From 64412957ff9b741b3b57bce19124f204caf5e117 Mon Sep 17 00:00:00 2001
From: Rafael Fernandez <github@rafaelfernandez.dev>
Date: Sat, 28 Mar 2026 19:31:42 +0100
Subject: [PATCH] [SPARK-53313] Add JDBC Reader and Writer support

---
 crates/connect/src/readwriter.rs | 55 ++++++++++++++++++++++++++++++++
 1 file changed, 55 insertions(+)

diff --git a/crates/connect/src/readwriter.rs b/crates/connect/src/readwriter.rs
index 5d1a78e..1851191 100644
--- a/crates/connect/src/readwriter.rs
+++ b/crates/connect/src/readwriter.rs
@@ -382,6 +382,47 @@ define_file_options! {
     }
 }
 
+define_file_options! {
+    /// Options for configuring JDBC data source reads and writes.
+    ///
+    /// # Options
+    ///
+    /// - `url`: JDBC connection URL (e.g., `jdbc:postgresql://host:port/database`).
+    /// - `dbtable`: The JDBC table to read from or write to. Can also be a subquery in parentheses.
+    /// - `driver`: JDBC driver class name (e.g., `org.postgresql.Driver`).
+    /// - `user`: Database user name for authentication.
+    /// - `password`: Database password for authentication.
+    /// - `query`: A SQL query to use as the data source (alternative to `dbtable`).
+    /// - `partition_column`: Column used for partitioning reads.
+    /// - `lower_bound`: Lower bound of the partition column for parallel reads.
+    /// - `upper_bound`: Upper bound of the partition column for parallel reads.
+    /// - `num_partitions`: Number of partitions for parallel reads/writes.
+    /// - `fetch_size`: Number of rows to fetch per round trip.
+    /// - `batch_size`: Number of rows to insert per batch during writes.
+    /// - `isolation_level`: Transaction isolation level (e.g., `READ_COMMITTED`).
+    /// - `truncate`: Whether to truncate the table before writing (instead of dropping and recreating).
+    /// - `create_table_options`: Options appended to the CREATE TABLE statement.
+    /// - `create_table_column_types`: Column data types to use when creating the table.
+    pub struct JdbcOptions {
+        url : String, camel_case = "url"
+        dbtable : String, camel_case = "dbtable"
+        driver : String, camel_case = "driver"
+        user : String, camel_case = "user"
+        password : String, camel_case = "password"
+        query : String, camel_case = "query"
+        partition_column : String, camel_case = "partitionColumn"
+        lower_bound : String, camel_case = "lowerBound"
+        upper_bound : String, camel_case = "upperBound"
+        num_partitions : i32, camel_case = "numPartitions"
+        fetch_size : i32, camel_case = "fetchSize"
+        batch_size : i32, camel_case = "batchSize"
+        isolation_level : String, camel_case = "isolationLevel"
+        truncate : bool, camel_case = "truncate"
+        create_table_options : String, camel_case = "createTableOptions"
+        create_table_column_types : String, camel_case = "createTableColumnTypes"
+    }
+}
+
 /// DataFrameReader represents the entrypoint to create a DataFrame
 /// from a specific file format.
 #[derive(Clone, Debug)]
@@ -558,6 +599,13 @@ impl DataFrameReader {
         self.read_options.extend(config.to_options());
         self.load(paths)
     }
+
+    /// Reads data from a JDBC data source with the specified options.
+    pub fn jdbc<C: ConfigOpts>(mut self, config: C) -> Result<DataFrame, SparkError> {
+        self.format = Some("jdbc".to_string());
+        self.read_options.extend(config.to_options());
+        self.load(Vec::<&str>::new())
+    }
 }
 
 /// DataFrameWriter provides the ability to output a [DataFrame]
@@ -764,6 +812,13 @@ impl DataFrameWriter {
         self.write_options.extend(config.to_options());
         self.save(path).await
     }
+
+    /// Writes data to a JDBC data source with the specified options.
+    pub async fn jdbc<C: ConfigOpts>(mut self, config: C) -> Result<(), SparkError> {
+        self.format = Some("jdbc".to_string());
+        self.write_options.extend(config.to_options());
+        self.save("").await
+    }
 }
 
 pub struct DataFrameWriterV2 {