Github user KanakaKumar commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/2804#discussion_r231061128
--- Diff:
store/sdk/src/main/java/org/apache/carbondata/sdk/file/CarbonSchemaReader.java
---
@@ -61,14 +65,122 @@ public static Schema readSchemaInSchemaFile(String
schemaFilePath) throws IOExce
return new Schema(schemaList);
}
+ /**
+ * get carbondata/carbonindex file in path
+ *
+ * @param path carbon file path
+ * @return CarbonFile array
+ */
+ private static CarbonFile[] getCarbonFile(String path, final String
extension)
+ throws IOException {
+ String dataFilePath = path;
+ if (!(dataFilePath.contains(extension))) {
+ CarbonFile[] carbonFiles = FileFactory
+ .getCarbonFile(path)
+ .listFiles(new CarbonFileFilter() {
+ @Override
+ public boolean accept(CarbonFile file) {
+ if (file == null) {
+ return false;
+ }
+ return file.getName().endsWith(extension);
+ }
+ });
+ if (carbonFiles == null || carbonFiles.length < 1) {
+ throw new IOException("Carbon file not exists.");
+ }
+ return carbonFiles;
+ }
+ return null;
+ }
+
+ /**
+ * read schema from path,
+ * path can be folder path, carbonindex file path, and carbondata file
path
+ * and will not check all files schema
+ *
+ * @param path file/folder path
+ * @return schema
+ * @throws IOException
+ */
+ public static Schema readSchema(String path) throws IOException {
+ return readSchema(path, false);
+ }
+
+ /**
+ * read schema from path,
+ * path can be folder path, carbonindex file path, and carbondata file
path
+ * and user can decide whether check all files schema
+ *
+ * @param path file/folder path
+ * @param validateSchema whether check all files schema
+ * @return schema
+ * @throws IOException
+ */
+ public static Schema readSchema(String path, boolean validateSchema)
throws IOException {
+ if (path.endsWith(INDEX_FILE_EXT)) {
+ return readSchemaFromIndexFile(path);
+ } else if (path.endsWith(CARBON_DATA_EXT)) {
+ return readSchemaFromDataFile(path);
+ } else if (validateSchema) {
+ CarbonFile[] carbonIndexFiles = getCarbonFile(path, INDEX_FILE_EXT);
+ Schema schema;
+ if (carbonIndexFiles != null && carbonIndexFiles.length != 0) {
+ schema =
readSchemaFromIndexFile(carbonIndexFiles[0].getAbsolutePath());
+ for (int i = 1; i < carbonIndexFiles.length; i++) {
+ Schema schema2 =
readSchemaFromIndexFile(carbonIndexFiles[i].getAbsolutePath());
+ if (schema != schema2) {
+ throw new CarbonDataLoadingException("Schema is different
between different files.");
+ }
+ }
+ CarbonFile[] carbonDataFiles = getCarbonFile(path,
CARBON_DATA_EXT);
+ for (int i = 0; i < carbonDataFiles.length; i++) {
+ Schema schema2 =
readSchemaFromDataFile(carbonDataFiles[i].getAbsolutePath());
+ if (!schema.equals(schema2)) {
+ throw new CarbonDataLoadingException("Schema is different
between different files.");
+ }
+ }
+ return schema;
+ } else {
+ throw new CarbonDataLoadingException("No carbonindex file in this
path.");
+ }
+ } else {
+ String indexFilePath = getCarbonFile(path,
INDEX_FILE_EXT)[0].getAbsolutePath();
+ if (indexFilePath != null) {
--- End diff --
I think this null check is not required. Is there any chance the absolute
path can be null ?
---
