Revert "Revert "Fix unexpected behavior with FORMAT and SETTINGS parsing""#68614
Conversation
|
This is an automated comment for commit 0b7cba6 with description of existing statuses. It's updated for the latest CI running ❌ Click here to open a full report in a separate page
Successful checks
|
| apply_query_settings(*query_with_output->settings_ast); | ||
| else if (insert && insert->settings_ast) | ||
| apply_query_settings(*insert->settings_ast); | ||
| client_context->setSettings(old_settings); |
There was a problem hiding this comment.
After some debugging, I suspect 00653_verification_monotonic_data_load timeouts under tsan caused by this line, client hung here for some reason. However, no explanation why it reproduced only at one test and didn't reproduce before, even though we had setSettings under if (old_settings)
There was a problem hiding this comment.
Also, the issue might be related to grep -q, such as early termination of the client leading to that
There was a problem hiding this comment.
There’s a data race that this PR reveals
#68626 (comment)
We can merge this PR after fixing it by #68791
314f2b3 to
bf7ec7d
Compare
|
@alexey-milovidov PTAL 👀 |
…-68608-revert-60197-format-settings-parsing
|
FastTest is still 🔴. |
…608-revert-60197-format-settings-parsing
That's werid, like data race still reproduces... I'm going to take a look... |
|
@vdimir, at least it is not related to these changes. |
|
Seems like this might be causing |
|
Yes, it breaks the test just like the original PR: CI DB |
|
One more attempt to fix #71135 |
Introduced in ClickHouse#38900, removed in ClickHouse#60197 (ClickHouse#68614)
Reverts #68608
Changelog category (leave one):
Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):
Allow to write
SETTINGSbeforeFORMATin a chain of queries withUNIONwhen subqueries are inside parentheses. This closes #39712. Change the behavior when a query has the SETTINGS clause specified twice in a sequence. The closest SETTINGS clause will have a preference for the corresponding subquery. In the previous versions, the outermost SETTINGS clause could take a preference over the inner one.CI Settings (Only check the boxes if you know what you are doing):