chenjunjiedada commented on a change in pull request #830: Support name mapping
resolution for parquet
URL: https://github.com/apache/incubator-iceberg/pull/830#discussion_r390070742
##########
File path: parquet/src/main/java/org/apache/iceberg/parquet/ReadConf.java
##########
@@ -72,29 +74,31 @@
this.options = options;
this.reader = newReader(file, options);
MessageType fileSchema = reader.getFileMetaData().getSchema();
-
boolean hasIds = ParquetSchemaUtil.hasIds(fileSchema);
- MessageType typeWithIds = hasIds ? fileSchema :
ParquetSchemaUtil.addFallbackIds(fileSchema);
this.projection = hasIds ?
- ParquetSchemaUtil.pruneColumns(fileSchema, expectedSchema) :
- ParquetSchemaUtil.pruneColumnsFallback(fileSchema, expectedSchema);
+ ParquetSchemaUtil.pruneColumns(fileSchema, expectedSchema) :
+ ParquetSchemaUtil.pruneColumnsByName(fileSchema, expectedSchema);
Review comment:
Here is a reversed way. I build name mapping based on `fileSchema`, and then
project ids for expected schema according to names. The reason is the metrics
are built from parquet footer so we have to bind the expression to `fileSchema`
instead of expected schema in `ParquetMetricsRowGroupFilter` and
`ParquetDictionaryRowGroupFilter`.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
