org.apache.spark.sql.execution.datasources
HadoopFsRelation
Related Doc:
package datasources
A FileIndex that can enumerate the locations of all the files that comprise this relation.
The schema of the columns (if any) that are used to partition the relation
The schema of any remaining columns. Note that if any partition columns are present in the actual data files as well, they are preserved.
Describes the bucketing (hash-partitioning of the files by some column values).
A file format that can be used to read and write the data in files.
Configuration used when reading / writing data.
Describes the bucketing (hash-partitioning of the files by some column values).
The schema of any remaining columns.
The schema of any remaining columns. Note that if any partition columns are present in the actual data files as well, they are preserved.
A file format that can be used to read and write the data in files.
Returns the list of files that will be read when scanning this relation.
Returns the list of files that will be read when scanning this relation.
A FileIndex that can enumerate the locations of all the files that comprise this relation.
Whether does it need to convert the objects in Row to internal representation, for example: java.lang.String to UTF8String java.lang.Decimal to Decimal
Whether does it need to convert the objects in Row to internal representation, for example: java.lang.String to UTF8String java.lang.Decimal to Decimal
If needConversion is false, buildScan() should return an RDD of InternalRow
1.4.0
The internal representation is not stable across releases and thus data sources outside of Spark SQL should leave this as true.
Configuration used when reading / writing data.
The schema of the columns (if any) that are used to partition the relation
Returns an estimated size of this relation in bytes.
Returns an estimated size of this relation in bytes. This information is used by the planner to decide when it is safe to broadcast a relation and can be overridden by sources that know the size ahead of time. By default, the system will assume that tables are too large to broadcast. This method will be called multiple times during query planning and thus should not perform expensive operations for each invocation.
1.3.0
It is always better to overestimate size than underestimate, because underestimation could lead to execution plans that are suboptimal (i.e. broadcasting a very large table).
Returns the list of Filters that this datasource may not be able to handle.
Returns the list of Filters that this datasource may not be able to handle. These returned Filters will be evaluated by Spark SQL after data is output by a scan. By default, this function will return all filters, as it is always safe to double evaluate a Filter. However, specific implementations can override this function to avoid double filtering when they are capable of processing a filter internally.
1.6.0
Acts as a container for all of the metadata required to read from a datasource. All discovery, resolution and merging logic for schemas and partitions has been removed.
A FileIndex that can enumerate the locations of all the files that comprise this relation.
The schema of the columns (if any) that are used to partition the relation
The schema of any remaining columns. Note that if any partition columns are present in the actual data files as well, they are preserved.
Describes the bucketing (hash-partitioning of the files by some column values).
A file format that can be used to read and write the data in files.
Configuration used when reading / writing data.