8000 Truncate and log unbounded numerics that don't fit into destination type by ilidemi · Pull Request #3069 · PeerDB-io/peerdb · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Truncate and log unbounded numerics that don't fit into destination type #3069

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 25 commits into from
Jun 25, 2025

Conversation

ilidemi
Copy link
Contributor
@ilidemi ilidemi commented Jun 19, 2025

While at it, unified the four places where we were deciding on which type to use at the destination.

Attempted to support truncation from wider to narrower decimals in case the customer creates such columns, but we seem to royally not support diverging schemas for numerics. If anyone in the future has to implement this, hit me up for a bunch of test cases.

@ilidemi
Copy link
Contributor Author
ilidemi commented Jun 21, 2025

Are ColumnSettings with DestinationName different from SourceName supported? Seeing tests fail when I'm trying those but maybe I'm doing it wrong

Edit: not intended to be supported right now

@ilidemi ilidemi requested review from serprex and iamKunalGupta June 24, 2025 07:07
ilidemi and others added 4 commits June 24, 2025 07:27
Co-authored-by: Philip Dubé <serprex@users.noreply.github.com>
…#3066)

this avoids breaking existing setups build on pgvector types coming through as strings

this system can be used in the future to maintain a chronology of breaking changes
& not apply new behavior to old mirrors

it's meant to be a single dimension,
so there isn't the exponential explosion of feature combinations we have with the dynconf system
We've previously tried to leverage by having breaking changes only be introduced as optional features,
but this hits two problems: old behavior has to be default,
& it's a complicated mechanism designed for flexible configuration,
rather than a means to gate backwards incompatible changes from existing mirrors

this represents a point in PeerDB's life where we have people who just want the thing they tested (+ security fixes, etc)

infra for managing this system will be future work, outlined in #3095
@ilidemi ilidemi merged commit f7f1437 into main Jun 25, 2025
9 checks passed
@ilidemi ilidemi deleted the truncate-numerics branch June 25, 2025 18:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants
0