Solve CDC ordering issues #21
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jbfbell
approved these changes
Jul 10, 2023
edgao
approved these changes
Jul 10, 2023
one-table-bigquery.sql
Outdated
| ) | ||
| WHERE | ||
| JSON_VALUE(`_airbyte_data`, '$._ab_cdc_deleted_at') IS NOT NULL | ||
| OR JSON_TYPE(JSON_QUERY(`_airbyte_data`, '$._ab_cdc_deleted_at')) = 'null' |
There was a problem hiding this comment.
I think this needs to be flipped?
Suggested change
| OR JSON_TYPE(JSON_QUERY(`_airbyte_data`, '$._ab_cdc_deleted_at')) = 'null' | |
| OR JSON_TYPE(JSON_QUERY(`_airbyte_data`, '$._ab_cdc_deleted_at')) != 'null' |
... maybe we can delete the entire JSON_VALUE thing
There was a problem hiding this comment.
It turns out, in airbytehq/airbyte#28029 I removed all the 'null'(string) checks, because yep ^ that's what we wanted all along. - 7fad41a
There was a problem hiding this comment.
Why are both needed today?
Oops I think I left this comment before you responded, Evan! You can ignore since the code was removed.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR does better than #20 in that it solves out-of-order CDC inserts and deletes, even in the case when the PK should come back after a delete.
The main insight is that to keep this performant, we need to do all logical comparisons based on the order of things, by cursor, within the final table. So, we temporarily add back each remaining CDC-deleted row which still has an entry in the raw table so we can consider wether it is newer or older than new records. Then, like always, we'll remove any
deleted_at=truerecords at the end of the transaction. We still need to watch out for airbytehq/airbyte#27923, which means we cannot do any joins or cross-compares between the raw and final tables.This all remains fine, because we do all of the above within a transaction the user will never see "for real"