com.datastax.driver.core

Class TokenRange

java.lang.Object

com.datastax.driver.core.TokenRange

All Implemented Interfaces:

Comparable<TokenRange>

public final class TokenRange
extends Object
implements Comparable<TokenRange>

A range of tokens on the Cassandra ring.

A range is start-exclusive and end-inclusive. It is empty when start and end are the same token, except if that is the minimum token, in which case the range covers the whole ring (this is consistent with the behavior of CQL range queries).

Note that CQL does not handle wrapping. To query all partitions in a range, see unwrap().

Method Summary

Modifier and Type	Method and Description
int	compareTo(TokenRange other)
boolean	contains(Token token) Checks whether this range contains a given token.
boolean	equals(Object other)
Token	getEnd() Return the end of the range.
Token	getStart() Return the start of the range.
int	hashCode()
boolean	intersects(TokenRange that) Returns whether this range intersects another one.
List<TokenRange>	intersectWith(TokenRange that) Computes the intersection of this range with another one.
boolean	isEmpty() Returns whether this range is empty.
boolean	isWrappedAround() Returns whether this range wraps around the end of the ring.
TokenRange	mergeWith(TokenRange that) Merges this range with another one.
List<TokenRange>	splitEvenly(int numberOfSplits) Splits this range into a number of smaller ranges of equal "size" (referring to the number of tokens, not the actual amount of data).
String	toString()
List<TokenRange>	unwrap() Splits this range into a list of two non-wrapping ranges.

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Method Detail

getStart

public Token getStart()

Return the start of the range.

Returns:

the start of the range (exclusive).

getEnd

public Token getEnd()

Return the end of the range.

Returns:

the end of the range (inclusive).

splitEvenly

public List<TokenRange> splitEvenly(int numberOfSplits)

Splits this range into a number of smaller ranges of equal "size" (referring to the number of tokens, not the actual amount of data).

Splitting an empty range is not permitted. But note that, in edge cases, splitting a range might produce one or more empty ranges.

Parameters:

numberOfSplits - the number of splits to create.

Returns:

the splits.

Throws:

IllegalArgumentException - if the range is empty or if numberOfSplits < 1.

isEmpty

public boolean isEmpty()

Returns whether this range is empty.

A range is empty when start and end are the same token, except if that is the minimum token, in which case the range covers the whole ring (this is consistent with the behavior of CQL range queries).

Returns:

whether the range is empty.

isWrappedAround

public boolean isWrappedAround()

Returns whether this range wraps around the end of the ring.

Returns:

whether this range wraps around.

unwrap

public List<TokenRange> unwrap()

Splits this range into a list of two non-wrapping ranges. This will return the range itself if it is non-wrapping, or two ranges otherwise.

For example:

• ]1,10] unwraps to itself;
• ]10,1] unwraps to ]10,min_token] and ]min_token,1].

This is useful for CQL range queries, which do not handle wrapping:

 
 List<Row> rows = new ArrayList<Row>();
 for (TokenRange subRange : range.unwrap()) {
     ResultSet rs = session.execute("SELECT * FROM mytable WHERE token(pk) > ? and token(pk) <= ?",
                                    subRange.getStart(), subRange.getEnd());
     rows.addAll(rs.all());
 }
 

Returns:

the list of non-wrapping ranges.

intersects

public boolean intersects(TokenRange that)

Returns whether this range intersects another one.

For example:

• ]3,5] intersects ]1,4], ]4,5]...
• ]3,5] does not intersect ]1,2], ]2,3], ]5,7]...

Parameters:

that - the other range.

Returns:

whether they intersect.

intersectWith

public List<TokenRange> intersectWith(TokenRange that)

Computes the intersection of this range with another one.

If either of these ranges overlap the the ring, they are unwrapped and the unwrapped tokens are compared with one another.

This call will fail if the two ranges do not intersect, you must check by calling intersects(TokenRange) beforehand.

Parameters:

that - the other range.

Returns:

the range(s) resulting from the intersection.

Throws:

IllegalArgumentException - if the ranges do not intersect.

contains

public boolean contains(Token token)

Checks whether this range contains a given token.

Parameters:

token - the token to check for.

Returns:

whether this range contains the token, i.e. range.start < token <= range.end.

mergeWith

public TokenRange mergeWith(TokenRange that)

Merges this range with another one.

The two ranges should either intersect or be adjacent; in other words, the merged range should not include tokens that are in neither of the original ranges.

For example:

• merging ]3,5] with ]4,7] produces ]3,7];
• merging ]3,5] with ]4,5] produces ]3,5];
• merging ]3,5] with ]5,8] produces ]3,8];
• merging ]3,5] with ]6,8] fails.

Parameters:

that - the other range.

Returns:

the resulting range.

Throws:

IllegalArgumentException - if the ranges neither intersect nor are adjacent.

equals

public boolean equals(Object other)

Overrides:

equals in class Object

hashCode

public int hashCode()

Overrides:

hashCode in class Object

toString

public String toString()

Overrides:

toString in class Object

compareTo

public int compareTo(TokenRange other)

Specified by:

compareTo in interface Comparable<TokenRange>