Claudenw commented on a change in pull request #140: Bloomfilter updates2
URL:
https://github.com/apache/commons-collections/pull/140#discussion_r391772544
##########
File path:
src/main/java/org/apache/commons/collections4/bloomfilter/hasher/Shape.java
##########
@@ -124,103 +128,188 @@ public Shape(final HashFunctionIdentity
hashFunctionIdentity, final double proba
// similarly we can not produce a number greater than numberOfBits so
we
// do not have to check for Integer.MAX_VALUE either.
this.numberOfItems = (int) n;
- this.hashCode = generateHashCode();
// check that probability is within range
- getProbability();
+ checkCalculatedProbability(getProbability());
+ this.hashCode = generateHashCode();
}
/**
- * Constructs a filter configuration with the specified number of items and
- * probability. <p> The actual probability will be approximately equal to
the
- * desired probability but will be dependent upon the calculated bloom
filter size
- * and function count. </p>
+ * Constructs a filter configuration with the specified number of items
({@code n}) and
+ * desired false-positive probability ({@code p}).
+ *
+ * <p>The number of bits ({@code m}) for the filter is computed.
+ * <pre>m = ceil(n * log(p) / log(1 / 2^log(2)))</pre>
*
- * @param hashFunctionIdentity The HashFunctionIdentity of the hash
function this shape uses.
- * @param numberOfItems Number of items to be placed in the filter.
- * @param probability The desired probability of duplicates. Must be in
the range
- * (0.0,1.0).
+ * <p>The optimal number of hash functions ({@code k}) is computed.
+ * <pre>k = round((m / n) * log(2))</pre>
+ *
+ * <p>The actual probability will be approximately equal to the
+ * desired probability but will be dependent upon the calculated number of
bits and hash
+ * functions. An exception is raised if this is greater than or equal to 1
(i.e. the
+ * shape is invalid for use as a Bloom filter).
+ *
+ * @param hashFunctionIdentity The identity of the hash function this
shape uses
+ * @param numberOfItems Number of items to be placed in the filter
+ * @param probability The desired false-positive probability in the range
{@code (0, 1)}
+ * @throws NullPointerException if the hash function identity is null
+ * @throws IllegalArgumentException if the number of items is not strictly
positive
+ * @throws IllegalArgumentException if the desired probability is not in
the range {@code (0, 1)}
+ * @throws IllegalArgumentException if the calculated probability is not
below 1
+ * @see #getProbability()
*/
public Shape(final HashFunctionIdentity hashFunctionIdentity, final int
numberOfItems, final double probability) {
- Objects.requireNonNull(hashFunctionIdentity, "hashFunctionIdentity");
- if (numberOfItems < 1) {
- throw new IllegalArgumentException("Number of Items must be
greater than 0");
- }
- if (probability <= 0.0) {
- throw new IllegalArgumentException("Probability must be greater
than 0.0");
- }
- if (probability >= 1.0) {
- throw new IllegalArgumentException("Probability must be less than
1.0");
- }
- this.hashFunctionIdentity = hashFunctionIdentity;
- this.numberOfItems = numberOfItems;
- /*
- * number of bits is called "m" in most mathematical statement
describing
- * bloom filters so we use it here.
- */
+ this.hashFunctionIdentity =
Objects.requireNonNull(hashFunctionIdentity, "hashFunctionIdentity");
+ this.numberOfItems = checkNumberOfItems(numberOfItems);
+ checkProbability(probability);
+
+ // Number of bits (m)
final double m = Math.ceil(numberOfItems * Math.log(probability) /
DENOMINATOR);
if (m > Integer.MAX_VALUE) {
- throw new IllegalArgumentException("Resulting filter has more than
" + Integer.MAX_VALUE + " bits");
+ throw new IllegalArgumentException("Resulting filter has more than
" + Integer.MAX_VALUE + " bits: " + m);
}
this.numberOfBits = (int) m;
+
this.numberOfHashFunctions =
calculateNumberOfHashFunctions(numberOfItems, numberOfBits);
- this.hashCode = generateHashCode();
// check that probability is within range
- getProbability();
+ checkCalculatedProbability(getProbability());
+ this.hashCode = generateHashCode();
}
/**
- * Constructs a filter configuration with the specified number of items and
- * probability.
+ * Constructs a filter configuration with the specified number of items
({@code n}) and
+ * bits ({@code m}).
*
- * @param hashFunctionIdentity The HashFunctionIdentity of the hash
function this shape uses.
- * @param numberOfItems Number of items to be placed in the filter.
- * @param numberOfBits The number of bits in the filter.
+ * <p>The optimal number of hash functions ({@code k}) is computed.
+ * <pre>k = round((m / n) * log(2))</pre>
+ *
+ * <p>The false-positive probability is computed using the number of
items, bits and hash
+ * functions. An exception is raised if this is greater than or equal to 1
(i.e. the
+ * shape is invalid for use as a Bloom filter).
+ *
+ * @param hashFunctionIdentity The identity of the hash function this
shape uses
+ * @param numberOfItems Number of items to be placed in the filter
+ * @param numberOfBits The number of bits in the filter
+ * @throws NullPointerException if the hash function identity is null
+ * @throws IllegalArgumentException if the number of items is not strictly
positive
+ * @throws IllegalArgumentException if the number of bits is not above 8
Review comment:
again >= 8
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
