fix(sl): correct reserved-word/week-grain SQL and classify sl_query errors (#340)

Reserved-word columns (like, default, ...) referenced as source.col were
quoted with postgres double quotes even on BigQuery/MySQL, where a
double-quoted token is a string literal, not an identifier -- the
"Unexpected string literal" semantic-layer errors. quote_reserved_identifiers
now uses the identifier quote char of the dialect it will be parsed in
(backtick for BigQuery/MySQL), threaded through the planner and generator
parse sites; week_<weekday> granularity now emits WEEK(<weekday>) on BigQuery
instead of the invalid WEEK_MONDAY.

On the telemetry side, warehouse rejections from the sl_query execution path
are classified as expected (KtxQueryError) via a new shared markExpected()
helper, so routine agent/warehouse query failures stop reaching PostHog Error
Tracking as ktx faults; the sql_execution catch is refactored onto the same
helper. The daemon-compile boundary is deliberately left unclassified here so
genuine daemon crashes stay visible.
This commit is contained in:
Andrey Avtomonov 2026-07-03 23:19:33 +02:00 committed by GitHub
parent a0d19ba26f
commit 4ebce75449
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 156 additions and 21 deletions

View file

@ -254,7 +254,7 @@ class JoinGraph:
from sqlglot import exp as _exp
from semantic_layer.parser import quote_reserved_identifiers
quoted = quote_reserved_identifiers(on_clause)
quoted = quote_reserved_identifiers(on_clause, self.dialect)
tree = sqlglot.parse_one(
f"SELECT 1 FROM _a JOIN _b ON {quoted}", read=self.dialect
)