[GitHub] [hudi] danny0405 commented on a diff in pull request #7355: [HUDI-5308] Hive3 query returns null when the where clause has a partition field

["danny0405 (via GitHub)" <gi...@apache.org>]

danny0405 commented on code in PR #7355:
URL: https://github.com/apache/hudi/pull/7355#discussion_r1183441571


##########
hudi-hadoop-mr/src/main/java/org/apache/hudi/hadoop/utils/HoodieRealtimeInputFormatUtils.java:
##########
@@ -84,6 +84,18 @@ private static Configuration addProjectionField(Configuration conf, String field
     return conf;
   }
 
+  public static void addProjectionField(Configuration conf, Option<String[]> fieldName) {
+    if (fieldName.isPresent()) {
+      List<String> columnNameList = Arrays.stream(conf.get(serdeConstants.LIST_COLUMNS).split(",")).collect(Collectors.toList());
+      Arrays.stream(fieldName.get()).forEach(field -> {

Review Comment:
   Can you explain what this value represents
   
   ```java
   conf.get(serdeConstants.LIST_COLUMNS
   ```



##########
hudi-hadoop-mr/src/main/java/org/apache/hudi/hadoop/HoodieParquetInputFormat.java:
##########
@@ -84,6 +90,13 @@ public RecordReader<NullWritable, ArrayWritable> getRecordReader(final InputSpli
       LOG.debug("EMPLOYING DEFAULT RECORD READER - " + split);
     }
 
+    Path inputPath = ((FileSplit)split).getPath();
+    FileSystem fs =  inputPath.getFileSystem(job);
+    Option<Path> tablePath = TablePathUtils.getTablePath(fs, inputPath);
+    HoodieTableMetaClient metaClient = HoodieTableMetaClient.builder().setConf(job).setBasePath(tablePath.get().toString()).build();

Review Comment:
   Does this code work here:
   
   ```java
   HoodieTableMetaClient metaClient = HoodieTableMetaClient.builder().setConf(jobConf).setBasePath(realtimeSplit.getBasePath()).build();
   ```



##########
hudi-hadoop-mr/src/main/java/org/apache/hudi/hadoop/utils/HoodieRealtimeRecordReaderUtils.java:
##########
@@ -257,10 +257,8 @@ public static List<String> orderFields(String fieldNameCsv, String fieldOrderCsv
     String[] fieldOrdersWithDups = fieldOrderCsv.isEmpty() ? new String[0] : fieldOrderCsv.split(",");
     Set<String> fieldOrdersSet = new LinkedHashSet<>(Arrays.asList(fieldOrdersWithDups));
     String[] fieldOrders = fieldOrdersSet.toArray(new String[0]);
-    List<String> fieldNames = fieldNameCsv.isEmpty() ? new ArrayList<>() : Arrays.stream(fieldNameCsv.split(","))
-        .filter(fn -> !partitioningFields.contains(fn)).collect(Collectors.toList());
+    List<String> fieldNames = fieldNameCsv.isEmpty() ? new ArrayList<>() : Arrays.stream(fieldNameCsv.split(",")).collect(Collectors.toList());

Review Comment:
   Does the removal of partition fiels affect the behavior of Hive 2.x?



-- 
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.

To unsubscribe, e-mail: commits-unsubscribe@hudi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org