private boolean positionEqualsPosition(int leftPosition, int rightPosition) {
long leftPageAddress = addresses.getLong(leftPosition);
int leftBlockIndex = decodeSliceIndex(leftPageAddress);
int leftBlockPosition = decodePosition(leftPageAddress);
long rightPageAddress = addresses.getLong(rightPosition);
int rightBlockIndex = decodeSliceIndex(rightPageAddress);
int rightBlockPosition = decodePosition(rightPageAddress);
return pagesHashStrategy.positionEqualsPosition(
leftBlockIndex, leftBlockPosition, rightBlockIndex, rightBlockPosition);
}
private boolean positionEqualsCurrentRow(int position, BlockCursor... cursors) {
long pageAddress = addresses.getLong(position);
int blockIndex = decodeSliceIndex(pageAddress);
int blockPosition = decodePosition(pageAddress);
return pagesHashStrategy.positionEqualsCursors(blockIndex, blockPosition, cursors);
}
private int hashPosition(int position) {
long pageAddress = addresses.getLong(position);
int blockIndex = decodeSliceIndex(pageAddress);
int blockPosition = decodePosition(pageAddress);
return pagesHashStrategy.hashPosition(blockIndex, blockPosition);
}
public void appendTo(long position, PageBuilder pageBuilder, int outputChannelOffset) {
long pageAddress = addresses.getLong(Ints.checkedCast(position));
int blockIndex = decodeSliceIndex(pageAddress);
int blockPosition = decodePosition(pageAddress);
pagesHashStrategy.appendTo(blockIndex, blockPosition, pageBuilder, outputChannelOffset);
}
public InMemoryJoinHash(
LongArrayList addresses,
PagesHashStrategy pagesHashStrategy,
OperatorContext operatorContext) {
this.addresses = checkNotNull(addresses, "addresses is null");
this.pagesHashStrategy = checkNotNull(pagesHashStrategy, "pagesHashStrategy is null");
this.channelCount = pagesHashStrategy.getChannelCount();
checkNotNull(operatorContext, "operatorContext is null");
// reserve memory for the arrays
int hashSize = HashCommon.arraySize(addresses.size(), 0.75f);
operatorContext.reserveMemory(sizeOfIntArray(hashSize) + sizeOfIntArray(addresses.size()));
mask = hashSize - 1;
key = new int[hashSize];
Arrays.fill(key, -1);
this.positionLinks = new int[addresses.size()];
Arrays.fill(positionLinks, -1);
// index pages
for (int position = 0; position < addresses.size(); position++) {
int pos = ((int) Murmur3.hash64(hashPosition(position))) & mask;
// look for an empty slot or a slot containing this key
while (key[pos] != -1) {
int currentKey = key[pos];
if (positionEqualsPosition(currentKey, position)) {
// found a slot for this key
// link the new key position to the current key position
positionLinks[position] = currentKey;
// key[pos] updated outside of this loop
break;
}
// increment position and mask to handler wrap around
pos = (pos + 1) & mask;
}
key[pos] = position;
}
}
/**
* Check that we score items but do not provide scores for items the user has previously rated.
* User 5 has rated only item 8 previously.
*/
@Test
public void testItemScorerNoRating() {
long[] items = {7, 8};
ItemItemScorer scorer = session.get(ItemItemScorer.class);
assertThat(scorer, notNullValue());
SparseVector scores = scorer.score(5, LongArrayList.wrap(items));
assertThat(scores, notNullValue());
assertThat(scores.size(), equalTo(1));
assertThat(scores.get(7), not(notANumber()));
assertThat(scores.containsKey(8), equalTo(false));
}
/**
* Check that we score items but do not provide scores for items the user has previously rated.
* User 5 has rated only item 8 previously.
*/
@Test
public void testItemScorerChannels() {
long[] items = {7, 8};
ItemItemScorer scorer = session.get(ItemItemScorer.class);
assertThat(scorer, notNullValue());
SparseVector scores = scorer.score(5, LongArrayList.wrap(items));
assertThat(scores, notNullValue());
assertThat(scores.size(), equalTo(1));
assertThat(scores.get(7), not(notANumber()));
assertThat(
scores.getChannelVector(ItemItemScorer.NEIGHBORHOOD_SIZE_SYMBOL).get(7),
closeTo(1.0, 1.0e-5));
assertThat(scores.containsKey(8), equalTo(false));
long[] items2 = {7, 8, 9};
scorer = session.get(ItemItemScorer.class);
assertThat(scorer, notNullValue());
scores = scorer.score(2, LongArrayList.wrap(items2));
assertThat(
scores.getChannelVector(ItemItemScorer.NEIGHBORHOOD_SIZE_SYMBOL).get(9),
closeTo(3.0, 1.0e-5)); // 1, 7, 8
}
/**
* Get the keys of this vector sorted by the value of the items stored for each key.
*
* @param decreasing If {@code true}, sort in decreasing order.
* @return The sorted list of keys of this vector.
*/
public LongArrayList keysByValue(boolean decreasing) {
long[] skeys = keySet().toLongArray();
LongComparator cmp;
// Set up the comparator. We use the key as a secondary comparison to get
// a reproducible sort irrespective of sorting algorithm.
if (decreasing) {
cmp =
new AbstractLongComparator() {
@Override
public int compare(long k1, long k2) {
int c = Double.compare(get(k2), get(k1));
if (c != 0) {
return c;
} else {
return Longs.compare(k1, k2);
}
}
};
} else {
cmp =
new AbstractLongComparator() {
@Override
public int compare(long k1, long k2) {
int c = Double.compare(get(k1), get(k2));
if (c != 0) {
return c;
} else {
return Longs.compare(k1, k2);
}
}
};
}
LongArrays.quickSort(skeys, cmp);
return LongArrayList.wrap(skeys);
}
