From 7da229e343f2fa698c323b5155097f88b158391f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Edgar=20Ram=C3=ADrez=20Mondrag=C3=B3n?=
 <edgarrm358@gmail.com>
Date: Mon, 27 Jan 2025 17:21:29 -0600
Subject: [PATCH] feat: Support `x-sql-datatype` for integer types

---
 README.md                           | 25 +++++++++++++++++++++++++
 target_postgres/connector.py        |  3 +++
 target_postgres/tests/test_types.py |  8 ++++++++
 3 files changed, 36 insertions(+)

diff --git a/README.md b/README.md
index 5cbb6e2..9f1b413 100644
--- a/README.md
+++ b/README.md
@@ -302,6 +302,18 @@ If a column has multiple jsonschema types, the following order is using to order
 - BOOLEAN
 - NOTYPE
 
+### `x-sql-datatype` extension
+
+This target supports the [`x-sql-datatype` extension](https://sdk.meltano.com/en/latest/guides/sql-target.html#use-the-x-sql-datatype-json-schema-extension) to the JSON schema. This extension allows you to specify the Postgres data type that should be used for a given field. This can be useful when the default mapping is not what you want.
+
+<!-- insert a table with the mapping -->
+
+| `x-sql-datatype` | Postgres | Description                                                        |
+| :--------------- | :------- | :----------------------------------------------------------------- |
+| smallint         | smallint | small-range integer (-32768 to +32767)                             |
+| integer          | integer  | typical choice for integer (-2147483648 to +2147483647)            |
+| bigint           | bigint   | large-range integer (-9223372036854775808 to +9223372036854775807) |
+
 ### Using the Singer catalog to narrow down the Postgres data types
 
 You can use [Singer catalog's schema](https://github.com/singer-io/getting-started/blob/master/docs/DISCOVERY_MODE.md#schemas) to override the data types coming from the tap. The easiest way to do this is to use Meltano and its [`schema` setting](https://docs.meltano.com/concepts/plugins/#schema-extra) for the tap:
@@ -320,6 +332,19 @@ plugins:
           maximum: 1000
 ```
 
+Or to use the `x-sql-datatype` extension:
+
+```yaml
+# meltano.yml
+plugins:
+  extractors:
+  - name: tap-my-tap
+    schema:
+      some_stream_id:
+        my_column:
+          x-sql-datatype: smallint
+```
+
 ## Content Encoding Support
 
 Json Schema supports the [`contentEncoding` keyword](https://datatracker.ietf.org/doc/html/rfc4648#section-8), which can be used to specify the encoding of input string types.
diff --git a/target_postgres/connector.py b/target_postgres/connector.py
index 228a3f0..75bfa6f 100644
--- a/target_postgres/connector.py
+++ b/target_postgres/connector.py
@@ -310,6 +310,9 @@ def jsonschema_to_sql(self) -> JSONSchemaToSQL:
         to_sql.register_format_handler("hostname", TEXT)
         to_sql.register_format_handler("ipv4", TEXT)
         to_sql.register_format_handler("ipv6", TEXT)
+        to_sql.register_sql_datatype_handler("smallint", SMALLINT)
+        to_sql.register_sql_datatype_handler("integer", INTEGER)
+        to_sql.register_sql_datatype_handler("bigint", BIGINT)
         return to_sql
 
     def to_sql_type(self, jsonschema_type: dict) -> sa.types.TypeEngine:
diff --git a/target_postgres/tests/test_types.py b/target_postgres/tests/test_types.py
index 065ca32..436aa8d 100644
--- a/target_postgres/tests/test_types.py
+++ b/target_postgres/tests/test_types.py
@@ -93,6 +93,14 @@ def test_datetime_string(self, to_postgres: JSONSchemaToPostgres):
                 BIGINT,
                 id="bigint",
             ),
+            pytest.param(
+                {
+                    "type": "integer",
+                    "x-sql-datatype": "smallint",
+                },
+                SMALLINT,
+                id="x-sql-datatype-smallint",
+            ),
         ],
     )
     def test_integers(