feat(electric): Enable introspection of electrified enum types in the proxy

[alco]

Fixes VAX-1042.

[linear]

VAX-1042 Pass the enum type info to the client (via the generator)

[davidmartos96]

@alco I'm trying out the PR with the an enum migration but the introspection fails with the following error.
Am I doing something wrong? I'm running the Electric service built in latest commit from this branch.

...

CREATE TYPE color AS ENUM ('RED', 'GREEN', 'BLUE');
CREATE TABLE enums (
    id TEXT PRIMARY KEY,
    c color
);
ALTER TABLE enums ENABLE ELECTRIC;

16:03:44.955 pid=<0.3106.0> pg_producer=postgres_1 [info] Migrating version <nil> -> 1
16:03:44.955 pid=<0.3106.0> pg_producer=postgres_1 [info] Applying migration 1: "CREATE TABLE items (\n    value text NOT NULL,\n    CONSTRAINT items_pkey PRIMARY KEY (value)\n);\n\n\n"
16:03:44.955 pid=<0.3106.0> pg_producer=postgres_1 [info] CREATE TABLE items (...)
16:03:44.956 pid=<0.3106.0> pg_producer=postgres_1 [info] Applying migration 1: "CREATE TYPE color AS ENUM (\n 'RED',\n 'GREEN',\n 'BLUE'\n);\n\n\nCREATE TABLE enums (\n    id text NOT NULL,\n    c color,\n    CONSTRAINT enums_pkey PRIMARY KEY (id)\n);\n\n\n"
16:03:44.956 pid=<0.3106.0> pg_producer=postgres_1 [info] CREATE ENUM "public"."color" WITH VALUES ["RED", "GREEN", "BLUE"]
16:03:44.956 pid=<0.3106.0> pg_producer=postgres_1 [info] CREATE TABLE enums (...)
16:03:44.957 pid=<0.3106.0> pg_producer=postgres_1 [info] Saving schema version 1 /{Electric.Postgres.Extension.SchemaCache, "postgres_1"}/
16:03:44.961 pid=<0.3100.0> pg_producer=postgres_1 [info] Saved schema version 1
16:03:44.962 pid=<0.3106.0> pg_producer=postgres_1 [info] Electric.Replication.Postgres.Client.with_conn(%{database: ~c"web_wa_sqlite", host: ~c"postgres", ip_addr: ~c"172.29.0.2", ipv6: true, nulls: [nil, :null, :undefined], password: ~c"******", port: 5432, ssl: true, timeout: 5000, username: ~c"postgres"})
16:17:23.781 pid=<0.3435.0> [info] GET /api/migrations
16:17:23.801 pid=<0.3435.0> [info] Sent 200 in 19ms
16:17:24.347 pid=<0.3437.0> proxy_session_id=6 [warning] Not validating user "prisma"
16:17:24.347 pid=<0.3437.0> proxy_session_id=6 [info] Initialising injector in capture mode %Electric.Postgres.Proxy.Injector.Prisma{config: %Electric.Postgres.Proxy.Prisma{server_version: {"14.9", 140009}}, prepared_statements: %{}, active_statement: nil, portals: %{}}
16:17:24.350 pid=<0.3437.0> proxy_session_id=6 [info] Upstream connection is ready to accept queries
16:17:24.352 pid=<0.3437.0> proxy_session_id=6 [info] Matched prisma introspection query to Elixir.Electric.Postgres.Proxy.Prisma.Query.NamespaceV5_2
16:17:24.352 pid=<0.3437.0> proxy_session_id=6 [info] Matched prisma introspection query to Elixir.Electric.Postgres.Proxy.Prisma.Query.TableListV4_8
16:17:24.353 pid=<0.3437.0> proxy_session_id=6 [info] Matched prisma introspection query to Elixir.Electric.Postgres.Proxy.Prisma.Query.TypeV4_8
16:17:24.353 pid=<0.3437.0> proxy_session_id=6 [error] GenServer #PID<0.3437.0> terminating
** (ArgumentError) errors were found at the given arguments:

  * 1st argument: not a bitstring

    :erlang.byte_size(["RED", "GREEN", "BLUE"])
    (pg_protocol 0.1.0) lib/pg_protocol/message.ex:483: anonymous fn/1 in PgProtocol.Encoder.PgProtocol.Message.DataRow.encode/1
    (elixir 1.15.4) lib/enum.ex:1693: Enum."-map/2-lists^map/1-1-"/2
    (elixir 1.15.4) lib/enum.ex:1693: Enum."-map/2-lists^map/1-1-"/2
    (pg_protocol 0.1.0) lib/pg_protocol/message.ex:481: PgProtocol.Encoder.PgProtocol.Message.DataRow.encode/1
    (elixir 1.15.4) lib/enum.ex:1693: Enum."-map/2-lists^map/1-1-"/2
    (elixir 1.15.4) lib/enum.ex:1693: Enum."-map/2-lists^map/1-1-"/2
    (pg_protocol 0.1.0) lib/pg_protocol/encoder.ex:9: PgProtocol.Encoder.List.encode/1
Last message: {:tcp, #Port<0.29>, <<66, 0, 0, 0, 52, 0, 115, 51, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 30, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 19, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 6, 112, 117, 98, 108, ...>>}

[alco]

@davidmartos96 I haven't tested it after the redesign of internal representation of enums done in #804

Thanks for sharing your log! I'll run some tests tomorrow and will get back to you.

[alco]

@davidmartos96 There was a bug in the implementation. Please try again using the latest commit here.

[magnetised]

Nice.