Github user sounakr commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/2345#discussion_r191328391
--- Diff:
store/sdk/src/main/java/org/apache/carbondata/sdk/file/CarbonReader.java ---
@@ -101,6 +108,45 @@ public static CarbonReaderBuilder builder(String
tablePath, String tableName) {
return reader.readSchema();
}
+ /**
+ * Read carbonindex file and return the schema.
+ * This API is better use when index file is present,
+ * as It is faster compared to readSchemaInDataFile when carbondata file
is big in size
+ *
+ * @param indexFilePath complete path including index file name
+ * @return List<ColumnSchema> from the index file.
+ * @throws IOException
+ */
+ public static List<ColumnSchema> readSchemaInIndexFile(String
indexFilePath) throws IOException {
--- End diff --
In case we have a performance result for comparison between reading from
CarbonIndexFile and CarbonDataFile, and if reading from CarbonIndexFile Shows
better output, then better to read from Index File in all places, i.e. whenever
internally we infer schema.
---
