OpenOpened commented on a change in pull request #1200: [HUDI-514] A schema
provider to get metadata through Jdbc
URL: https://github.com/apache/incubator-hudi/pull/1200#discussion_r378134572
##########
File path:
hudi-utilities/src/main/java/org/apache/hudi/utilities/UtilHelpers.java
##########
@@ -235,4 +248,57 @@ public static TypedProperties readConfig(InputStream in)
throws IOException {
defaults.load(in);
return defaults;
}
+
+ /***
+ * call spark function get the schema through jdbc.
+ * The code logic implementation refers to spark 2.4.x and spark 3.x.
+ * @param options
+ * @return
+ * @throws Exception
+ */
+ public static Schema getJDBCSchema(Map<String, String> options) throws
Exception {
+ scala.collection.immutable.Map<String, String> ioptions =
toScalaImmutableMap(options);
+ JDBCOptions jdbcOptions = new JDBCOptions(ioptions);
+ Connection conn = JdbcUtils.createConnectionFactory(jdbcOptions).apply();
+ String url = jdbcOptions.url();
+ String table = jdbcOptions.tableOrQuery();
+ JdbcOptionsInWrite jdbcOptionsInWrite = new JdbcOptionsInWrite(ioptions);
+ boolean tableExists = JdbcUtils.tableExists(conn, jdbcOptionsInWrite);
+
+ if (tableExists) {
+ JdbcDialect dialect = JdbcDialects.get(url);
+ try (PreparedStatement statement =
conn.prepareStatement(dialect.getSchemaQuery(table))) {
+ statement.setQueryTimeout(Integer.parseInt(options.get("timeout")));
+ try (ResultSet rs = statement.executeQuery()) {
+ StructType structType;
+ if (Boolean.parseBoolean(ioptions.get("nullable").get())) {
+ structType = JdbcUtils.getSchema(rs, dialect, true);
+ } else {
+ structType = JdbcUtils.getSchema(rs, dialect, false);
+ }
+ return AvroConversionUtils.convertStructTypeToAvroSchema(structType,
table, "hoodie." + table);
+ }
+ }
+ } else {
+ throw new HoodieException(String.format("%s table does not exists!",
table));
+ }
+ }
+
+ /**
+ * Replace java map with scala immutable map.
+ * refers:
https://stackoverflow.com/questions/11903167/convert-java-util-hashmap-to-scala-collection-immutable-map-in-java/11903737#11903737
Review comment:
@vinothchandar @leesf I reimplemented the relevant logic using the second
approach.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
