|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Objectorg.apache.hadoop.hbase.regionserver.NonLazyKeyValueScanner
@InterfaceAudience.Private public abstract class NonLazyKeyValueScanner
A "non-lazy" scanner which always does a real seek operation. Most scanners are inherited from this class.
Constructor Summary | |
---|---|
NonLazyKeyValueScanner()
|
Method Summary | |
---|---|
static boolean |
doRealSeek(KeyValueScanner scanner,
KeyValue kv,
boolean forward)
|
void |
enforceSeek()
Does the real seek operation in case it was skipped by seekToRowCol(KeyValue, boolean) (TODO: Whats this?). |
byte[] |
getNextIndexedKey()
|
boolean |
isFileScanner()
|
boolean |
realSeekDone()
We optimize our store scanners by checking the most recent store file first, so we sometimes pretend we have done a seek but delay it until the store scanner bubbles up to the top of the key-value heap. |
boolean |
requestSeek(KeyValue kv,
boolean forward,
boolean useBloom)
Similar to KeyValueScanner.seek(org.apache.hadoop.hbase.KeyValue) (or KeyValueScanner.reseek(org.apache.hadoop.hbase.KeyValue) if forward is true) but only
does a seek operation after checking that it is really necessary for the
row/column combination specified by the kv parameter. |
boolean |
shouldUseScanner(Scan scan,
SortedSet<byte[]> columns,
long oldestUnexpiredTS)
Allows to filter out scanners (both StoreFile and memstore) that we don't want to use based on criteria such as Bloom filters and timestamp ranges. |
Methods inherited from class java.lang.Object |
---|
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
Methods inherited from interface org.apache.hadoop.hbase.regionserver.KeyValueScanner |
---|
backwardSeek, close, getSequenceID, next, peek, reseek, seek, seekToLastRow, seekToPreviousRow |
Constructor Detail |
---|
public NonLazyKeyValueScanner()
Method Detail |
---|
public boolean requestSeek(KeyValue kv, boolean forward, boolean useBloom) throws IOException
KeyValueScanner
KeyValueScanner.seek(org.apache.hadoop.hbase.KeyValue)
(or KeyValueScanner.reseek(org.apache.hadoop.hbase.KeyValue)
if forward is true) but only
does a seek operation after checking that it is really necessary for the
row/column combination specified by the kv parameter. This function was
added to avoid unnecessary disk seeks by checking row-column Bloom filters
before a seek on multi-column get/scan queries, and to optimize by looking
up more recent files first.
requestSeek
in interface KeyValueScanner
forward
- do a forward-only "reseek" instead of a random-access seekuseBloom
- whether to enable multi-column Bloom filter optimization
IOException
public boolean realSeekDone()
KeyValueScanner
realSeekDone
in interface KeyValueScanner
public void enforceSeek() throws IOException
KeyValueScanner
KeyValueScanner.realSeekDone()
first.
enforceSeek
in interface KeyValueScanner
IOException
public static boolean doRealSeek(KeyValueScanner scanner, KeyValue kv, boolean forward) throws IOException
IOException
public boolean shouldUseScanner(Scan scan, SortedSet<byte[]> columns, long oldestUnexpiredTS)
KeyValueScanner
shouldUseScanner
in interface KeyValueScanner
scan
- the scan that we are selecting scanners forcolumns
- the set of columns in the current column family, or null if
not specified by the scanoldestUnexpiredTS
- the oldest timestamp we are interested in for
this query, based on TTL
public boolean isFileScanner()
isFileScanner
in interface KeyValueScanner
public byte[] getNextIndexedKey()
getNextIndexedKey
in interface KeyValueScanner
|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |