flux-pr-1501
sqlparser-rs (Rust) · W1 · GPT-5.1 Codex Mini
Tests passed. 1/1 commands passed. Strength: strong.
env PATH=/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin cargo test --all-featuresPartial score: 1/1
Trajectory
unknown · partial order onlyCanonical trajectory missing; showing coarse derived order only.
Quality
Equivalence Reasoning
behavioral
The patch appears to miss core SHOW behavior integration: in the shown parser changes, `parse_show()` still consumes `DATABASES/SCHEMAS` before calling `parse_show_databases()/parse_show_schemas()`, but those functions now also `expect` the same keyword, which is a functional mismatch. The visible changes also don’t clearly show the required dialect hook for LIKE-before-IN ordering and Snowflake modifiers wired consistently across all SHOW variants, so intent is not fully satisfied.
Code Review
The patch does not likely satisfy the intended change: it partially reshapes SHOW AST/options but introduces high-probability parser regressions in keyword consumption/order and likely drops some SHOW variants.
parse_show already consumes DATABASES/SCHEMAS before delegating, but parse_show_databases/parse_show_schemas then call expect_keyword for DATABASES/SCHEMAS again. This will fail on valid input because the token has advanced past the object type.
TERSE is parsed inside parse_show_databases/parse_show_schemas, but these functions are entered only after DATABASES/SCHEMAS has already been matched. Valid forms like SHOW TERSE DATABASES cannot be parsed correctly with this ordering.
Previous logic explicitly handled MATERIALIZED VIEWS, but new dispatch only checks a single peek_show_keyword path to VIEWS and calls parse_show_views() without the materialized flag. This risks rejecting or mis-parsing SHOW MATERIALIZED VIEWS.