All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

• Requires localize ~> 1.4, which bundles CLDR 49 (Unicode 18.0).

• Localize.Territory.Ecto.Type casts a CLDR territory alias to its code, as Localize.validate_territory/1 does: "840" casts to :US and "UK" to :GB.

• mix localize.ecto.audit no longer fails on a Unicode version gap between the server's ICU and the application's CLDR; it prints the warning and succeeds, as documented.

• The time zone audit no longer reports CLDR's Etc/Unknown placeholder as unknown to the server, which made the audit fail on every server.

• The CLDR-to-Unicode version table behind Localize.Ecto.Audit.unicode_versions/1 recorded CLDR 48 as Unicode 16.0 and CLDR 49 as 17.0; the correct values are 17.0 and 18.0. mix localize.ecto.audit no longer reports spurious drift against a PostgreSQL 18 server linked against ICU 77, which is also Unicode 17.

• The collation guide now recommends the builtin providers over C.UTF-8 as a version ladder — PG_UNICODE_FAST on PostgreSQL 18+, PG_C_UTF8 on 17, C earlier — and documents their equivalence to Elixir's own string ordering. PG_UNICODE_FAST keeps the prefix-LIKE index optimization that ICU collations forgo.

Renamed from localize_ecto to localize_sql — the package is about SQL databases (collations, serialization, migrations), not Ecto in general, which can be used without a database. The Localize.Ecto.* module namespace and the mix localize.ecto.audit task are unchanged; the app, config key (:localize_sql), env var (LOCALIZE_SQL_SQLITE_ICU) and priv/localize_sql/ path are renamed.

localize_sql grows from locale-aware collation into the SQL home for every Localize type — serializing the full set of Localize and Elixir data types and generating the tagged-decimal aggregate machinery ex_money_sql builds on.

• Localize.Ecto.TaggedDecimal and its DDL generate a PostgreSQL composite type with tag-guarded sum/avg/min/max aggregates and +/- operators for any tag-plus-decimal value; a mismatched tag (euros plus yen) raises in the database. ex_money_sql now builds money_with_currency on this rather than its own SQL.

• Unit serialization, folded in from localize_units_sql: Localize.Unit.Ecto.{Composite,Map}.Type and the Localize.UnitWithUsage variants, Localize.Unit.DDL (now sum/min/max/avg for cldr_unit and cldr_unit_with_usage), and mix localize.unit.gen.migration.

• Ecto types for the remaining data types: Localize.Currency.Ecto.Type, Localize.LanguageTag.Ecto.Type, Localize.Territory.Ecto.Type and Localize.Script.Ecto.Type (text); Localize.Ecto.Type.IntegerRange and .DateRange (int8range/daterange); Localize.Duration.Ecto.Type and Localize.Ecto.Type.Duration (interval).

• Localize.Ecto.Range provides query macros for PostgreSQL range columns — containment, overlap, adjacency, the bound accessors and the range aggregates — and Localize.Ecto.Range.Migration creates a time range type, the btree_gist extension and the no_overlap/3 exclusion constraint that forbids overlapping bookings.

• Localize.Ecto.Migration.Generator renders and writes migrations for the SQL-generating mix tasks.

• Localize.UnitWithUsage.Ecto.Composite.Type.cast/1 now accepts a string-keyed map with no "usage" key, casting it as a unit without a usage. Only the atom-keyed form tolerated the key's absence, so an HTML form — which submits string keys — could not cast a unit unless it also submitted a usage.

• Requires localize ~> 1.0. postgrex is an optional dependency; the range, duration and tagged-decimal composite types compile only when it is present, so a SQLite-only or collation-only consumer carries nothing extra.

• The localize_units_sql package is superseded — its unit serialization is folded in here, and its sum-only aggregates are replaced by sum, min, max and avg for both cldr_unit and cldr_unit_with_usage.

The 1.0 release candidate. localize_ecto now supports SQLite as well as PostgreSQL, with the same locale resolving to the same collation on both.

• SQLite support via localize_icu, a bundled SQLite loadable extension that registers ICU collations under the same names PostgreSQL uses. The same locale sorts identically on both databases.

• Localize.Ecto.SQLite3 provides collate/1,2, lower/1,2, upper/1,2 and initcap/1,2 for SQLite. BCP 47 tailorings such as -u-co-phonebk need no migration; collations are registered on demand.

• Localize.Ecto.SQLite3.Extension locates the compiled extension for the :load_extensions option of Exqlite and Ecto.Adapters.SQLite3.

• Localize.Ecto.Postgres names the PostgreSQL macros explicitly, alongside the new SQLite module.

• SQLite compares canonically equivalent strings (the NFC and NFD forms of café) as equal where PostgreSQL, using deterministic collations, compares them as unequal. Ordering agrees on both; normalizing on write removes the difference.

• The SQLite extension is an opt-in build (config :localize_ecto, :sqlite_icu, true, or LOCALIZE_ECTO_SQLITE_ICU=true). Without it the package stays pure Elixir and PostgreSQL users are unaffected.

• Query macros emit Ecto's identifier/1 rather than the deprecated literal/1. The generated SQL is unchanged.

• Collation resolution carries all collation-affecting BCP 47 -u- keywords (kn, ks, kc, kf, ka, kb, kr, kv) into the collation name, so collate(f.name, "en-u-kn-true") finds a natural-sort collation created in a migration.

• create_collation/2 accepts tailoring options (:strength, :numeric, :case_level, :case_first, :alternate, :backwards) using the Localize.Collation.Options vocabulary. Primary and secondary strengths default to a nondeterministic collation, enabling case/accent-insensitive equality and unique indexes.

• mix localize.ecto.audit and Localize.Ecto.Audit report collation version drift with per-index remediation SQL, database default collation drift, server-versus-application Unicode/ICU/CLDR versions, and time zone inventory differences.

• Localize.Ecto.Type.TimeZone is an Ecto type for IANA time zone identifiers, canonicalizing aliases and BCP 47 short zone ids against the CLDR inventory; at_time_zone/2 applies a validated AT TIME ZONE.

• Localize.Ecto.TextSearch.config_for/2 resolves a locale to the best PostgreSQL text search configuration, and ts_match/2,3 builds a locale-aware to_tsvector … @@ websearch_to_tsquery match.

• lower/1,2, upper/1,2 and initcap/1,2 apply locale-aware case mapping via the locale's collation ("INDIGO" lowercases to "ındıgo" under "tr").

• Requires Localize ~> 1.0-rc.1 for the CLDR time zone inventory.

• Requires Localize ~> 1.0-rc

Initial release.

• Localize.Ecto.collate/1,2 query macros applying PostgreSQL ICU COLLATE clauses to query expressions and comparisons, resolved from a locale or the current Localize locale.

• Locale-to-collation resolution via CLDR language matching, including BCP 47 -u-co- collation types and direct collation names.

• Localize.Ecto.Migration.create_collation/2 and drop_collation/2 for creating ICU collations reversibly in migrations, and collated/2 for building collated indexes with Ecto.Migration.index/3.

See the README for full documentation.