🐛 BUG-4307: Fix incorrect handling of LONG RAW data type (#7125)

Author: alexandertsukanov

airbyte-config/init/src/main/resources/config/STANDARD_SOURCE_DEFINITION/b39a7370-74c3-45a6-ac3a-380d48520a83.json

@@ -2,6 +2,6 @@
  "sourceDefinitionId": "b39a7370-74c3-45a6-ac3a-380d48520a83",
  "name": "Oracle DB",
  "dockerRepository": "airbyte/source-oracle",
- "dockerImageTag": "0.3.7",
+ "dockerImageTag": "0.3.8",
  "documentationUrl": "https://docs.airbyte.io/integrations/sources/oracle"
 }

airbyte-config/init/src/main/resources/seed/source_definitions.yaml

@@ -332,7 +332,7 @@
 - sourceDefinitionId: b39a7370-74c3-45a6-ac3a-380d48520a83
  name: Oracle DB
  dockerRepository: airbyte/source-oracle
- dockerImageTag: 0.3.7
+ dockerImageTag: 0.3.8
  documentationUrl: https://docs.airbyte.io/integrations/sources/oracle
  sourceType: database
 - sourceDefinitionId: c8630570-086d-4a40-99ae-ea5b18673071

airbyte-integrations/connectors/source-oracle/BOOTSTRAP.md

@@ -0,0 +1,10 @@
+# Oracle Source
+The Oracle source connector allows syncing the data from the Oracle DB. The current source connector supports Oracle 11g or above.
+The connector uses *ojdbc8* driver underneath to establish the connection. The Oracle source does not alter the schema present in your database.
+
+### Important details
+Connector works with `useFetchSizeWithLongColumn=true` property, which required to select the data from `LONG` or `LONG RAW` type columns.
+Oracle recommends avoiding LONG and LONG RAW columns. Use LOB instead. They are included in Oracle only for legacy reasons.
+THIS IS A THIN ONLY PROPERTY. IT SHOULD NOT BE USED WITH ANY OTHER DRIVERS.
+
+See [this](https://docs.airbyte.io/integrations/sources/oracle) link for the nuances about the connector.

airbyte-integrations/connectors/source-oracle/Dockerfile

@@ -8,5 +8,5 @@ ENV TZ UTC
 COPY build/distributions/${APPLICATION}*.tar ${APPLICATION}.tar
 RUN tar xf ${APPLICATION}.tar --strip-components=1
 
-LABEL io.airbyte.version=0.3.7
+LABEL io.airbyte.version=0.3.8
 LABEL io.airbyte.name=airbyte/source-oracle

airbyte-integrations/connectors/source-oracle/README.md

@@ -0,0 +1,33 @@
+# Oracle Source
+
+## Documentation
+This is the repository for the Oracle only source connector in Java.
+For information about how to use this connector within Airbyte, see [User Documentation](https://docs.airbyte.io/integrations/sources/oracle)
+
+## Local development
+
+#### Building via Gradle
+From the Airbyte repository root, run:
+```
+./gradlew :airbyte-integrations:connectors:source-oracle:build
+```
+
+### Locally running the connector docker image
+
+#### Build
+Build the connector image via Gradle:
+```
+./gradlew :airbyte-integrations:connectors:source-oracle:airbyteDocker
+```
+When building via Gradle, the docker image name and tag, respectively, are the values of the `io.airbyte.name` and `io.airbyte.version` `LABEL`s in
+the Dockerfile.
+
+## Testing
+We use `JUnit` for Java tests.
+
+### Test Configuration
+#### Acceptance Tests
+To run acceptance and custom integration tests:
+```
+./gradlew :airbyte-integrations:connectors:source-oracle:integrationTest
+```

airbyte-integrations/connectors/source-oracle/src/main/java/io/airbyte/integrations/source/oracle/OracleSource.java

@@ -55,6 +55,15 @@ public static Source sshWrappedSource() {
  public JsonNode toDatabaseConfig(final JsonNode config) {
  final List<String> additionalParameters = new ArrayList<>();
 
+ /*
+ The property useFetchSizeWithLongColumn required to select LONG or LONG RAW columns.
+ Oracle recommends avoiding LONG and LONG RAW columns. Use LOB instead. They are included in Oracle only for legacy reasons.
+ THIS IS A THIN ONLY PROPERTY. IT SHOULD NOT BE USED WITH ANY OTHER DRIVERS.
+ See https://docs.oracle.com/cd/E11882_01/appdev.112/e13995/oracle/jdbc/OracleDriver.html
+ https://docs.oracle.com/cd/B19306_01/java.102/b14355/jstreams.htm#i1014085
+ */
+ additionalParameters.add("oracle.jdbc.useFetchSizeWithLongColumn=true");
+
  final Protocol protocol = config.has("encryption")
  ? obtainConnectionProtocol(config.get("encryption"), additionalParameters)
  : Protocol.TCP;

airbyte-integrations/connectors/source-oracle/src/test-integration/java/io/airbyte/integrations/source/oracle/OracleSourceDatatypeTest.java

@@ -174,7 +174,7 @@ protected void initTests() {
  TestDataHolder.builder()
  .sourceType("BINARY_DOUBLE")
  .airbyteType(JsonSchemaPrimitive.NUMBER)
- .addInsertValues("126.45d", "2.22507485850720E-308", "TO_BINARY_DOUBLE('1.79769313486231E+308')", "BINARY_DOUBLE_INFINITY")
+ .addInsertValues("126.45d", "2.22507485850720E-308", "1.79769313486231E+308d", "BINARY_DOUBLE_INFINITY")
  .addExpectedValues("126.45", "0.0", "1.79769313486231E308", "Infinity")
  .build());
 
@@ -266,8 +266,8 @@ protected void initTests() {
  .sourceType("LONG")
  .airbyteType(JsonSchemaPrimitive.STRING)
  .fullSourceDataType("LONG RAW")
- // @TODO stream fails when reading data back
- // .addInsertValues("utl_raw.cast_to_raw('some content here')", "null")
+  .addInsertValues("utl_raw.cast_to_raw('some content here')", "null")
+ .addExpectedValues("c29tZSBjb250ZW50IGhlcmU=", null)
  .build());
 
  addDataTypeTestData(

docs/integrations/sources/oracle.md

@@ -132,6 +132,7 @@ Airbite has the ability to connect to the Oracle source with 3 network connectiv
 
 | Version | Date | Pull Request | Subject |
 | :--- | :--- | :--- | :--- |
+| 0.3.8 | 2021-10-13 | [7125](https://github.com/airbytehq/airbyte/pull/7125) | Fix incorrect handling of LONG RAW data type |
 | 0.3.7 | 2021-10-01 | [6616](https://github.com/airbytehq/airbyte/pull/6616) | Added network encryption options |
 | 0.3.6 | 2021-09-30 | [6585](https://github.com/airbytehq/airbyte/pull/6585) | Improved SSH Tunnel key generation steps |
 | 0.3.5 | 2021-09-22 | [6356](https://github.com/airbytehq/airbyte/pull/6356) | Added option to connect to DB via SSH. |