lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-4003] Try to read all the log file to parse schema (#5473)

Browse Source
This commit is contained in:
兰缘小妖
2022-05-11 06:45:53 +08:00
committed by GitHub
parent 6fd21d0f10
commit 4258a71517
1 changed files with 19 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
hudi-common/src/main/java/org/apache/hudi/common/table/TableSchemaResolver.java
View File

@@ -61,6 +61,7 @@ import org.apache.parquet.schema.MessageType;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Iterator;
import java.util.List;
import static org.apache.hudi.avro.AvroSchemaUtils.appendFieldsToSchema;
@@ -98,8 +99,8 @@ public class TableSchemaResolver {
// For COW table, the file has data written must be in parquet or orc format currently.
if (instantAndCommitMetadata.isPresent()) {
HoodieCommitMetadata commitMetadata = instantAndCommitMetadata.get().getRight();
String filePath = commitMetadata.getFileIdAndFullPaths(metaClient.getBasePath()).values().stream().findAny().get();
return readSchemaFromBaseFile(filePath);
Iterator<String> filePaths = commitMetadata.getFileIdAndFullPaths(metaClient.getBasePath()).values().iterator();
return fetchSchemaFromFiles(filePaths);
} else {
throw new IllegalArgumentException("Could not find any data file written for commit, "
+ "so could not get schema for table " + metaClient.getBasePath());
@@ -109,13 +110,8 @@ public class TableSchemaResolver {
// Determine the file format based on the file name, and then extract schema from it.
if (instantAndCommitMetadata.isPresent()) {
HoodieCommitMetadata commitMetadata = instantAndCommitMetadata.get().getRight();
String filePath = commitMetadata.getFileIdAndFullPaths(metaClient.getBasePath()).values().stream().findAny().get();
if (filePath.contains(HoodieFileFormat.HOODIE_LOG.getFileExtension())) {
// this is a log file
return readSchemaFromLogFile(new Path(filePath));
} else {
return readSchemaFromBaseFile(filePath);
}
Iterator<String> filePaths = commitMetadata.getFileIdAndFullPaths(metaClient.getBasePath()).values().iterator();
return fetchSchemaFromFiles(filePaths);
} else {
throw new IllegalArgumentException("Could not find any data file written for commit, "
+ "so could not get schema for table " + metaClient.getBasePath());
@@ -129,6 +125,20 @@ public class TableSchemaResolver {
}
}
private MessageType fetchSchemaFromFiles(Iterator<String> filePaths) throws IOException {
MessageType type = null;
while (filePaths.hasNext() && type == null) {
String filePath = filePaths.next();
if (filePath.contains(HoodieFileFormat.HOODIE_LOG.getFileExtension())) {
// this is a log file
type = readSchemaFromLogFile(new Path(filePath));
} else {
type = readSchemaFromBaseFile(filePath);
}
}
return type;
}
private MessageType readSchemaFromBaseFile(String filePath) throws IOException {
if (filePath.contains(HoodieFileFormat.PARQUET.getFileExtension())) {
// this is a parquet file