commons-collections
commons-collections copied to clipboard
apache
COLLECTIONS-873: create PatriciaTrieSet
According to https://issues.apache.org/jira/browse/COLLECTIONS-873
MR is basically a draft or discussion of what to do.
I make PatriciaTrieSet based on PatriciaTrie under the hood.
Below I type what I don't like. Please give some advice.
The main problem is when I access PatriciaTrie, I have to return SortedSet. But all collections under the hood of PatriciaTrie don't use SortedSet. Thus, I wrap call under TreeSet.
For example
@Override
public SortedSet<String> headSet(String toElement) {
return new TreeSet<>(trie.headMap(toElement).keySet());
}
This is overwhelming because the keySet is already sorted, and adding to a TreeSet also takes time.
To improve the situation, I make two approaches:
- Fully rewrite PatriciaTrieSet, copying logic with bit operations to the corresponding class. The drawback is that a change (bug) in bit operations leads to a rewrite in PatriciaTrieSet and PatriciaTrie independently.
- Extend the internal PatricaiTrie class to implement NavigableMap (PrefixRangeMap and RangeEntryMap + some other classes).
public interface NavigableMap<K,V> extends SortedMap<K,V>From NavigableMap, I can get navigableKeySet and work with it in PatriciaTrieSet.
Hello @ekuvardin The build fails, it seems you did not read or follow the steps in https://github.com/apache/commons-collections/blob/master/.github/pull_request_template.md
@garydgregory Fix build. My fault forget for Apache license header and manually disable rat check
Now:
- Localy run mvn with no errors
- Checkstyle doesn't add new warning on new files
- Make more unit test
- Make benchmarks. I don't find JMH benchmarks in the project and write on my own. https://github.com/ekuvardin/apache-common-collections-benchmarks Goal: Approve that set, give some additional time to work, and performance on each method doesn't degrade dramatically.
Benchmarks results
Benchmark (sizeFill) Mode Cnt Score Error Units
PatriciaBenchmarkAddExistingValue.addPatriciaTrie 10000 avgt 10 132,648 ? 1,718 ns/op
PatriciaBenchmarkAddExistingValue.addPatriciaTrieSet 10000 avgt 10 141,382 ? 2,160 ns/op
Benchmark (sizeFill) (sizeTryToAdd) Mode Cnt Score Error Units
PatriciaBenchmarkAddNewValue.addPatriciaTrie 10000 100000 avgt 5 158,198 ? 1,014 ns/op
PatriciaBenchmarkAddNewValue.addPatriciaTrieSet 10000 100000 avgt 5 175,539 ? 3,196 ns/op
Benchmark (sizeFill) (sizeTryToRemove) Mode Cnt Score Error Units
PatriciaBenchmarkRemove.removePatriciaTrie 100000 100000 avgt 5 133,199 ? 0,989 ns/op
PatriciaBenchmarkRemove.removePatriciaTrieSet 100000 100000 avgt 5 134,057 ? 0,960 ns/op
Benchmark (size) Mode Cnt Score Error Units
PatriciaBenchmarkSimpleOperation.containsPatriciaTrie 1000 avgt 5 48,596 ? 1,230 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrie 10000 avgt 5 51,476 ? 0,824 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrie 100000 avgt 5 51,834 ? 0,413 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrieSet 1000 avgt 5 48,540 ? 1,174 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrieSet 10000 avgt 5 52,369 ? 0,398 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrieSet 100000 avgt 5 52,190 ? 0,547 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrie 1000 avgt 5 6,270 ? 0,093 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrie 10000 avgt 5 6,262 ? 0,061 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrie 100000 avgt 5 6,230 ? 0,187 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrieSet 1000 avgt 5 6,817 ? 0,066 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrieSet 10000 avgt 5 6,751 ? 0,088 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrieSet 100000 avgt 5 6,737 ? 0,085 ns/op
PatriciaBenchmarkSimpleOperation.headMapPatriciaTrie 1000 avgt 5 6,687 ? 0,023 ns/op
PatriciaBenchmarkSimpleOperation.headMapPatriciaTrie 10000 avgt 5 6,808 ? 0,231 ns/op
PatriciaBenchmarkSimpleOperation.headMapPatriciaTrie 100000 avgt 5 6,819 ? 0,319 ns/op
PatriciaBenchmarkSimpleOperation.headSetPatriciaTrieSet 1000 avgt 5 9,267 ? 0,075 ns/op
PatriciaBenchmarkSimpleOperation.headSetPatriciaTrieSet 10000 avgt 5 9,343 ? 0,211 ns/op
PatriciaBenchmarkSimpleOperation.headSetPatriciaTrieSet 100000 avgt 5 9,330 ? 0,154 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrie 1000 avgt 5 0,792 ? 0,019 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrie 10000 avgt 5 0,799 ? 0,005 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrie 100000 avgt 5 0,889 ? 0,060 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrieSet 1000 avgt 5 1,021 ? 0,180 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrieSet 10000 avgt 5 1,186 ? 0,195 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrieSet 100000 avgt 5 1,024 ? 0,155 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrie 1000 avgt 5 10405,896 ? 202,444 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrie 10000 avgt 5 123151,828 ? 34419,185 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrie 100000 avgt 5 1727588,628 ? 25520,548 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrieSet 1000 avgt 5 10049,078 ? 130,071 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrieSet 10000 avgt 5 141096,432 ? 1309,961 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrieSet 100000 avgt 5 1664385,856 ? 8252,919 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrie 1000 avgt 5 6866,638 ? 235,368 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrie 10000 avgt 5 27320,421 ? 359,976 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrie 100000 avgt 5 718470,198 ? 16591,016 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrieSet 1000 avgt 5 7169,227 ? 51,483 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrieSet 10000 avgt 5 27232,041 ? 819,069 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrieSet 100000 avgt 5 742578,217 ? 5893,876 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrie 1000 avgt 5 16,504 ? 0,077 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrie 10000 avgt 5 16,445 ? 0,250 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrie 100000 avgt 5 23,058 ? 0,407 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrieSet 1000 avgt 5 18,047 ? 0,138 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrieSet 10000 avgt 5 17,766 ? 0,502 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrieSet 100000 avgt 5 24,725 ? 0,513 ns/op
PatriciaBenchmarkSimpleOperation.prefixMapPatriciaTrie 1000 avgt 5 8,907 ? 0,046 ns/op
PatriciaBenchmarkSimpleOperation.prefixMapPatriciaTrie 10000 avgt 5 8,943 ? 0,153 ns/op
PatriciaBenchmarkSimpleOperation.prefixMapPatriciaTrie 100000 avgt 5 8,887 ? 0,013 ns/op
PatriciaBenchmarkSimpleOperation.prefixSetPatriciaTrieSet 1000 avgt 5 11,372 ? 0,046 ns/op
PatriciaBenchmarkSimpleOperation.prefixSetPatriciaTrieSet 10000 avgt 5 11,398 ? 0,036 ns/op
PatriciaBenchmarkSimpleOperation.prefixSetPatriciaTrieSet 100000 avgt 5 11,341 ? 0,012 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrie 1000 avgt 5 0,948 ? 0,119 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrie 10000 avgt 5 0,803 ? 0,052 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrie 100000 avgt 5 0,889 ? 0,073 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrieSet 1000 avgt 5 1,038 ? 0,137 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrieSet 10000 avgt 5 1,102 ? 0,201 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrieSet 100000 avgt 5 1,250 ? 0,248 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrie 1000 avgt 5 8,460 ? 0,051 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrie 10000 avgt 5 8,982 ? 0,036 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrie 100000 avgt 5 8,413 ? 0,044 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieContains 1000 avgt 5 25,776 ? 0,592 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieContains 10000 avgt 5 26,233 ? 0,247 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieContains 100000 avgt 5 26,073 ? 0,307 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieSetContains 1000 avgt 5 26,864 ? 0,812 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieSetContains 10000 avgt 5 27,708 ? 0,398 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieSetContains 100000 avgt 5 26,643 ? 0,286 ns/op
PatriciaBenchmarkSimpleOperation.subSetPatriciaTrieSet 1000 avgt 5 11,157 ? 0,400 ns/op
PatriciaBenchmarkSimpleOperation.subSetPatriciaTrieSet 10000 avgt 5 11,364 ? 0,049 ns/op
PatriciaBenchmarkSimpleOperation.subSetPatriciaTrieSet 100000 avgt 5 11,009 ? 0,293 ns/op
PatriciaBenchmarkSimpleOperation.tailMapPatriciaTrie 1000 avgt 5 6,842 ? 0,057 ns/op
PatriciaBenchmarkSimpleOperation.tailMapPatriciaTrie 10000 avgt 5 6,839 ? 0,034 ns/op
PatriciaBenchmarkSimpleOperation.tailMapPatriciaTrie 100000 avgt 5 6,850 ? 0,029 ns/op
PatriciaBenchmarkSimpleOperation.tailSetPatriciaTrieSet 1000 avgt 5 9,305 ? 0,007 ns/op
PatriciaBenchmarkSimpleOperation.tailSetPatriciaTrieSet 10000 avgt 5 9,508 ? 0,196 ns/op
PatriciaBenchmarkSimpleOperation.tailSetPatriciaTrieSet 100000 avgt 5 9,338 ? 0,036 ns/op
Summary
How it works
The main idea is to build logic on the above existing PatriciaTrie not reinventing a wheel.
To do this, I create a stub Object and use it as a value in PatriciaTrie
static final Object PRESENT = new Object();
All put methods are similar to
patriciaTrie .put(e, PRESENT)
All new classes are put on a picture below.
TrieSet
New interface TrieSet consists of methods.
SortedSet<K> prefixSet(K key);
and extends
SortedSet
PatriciaTrieSet
New class PatriciaTrieSet
implements
TrieSet<String>
extends
AbstractSet<String>
Serializable
RangePatriciaTrieSortedSet User will want to work with results of methods below in a SortedSet manner.
SortedSet<E> subSet(E fromElement, E toElement);
SortedSet<E> headSet(E toElement);
SortedSet<E> tailSet(E fromElement);
SortedSet<K> prefixSet(K key);
To do so, I create an inner class RangePatriciaTrieSortedSet in PatriciaTrieSet
extends
AbstractSet<String>
implements
SortedSet<String>
RangePatriciaTrieSortedSet doesn't immediately construct SortedSet; it's a wrapper under PatriciaSet.RangeEntryMap and PatriciaSet.PrefixRangeMap works in the same manner.
Objects in this SortedMap are sorted according to lexicographical order. Objects in SortedSet just ignore value.
Benchmarks Also write some benchmarks to prove that Set doesn't degrade dramatically. Make benchmarks. I don't find JMH benchmarks in the project and write on my own. https://github.com/ekuvardin/apache-common-collections-benchmarks
Benchmarks results
Benchmark (sizeFill) Mode Cnt Score Error Units
PatriciaBenchmarkAddExistingValue.addPatriciaTrie 10000 avgt 10 132,648 ? 1,718 ns/op
PatriciaBenchmarkAddExistingValue.addPatriciaTrieSet 10000 avgt 10 141,382 ? 2,160 ns/op
Benchmark (sizeFill) (sizeTryToAdd) Mode Cnt Score Error Units
PatriciaBenchmarkAddNewValue.addPatriciaTrie 10000 100000 avgt 5 158,198 ? 1,014 ns/op
PatriciaBenchmarkAddNewValue.addPatriciaTrieSet 10000 100000 avgt 5 175,539 ? 3,196 ns/op
Benchmark (sizeFill) (sizeTryToRemove) Mode Cnt Score Error Units
PatriciaBenchmarkRemove.removePatriciaTrie 100000 100000 avgt 5 133,199 ? 0,989 ns/op
PatriciaBenchmarkRemove.removePatriciaTrieSet 100000 100000 avgt 5 134,057 ? 0,960 ns/op
Benchmark (size) Mode Cnt Score Error Units
PatriciaBenchmarkSimpleOperation.containsPatriciaTrie 1000 avgt 5 48,596 ? 1,230 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrie 10000 avgt 5 51,476 ? 0,824 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrie 100000 avgt 5 51,834 ? 0,413 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrieSet 1000 avgt 5 48,540 ? 1,174 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrieSet 10000 avgt 5 52,369 ? 0,398 ns/op
PatriciaBenchmarkSimpleOperation.containsPatriciaTrieSet 100000 avgt 5 52,190 ? 0,547 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrie 1000 avgt 5 6,270 ? 0,093 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrie 10000 avgt 5 6,262 ? 0,061 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrie 100000 avgt 5 6,230 ? 0,187 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrieSet 1000 avgt 5 6,817 ? 0,066 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrieSet 10000 avgt 5 6,751 ? 0,088 ns/op
PatriciaBenchmarkSimpleOperation.firstPatriciaTrieSet 100000 avgt 5 6,737 ? 0,085 ns/op
PatriciaBenchmarkSimpleOperation.headMapPatriciaTrie 1000 avgt 5 6,687 ? 0,023 ns/op
PatriciaBenchmarkSimpleOperation.headMapPatriciaTrie 10000 avgt 5 6,808 ? 0,231 ns/op
PatriciaBenchmarkSimpleOperation.headMapPatriciaTrie 100000 avgt 5 6,819 ? 0,319 ns/op
PatriciaBenchmarkSimpleOperation.headSetPatriciaTrieSet 1000 avgt 5 9,267 ? 0,075 ns/op
PatriciaBenchmarkSimpleOperation.headSetPatriciaTrieSet 10000 avgt 5 9,343 ? 0,211 ns/op
PatriciaBenchmarkSimpleOperation.headSetPatriciaTrieSet 100000 avgt 5 9,330 ? 0,154 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrie 1000 avgt 5 0,792 ? 0,019 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrie 10000 avgt 5 0,799 ? 0,005 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrie 100000 avgt 5 0,889 ? 0,060 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrieSet 1000 avgt 5 1,021 ? 0,180 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrieSet 10000 avgt 5 1,186 ? 0,195 ns/op
PatriciaBenchmarkSimpleOperation.isEmptyPatriciaTrieSet 100000 avgt 5 1,024 ? 0,155 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrie 1000 avgt 5 10405,896 ? 202,444 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrie 10000 avgt 5 123151,828 ? 34419,185 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrie 100000 avgt 5 1727588,628 ? 25520,548 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrieSet 1000 avgt 5 10049,078 ? 130,071 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrieSet 10000 avgt 5 141096,432 ? 1309,961 ns/op
PatriciaBenchmarkSimpleOperation.iteratorPatriciaTrieSet 100000 avgt 5 1664385,856 ? 8252,919 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrie 1000 avgt 5 6866,638 ? 235,368 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrie 10000 avgt 5 27320,421 ? 359,976 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrie 100000 avgt 5 718470,198 ? 16591,016 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrieSet 1000 avgt 5 7169,227 ? 51,483 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrieSet 10000 avgt 5 27232,041 ? 819,069 ns/op
PatriciaBenchmarkSimpleOperation.iteratorSubSetPatriciaTrieSet 100000 avgt 5 742578,217 ? 5893,876 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrie 1000 avgt 5 16,504 ? 0,077 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrie 10000 avgt 5 16,445 ? 0,250 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrie 100000 avgt 5 23,058 ? 0,407 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrieSet 1000 avgt 5 18,047 ? 0,138 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrieSet 10000 avgt 5 17,766 ? 0,502 ns/op
PatriciaBenchmarkSimpleOperation.lastPatriciaTrieSet 100000 avgt 5 24,725 ? 0,513 ns/op
PatriciaBenchmarkSimpleOperation.prefixMapPatriciaTrie 1000 avgt 5 8,907 ? 0,046 ns/op
PatriciaBenchmarkSimpleOperation.prefixMapPatriciaTrie 10000 avgt 5 8,943 ? 0,153 ns/op
PatriciaBenchmarkSimpleOperation.prefixMapPatriciaTrie 100000 avgt 5 8,887 ? 0,013 ns/op
PatriciaBenchmarkSimpleOperation.prefixSetPatriciaTrieSet 1000 avgt 5 11,372 ? 0,046 ns/op
PatriciaBenchmarkSimpleOperation.prefixSetPatriciaTrieSet 10000 avgt 5 11,398 ? 0,036 ns/op
PatriciaBenchmarkSimpleOperation.prefixSetPatriciaTrieSet 100000 avgt 5 11,341 ? 0,012 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrie 1000 avgt 5 0,948 ? 0,119 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrie 10000 avgt 5 0,803 ? 0,052 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrie 100000 avgt 5 0,889 ? 0,073 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrieSet 1000 avgt 5 1,038 ? 0,137 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrieSet 10000 avgt 5 1,102 ? 0,201 ns/op
PatriciaBenchmarkSimpleOperation.sizePatriciaTrieSet 100000 avgt 5 1,250 ? 0,248 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrie 1000 avgt 5 8,460 ? 0,051 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrie 10000 avgt 5 8,982 ? 0,036 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrie 100000 avgt 5 8,413 ? 0,044 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieContains 1000 avgt 5 25,776 ? 0,592 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieContains 10000 avgt 5 26,233 ? 0,247 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieContains 100000 avgt 5 26,073 ? 0,307 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieSetContains 1000 avgt 5 26,864 ? 0,812 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieSetContains 10000 avgt 5 27,708 ? 0,398 ns/op
PatriciaBenchmarkSimpleOperation.subMapPatriciaTrieSetContains 100000 avgt 5 26,643 ? 0,286 ns/op
PatriciaBenchmarkSimpleOperation.subSetPatriciaTrieSet 1000 avgt 5 11,157 ? 0,400 ns/op
PatriciaBenchmarkSimpleOperation.subSetPatriciaTrieSet 10000 avgt 5 11,364 ? 0,049 ns/op
PatriciaBenchmarkSimpleOperation.subSetPatriciaTrieSet 100000 avgt 5 11,009 ? 0,293 ns/op
PatriciaBenchmarkSimpleOperation.tailMapPatriciaTrie 1000 avgt 5 6,842 ? 0,057 ns/op
PatriciaBenchmarkSimpleOperation.tailMapPatriciaTrie 10000 avgt 5 6,839 ? 0,034 ns/op
PatriciaBenchmarkSimpleOperation.tailMapPatriciaTrie 100000 avgt 5 6,850 ? 0,029 ns/op
PatriciaBenchmarkSimpleOperation.tailSetPatriciaTrieSet 1000 avgt 5 9,305 ? 0,007 ns/op
PatriciaBenchmarkSimpleOperation.tailSetPatriciaTrieSet 10000 avgt 5 9,508 ? 0,196 ns/op
PatriciaBenchmarkSimpleOperation.tailSetPatriciaTrieSet 100000 avgt 5 9,338 ? 0,036 ns/op
Merge with master and fix new checkstyle warnings that come with the latest master commits.