test(statement): cover null-typed bind parameter columns - #5
Closed
fornwall wants to merge 1 commit into
Closed
Conversation
fornwall
added a commit
to fornwall/adbc-spanner
that referenced
this pull request
Jul 14, 2026
…a + accept Null-typed bind columns (#279) * feat(statement): infer real parameter types in get_parameter_schema via a PLAN probe get_parameter_schema used to type every @name parameter as Null, claiming Spanner "exposes no way to introspect them beforehand". It does: a QueryMode::Plan probe returns the statement's undeclared parameters in ResultSetMetadata with the types the surrounding SQL implies (the same mechanism PGAdapter uses for PostgreSQL Describe). Queries plan in a single-use read-only transaction (the execute_schema surface); DML can only be planned in a read/write transaction, so it runs through the transaction runner - the plan executes nothing and the transaction commits empty. Types map through the existing arrow_field mapping, so a JSON parameter carries the arrow.json extension tag the bind path understands. A parameter the probe cannot type - DDL, DML on a read-only connection, a type the SQL context doesn't pin down, or a failed probe - stays Null, ADBC's convention for "type cannot be determined" (AdbcStatementGetParameterSchema in adbc.h). Emulator-verified: undeclared_parameters comes back for both query and DML plans; integration test now asserts Int64/Utf8 inference on both paths. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0192G976j7A1L8qFbfAJaU6N * fix(bind): accept Null-typed bind parameter columns (CONV-1) scalar_binder had no arm for Arrow's Null type, so a bind batch with a Null-typed column failed InvalidArguments ("unsupported Arrow type Null") - contradicting the driver's own get_parameter_schema, which types an undetermined parameter Null per adbc.h's AdbcStatementGetParameterSchema contract. A client building its bind batch from the reported schema (or pyarrow inferring null for an all-None parameter set) hit the rejection. Every cell of a Null-typed column now binds as an untyped NULL - the same wire shape as a NULL cell of any typed column, since add_param declares no parameter types and Spanner infers them from the SQL context. Not gated on Array::is_null: a NullArray carries no validity buffer, so the physical is_null reports false for its all-null cells. Verified against the emulator (integration test inserts via a Null-typed column and reads NULL back) and by the upstream validation suite's test_parameter_null_typed (fornwall/validation#5), which fails on main and passes with this fix. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0192G976j7A1L8qFbfAJaU6N --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
fornwall
force-pushed
the
test/bind-null-typed
branch
2 times, most recently
from
July 14, 2026 09:49
b204b30 to
326f697
Compare
Adds a test that binds a batch whose parameter column has Arrow type null and asserts NULL is bound per row. Signed-off-by: Fredrik Fornwall <fredrik@fornwall.net>
fornwall
force-pushed
the
test/bind-null-typed
branch
from
July 14, 2026 09:50
326f697 to
e363d8e
Compare
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.
Split out of #4 (one PR per test). Adds one generic, driver-agnostic test for a gap found while reviewing an ADBC driver (the adbc-spanner Rust driver) against this suite; candidate for upstreaming to adbc-drivers/validation later.
TestStatement.test_parameter_null_typed(gated onstatement_bind)Binds a batch whose parameter column has Arrow type
nulland asserts NULL is bound per row (insert + read-back through the existingsample_tablefixture / itsquery_overridehook).get_parameter_schema).AdbcStatementGetParameterSchemadoc comment inadbc.h: "If the type cannot be determined, the type of the corresponding field will be NA (NullType)." A client that builds its bind batch from the driver's own reported parameter schema therefore produces null-typed columns; pyarrow also infersnullfor an all-Noneparameter set (the DBAPIexecutemanyshape). A driver advertising bind support that rejects the type its ownGetParameterSchemahands out contradicts itself.Results on real drivers
INVALID_ARGUMENT: cannot bind parameter "p2": unsupported Arrow type Null.Framework checks
uv run pytest tests/(210 passed, 1 skipped),uv run ty check(clean),pre-commit run --all-files(all hooks pass) — all on this branch independently.🤖 Generated with Claude Code
https://claude.ai/code/session_01LGf8PVEe2tYkw8Q6Pd95tq