leerho commented on code in PR #676:
URL: https://github.com/apache/datasketches-java/pull/676#discussion_r2229750207
##########
src/main/java/org/apache/datasketches/count/CountMinSketch.java:
##########
@@ -57,30 +65,59 @@ int mask() {
* @param seed The base hash seed
*/
CountMinSketch(final byte numHashes, final int numBuckets, final long seed) {
- numHashes_ = numHashes;
- numBuckets_ = numBuckets;
- seed_ = seed;
- hashSeeds_ = new long[numHashes];
- sketchArray_ = new long[numHashes * numBuckets];
- totalWeight_ = 0;
+ // Validate numHashes
+ if (numHashes <= 0) {
+ throw new SketchesArgumentException("Number of hash functions must be
positive, got: " + numHashes);
+ }
+ // Validate numBuckets with clear mathematical justification
+ if (numBuckets <= 0) {
+ throw new SketchesArgumentException("Number of buckets must be positive,
got: " + numBuckets);
+ }
if (numBuckets < 3) {
- throw new SketchesArgumentException("Using fewer than 3 buckets incurs
relative error greater than 1.");
+ throw new SketchesArgumentException("Number of buckets must be at least
3 to ensure relative error ≤ 1.0. " +
+ "With " + numBuckets + " buckets, relative error would be " +
String.format("%.3f", Math.exp(1.0) / numBuckets));
+ }
+
+ // Check for potential overflow in array size calculation
+ // Use long arithmetic to detect overflow before casting
+ final long totalSize = (long) numHashes * (long) numBuckets;
+ if (totalSize > Integer.MAX_VALUE) {
+ throw new SketchesArgumentException("Sketch array size would overflow: "
+ numHashes + " * " + numBuckets +
+ " = " + totalSize + " > " + Integer.MAX_VALUE);
}
// This check is to ensure later compatibility with a Java implementation
whose maximum size can only
// be 2^31-1. We check only against 2^30 for simplicity.
- if (numBuckets * numHashes >= 1 << 30) {
- throw new SketchesArgumentException("These parameters generate a sketch
that exceeds 2^30 elements. \n" +
- "Try reducing either the number of buckets or the number of hash
functions.");
+ if (totalSize >= (1L << 30)) {
+ throw new SketchesArgumentException("Sketch would require excessive
memory: " + numHashes + " * " + numBuckets +
+ " = " + totalSize + " elements (~" + String.format("%.1f", totalSize
* 8.0 / (1024 * 1024 * 1024)) + " GB). " +
+ "Consider reducing numHashes or numBuckets.");
}
+ numHashes_ = numHashes;
+ numBuckets_ = numBuckets;
+ seed_ = seed;
+ hashSeeds_ = new long[numHashes];
+ sketchArray_ = new long[(int) totalSize];
+ totalWeight_ = 0;
+
Random rand = new Random(seed);
for (int i = 0; i < numHashes; i++) {
Review Comment:
missing "final" for Random rand...
--
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.
To unsubscribe, e-mail: dev-unsubscr...@datasketches.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@datasketches.apache.org
For additional commands, e-mail: dev-h...@datasketches.apache.org
