You are viewing a plain text version of this content. The canonical link for it is here.
Posted to pr@cassandra.apache.org by "dcapwell (via GitHub)" <gi...@apache.org> on 2024/03/04 20:03:41 UTC
[PR] CASSANDRA-19355: Refactort CommandsForRange to work based off disk state rather than in-memory state [cassandra-accord]
dcapwell opened a new pull request, #83:
URL: https://github.com/apache/cassandra-accord/pull/83
(no comment)
--
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: pr-unsubscribe@cassandra.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org
Re: [PR] CASSANDRA-19355: Refactort CommandsForRange to work based off disk state rather than in-memory state [cassandra-accord]
Posted by "dcapwell (via GitHub)" <gi...@apache.org>.
dcapwell commented on code in PR #83:
URL: https://github.com/apache/cassandra-accord/pull/83#discussion_r1511738613
##########
accord-core/src/test/java/accord/utils/Gens.java:
##########
@@ -116,6 +125,253 @@ public static Gen.IntGen pickZipf(int[] array)
};
}
+ public static Gen.LongGen pickZipf(long[] array)
+ {
+ if (array == null || array.length == 0)
+ throw new IllegalArgumentException("Empty array given");
+ if (array.length == 1)
+ return ignore -> array[0];
+ BigDecimal[] weights = new BigDecimal[array.length];
+ BigDecimal base = BigDecimal.valueOf(Math.pow(2, array.length));
+ weights[0] = base;
+ for (int i = 1; i < array.length; i++)
+ weights[i] = base.divide(BigDecimal.valueOf(i + 1), RoundingMode.UP);
+ BigDecimal totalWeights = Stream.of(weights).reduce(BigDecimal.ZERO, BigDecimal::add);
+
+ return rs -> {
+ BigDecimal value = BigDecimal.valueOf(rs.nextDouble()).multiply(totalWeights);
+ for (int i = 0; i < weights.length; i++)
+ {
+ value = value.subtract(weights[i]);
+ if (value.compareTo(BigDecimal.ZERO) <= 0)
+ return array[i];
+ }
+ return array[array.length - 1];
+ };
+ }
+
+ public static <T> Gen<T> pickZipf(T... array)
+ {
+ return pickZipf(Arrays.asList(array));
+ }
+
+ public static <T> Gen<T> pickZipf(List<T> array)
+ {
+ if (array == null || array.isEmpty())
+ throw new IllegalArgumentException("Empty array given");
+ if (array.size() == 1)
+ return ignore -> array.get(0);
+ BigDecimal[] weights = new BigDecimal[array.size()];
+ BigDecimal base = BigDecimal.valueOf(Math.pow(2, array.size()));
+ weights[0] = base;
+ for (int i = 1; i < array.size(); i++)
+ weights[i] = base.divide(BigDecimal.valueOf(i + 1), RoundingMode.UP);
+ BigDecimal totalWeights = Stream.of(weights).reduce(BigDecimal.ZERO, BigDecimal::add);
+
+ return rs -> {
+ BigDecimal value = BigDecimal.valueOf(rs.nextDouble()).multiply(totalWeights);
+ for (int i = 0; i < weights.length; i++)
+ {
+ value = value.subtract(weights[i]);
+ if (value.compareTo(BigDecimal.ZERO) <= 0)
+ return array.get(i);
+ }
+ return array.get(array.size() - 1);
+ };
+ }
+
+ public static Gen<Gen.IntGen> randomWeights(int[] array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.length);
+ return r -> array[index(r, weights)];
+ };
+ }
+
+ public static Gen<Gen.LongGen> randomWeights(long[] array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.length);
+ return r -> array[index(r, weights)];
+ };
+ }
+
+ public static <T> Gen<Gen<T>> randomWeights(T[] array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.length);
+ return r -> array[index(r, weights)];
+ };
+ }
+
+ public static <T> Gen<Gen<T>> randomWeights(List<T> array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.size());
+ return r -> array.get(index(r, weights));
+ };
+ }
+
+ private static int index(RandomSource rs, float[] weights)
+ {
+ int i = Arrays.binarySearch(weights, rs.nextFloat());
+ if (i < 0) i = -1 - i;
+ return i;
+ }
+
+ public static Gen<Gen.IntGen> mixedDistribution(int minInclusive, int maxExclusive)
+ {
+ int domainSize = (maxExclusive - minInclusive + 1);
+ if (domainSize < 0)
+ throw new IllegalArgumentException("Range is too large; min=" + minInclusive + ", max=" + maxExclusive);
+ int[] array, indexes;
+ if (domainSize > 200) // randomly selected
+ {
+ int numBuckets = 10;
Review Comment:
when it comes to `randomWeight` and `zipF` having larger arrays doesn't make too much sense, so this puts a bias into a smaller array and generates values within that section of the range.
I could make this a config, but I didn't need so can refactor later if needed
--
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: pr-unsubscribe@cassandra.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org
Re: [PR] CASSANDRA-19355: Refactort CommandsForRange to work based off disk state rather than in-memory state [cassandra-accord]
Posted by "dcapwell (via GitHub)" <gi...@apache.org>.
dcapwell commented on code in PR #83:
URL: https://github.com/apache/cassandra-accord/pull/83#discussion_r1511767120
##########
accord-core/src/test/java/accord/utils/SearchableRangeListTest.java:
##########
@@ -0,0 +1,116 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package accord.utils;
+
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import java.util.function.BiConsumer;
+
+import org.junit.jupiter.api.Test;
+
+import accord.impl.IntKey;
+import accord.primitives.Range;
+import org.assertj.core.api.Assertions;
+
+import static accord.utils.Property.qt;
+
+class SearchableRangeListTest
Review Comment:
this is kinda duplicated in `org.apache.cassandra.utils.RTreeTest#test` but has better coverage... I wrote this class first and feel its always best to have tests at this level / project rather than depending on the C* test...
--
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: pr-unsubscribe@cassandra.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org
Re: [PR] CASSANDRA-19355: Refactort CommandsForRange to work based off disk state rather than in-memory state [cassandra-accord]
Posted by "dcapwell (via GitHub)" <gi...@apache.org>.
dcapwell commented on code in PR #83:
URL: https://github.com/apache/cassandra-accord/pull/83#discussion_r1511767730
##########
buildSrc/src/main/groovy/accord.java-conventions.gradle:
##########
@@ -29,7 +29,7 @@ repositories {
}
compileJava {
- sourceCompatibility = JavaVersion.VERSION_1_8
+ sourceCompatibility = JavaVersion.VERSION_11
Review Comment:
we actually don't compile on jdk8... so this just makes it so we can use JDK 11 features again
--
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: pr-unsubscribe@cassandra.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org
Re: [PR] CASSANDRA-19355: Refactort CommandsForRange to work based off disk state rather than in-memory state [cassandra-accord]
Posted by "dcapwell (via GitHub)" <gi...@apache.org>.
dcapwell commented on code in PR #83:
URL: https://github.com/apache/cassandra-accord/pull/83#discussion_r1511736450
##########
accord-core/src/test/java/accord/utils/Gens.java:
##########
@@ -116,6 +125,253 @@ public static Gen.IntGen pickZipf(int[] array)
};
}
+ public static Gen.LongGen pickZipf(long[] array)
+ {
+ if (array == null || array.length == 0)
+ throw new IllegalArgumentException("Empty array given");
+ if (array.length == 1)
+ return ignore -> array[0];
+ BigDecimal[] weights = new BigDecimal[array.length];
+ BigDecimal base = BigDecimal.valueOf(Math.pow(2, array.length));
+ weights[0] = base;
+ for (int i = 1; i < array.length; i++)
+ weights[i] = base.divide(BigDecimal.valueOf(i + 1), RoundingMode.UP);
+ BigDecimal totalWeights = Stream.of(weights).reduce(BigDecimal.ZERO, BigDecimal::add);
+
+ return rs -> {
+ BigDecimal value = BigDecimal.valueOf(rs.nextDouble()).multiply(totalWeights);
+ for (int i = 0; i < weights.length; i++)
+ {
+ value = value.subtract(weights[i]);
+ if (value.compareTo(BigDecimal.ZERO) <= 0)
+ return array[i];
+ }
+ return array[array.length - 1];
+ };
+ }
+
+ public static <T> Gen<T> pickZipf(T... array)
+ {
+ return pickZipf(Arrays.asList(array));
+ }
+
+ public static <T> Gen<T> pickZipf(List<T> array)
+ {
+ if (array == null || array.isEmpty())
+ throw new IllegalArgumentException("Empty array given");
+ if (array.size() == 1)
+ return ignore -> array.get(0);
+ BigDecimal[] weights = new BigDecimal[array.size()];
+ BigDecimal base = BigDecimal.valueOf(Math.pow(2, array.size()));
+ weights[0] = base;
+ for (int i = 1; i < array.size(); i++)
+ weights[i] = base.divide(BigDecimal.valueOf(i + 1), RoundingMode.UP);
+ BigDecimal totalWeights = Stream.of(weights).reduce(BigDecimal.ZERO, BigDecimal::add);
+
+ return rs -> {
+ BigDecimal value = BigDecimal.valueOf(rs.nextDouble()).multiply(totalWeights);
+ for (int i = 0; i < weights.length; i++)
+ {
+ value = value.subtract(weights[i]);
+ if (value.compareTo(BigDecimal.ZERO) <= 0)
+ return array.get(i);
+ }
+ return array.get(array.size() - 1);
+ };
+ }
+
+ public static Gen<Gen.IntGen> randomWeights(int[] array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.length);
+ return r -> array[index(r, weights)];
+ };
+ }
+
+ public static Gen<Gen.LongGen> randomWeights(long[] array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.length);
+ return r -> array[index(r, weights)];
+ };
+ }
+
+ public static <T> Gen<Gen<T>> randomWeights(T[] array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.length);
+ return r -> array[index(r, weights)];
+ };
+ }
+
+ public static <T> Gen<Gen<T>> randomWeights(List<T> array)
+ {
+ return rs -> {
+ float[] weights = Picker.randomWeights(rs, array.size());
+ return r -> array.get(index(r, weights));
+ };
+ }
+
+ private static int index(RandomSource rs, float[] weights)
+ {
+ int i = Arrays.binarySearch(weights, rs.nextFloat());
+ if (i < 0) i = -1 - i;
+ return i;
+ }
+
+ public static Gen<Gen.IntGen> mixedDistribution(int minInclusive, int maxExclusive)
+ {
+ int domainSize = (maxExclusive - minInclusive + 1);
+ if (domainSize < 0)
+ throw new IllegalArgumentException("Range is too large; min=" + minInclusive + ", max=" + maxExclusive);
Review Comment:
I could support but figured its best to keep this logic simpler until thats needed... if you want `mixDistribution(Long.MIN_VALUE, Long.MAX_VALUE)` we need `BigInteger` to handle this...
--
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: pr-unsubscribe@cassandra.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org