wgtmac commented on code in PR #1273:
URL: https://github.com/apache/parquet-mr/pull/1273#discussion_r1557919070
##########
parquet-hadoop/src/main/java/org/apache/parquet/hadoop/rewrite/ParquetRewriter.java:
##########
@@ -904,4 +953,171 @@ public byte[] getDictPageAAD() {
return this.dictPageAAD;
}
}
+
+ private static class RightColumnWriter {
+ private final Queue<TransParquetFileReader> inputFiles;
+ private final ParquetRewriter parquetRewriter;
+ private final ParquetFileWriter writer;
+ private final MessageType schema;
+ private final Map<ColumnPath, ColumnDescriptor> descriptorsMap;
+ private final Map<ColumnDescriptor, ColumnReader> colReaders = new
HashMap<>();
+ private final Map<ColumnDescriptor, ColumnChunkPageWriteStore> cPageStores
= new HashMap<>();
+ private final Map<ColumnDescriptor, ColumnWriteStore> cStores = new
HashMap<>();
+ private final Map<ColumnDescriptor, ColumnWriter> cWriters = new
HashMap<>();
+ private int rowGroupIdxIn = 0;
+ private int rowGroupIdxOut = 0;
+ private int writtenFromBlock = 0;
+
+ public RightColumnWriter(Queue<TransParquetFileReader> inputFiles,
ParquetRewriter parquetRewriter)
+ throws IOException {
+ this.inputFiles = inputFiles;
+ this.parquetRewriter = parquetRewriter;
+ this.writer = parquetRewriter.writer;
+ this.schema =
inputFiles.peek().getFooter().getFileMetaData().getSchema();
+ this.descriptorsMap = this.schema.getColumns().stream()
+ .collect(Collectors.toMap(x -> ColumnPath.get(x.getPath()), x -> x));
+ initReaders();
+ initWriters();
+ }
+
+ public void writeRows(int rowGroupIdx, long rowsToWrite) throws
IOException {
+ if (rowGroupIdxIn != rowGroupIdx) {
Review Comment:
Yes, that sounds good.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
