chore: release v0.1.6

[github-actions]

🤖 New release

• drizzle-types: 0.1.5 -> 0.1.6
• drizzle-core: 0.1.5 -> 0.1.6
• drizzle-migrations: 0.1.5 -> 0.1.6
• drizzle-postgres: 0.1.5 -> 0.1.6
• drizzle-sqlite: 0.1.5 -> 0.1.6
• drizzle-macros: 0.1.5 -> 0.1.6
• drizzle-seed: 0.1.5 -> 0.1.6
• drizzle-cli: 0.1.5 -> 0.1.6
• drizzle: 0.1.5 -> 0.1.6

Changelog

drizzle

0.1.6 - 2026-05-16

Added

• db.migrate returns MigrateOutcome across all drivers
• (migrations) build::Config::from_toml + getters + auto-watch
• (drizzle) re-export Casing at the crate root
• (postgres,core) Postgres COLLATE + Constraint enum + expr-level collate()
• (sqlite) column-level COLLATE attribute
• align postgres codecs with drizzle rc1
• (postgres) AWS Aurora Data API driver
• (sqlite) Durable Objects transactions + savepoints, doc refresh
• (core) add sqlcommenter-style .comment() and .comment_tags()
• (sqlite) scaffold Cloudflare Durable Objects SQL driver (WASM-only)
• (sqlite) scaffold Cloudflare D1 driver (WASM-only)
• (builder) typestate deduplication and compile-time GROUP BY validation
• (types) improve SQL type system correctness
• (migrations) align folder layout and tracking upgrades
• (migrations) redesign runtime and build migration APIs
• (core) support &ZST tables/columns across macros and traits
• add percent_rank/cume_dist window functions and improve expr docs
• many-to-many relations through junction tables
• add .first() sugar on RelationHandle
• handle BLOB and boolean columns in query API JSON serialization
• move enum value conversions from table macro to enum derive
• compile-time view query DSL with const SQL generation
• support foreign key references on views for relational queries
• add relational query API with nested relation loading
• (builder) expand transaction builders with group_by, set ops, CTEs, and select_distinct
• (core) add tag! macro for defining aliases in one line
• (builder) add .value() shorthand for single-row inserts
• (builder) expand DrizzleBuilder with IntoSelect, set operations, group_by, having, and PG joins
• (prepared) add debug_assert for parameter count validation
• (core) constrain PostgreSQL cast() with Compatible bound
• (core) enforce Textual bound on group_concat()
• (core) enforce BooleanLike constraint on WHERE, HAVING, and logical operators
• close type propagation gaps for SQLite and PostgreSQL value layers
• (prepared) use const-generic arrays for parameter bindings
• (postgres) add time crate date/time type support
• (prepared) validate parameter bindings and add ParamSet
• (postgres) add rust-decimal Numeric value support
• (types) add Assignable trait and nullable typed placeholders
• (bind) add typed placeholders and remove params macros
• (expr) add typed subqueries with CASE/window support
• (types) strengthen marker inference and prevent Any widening
• (postgres) add dedicated markers for extended postgres types
• (expr) enforce dialect-aware scalar typing and strict decode
• (sqlite) reject implicit Any fallback in type inference
• (select) block untyped raw expressions in strict decode
• (cast) enforce sqlite compatibility for typed cast targets
• (expr) add dialect-aware aggregate result policy
• (sqlite) enforce dialect cast targets and strict affinity guards
• (expr) infer coalesce_many nullability for strict decode
• (expr) accept dialect marker values in cast targets
• (types) add sqlite/postgres type alias modules
• (select) enforce typed row-shape matching for strict decode
• (select) enforce strict row-shape checks for strict decode
• (dml) add typed returning markers with scoped validation
• (row) enforce selector table scopes at compile time
• (alias) introduce Tag-based aliases for tables and CTEs
• (types) improve row type inference across builders and macros
• (push) scope and normalize live snapshot for idempotent push
• (push) execute statements directly instead of returning them
• (postgres) add introspect and push to Postgres drivers
• (sqlite) add introspect and push to all SQLite drivers
• (core) add asc and desc order-by helpers
• (cli) add JSON import/export to drizzle new
• (cli) add drizzle new interactive schema builder
• (test) capture SQL, params, and source expression in test failure reports
• add no_std/alloc support to sqlite, postgres, and mysql crates
• add executor traits for async SQLite prepared statements
• (tests) add prepared statement transaction tests
• add SQL::map_params() and ToSQL for OwnedPreparedStatement
• (tests) add async edge case tests for cloned Drizzle across tasks
• make Drizzle cloneable for sharing across async tasks
• auto-derive Debug for schema types in macros
• store schema in Transaction structs and add schema() accessor
• (bench) add tokio-postgres benchmarks and split CI runs
• (tests) add postgres transaction and savepoint tests
• store schema in Drizzle struct and derive Copy for schema types
• (ci) extract structured failure reports from turso test output
• (examples) add turso async driver example
• generate reverse Joinable impls for auto-FK chained joins
• add JoinedTable associated type to JoinArg for chained joins
• (tests) add many-to-many join tests and FK references to junction tables
• (joins) auto-derive join conditions from FK metadata
• (constraints) compile-time FK, PK, and constraint type system
• (seed) add deterministic database seeder with type-safe config
• (relations) auto-derive relation metadata from FK attributes
• (savepoints) add nested transaction support via SQL savepoints
• (tracing) add feature-gated tracing integration alongside puffin
• (types) add smallserial parity and flexible text/blob wrappers
• add datetime, math, and PG sequence expressions
• add aggregate expressions (total, every, json_object_agg)
• add string expressions (concat_ws, char_length, translate, regexp)
• add comparison expressions and dialect-aware greatest/least
• (core) propagate aggregate markers and enforce type constraints
• (core) add AggOr combinator trait for aggregate propagation
• (sqlite) add serde_json::Value conversions
• (bind) expand ValueTypeForDialect for all supported Rust types
• (update) add Empty/NonEmpty typestate to prevent empty updates and improve FK/PK error spans
• compile-time SQL generation for PostgreSQL tables, indexes, and enums
• (procmacros) add PostgreSQL enum FromSql/ToSql and feature-gate cleanup
• (types) add distinct sqlite/postgres marker modules
• (postgres) align rc2 type normalization
• (postgres) add Enum variant to OwnedPostgresValue
• (postgres) fill AWS Data API row impls for chrono FixedOffset/Duration and time crate
• (migrations) add postgres introspection queries to migrations crate
• (migrations) add generate() for programmatic migration diffing
• (cli) add parity overrides and migration planning modes
• (types) lift EnvOr/EnvOrError into drizzle-types::migration
• (cli) add cargo-binstall metadata
• (cli) Cloudflare D1 HTTP driver for the CLI
• (cli) emit migrations.js bundle index on generate when configured
• (cli) add JSON Schema derives for config types

Changed

• (tx) centralize savepoint logic in shared module
• (builder,tx) collapse DrizzleBuilder + TransactionBuilder into one
• (tx) collapse per-driver Postgres TransactionBuilder behind generic
• (tx) collapse per-driver SQLite TransactionBuilder behind generic
• (tx) macro-driven typestate for SQLite TransactionBuilder delete
• (procmacros,tests) migrate test macros to #[drizzle::test] attribute with DI
• (core,procmacros) pack ColumnRef metadata into ColumnFlags byte
• (migrations) identify migrations by name, not created_at
• simplify migration APIs
• (sqlite) remove unchecked decode feature path
• convert query extension traits to inherent methods, add LATERAL fix
• rename .select_columns() to .columns() on table ZSTs
• replace dyn trait objects with const Copy structs for schema metadata
• use file-based temp DBs for fully parallel SQLite tests
• schema-centric architecture with compile-time SQL and metadata traits
• add Clauses typestate to prevent duplicate query clauses
• deduplicate transaction query builder constructors with macros
• replace row() wrapper with direct tuple support in in_subquery
• restructure justfile with test module system
• (tests) replace manual result structs with SelectSimple and add DB validation
• (types) normalize impl ordering to canonical type order
• (types) replace core SQL type markers with dialect-resolved types
• (alias) remove runtime alias_named in favor of tag aliases
• (postgres) consolidate snapshot push API and deduplicate nextval parsing
• rename expressions() path helper and clean up prelude imports
• consolidate expressions module into expr
• (test) remove DDL from EXECUTED STATEMENTS
• (test) migrate postgres tests to drizzle_exec! => pattern
• (test) migrate sqlite tests to drizzle_exec! => pattern
• make prepared statement inner fields pub(crate)
• rename connection accessors to conn()/conn_mut() and inner()
• rename SQLSchema::sql() to ddl() for clarity
• (tokio-postgres) use collect for param_refs and restructure profiling scopes
• (examples) move imports inside main and add clone demo
• migrate transaction closures from Box::pin to AsyncFnOnce
• (tests) improve assertions and remove debug output
• (tests) fix feature-gate warnings and clean up imports
• (tests) consolidate feature-gated schema definitions
• (tests) clean up test infrastructure and remove dead code
• (tests) migrate compile-fail tests from trybuild to doc tests
• (seed) feature-gate backends and refine public API
• (insert) restructure ON CONFLICT with builder pattern
• (core/row) tighten row-decoding for sqlite + postgres
• (core) unify libsql + turso row decoding behind SqliteValueRow
• (no_std) full workspace audit, const fn, and prelude cleanup
• (core) macro-ify ValueTypeForDialect impls in bind.rs
• (core) share checked_float_to_int helper across dialects
• (core) add ParamStyle to decouple placeholder syntax from dialect
• (types) derive expr markers from resolved column types
• (types) move SQL marker system into drizzle-types
• (alias) switch typed aliases to generic Tagged wrapper
• (shared) consolidate CTE and JoinArg macro implementations
• (core) remove dead markers and unused migration helpers
• (core) replace manual tuple impls with recursive accumulator macros and add diagnostic error messages
• (macros) factor model-marker generation, slim view + insert codegen
• (macros) drop is_primary/is_unique bools from FieldInfo
• (macros) Constraint enum for SQLite primary/unique state
• (macros) unify Postgres DDL emission with SQLite pattern
• (macros) unify SQLite DDL emission into single fragment emitter
• (procmacros) dedupe view flag packing and test assertion macros
• use const SQL in SQLite schema create_statements for consistency
• clean up proc macro compile error messages
• deduplicate constraint generation between SQLite and PostgreSQL proc macros
• (macros) replace too-many-args with config structs
• (migrations) extract Snapshot + canonical EntityCollection
• make migration version helpers const fn
• remove unused VersionLt trait
• (types) share casing and migration tracking config
• apply let-chains and match-style rewrites (MSRV 1.95)
• zero-copy seed crate by replacing String keys with &'static str
• (cli) centralize command harness boilerplate
• (cli) use canonical queries from migrations crate

Documentation

• (readme) comprehensive cleanup + runnable smoke test
• (readme) regroup query features under single Querying H2, lift Migrations out of Getting Started, compress duplicated sections
• (readme) update group_by syntax and add condition examples
• (readme) clarify migration workflow
• tighten README flow
• (doctest) convert ignored fragments to runnable rust examples
• add query type alias example to README
• clean up README section ordering and remove internal details
• add trait import note for relational queries
• add relational query API to README
• add expressions, group by, and set operations to README
• clean up jargon, add missing doc comments, hide internal types
• overhaul README with generated models, CLI reference, and runtime migrations
• restructure README with getting started guide and CLI workflow
• add transaction, savepoint, and prepared statement examples to driver modules
• add transactions and savepoints section to README
• point install instructions to git repo instead of crates.io
• (readme) update join syntax and add order_by/limit/offset
• move project instructions to AGENTS.md and add git hunk tips
• (readme) update to 0.1.5 and align postgres schema example
• (types) add diagnostic::on_unimplemented to CountPolicy and ArithmeticOutput
• replace todo/ignore in prepared statement doctests with compilable scaffolds
• (postgres) add missing types module to doctest scaffolds
• (migrations) update crate-level docs with compilable examples

Fixed

• (docs) repair pre-existing broken postgres doc tests
• address ci benchmark and lint failures
• (transaction) decouple query lifetimes from tx borrows
• (tests) drop unneeded wildcard fields covered by .. in postgres/query
• (ci) drop libsql from Linux benchmarks too
• (ci) drop libsql from macOS benchmarks
• (bench) skip turso RETURNING benchmark
• (ci) use per-OS feature matrix for sqlite benchmarks
• (test) update sqlite returning doctest
• (macros) resolve optional-feature schema and helper compile errors
• satisfy pre-commit checks for ref trait impls
• update compile_fail stderr files for Deref type path changes
• make greatest/least PostgreSQL-only instead of mapping to SQLite MAX/MIN
• filter views query by schema during push to avoid concurrent DDL failures
• regenerate trybuild .stderr snapshots for current compiler
• query API param ordering, codegen, and error improvements
• increase prepared statement perf test tolerance to 5x
• feature-gate trybuild tests requiring uuid extension types
• use TransactionError and From for PG transaction error handling
• filter null elements from json_group_array and improve query API docs
• (ci) gate raw_sql_ui trybuild test on uuid feature
• resolve CI failures from missing ToSQL import and cross-driver catch_unwind
• use drizzle_catch_unwind! macro for cross-driver prepared statement tests, correct sign() return type
• (builder) enforce BooleanLike on ON CONFLICT WHERE clauses
• (macros) use specific driver features for postgres proc macro codegen
• (postgres) decode optional select models via NullProbeRow
• (expr) infer non-null coalesce/ifnull result nullability
• (test) enable foreign_keys pragma in SQLite test setup
• (test) capture Turso failures in structured reports
• (test) improve failure report readability
• (tests) wrap async prepared calls in drizzle_try!() for transactions
• (tests) update transaction tests for AsyncFnOnce closure syntax
• (tokio-postgres) remove Send bound from transaction future closures
• (tests) improve wrap_text and long test name rendering in failure reports
• (examples) feature-gate rusqlite example imports
• (postgres) convert doc tests from ignore to no_run with mod drizzle shim
• (tests) correct FK type mismatch in Post.author_id
• (parity) align sqlite metadata DDL and serde-gated test structs
• (migrations) track applied entries by created_at
• (postgres) improve alias ergonomics and add regressions
• (sqlite) decode named FromRow structs by column
• (fromrow) decode named postgres structs by column
• (ci) add workflow_dispatch trigger to release workflow
• (core) escape embedded double-quotes in SQL identifiers
• correct nullability in greatest/least and setval, fix sign doc
• remove extraneous Mo::Nullable bound in make_timestamp
• propagate nullability and aggregate kind in make_timestamp and date_bin
• rename duplicate 'Distinct Wrapper' section header in agg.rs
• ensure inner subquery exposes FK columns needed by nested relations
• emit compile error for invalid composite FK fields instead of silent fallback
• correct PG param ordering, UTF-8 safety, and query API robustness
• eliminate unnecessary string allocations in SQL generation
• (core) detect f32 overflow in SQLite FromRow conversions
• (types) complete tuple type propagation support
• use impl Iterator for create_statements to auto-derive Send
• (macros) resolve const_format path via proc-macro-crate
• expose postgres driver types in doctests
• (ci) align macro doctest feature wiring and time ISO formatting
• (doctest) wire feature-gated macro docs across drivers
• prevent duplicate From impls for #[derive(PostgresEnum)] fields
• use unwrapped base type for partial select model fields
• (fromrow) generate offset-aware row decoding in macros
• (postgres) prevent SERIAL columns from getting GENERATED AS IDENTITY
• (test) capture prepared statement SQL and params in reports
• (fromrow) fix turso JSON field temporary lifetime error
• (ddl) harden SQL generation and migration execution
• (sqlite) add shim to on_conflict doc test so it compiles with no_run
• (postgres) propagate scope markers in join_using methods
• (postgres) move schema() to SQLTableInfo impl in refresh test
• (lint) remove redundant .into_iter() in topological sort
• (postgres) fix PgTypeCategory misclassifying integer as serial
• (postgres) harden sequence and index introspection queries
• (codegen) map BOOLEAN SQL type to bool in SQLite introspection
• (types) always serialize Option fields in DDL snapshot structs
• (seed) split dialect imports so single-feature builds resolve
• (cli) bind view schema filter when introspecting postgres views
• remove spurious bind parameter from VIEWS_QUERY calls
• (cli) cfg-gate driver-dependent code and suppress warnings

Other

• (sqlite) add raw prepared parity across drivers
• add postgres suite and puffin scope reporting

Performance

• (sqlite) preallocate async prepared bind params
• (drivers) optimize decode paths and add fluent rows cursors
• (core) reduce SQL render allocations and bind overhead

---

This PR was generated with release-plz.

[themixednuts]

Closing to re-trigger CI under the new ci.yml that runs on release PRs. Reopening immediately.

[themixednuts]

Reopened — CI should now run on this PR's head commit.