diff --git a/project/TastyMiMaFilters.scala b/project/TastyMiMaFilters.scala index d0533c6e464e..fd9431c05c45 100644 --- a/project/TastyMiMaFilters.scala +++ b/project/TastyMiMaFilters.scala @@ -65,11 +65,8 @@ object TastyMiMaFilters { // Problem: Inferred result type of non-private member differs ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.collection.convert.JavaCollectionWrappers.IterableWrapperTrait.iterator"), - ProblemMatcher.make(ProblemKind.IncompatibleTypeChange, "scala.util.matching.Regex.MatchIterator.replacementData"), // before: scala.Any; after: scala.collection.AbstractIterator[scala.util.matching.Regex] & scala.util.matching.Regex.Replacement - ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.util.hashing.Hashing.fromFunction"), - ProblemMatcher.make(ProblemKind.IncompatibleTypeChange, "scala.collection.mutable.UnrolledBuffer.classTagCompanion"), ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.collection.convert.JavaCollectionWrappers.SetWrapper.iterator"), - ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.collection.concurrent.FailedNode.string"), + ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.util.hashing.Hashing.fromFunction"), // Problem: implicit class ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.collection.generic.IsIterableLowPriority.is*LikeIsIterable"), @@ -143,6 +140,7 @@ object TastyMiMaFilters { ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.collection.package.+:.unapply"), ProblemMatcher.make(ProblemKind.MissingTermMember, "scala.collection.convert.impl.*StepperBase.semiclone"), ProblemMatcher.make(ProblemKind.NewAbstractMember, "scala.collection.convert.impl.*StepperBase.semiclone"), + ProblemMatcher.make(ProblemKind.IncompatibleTypeChange, "scala.util.matching.Regex.MatchIterator.replacementData"), // before: scala.Any; after: scala.collection.AbstractIterator[scala.util.matching.Regex] & scala.util.matching.Regex.Replacement // TASTy-MiMa bug? Wildcards in self type ProblemMatcher.make(ProblemKind.MissingTypeMember, "scala.collection.generic.DefaultSerializable._$1"), @@ -150,6 +148,7 @@ object TastyMiMaFilters { // TASTy-MiMa bug? module classes ProblemMatcher.make(ProblemKind.IncompatibleTypeChange, "scala.collection.immutable.BitSet.bitSetFactory"), // The symbol scala.collection.immutable.BitSet.bitSetFactory has an incompatible type in current version: before: scala.collection.immutable.BitSet$; after: scala.collection.immutable.BitSet.type ProblemMatcher.make(ProblemKind.IncompatibleTypeChange, "scala.collection.mutable.BitSet.bitSetFactory"), // The symbol scala.collection.mutable.BitSet.bitSetFactory has an incompatible type in current version: before: scala.collection.mutable.BitSet$; after: scala.collection.mutable.BitSet.type + ProblemMatcher.make(ProblemKind.IncompatibleTypeChange, "scala.collection.mutable.UnrolledBuffer.classTagCompanion"), // The symbol scala.collection.mutable.UnrolledBuffer.classTagCompanion has an incompatible type in current version: before: scala.collection.mutable.UnrolledBuffer$; after: scala.collection.mutable.UnrolledBuffer.type // TASTy-MiMa bugs ProblemMatcher.make(ProblemKind.InternalError, "scala.collection.SeqView.appendedAll"), diff --git a/stdlib-bootstrapped/src/scala/collection/concurrent/TrieMap.scala b/stdlib-bootstrapped/src/scala/collection/concurrent/TrieMap.scala new file mode 100644 index 000000000000..7474d56cf311 --- /dev/null +++ b/stdlib-bootstrapped/src/scala/collection/concurrent/TrieMap.scala @@ -0,0 +1,1202 @@ +/* + * Scala (https://www.scala-lang.org) + * + * Copyright EPFL and Lightbend, Inc. + * + * Licensed under Apache License 2.0 + * (http://www.apache.org/licenses/LICENSE-2.0). + * + * See the NOTICE file distributed with this work for + * additional information regarding copyright ownership. + */ + +package scala +package collection +package concurrent + +import java.util.concurrent.atomic._ +import scala.{unchecked => uc} +import scala.annotation.tailrec +import scala.collection.concurrent.TrieMap.RemovalPolicy +import scala.collection.generic.DefaultSerializable +import scala.collection.immutable.{List, Nil} +import scala.collection.mutable.GrowableBuilder +import scala.util.Try +import scala.util.hashing.Hashing + +private[collection] final class INode[K, V](bn: MainNode[K, V], g: Gen, equiv: Equiv[K]) extends INodeBase[K, V](g) { + import INodeBase._ + + WRITE(bn) + + def this(g: Gen, equiv: Equiv[K]) = this(null, g, equiv) + + def WRITE(nval: MainNode[K, V]) = INodeBase.updater.set(this, nval) + + def CAS(old: MainNode[K, V], n: MainNode[K, V]) = INodeBase.updater.compareAndSet(this, old, n) + + def gcasRead(ct: TrieMap[K, V]): MainNode[K, V] = GCAS_READ(ct) + + def GCAS_READ(ct: TrieMap[K, V]): MainNode[K, V] = { + val m = /*READ*/mainnode + val prevval = /*READ*/m.prev + if (prevval eq null) m + else GCAS_Complete(m, ct) + } + + @tailrec private def GCAS_Complete(m: MainNode[K, V], ct: TrieMap[K, V]): MainNode[K, V] = if (m eq null) null else { + // complete the GCAS + val prev = /*READ*/m.prev + val ctr = ct.readRoot(abort = true) + + prev match { + case null => + m + case fn: FailedNode[_, _] => // try to commit to previous value + if (CAS(m, fn.prev)) fn.prev + else GCAS_Complete(/*READ*/mainnode, ct) + case vn: MainNode[_, _] => + // Assume that you've read the root from the generation G. + // Assume that the snapshot algorithm is correct. + // ==> you can only reach nodes in generations <= G. + // ==> `gen` is <= G. + // We know that `ctr.gen` is >= G. + // ==> if `ctr.gen` = `gen` then they are both equal to G. + // ==> otherwise, we know that either `ctr.gen` > G, `gen` < G, + // or both + if ((ctr.gen eq gen) && ct.nonReadOnly) { + // try to commit + if (m.CAS_PREV(prev, null)) m + else GCAS_Complete(m, ct) + } else { + // try to abort + m.CAS_PREV(prev, new FailedNode(prev)) + GCAS_Complete(/*READ*/mainnode, ct) + } + } + } + + def GCAS(old: MainNode[K, V], n: MainNode[K, V], ct: TrieMap[K, V]): Boolean = { + n.WRITE_PREV(old) + if (CAS(old, n)) { + GCAS_Complete(n, ct) + /*READ*/n.prev eq null + } else false + } + + private def equal(k1: K, k2: K, ct: TrieMap[K, V]) = ct.equality.equiv(k1, k2) + + private def inode(cn: MainNode[K, V]) = { + val nin = new INode[K, V](gen, equiv) + nin.WRITE(cn) + nin + } + + def copyToGen(ngen: Gen, ct: TrieMap[K, V]) = { + val nin = new INode[K, V](ngen, equiv) + val main = GCAS_READ(ct) + nin.WRITE(main) + nin + } + + /** Inserts a key value pair, overwriting the old pair if the keys match. + * + * @return true if successful, false otherwise + */ + @tailrec def rec_insert(k: K, v: V, hc: Int, lev: Int, parent: INode[K, V], startgen: Gen, ct: TrieMap[K, V]): Boolean = { + val m = GCAS_READ(ct) // use -Yinline! + + m match { + case cn: CNode[K, V] => // 1) a multiway node + val idx = (hc >>> lev) & 0x1f + val flag = 1 << idx + val bmp = cn.bitmap + val mask = flag - 1 + val pos = Integer.bitCount(bmp & mask) + if ((bmp & flag) != 0) { + // 1a) insert below + cn.array(pos) match { + case in: INode[K, V] @uc => + if (startgen eq in.gen) in.rec_insert(k, v, hc, lev + 5, this, startgen, ct) + else { + if (GCAS(cn, cn.renewed(startgen, ct), ct)) rec_insert(k, v, hc, lev, parent, startgen, ct) + else false + } + case sn: SNode[K, V] @uc => + if (sn.hc == hc && equal(sn.k, k, ct)) GCAS(cn, cn.updatedAt(pos, new SNode(sn.k, v, hc), gen), ct) + else { + val rn = if (cn.gen eq gen) cn else cn.renewed(gen, ct) + val nn = rn.updatedAt(pos, inode(CNode.dual(sn, sn.hc, new SNode(k, v, hc), hc, lev + 5, gen, equiv)), gen) + GCAS(cn, nn, ct) + } + case basicNode => throw new MatchError(basicNode) + } + } else { + val rn = if (cn.gen eq gen) cn else cn.renewed(gen, ct) + val ncnode = rn.insertedAt(pos, flag, k, v, hc, gen) + GCAS(cn, ncnode, ct) + } + case tn: TNode[K, V] => + clean(parent, ct, lev - 5) + false + case ln: LNode[K, V] => // 3) an l-node + val nn = ln.inserted(k, v) + GCAS(ln, nn, ct) + case mainNode => throw new MatchError(mainNode) + } + } + + + + /** Inserts a new key value pair, given that a specific condition is met. + * + * @param cond KEY_PRESENT_OR_ABSENT - don't care if the key was there, insert or overwrite + * KEY_ABSENT - key wasn't there, insert only, do not overwrite + * KEY_PRESENT - key was there, overwrite only, do not insert + * other value `v` - only overwrite if the current value is this + * @param fullEquals whether to use reference or full equals when comparing `v` to the current value + * @param hc the hashcode of `k` + * + * @return null if unsuccessful, Option[V] otherwise (indicating previous value bound to the key) + */ + @tailrec def rec_insertif(k: K, v: V, hc: Int, cond: AnyRef, fullEquals: Boolean, lev: Int, parent: INode[K, V], startgen: Gen, ct: TrieMap[K, V]): Option[V] = { + val m = GCAS_READ(ct) // use -Yinline! + + m match { + case cn: CNode[K, V] => // 1) a multiway node + val idx = (hc >>> lev) & 0x1f + val flag = 1 << idx + val bmp = cn.bitmap + val mask = flag - 1 + val pos = Integer.bitCount(bmp & mask) + if ((bmp & flag) != 0) { + // 1a) insert below + cn.array(pos) match { + case in: INode[K, V] @uc => + if (startgen eq in.gen) in.rec_insertif(k, v, hc, cond, fullEquals, lev + 5, this, startgen, ct) + else { + if (GCAS(cn, cn.renewed(startgen, ct), ct)) rec_insertif(k, v, hc, cond, fullEquals, lev, parent, startgen, ct) + else null + } + case sn: SNode[K, V] @uc => cond match { + case INode.KEY_PRESENT_OR_ABSENT => + if (sn.hc == hc && equal(sn.k, k, ct)) { + if (GCAS(cn, cn.updatedAt(pos, new SNode(sn.k, v, hc), gen), ct)) Some(sn.v) else null + } else { + val rn = if (cn.gen eq gen) cn else cn.renewed(gen, ct) + val nn = rn.updatedAt(pos, inode(CNode.dual(sn, sn.hc, new SNode(k, v, hc), hc, lev + 5, gen, equiv)), gen) + if (GCAS(cn, nn, ct)) None + else null + } + case INode.KEY_ABSENT => + if (sn.hc == hc && equal(sn.k, k, ct)) Some(sn.v) + else { + val rn = if (cn.gen eq gen) cn else cn.renewed(gen, ct) + val nn = rn.updatedAt(pos, inode(CNode.dual(sn, sn.hc, new SNode(k, v, hc), hc, lev + 5, gen, equiv)), gen) + if (GCAS(cn, nn, ct)) None + else null + } + case INode.KEY_PRESENT => + if (sn.hc == hc && equal(sn.k, k, ct)) { + if (GCAS(cn, cn.updatedAt(pos, new SNode(k, v, hc), gen), ct)) Some(sn.v) else null + } else None + case otherv => + if (sn.hc == hc && equal(sn.k, k, ct) && (if (fullEquals) sn.v == otherv else sn.v.asInstanceOf[AnyRef] eq otherv)) { + if (GCAS(cn, cn.updatedAt(pos, new SNode(k, v, hc), gen), ct)) Some(sn.v) else null + } else None + } + case basicNode => throw new MatchError(basicNode) + } + } else cond match { + case INode.KEY_PRESENT_OR_ABSENT | INode.KEY_ABSENT => + val rn = if (cn.gen eq gen) cn else cn.renewed(gen, ct) + val ncnode = rn.insertedAt(pos, flag, k, v, hc, gen) + if (GCAS(cn, ncnode, ct)) None else null + case INode.KEY_PRESENT => None + case otherv => None + } + case sn: TNode[K, V] => + clean(parent, ct, lev - 5) + null + case ln: LNode[K, V] => // 3) an l-node + def insertln() = { + val nn = ln.inserted(k, v) + GCAS(ln, nn, ct) + } + cond match { + case INode.KEY_PRESENT_OR_ABSENT => + val optv = ln.get(k) + if (insertln()) optv else null + case INode.KEY_ABSENT => + ln.get(k) match { + case None => if (insertln()) None else null + case optv => optv + } + case INode.KEY_PRESENT => + ln.get(k) match { + case Some(v0) => if (insertln()) Some(v0) else null + case None => None + } + case otherv => + ln.get(k) match { + case Some(v0) if (if (fullEquals) v0 == otherv else v0.asInstanceOf[AnyRef] eq otherv) => + if (insertln()) Some(otherv.asInstanceOf[V]) else null + case _ => None + } + } + case mainNode => throw new MatchError(mainNode) + } + } + + /** Looks up the value associated with the key. + * + * @param hc the hashcode of `k` + * + * @return NO_SUCH_ELEMENT_SENTINEL if no value has been found, RESTART if the operation wasn't successful, + * or any other value otherwise + */ + @tailrec def rec_lookup(k: K, hc: Int, lev: Int, parent: INode[K, V], startgen: Gen, ct: TrieMap[K, V]): AnyRef = { + val m = GCAS_READ(ct) // use -Yinline! + + m match { + case cn: CNode[K, V] => // 1) a multinode + val idx = (hc >>> lev) & 0x1f + val flag = 1 << idx + val bmp = cn.bitmap + if ((bmp & flag) == 0) NO_SUCH_ELEMENT_SENTINEL // 1a) bitmap shows no binding + else { // 1b) bitmap contains a value - descend + val pos = if (bmp == 0xffffffff) idx else Integer.bitCount(bmp & (flag - 1)) + val sub = cn.array(pos) + sub match { + case in: INode[K, V] @uc => + if (ct.isReadOnly || (startgen eq in.gen)) in.rec_lookup(k, hc, lev + 5, this, startgen, ct) + else { + if (GCAS(cn, cn.renewed(startgen, ct), ct)) rec_lookup(k, hc, lev, parent, startgen, ct) + else RESTART + } + case sn: SNode[K, V] @uc => // 2) singleton node + if (sn.hc == hc && equal(sn.k, k, ct)) sn.v.asInstanceOf[AnyRef] + else NO_SUCH_ELEMENT_SENTINEL + case basicNode => throw new MatchError(basicNode) + } + } + case tn: TNode[_, _] => // 3) non-live node + def cleanReadOnly(tn: TNode[K, V]) = if (ct.nonReadOnly) { + clean(parent, ct, lev - 5) + RESTART + } else { + if (tn.hc == hc && tn.k == k) tn.v.asInstanceOf[AnyRef] + else NO_SUCH_ELEMENT_SENTINEL + } + cleanReadOnly(tn) + case ln: LNode[K, V] => // 5) an l-node + ln.get(k).asInstanceOf[Option[AnyRef]].getOrElse(NO_SUCH_ELEMENT_SENTINEL) + case mainNode => throw new MatchError(mainNode) + } + } + + /** Removes the key associated with the given value. + * + * @param hc the hashcode of `k` + * + * @param removalPolicy policy deciding whether to remove `k` based on `v` and the + * current value associated with `k` (Always, FullEquals, or ReferenceEq) + * + * @return null if not successful, an Option[V] indicating the previous value otherwise + */ + def rec_remove( + k: K, + v: V, + removalPolicy: Int, + hc: Int, + lev: Int, + parent: INode[K, V], + startgen: Gen, + ct: TrieMap[K, V]): Option[V] = { + + val m = GCAS_READ(ct) // use -Yinline! + + m match { + case cn: CNode[K, V] => + val idx = (hc >>> lev) & 0x1f + val bmp = cn.bitmap + val flag = 1 << idx + if ((bmp & flag) == 0) None + else { + val pos = Integer.bitCount(bmp & (flag - 1)) + val sub = cn.array(pos) + val res = sub match { + case in: INode[K, V] @uc => + if (startgen eq in.gen) in.rec_remove(k, v, removalPolicy, hc, lev + 5, this, startgen, ct) + else { + if (GCAS(cn, cn.renewed(startgen, ct), ct)) rec_remove(k, v, removalPolicy, hc, lev, parent, startgen, ct) + else null + } + case sn: SNode[K, V] @uc => + if (sn.hc == hc && equal(sn.k, k, ct) && RemovalPolicy.shouldRemove(removalPolicy)(sn.v, v)) { + val ncn = cn.removedAt(pos, flag, gen).toContracted(lev) + if (GCAS(cn, ncn, ct)) Some(sn.v) else null + } else None + case basicNode => throw new MatchError(basicNode) + } + + if (res == None || (res eq null)) res + else { + @tailrec def cleanParent(nonlive: AnyRef): Unit = { + val pm = parent.GCAS_READ(ct) + pm match { + case cn: CNode[K, V] => + val idx = (hc >>> (lev - 5)) & 0x1f + val bmp = cn.bitmap + val flag = 1 << idx + if ((bmp & flag) == 0) {} // somebody already removed this i-node, we're done + else { + val pos = Integer.bitCount(bmp & (flag - 1)) + val sub = cn.array(pos) + if (sub eq this) (nonlive: @uc) match { + case tn: TNode[K, V] @uc => + val ncn = cn.updatedAt(pos, tn.copyUntombed, gen).toContracted(lev - 5) + if (!parent.GCAS(cn, ncn, ct)) + if (ct.readRoot().gen == startgen) cleanParent(nonlive) + } + } + case _ => // parent is no longer a cnode, we're done + } + } + + if (parent ne null) { // never tomb at root + val n = GCAS_READ(ct) + if (n.isInstanceOf[TNode[_, _]]) + cleanParent(n) + } + + res + } + } + case tn: TNode[K, V] => + clean(parent, ct, lev - 5) + null + case ln: LNode[K, V] => + if (removalPolicy == RemovalPolicy.Always) { + val optv = ln.get(k) + val nn = ln.removed(k, ct) + if (GCAS(ln, nn, ct)) optv else null + } else ln.get(k) match { + case optv @ Some(v0) if RemovalPolicy.shouldRemove(removalPolicy)(v, v0) => + val nn = ln.removed(k, ct) + if (GCAS(ln, nn, ct)) optv else null + case _ => None + } + case mainNode => throw new MatchError(mainNode) + } + } + + private def clean(nd: INode[K, V], ct: TrieMap[K, V], lev: Int): Unit = { + val m = nd.GCAS_READ(ct) + m match { + case cn: CNode[K, V] => nd.GCAS(cn, cn.toCompressed(ct, lev, gen), ct) + case _ => + } + } + + def isNullInode(ct: TrieMap[K, V]) = GCAS_READ(ct) eq null + + def cachedSize(ct: TrieMap[K, V]): Int = + GCAS_READ(ct).cachedSize(ct) + + def knownSize(ct: TrieMap[K, V]): Int = + GCAS_READ(ct).knownSize() + + /* this is a quiescent method! */ + def string(lev: Int) = "%sINode -> %s".format(" " * lev, mainnode match { + case null => "" + case tn: TNode[_, _] => "TNode(%s, %s, %d, !)".format(tn.k, tn.v, tn.hc) + case cn: CNode[_, _] => cn.string(lev) + case ln: LNode[_, _] => ln.string(lev) + case x => "".format(x) + }) + +} + + +private[concurrent] object INode { + //////////////////////////////////////////////////////////////////////////////////////////////////// + // Arguments for `cond` argument in TrieMap#rec_insertif + //////////////////////////////////////////////////////////////////////////////////////////////////// + final val KEY_PRESENT = new AnyRef + final val KEY_ABSENT = new AnyRef + final val KEY_PRESENT_OR_ABSENT = new AnyRef + + def newRootNode[K, V](equiv: Equiv[K]) = { + val gen = new Gen + val cn = new CNode[K, V](0, new Array(0), gen) + new INode[K, V](cn, gen, equiv) + } +} + + +private[concurrent] final class FailedNode[K, V](p: MainNode[K, V]) extends MainNode[K, V] { + WRITE_PREV(p) + + def string(lev: Int): Nothing = throw new UnsupportedOperationException + + def cachedSize(ct: AnyRef): Int = throw new UnsupportedOperationException + + def knownSize: Int = throw new UnsupportedOperationException + + override def toString = "FailedNode(%s)".format(p) +} + + +private[concurrent] trait KVNode[K, V] { + def kvPair: (K, V) +} + + +private[collection] final class SNode[K, V](final val k: K, final val v: V, final val hc: Int) + extends BasicNode with KVNode[K, V] { + def copy = new SNode(k, v, hc) + def copyTombed = new TNode(k, v, hc) + def copyUntombed = new SNode(k, v, hc) + def kvPair = (k, v) + def string(lev: Int) = (" " * lev) + "SNode(%s, %s, %x)".format(k, v, hc) +} + +// Tomb Node, used to ensure proper ordering during removals +private[collection] final class TNode[K, V](final val k: K, final val v: V, final val hc: Int) + extends MainNode[K, V] with KVNode[K, V] { + def copy = new TNode(k, v, hc) + def copyTombed = new TNode(k, v, hc) + def copyUntombed = new SNode(k, v, hc) + def kvPair = (k, v) + def cachedSize(ct: AnyRef): Int = 1 + def knownSize: Int = 1 + def string(lev: Int) = (" " * lev) + "TNode(%s, %s, %x, !)".format(k, v, hc) +} + +// List Node, leaf node that handles hash collisions +private[collection] final class LNode[K, V](val entries: List[(K, V)], equiv: Equiv[K]) + extends MainNode[K, V] { + + def this(k: K, v: V, equiv: Equiv[K]) = this((k -> v) :: Nil, equiv) + + def this(k1: K, v1: V, k2: K, v2: V, equiv: Equiv[K]) = + this(if (equiv.equiv(k1, k2)) (k2 -> v2) :: Nil else (k1 -> v1) :: (k2 -> v2) :: Nil, equiv) + + def inserted(k: K, v: V) = { + var k0: K = k + @tailrec + def remove(elems: List[(K, V)], acc: List[(K, V)]): List[(K, V)] = { + if (elems.isEmpty) acc + else if (equiv.equiv(elems.head._1, k)) { + k0 = elems.head._1 + acc ::: elems.tail + } else remove(elems.tail, elems.head :: acc) + } + val e = remove(entries, Nil) + new LNode((k0 -> v) :: e, equiv) + } + + def removed(k: K, ct: TrieMap[K, V]): MainNode[K, V] = { + val updmap = entries.filterNot(entry => equiv.equiv(entry._1, k)) + if (updmap.sizeIs > 1) new LNode(updmap, equiv) + else { + val (k, v) = updmap.iterator.next() + new TNode(k, v, ct.computeHash(k)) // create it tombed so that it gets compressed on subsequent accesses + } + } + + def get(k: K): Option[V] = entries.find(entry => equiv.equiv(entry._1, k)).map(_._2) + + def cachedSize(ct: AnyRef): Int = entries.size + + def knownSize: Int = -1 // shouldn't ever be empty, and the size of a list is not known + + def string(lev: Int) = (" " * lev) + "LNode(%s)".format(entries.mkString(", ")) + +} + +// Ctrie Node, contains bitmap and array of references to branch nodes +private[collection] final class CNode[K, V](val bitmap: Int, val array: Array[BasicNode], val gen: Gen) extends CNodeBase[K, V] { + // this should only be called from within read-only snapshots + def cachedSize(ct: AnyRef): Int = { + val currsz = READ_SIZE() + if (currsz != -1) currsz + else { + val sz = computeSize(ct.asInstanceOf[TrieMap[K, V]]) + while (READ_SIZE() == -1) CAS_SIZE(-1, sz) + READ_SIZE() + } + } + + def knownSize: Int = READ_SIZE() // this should only ever return -1 if unknown + + // lends itself towards being parallelizable by choosing + // a random starting offset in the array + // => if there are concurrent size computations, they start + // at different positions, so they are more likely to + // to be independent + private def computeSize(ct: TrieMap[K, V]): Int = { + var i = 0 + var sz = 0 + val offset = + if (array.length > 0) + //util.Random.nextInt(array.length) /* <-- benchmarks show that this causes observable contention */ + java.util.concurrent.ThreadLocalRandom.current.nextInt(0, array.length) + else 0 + while (i < array.length) { + val pos = (i + offset) % array.length + array(pos) match { + case sn: SNode[_, _] => sz += 1 + case in: INode[K, V] @uc => sz += in.cachedSize(ct) + case basicNode => throw new MatchError(basicNode) + } + i += 1 + } + sz + } + + def updatedAt(pos: Int, nn: BasicNode, gen: Gen) = { + val len = array.length + val narr = new Array[BasicNode](len) + Array.copy(array, 0, narr, 0, len) + narr(pos) = nn + new CNode[K, V](bitmap, narr, gen) + } + + def removedAt(pos: Int, flag: Int, gen: Gen) = { + val arr = array + val len = arr.length + val narr = new Array[BasicNode](len - 1) + Array.copy(arr, 0, narr, 0, pos) + Array.copy(arr, pos + 1, narr, pos, len - pos - 1) + new CNode[K, V](bitmap ^ flag, narr, gen) + } + + def insertedAt(pos: Int, flag: Int, k: K, v: V, hc: Int, gen: Gen) = { + val len = array.length + val bmp = bitmap + val narr = new Array[BasicNode](len + 1) + Array.copy(array, 0, narr, 0, pos) + narr(pos) = new SNode(k, v, hc) + Array.copy(array, pos, narr, pos + 1, len - pos) + new CNode[K, V](bmp | flag, narr, gen) + } + + /** Returns a copy of this cnode such that all the i-nodes below it are copied + * to the specified generation `ngen`. + */ + def renewed(ngen: Gen, ct: TrieMap[K, V]) = { + var i = 0 + val arr = array + val len = arr.length + val narr = new Array[BasicNode](len) + while (i < len) { + arr(i) match { + case in: INode[K, V] @uc => narr(i) = in.copyToGen(ngen, ct) + case bn: BasicNode => narr(i) = bn + } + i += 1 + } + new CNode[K, V](bitmap, narr, ngen) + } + + private def resurrect(inode: INode[K, V], inodemain: AnyRef): BasicNode = inodemain match { + case tn: TNode[_, _] => tn.copyUntombed + case _ => inode + } + + def toContracted(lev: Int): MainNode[K, V] = if (array.length == 1 && lev > 0) array(0) match { + case sn: SNode[K, V] @uc => sn.copyTombed + case _ => this + } else this + + // - if the branching factor is 1 for this CNode, and the child + // is a tombed SNode, returns its tombed version + // - otherwise, if there is at least one non-null node below, + // returns the version of this node with at least some null-inodes + // removed (those existing when the op began) + // - if there are only null-i-nodes below, returns null + def toCompressed(ct: TrieMap[K, V], lev: Int, gen: Gen) = { + val bmp = bitmap + var i = 0 + val arr = array + val tmparray = new Array[BasicNode](arr.length) + while (i < arr.length) { // construct new bitmap + val sub = arr(i) + sub match { + case in: INode[K, V] @uc => + val inodemain = in.gcasRead(ct) + assert(inodemain ne null) + tmparray(i) = resurrect(in, inodemain) + case sn: SNode[K, V] @uc => + tmparray(i) = sn + case basicNode => throw new MatchError(basicNode) + } + i += 1 + } + + new CNode[K, V](bmp, tmparray, gen).toContracted(lev) + } + + def string(lev: Int): String = "CNode %x\n%s".format(bitmap, array.map(_.string(lev + 1)).mkString("\n")) + + override def toString = { + def elems: Seq[String] = array.flatMap { + case sn: SNode[K, V] @uc => Iterable.single(sn.kvPair._2.toString) + case in: INode[K, V] @uc => Iterable.single(augmentString(in.toString).drop(14) + "(" + in.gen + ")") + case basicNode => throw new MatchError(basicNode) + } + f"CNode(sz: ${elems.size}%d; ${elems.sorted.mkString(", ")})" + } +} + +private[concurrent] object CNode { + + def dual[K, V](x: SNode[K, V], xhc: Int, y: SNode[K, V], yhc: Int, lev: Int, gen: Gen, equiv: Equiv[K]): MainNode[K, V] = if (lev < 35) { + val xidx = (xhc >>> lev) & 0x1f + val yidx = (yhc >>> lev) & 0x1f + val bmp = (1 << xidx) | (1 << yidx) + if (xidx == yidx) { + val subinode = new INode[K, V](gen, equiv)//(TrieMap.inodeupdater) + subinode.mainnode = dual(x, xhc, y, yhc, lev + 5, gen, equiv) + new CNode(bmp, Array(subinode), gen) + } else { + if (xidx < yidx) new CNode(bmp, Array(x, y), gen) + else new CNode(bmp, Array(y, x), gen) + } + } else { + new LNode(x.k, x.v, y.k, y.v, equiv) + } + +} + + +private[concurrent] case class RDCSS_Descriptor[K, V](old: INode[K, V], expectedmain: MainNode[K, V], nv: INode[K, V]) { + @volatile var committed = false +} + + +/** A concurrent hash-trie or TrieMap is a concurrent thread-safe lock-free + * implementation of a hash array mapped trie. It is used to implement the + * concurrent map abstraction. It has particularly scalable concurrent insert + * and remove operations and is memory-efficient. It supports O(1), atomic, + * lock-free snapshots which are used to implement linearizable lock-free size, + * iterator and clear operations. The cost of evaluating the (lazy) snapshot is + * distributed across subsequent updates, thus making snapshot evaluation horizontally scalable. + * + * For details, see: [[http://lampwww.epfl.ch/~prokopec/ctries-snapshot.pdf]] + */ +@SerialVersionUID(-5212455458703321708L) +final class TrieMap[K, V] private (r: AnyRef, rtupd: AtomicReferenceFieldUpdater[TrieMap[K, V], AnyRef], hashf: Hashing[K], ef: Equiv[K]) + extends scala.collection.mutable.AbstractMap[K, V] + with scala.collection.concurrent.Map[K, V] + with scala.collection.mutable.MapOps[K, V, TrieMap, TrieMap[K, V]] + with scala.collection.MapFactoryDefaults[K, V, TrieMap, mutable.Iterable] + with DefaultSerializable { + + private[this] var hashingobj = if (hashf.isInstanceOf[Hashing.Default[_]]) new TrieMap.MangledHashing[K] else hashf + private[this] var equalityobj = ef + @transient + private[this] var rootupdater = rtupd + def hashing = hashingobj + def equality = equalityobj + @volatile private var root = r + + def this(hashf: Hashing[K], ef: Equiv[K]) = this( + INode.newRootNode(ef), + AtomicReferenceFieldUpdater.newUpdater(classOf[TrieMap[K, V]], classOf[AnyRef], "root"), + hashf, + ef + ) + + def this() = this(Hashing.default, Equiv.universal) + + override def mapFactory: MapFactory[TrieMap] = TrieMap + + /* internal methods */ + + private def writeObject(out: java.io.ObjectOutputStream): Unit = { + out.writeObject(hashingobj) + out.writeObject(equalityobj) + + val it = iterator + while (it.hasNext) { + val (k, v) = it.next() + out.writeObject(k) + out.writeObject(v) + } + out.writeObject(TrieMapSerializationEnd) + } + + private def readObject(in: java.io.ObjectInputStream): Unit = { + root = INode.newRootNode(equality) + rootupdater = AtomicReferenceFieldUpdater.newUpdater(classOf[TrieMap[K, V]], classOf[AnyRef], "root") + + hashingobj = in.readObject().asInstanceOf[Hashing[K]] + equalityobj = in.readObject().asInstanceOf[Equiv[K]] + + var obj: AnyRef = in.readObject() + + while (obj != TrieMapSerializationEnd) { + obj = in.readObject() + if (obj != TrieMapSerializationEnd) { + val k = obj.asInstanceOf[K] + val v = in.readObject().asInstanceOf[V] + update(k, v) + } + } + } + + private def CAS_ROOT(ov: AnyRef, nv: AnyRef) = rootupdater.compareAndSet(this, ov, nv) + + private[collection] def readRoot(abort: Boolean = false): INode[K, V] = RDCSS_READ_ROOT(abort) + + private[concurrent] def RDCSS_READ_ROOT(abort: Boolean = false): INode[K, V] = { + val r = /*READ*/root + r match { + case in: INode[K, V] @uc => in + case desc: RDCSS_Descriptor[K, V] @uc => RDCSS_Complete(abort) + case x => throw new MatchError(x) + } + } + + @tailrec private def RDCSS_Complete(abort: Boolean): INode[K, V] = { + val v = /*READ*/root + v match { + case in: INode[K, V] @uc => in + case desc: RDCSS_Descriptor[K, V] @uc => + val RDCSS_Descriptor(ov, exp, nv) = desc + if (abort) { + if (CAS_ROOT(desc, ov)) ov + else RDCSS_Complete(abort) + } else { + val oldmain = ov.gcasRead(this) + if (oldmain eq exp) { + if (CAS_ROOT(desc, nv)) { + desc.committed = true + nv + } else RDCSS_Complete(abort) + } else { + if (CAS_ROOT(desc, ov)) ov + else RDCSS_Complete(abort) + } + } + case x => throw new MatchError(x) + } + } + + private def RDCSS_ROOT(ov: INode[K, V], expectedmain: MainNode[K, V], nv: INode[K, V]): Boolean = { + val desc = RDCSS_Descriptor(ov, expectedmain, nv) + if (CAS_ROOT(ov, desc)) { + RDCSS_Complete(abort = false) + /*READ*/desc.committed + } else false + } + + @tailrec private def inserthc(k: K, hc: Int, v: V): Unit = { + val r = RDCSS_READ_ROOT() + if (!r.rec_insert(k, v, hc, 0, null, r.gen, this)) inserthc(k, hc, v) + } + + @tailrec private def insertifhc(k: K, hc: Int, v: V, cond: AnyRef, fullEquals: Boolean): Option[V] = { + val r = RDCSS_READ_ROOT() + + val ret = r.rec_insertif(k, v, hc, cond, fullEquals, 0, null, r.gen, this) + if (ret eq null) insertifhc(k, hc, v, cond, fullEquals) + else ret + } + + /** Finds the value associated with this key + * + * @param k the key to look up + * @param hc the hashcode of `k` + * + * @return the value: V associated with `k`, if it exists. Otherwise, INodeBase.NO_SUCH_ELEMENT_SENTINEL + */ + @tailrec private def lookuphc(k: K, hc: Int): AnyRef = { + val r = RDCSS_READ_ROOT() + val res = r.rec_lookup(k, hc, 0, null, r.gen, this) + if (res eq INodeBase.RESTART) lookuphc(k, hc) + else res + } + + /** Removes a key-value pair from the map + * + * @param k the key to remove + * @param v the value compare with the value found associated with the key + * @param removalPolicy policy deciding whether to remove `k` based on `v` and the + * current value associated with `k` (Always, FullEquals, or ReferenceEq) + * @return an Option[V] indicating the previous value + */ + @tailrec private def removehc(k: K, v: V, removalPolicy: Int, hc: Int): Option[V] = { + val r = RDCSS_READ_ROOT() + val res = r.rec_remove(k, v, removalPolicy, hc, 0, null, r.gen, this) + if (res ne null) res + else removehc(k, v, removalPolicy, hc) + } + + + def string = RDCSS_READ_ROOT().string(0) + + /* public methods */ + + def isReadOnly = rootupdater eq null + + def nonReadOnly = rootupdater ne null + + /** Returns a snapshot of this TrieMap. + * This operation is lock-free and linearizable. + * + * The snapshot is lazily updated - the first time some branch + * in the snapshot or this TrieMap are accessed, they are rewritten. + * This means that the work of rebuilding both the snapshot and this + * TrieMap is distributed across all the threads doing updates or accesses + * subsequent to the snapshot creation. + */ + @tailrec def snapshot(): TrieMap[K, V] = { + val r = RDCSS_READ_ROOT() + val expmain = r.gcasRead(this) + if (RDCSS_ROOT(r, expmain, r.copyToGen(new Gen, this))) new TrieMap(r.copyToGen(new Gen, this), rootupdater, hashing, equality) + else snapshot() + } + + /** Returns a read-only snapshot of this TrieMap. + * This operation is lock-free and linearizable. + * + * The snapshot is lazily updated - the first time some branch + * of this TrieMap are accessed, it is rewritten. The work of creating + * the snapshot is thus distributed across subsequent updates + * and accesses on this TrieMap by all threads. + * Note that the snapshot itself is never rewritten unlike when calling + * the `snapshot` method, but the obtained snapshot cannot be modified. + * + * This method is used by other methods such as `size` and `iterator`. + */ + @tailrec def readOnlySnapshot(): scala.collection.Map[K, V] = { + val r = RDCSS_READ_ROOT() + val expmain = r.gcasRead(this) + if (RDCSS_ROOT(r, expmain, r.copyToGen(new Gen, this))) new TrieMap(r, null, hashing, equality) + else readOnlySnapshot() + } + + @tailrec override def clear(): Unit = { + val r = RDCSS_READ_ROOT() + if (!RDCSS_ROOT(r, r.gcasRead(this), INode.newRootNode[K, V](equality))) clear() + } + + def computeHash(k: K) = hashingobj.hash(k) + + @deprecated("Use getOrElse(k, null) instead.", "2.13.0") + def lookup(k: K): V = { + val hc = computeHash(k) + val lookupRes = lookuphc(k, hc) + val res = if (lookupRes == INodeBase.NO_SUCH_ELEMENT_SENTINEL) null else lookupRes + res.asInstanceOf[V] + } + + override def apply(k: K): V = { + val hc = computeHash(k) + val res = lookuphc(k, hc) + if (res eq INodeBase.NO_SUCH_ELEMENT_SENTINEL) throw new NoSuchElementException + else res.asInstanceOf[V] + } + + def get(k: K): Option[V] = { + val hc = computeHash(k) + val res = lookuphc(k, hc) + if (res eq INodeBase.NO_SUCH_ELEMENT_SENTINEL) None else Some(res).asInstanceOf[Option[V]] + } + + override def put(key: K, value: V): Option[V] = { + val hc = computeHash(key) + insertifhc(key, hc, value, INode.KEY_PRESENT_OR_ABSENT, fullEquals = false /* unused */) + } + + override def update(k: K, v: V): Unit = { + val hc = computeHash(k) + inserthc(k, hc, v) + } + + def addOne(kv: (K, V)) = { + update(kv._1, kv._2) + this + } + + override def remove(k: K): Option[V] = { + val hc = computeHash(k) + removehc(k = k, v = null.asInstanceOf[V], RemovalPolicy.Always, hc = hc) + } + + def subtractOne(k: K) = { + remove(k) + this + } + + def putIfAbsent(k: K, v: V): Option[V] = { + val hc = computeHash(k) + insertifhc(k, hc, v, INode.KEY_ABSENT, fullEquals = false /* unused */) + } + + // TODO once computeIfAbsent is added to concurrent.Map, + // move the comment there and tweak the 'at most once' part + /** If the specified key is not already in the map, computes its value using + * the given thunk `op` and enters it into the map. + * + * If the specified mapping function throws an exception, + * that exception is rethrown. + * + * Note: This method will invoke op at most once. + * However, `op` may be invoked without the result being added to the map if + * a concurrent process is also trying to add a value corresponding to the + * same key `k`. + * + * @param k the key to modify + * @param op the expression that computes the value + * @return the newly added value + */ + override def getOrElseUpdate(k: K, op: => V): V = { + val hc = computeHash(k) + lookuphc(k, hc) match { + case INodeBase.NO_SUCH_ELEMENT_SENTINEL => + val v = op + insertifhc(k, hc, v, INode.KEY_ABSENT, fullEquals = false /* unused */) match { + case Some(oldValue) => oldValue + case None => v + } + case oldValue => oldValue.asInstanceOf[V] + } + } + + def remove(k: K, v: V): Boolean = { + val hc = computeHash(k) + removehc(k, v, RemovalPolicy.FullEquals, hc).nonEmpty + } + + override private[collection] def removeRefEq(k: K, v: V): Boolean = { + val hc = computeHash(k) + removehc(k, v, RemovalPolicy.ReferenceEq, hc).nonEmpty + } + + def replace(k: K, oldvalue: V, newvalue: V): Boolean = { + val hc = computeHash(k) + insertifhc(k, hc, newvalue, oldvalue.asInstanceOf[AnyRef], fullEquals = true).nonEmpty + } + + override private[collection] def replaceRefEq(k: K, oldValue: V, newValue: V): Boolean = { + val hc = computeHash(k) + insertifhc(k, hc, newValue, oldValue.asInstanceOf[AnyRef], fullEquals = false).nonEmpty + } + + def replace(k: K, v: V): Option[V] = { + val hc = computeHash(k) + insertifhc(k, hc, v, INode.KEY_PRESENT, fullEquals = false /* unused */) + } + + def iterator: Iterator[(K, V)] = { + if (nonReadOnly) readOnlySnapshot().iterator + else new TrieMapIterator(0, this) + } + + //////////////////////////////////////////////////////////////////////////// + // + // scala/bug#10177 These methods need overrides as the inherited implementations + // call `.iterator` more than once, which doesn't guarantee a coherent + // view of the data if there is a concurrent writer + // Note that the we don't need overrides for keysIterator or valuesIterator + // TrieMapTest validates the behaviour. + override def values: Iterable[V] = { + if (nonReadOnly) readOnlySnapshot().values + else super.values + } + override def keySet: Set[K] = { + if (nonReadOnly) readOnlySnapshot().keySet + else super.keySet + } + + override def view: MapView[K, V] = if (nonReadOnly) readOnlySnapshot().view else super.view + + @deprecated("Use .view.filterKeys(f). A future version will include a strict version of this method (for now, .view.filterKeys(p).toMap).", "2.13.0") + override def filterKeys(p: K => Boolean): collection.MapView[K, V] = view.filterKeys(p) + + @deprecated("Use .view.mapValues(f). A future version will include a strict version of this method (for now, .view.mapValues(f).toMap).", "2.13.0") + override def mapValues[W](f: V => W): collection.MapView[K, W] = view.mapValues(f) + // END extra overrides + /////////////////////////////////////////////////////////////////// + + override def size: Int = + if (nonReadOnly) readOnlySnapshot().size + else RDCSS_READ_ROOT().cachedSize(this) + override def knownSize: Int = + if (nonReadOnly) -1 + else RDCSS_READ_ROOT().knownSize(this) + override def isEmpty: Boolean = + (if (nonReadOnly) readOnlySnapshot() else this).sizeIs == 0 // sizeIs checks knownSize + override protected[this] def className = "TrieMap" + + override def lastOption: Option[(K, V)] = if (isEmpty) None else Try(last).toOption +} + + +@SerialVersionUID(3L) +object TrieMap extends MapFactory[TrieMap] { + + def empty[K, V]: TrieMap[K, V] = new TrieMap[K, V] + + def from[K, V](it: IterableOnce[(K, V)]): TrieMap[K, V] = new TrieMap[K, V]() ++= it + + def newBuilder[K, V]: mutable.GrowableBuilder[(K, V), TrieMap[K, V]] = new GrowableBuilder(empty[K, V]) + + @transient + val inodeupdater: AtomicReferenceFieldUpdater[INodeBase[_, _], MainNode[_, _]] = AtomicReferenceFieldUpdater.newUpdater(classOf[INodeBase[_, _]], classOf[MainNode[_, _]], "mainnode") + + class MangledHashing[K] extends Hashing[K] { + def hash(k: K): Int = scala.util.hashing.byteswap32(k.##) + } + + private[concurrent] object RemovalPolicy { + final val Always = 0 + final val FullEquals = 1 + final val ReferenceEq = 2 + + def shouldRemove[V](removalPolicy: Int)(a: V, b: V): Boolean = + removalPolicy match { + case Always => true + case FullEquals => a == b + case ReferenceEq => a.asInstanceOf[AnyRef] eq b.asInstanceOf[AnyRef] + } + } +} + +// non-final as an extension point for parallel collections +private[collection] class TrieMapIterator[K, V](var level: Int, private var ct: TrieMap[K, V], mustInit: Boolean = true) extends AbstractIterator[(K, V)] { + private val stack = new Array[Array[BasicNode]](7) + private val stackpos = new Array[Int](7) + private var depth = -1 + private var subiter: Iterator[(K, V)] = null + private var current: KVNode[K, V] = null + + if (mustInit) initialize() + + def hasNext = (current ne null) || (subiter ne null) + + def next() = if (hasNext) { + var r: (K, V) = null + if (subiter ne null) { + r = subiter.next() + checkSubiter() + } else { + r = current.kvPair + advance() + } + r + } else Iterator.empty.next() + + private def readin(in: INode[K, V]) = in.gcasRead(ct) match { + case cn: CNode[K, V] => + depth += 1 + stack(depth) = cn.array + stackpos(depth) = -1 + advance() + case tn: TNode[K, V] => + current = tn + case ln: LNode[K, V] => + subiter = ln.entries.iterator + checkSubiter() + case null => + current = null + case mainNode => throw new MatchError(mainNode) + } + + private def checkSubiter() = if (!subiter.hasNext) { + subiter = null + advance() + } + + private def initialize(): Unit = { + assert(ct.isReadOnly) + + val r = ct.RDCSS_READ_ROOT() + readin(r) + } + + @tailrec + final def advance(): Unit = if (depth >= 0) { + val npos = stackpos(depth) + 1 + if (npos < stack(depth).length) { + stackpos(depth) = npos + stack(depth)(npos) match { + case sn: SNode[K, V] @uc => current = sn + case in: INode[K, V] @uc => readin(in) + case basicNode => throw new MatchError(basicNode) + } + } else { + depth -= 1 + advance() + } + } else current = null + + protected def newIterator(_lev: Int, _ct: TrieMap[K, V], _mustInit: Boolean): TrieMapIterator[K, V] = new TrieMapIterator[K, V](_lev, _ct, _mustInit) + + protected def dupTo(it: TrieMapIterator[K, V]): Unit = { + it.level = this.level + it.ct = this.ct + it.depth = this.depth + it.current = this.current + + // these need a deep copy + Array.copy(this.stack, 0, it.stack, 0, 7) + Array.copy(this.stackpos, 0, it.stackpos, 0, 7) + + // this one needs to be evaluated + if (this.subiter == null) it.subiter = null + else { + val lst = this.subiter.to(immutable.List) + this.subiter = lst.iterator + it.subiter = lst.iterator + } + } + + /** Returns a sequence of iterators over subsets of this iterator. + * It's used to ease the implementation of splitters for a parallel version of the TrieMap. + */ + protected def subdivide(): Seq[Iterator[(K, V)]] = if (subiter ne null) { + // the case where an LNode is being iterated + val it = newIterator(level + 1, ct, _mustInit = false) + it.depth = -1 + it.subiter = this.subiter + it.current = null + this.subiter = null + advance() + this.level += 1 + Seq(it, this) + } else if (depth == -1) { + this.level += 1 + Seq(this) + } else { + var d = 0 + while (d <= depth) { + val rem = stack(d).length - 1 - stackpos(d) + if (rem > 0) { + val (arr1, arr2) = stack(d).drop(stackpos(d) + 1).splitAt(rem / 2) + stack(d) = arr1 + stackpos(d) = -1 + val it = newIterator(level + 1, ct, _mustInit = false) + it.stack(0) = arr2 + it.stackpos(0) = -1 + it.depth = 0 + it.advance() // <-- fix it + this.level += 1 + return Seq(this, it) + } + d += 1 + } + this.level += 1 + Seq(this) + } + +} + +/** Only used for ctrie serialization. */ +@SerialVersionUID(3L) +private[concurrent] case object TrieMapSerializationEnd diff --git a/stdlib-bootstrapped/src/scala/collection/mutable/UnrolledBuffer.scala b/stdlib-bootstrapped/src/scala/collection/mutable/UnrolledBuffer.scala new file mode 100644 index 000000000000..874cbf9206d2 --- /dev/null +++ b/stdlib-bootstrapped/src/scala/collection/mutable/UnrolledBuffer.scala @@ -0,0 +1,442 @@ +/* + * Scala (https://www.scala-lang.org) + * + * Copyright EPFL and Lightbend, Inc. + * + * Licensed under Apache License 2.0 + * (http://www.apache.org/licenses/LICENSE-2.0). + * + * See the NOTICE file distributed with this work for + * additional information regarding copyright ownership. + */ + +package scala.collection +package mutable + +import scala.annotation.tailrec +import scala.collection.generic.DefaultSerializable +import scala.reflect.ClassTag +import scala.collection.immutable.Nil + +/** A buffer that stores elements in an unrolled linked list. + * + * Unrolled linked lists store elements in linked fixed size + * arrays. + * + * Unrolled buffers retain locality and low memory overhead + * properties of array buffers, but offer much more efficient + * element addition, since they never reallocate and copy the + * internal array. + * + * However, they provide `O(n/m)` complexity random access, + * where `n` is the number of elements, and `m` the size of + * internal array chunks. + * + * Ideal to use when: + * - elements are added to the buffer and then all of the + * elements are traversed sequentially + * - two unrolled buffers need to be concatenated (see `concat`) + * + * Better than singly linked lists for random access, but + * should still be avoided for such a purpose. + * + * @define coll unrolled buffer + * @define Coll `UnrolledBuffer` + * + */ +@SerialVersionUID(3L) +sealed class UnrolledBuffer[T](implicit val tag: ClassTag[T]) + extends AbstractBuffer[T] + with Buffer[T] + with Seq[T] + with SeqOps[T, UnrolledBuffer, UnrolledBuffer[T]] + with StrictOptimizedSeqOps[T, UnrolledBuffer, UnrolledBuffer[T]] + with EvidenceIterableFactoryDefaults[T, UnrolledBuffer, ClassTag] + with Builder[T, UnrolledBuffer[T]] + with DefaultSerializable { + + import UnrolledBuffer.Unrolled + + @transient private var headptr = newUnrolled + @transient private var lastptr = headptr + @transient private var sz = 0 + + private[collection] def headPtr = headptr + private[collection] def headPtr_=(head: Unrolled[T]) = headptr = head + private[collection] def lastPtr = lastptr + private[collection] def lastPtr_=(last: Unrolled[T]) = lastptr = last + private[collection] def size_=(s: Int) = sz = s + + protected def evidenceIterableFactory: UnrolledBuffer.type = UnrolledBuffer + protected def iterableEvidence: ClassTag[T] = tag + + override def iterableFactory: SeqFactory[UnrolledBuffer] = UnrolledBuffer.untagged + + protected def newUnrolled = new Unrolled[T](this) + + // The below would allow more flexible behavior without requiring inheritance + // that is risky because all the important internals are private. + // private var myLengthPolicy: Int => Int = x => x + // + // /** Specifies how the array lengths should vary. + // * + // * By default, `UnrolledBuffer` uses arrays of a fixed size. A length + // * policy can be given that changes this scheme to, for instance, an + // * exponential growth. + // * + // * @param nextLength computes the length of the next array from the length of the latest one + // */ + // def setLengthPolicy(nextLength: Int => Int): Unit = { myLengthPolicy = nextLength } + private[collection] def calcNextLength(sz: Int) = sz // myLengthPolicy(sz) + + def classTagCompanion: UnrolledBuffer.type = UnrolledBuffer + + /** Concatenates the target unrolled buffer to this unrolled buffer. + * + * The specified buffer `that` is cleared after this operation. This is + * an O(1) operation. + * + * @param that the unrolled buffer whose elements are added to this buffer + */ + def concat(that: UnrolledBuffer[T]) = { + // bind the two together + if (!lastptr.bind(that.headptr)) lastptr = that.lastPtr + + // update size + sz += that.sz + + // `that` is no longer usable, so clear it + // here we rely on the fact that `clear` allocates + // new nodes instead of modifying the previous ones + that.clear() + + // return a reference to this + this + } + + def addOne(elem: T) = { + lastptr = lastptr.append(elem) + sz += 1 + this + } + + def clear(): Unit = { + headptr = newUnrolled + lastptr = headptr + sz = 0 + } + + def iterator: Iterator[T] = new AbstractIterator[T] { + var pos: Int = -1 + var node: Unrolled[T] = headptr + scan() + + private def scan(): Unit = { + pos += 1 + while (pos >= node.size) { + pos = 0 + node = node.next + if (node eq null) return + } + } + def hasNext = node ne null + def next() = if (hasNext) { + val r = node.array(pos) + scan() + r + } else Iterator.empty.next() + } + + // this should be faster than the iterator + override def foreach[U](f: T => U) = headptr.foreach(f) + + def result() = this + + def length = sz + + override def knownSize: Int = sz + + def apply(idx: Int) = + if (idx >= 0 && idx < sz) headptr(idx) + else throw new IndexOutOfBoundsException(s"$idx is out of bounds (min 0, max ${sz-1})") + + def update(idx: Int, newelem: T) = + if (idx >= 0 && idx < sz) headptr(idx) = newelem + else throw new IndexOutOfBoundsException(s"$idx is out of bounds (min 0, max ${sz-1})") + + def mapInPlace(f: T => T): this.type = { + headptr.mapInPlace(f) + this + } + + def remove(idx: Int) = + if (idx >= 0 && idx < sz) { + sz -= 1 + headptr.remove(idx, this) + } else throw new IndexOutOfBoundsException(s"$idx is out of bounds (min 0, max ${sz-1})") + + @tailrec final def remove(idx: Int, count: Int): Unit = + if (count > 0) { + remove(idx) + remove(idx, count-1) + } + + def prepend(elem: T) = { + headptr = headptr prepend elem + sz += 1 + this + } + + def insert(idx: Int, elem: T): Unit = + insertAll(idx, elem :: Nil) + + def insertAll(idx: Int, elems: IterableOnce[T]): Unit = + if (idx >= 0 && idx <= sz) { + sz += headptr.insertAll(idx, elems, this) + } else throw new IndexOutOfBoundsException(s"$idx is out of bounds (min 0, max ${sz-1})") + + override def subtractOne(elem: T): this.type = { + if (headptr.subtractOne(elem, this)) { + sz -= 1 + } + this + } + + def patchInPlace(from: Int, patch: collection.IterableOnce[T], replaced: Int): this.type = { + remove(from, replaced) + insertAll(from, patch) + this + } + + private def writeObject(out: java.io.ObjectOutputStream): Unit = { + out.defaultWriteObject + out writeInt sz + for (elem <- this) out writeObject elem + } + + private def readObject(in: java.io.ObjectInputStream): Unit = { + in.defaultReadObject + + val num = in.readInt + + headPtr = newUnrolled + lastPtr = headPtr + sz = 0 + var i = 0 + while (i < num) { + this += in.readObject.asInstanceOf[T] + i += 1 + } + } + + override def clone(): UnrolledBuffer[T] = new UnrolledBuffer[T] ++= this + + override protected[this] def className = "UnrolledBuffer" +} + + +@SerialVersionUID(3L) +object UnrolledBuffer extends StrictOptimizedClassTagSeqFactory[UnrolledBuffer] { self => + + val untagged: SeqFactory[UnrolledBuffer] = new ClassTagSeqFactory.AnySeqDelegate(self) + + def empty[A : ClassTag]: UnrolledBuffer[A] = new UnrolledBuffer[A] + + def from[A : ClassTag](source: scala.collection.IterableOnce[A]): UnrolledBuffer[A] = newBuilder[A].addAll(source) + + def newBuilder[A : ClassTag]: UnrolledBuffer[A] = new UnrolledBuffer[A] + + final val waterline: Int = 50 + + final def waterlineDenom: Int = 100 + + @deprecated("Use waterlineDenom instead.", "2.13.0") + final val waterlineDelim: Int = waterlineDenom + + private[collection] val unrolledlength = 32 + + /** Unrolled buffer node. + */ + class Unrolled[T: ClassTag] private[collection] (var size: Int, var array: Array[T], var next: Unrolled[T], val buff: UnrolledBuffer[T] = null) { + private[collection] def this() = this(0, new Array[T](unrolledlength), null, null) + private[collection] def this(b: UnrolledBuffer[T]) = this(0, new Array[T](unrolledlength), null, b) + + private def nextlength = if (buff eq null) unrolledlength else buff.calcNextLength(array.length) + + // adds and returns itself or the new unrolled if full + @tailrec final def append(elem: T): Unrolled[T] = if (size < array.length) { + array(size) = elem + size += 1 + this + } else { + next = new Unrolled[T](0, new Array[T](nextlength), null, buff) + next append elem + } + def foreach[U](f: T => U): Unit = { + var unrolled = this + var i = 0 + while (unrolled ne null) { + val chunkarr = unrolled.array + val chunksz = unrolled.size + while (i < chunksz) { + val elem = chunkarr(i) + f(elem) + i += 1 + } + i = 0 + unrolled = unrolled.next + } + } + def mapInPlace(f: T => T): Unit = { + var unrolled = this + var i = 0 + while (unrolled ne null) { + val chunkarr = unrolled.array + val chunksz = unrolled.size + while (i < chunksz) { + val elem = chunkarr(i) + chunkarr(i) = f(elem) + i += 1 + } + i = 0 + unrolled = unrolled.next + } + } + @tailrec final def apply(idx: Int): T = + if (idx < size) array(idx) else next.apply(idx - size) + @tailrec final def update(idx: Int, newelem: T): Unit = + if (idx < size) array(idx) = newelem else next.update(idx - size, newelem) + @tailrec final def locate(idx: Int): Unrolled[T] = + if (idx < size) this else next.locate(idx - size) + def prepend(elem: T) = if (size < array.length) { + // shift the elements of the array right + // then insert the element + shiftright() + array(0) = elem + size += 1 + this + } else { + // allocate a new node and store element + // then make it point to this + val newhead = new Unrolled[T](buff) + newhead append elem + newhead.next = this + newhead + } + // shifts right assuming enough space + private def shiftright(): Unit = { + var i = size - 1 + while (i >= 0) { + array(i + 1) = array(i) + i -= 1 + } + } + // returns pointer to new last if changed + @tailrec final def remove(idx: Int, buffer: UnrolledBuffer[T]): T = + if (idx < size) { + // remove the element + // then try to merge with the next bucket + val r = array(idx) + shiftleft(idx) + size -= 1 + if (tryMergeWithNext()) buffer.lastPtr = this + r + } else next.remove(idx - size, buffer) + + @tailrec final def subtractOne(elem: T, buffer: UnrolledBuffer[T]): Boolean = { + var i = 0 + while (i < size) { + if(array(i) == elem) { + remove(i, buffer) + return true + } + i += 1 + } + if(next ne null) next.subtractOne(elem, buffer) else false + } + + // shifts left elements after `leftb` (overwrites `leftb`) + private def shiftleft(leftb: Int): Unit = { + var i = leftb + while (i < (size - 1)) { + array(i) = array(i + 1) + i += 1 + } + nullout(i, i + 1) + } + protected def tryMergeWithNext() = if (next != null && (size + next.size) < (array.length * waterline / waterlineDenom)) { + // copy the next array, then discard the next node + Array.copy(next.array, 0, array, size, next.size) + size = size + next.size + next = next.next + if (next eq null) true else false // checks if last node was thrown out + } else false + + @tailrec final def insertAll(idx: Int, t: scala.collection.IterableOnce[T], buffer: UnrolledBuffer[T]): Int = { + if (idx < size) { + // divide this node at the appropriate position and insert all into head + // update new next + val newnextnode = new Unrolled[T](0, new Array(array.length), null, buff) + Array.copy(array, idx, newnextnode.array, 0, size - idx) + newnextnode.size = size - idx + newnextnode.next = next + + // update this + nullout(idx, size) + size = idx + next = null + + // insert everything from iterable to this + var curr = this + var appended = 0 + for (elem <- t.iterator) { + curr = curr append elem + appended += 1 + } + curr.next = newnextnode + + // try to merge the last node of this with the newnextnode and fix tail pointer if needed + if (curr.tryMergeWithNext()) buffer.lastPtr = curr + else if (newnextnode.next eq null) buffer.lastPtr = newnextnode + appended + } + else if (idx == size || (next eq null)) { + var curr = this + var appended = 0 + for (elem <- t.iterator) { + curr = curr append elem + appended += 1 + } + appended + } + else next.insertAll(idx - size, t, buffer) + } + + private def nullout(from: Int, until: Int): Unit = { + var idx = from + while (idx < until) { + array(idx) = null.asInstanceOf[T] // TODO find a way to assign a default here!! + idx += 1 + } + } + + // assumes this is the last node + // `thathead` and `thatlast` are head and last node + // of the other unrolled list, respectively + def bind(thathead: Unrolled[T]) = { + assert(next eq null) + next = thathead + tryMergeWithNext() + } + + override def toString: String = + array.take(size).mkString("Unrolled@%08x".format(System.identityHashCode(this)) + "[" + size + "/" + array.length + "](", ", ", ")") + " -> " + (if (next ne null) next.toString else "") + } +} + +// This is used by scala.collection.parallel.mutable.UnrolledParArrayCombiner: +// Todo -- revisit whether inheritance is the best way to achieve this functionality +private[collection] class DoublingUnrolledBuffer[T](implicit t: ClassTag[T]) extends UnrolledBuffer[T]()(t) { + override def calcNextLength(sz: Int) = if (sz < 10000) sz * 2 else sz + protected override def newUnrolled = new UnrolledBuffer.Unrolled[T](0, new Array[T](4), null, this) +}