake, and we have not catched it so far until apache#10260 which begin to check if the `fromRowExpression` is resolved. Author: Wenchen Fan <wenchen@databricks.com> Closesapache#10263 from cloud-fan/encoder.
…patible schemas. Author: Nong Li <nong@databricks.com> Closesapache#10260 from nongli/spark-11271.
No description provided.