Set prefer_column_name_to_alias on ClickHouse connections - #948
Open
smoleyxd wants to merge 1 commit into
Open
Conversation
ClickHouse resolves identifiers to select-list aliases before table columns by default, so lookup queries that reuse a column name as an alias (e.g. 'data as metadata, 0 as data' in the item lookup) return the alias literal instead of the column. Item lookups then return metadata as UInt8 0, mixed lookup unions produce Variant(Array(Int8), UInt8), and the JDBC client fails every /co lookup with 'Column is not of array type'. Setting prefer_column_name_to_alias=1 restores the alias semantics these queries rely on under SQLite and MySQL. Fixes PlayPro#947
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #947.
ClickHouse resolves identifiers to select-list aliases before table columns by default (
prefer_column_name_to_alias = 0), so lookup queries that reuse a real column name as an alias mis-resolve. The item lookup builds... data as metadata, 0 as data ... FROM co_item, wheredataindata as metadatabinds to the0 as dataalias instead of the column:Standalone item lookups return
metadataas the literal0; mixed lookup unions end up asVariant(Array(Int8), UInt8)on ClickHouse 26.x. Either way the JDBC client throwsColumn is not of array typeinDatabaseUtils.getBytesand every/co lookup//co inspecttouching item rows returns nothing (details and full stack trace in #947).This sets
prefer_column_name_to_alias = 1on the connection, which makes ClickHouse resolve columns first — the same semantics SQLite and MySQL give these queries. I audited the generated queries for the inverse hazard (a query needing the alias where a real column with that name exists) and found none: the setting only changes resolution when both an alias and a column share a name, and the only such collision is thedataone this fixes.Running in production on two servers (ClickHouse 26.7.1, ~85M events) since 2026-08-09: item metadata comes back as the correct blobs, mixed unions unify to
Array(Int8), and lookups work again. TheWHERE notEmpty(data)-style mis-resolutions on shadowed columns are fixed by the same setting.