sunchao commented on a change in pull request #925:
URL: https://github.com/apache/parquet-mr/pull/925#discussion_r704828659
##########
File path:
parquet-hadoop/src/main/java/org/apache/parquet/format/converter/ParquetMetadataConverter.java
##########
@@ -1226,14 +1239,31 @@ public ParquetMetadata readParquetMetadata(InputStream
from) throws IOException
static FileMetaData filterFileMetaDataByMidpoint(FileMetaData metaData,
RangeMetadataFilter filter) {
List<RowGroup> rowGroups = metaData.getRow_groups();
List<RowGroup> newRowGroups = new ArrayList<RowGroup>();
+ long preStartIndex = 0;
+ long preCompressedSize = 0;
+ boolean firstColumnWithMetadata = true;
+ if (rowGroups != null && rowGroups.size() > 0) {
+ firstColumnWithMetadata =
rowGroups.get(0).getColumns().get(0).isSetMeta_data();
+ }
for (RowGroup rowGroup : rowGroups) {
long totalSize = 0;
long startIndex;
+ ColumnChunk columnChunk = rowGroup.getColumns().get(0);
+ if (firstColumnWithMetadata) {
+ startIndex = getOffset(columnChunk);
+ } else {
+ assert rowGroup.isSetFile_offset();
+ assert rowGroup.isSetTotal_compressed_size();
- if (rowGroup.isSetFile_offset()) {
+ //the file_offset of first block always holds the truth, while other
blocks don't :
Review comment:
I think this is no longer true with the issue we found in parquet-cpp.
--
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]
