Implement async/batch lookup of object data
An ObjectReader implementation may be very slow for a single object,
but yet support bulk queries efficiently by batching multiple small
requests into a single larger request. This easily happens when the
reader is built on top of a database that is stored on another host,
as the network round-trip time starts to dominate the operation cost.
RevWalk, ObjectWalk, UploadPack and PackWriter are the first major
users of this new bulk interface, with the goal being to support an
efficient way to pack a repository for a fetch/clone client when the
source repository is stored in a high-latency storage system.
Processing the want/have lists is now done in bulk, to remove
the high costs associated with common ancestor negotiation.
PackWriter already performs object reuse selection in bulk, but it
now can also do the object size lookup and object counting phases
with higher efficiency. Actual object reuse, deltification, and
final output are still doing sequential lookups, making them a bit
more expensive to perform.
Change-Id: I4c966f84917482598012074c370b9831451404ee
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
14 years ago Implement async/batch lookup of object data
An ObjectReader implementation may be very slow for a single object,
but yet support bulk queries efficiently by batching multiple small
requests into a single larger request. This easily happens when the
reader is built on top of a database that is stored on another host,
as the network round-trip time starts to dominate the operation cost.
RevWalk, ObjectWalk, UploadPack and PackWriter are the first major
users of this new bulk interface, with the goal being to support an
efficient way to pack a repository for a fetch/clone client when the
source repository is stored in a high-latency storage system.
Processing the want/have lists is now done in bulk, to remove
the high costs associated with common ancestor negotiation.
PackWriter already performs object reuse selection in bulk, but it
now can also do the object size lookup and object counting phases
with higher efficiency. Actual object reuse, deltification, and
final output are still doing sequential lookups, making them a bit
more expensive to perform.
Change-Id: I4c966f84917482598012074c370b9831451404ee
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
14 years ago Implement async/batch lookup of object data
An ObjectReader implementation may be very slow for a single object,
but yet support bulk queries efficiently by batching multiple small
requests into a single larger request. This easily happens when the
reader is built on top of a database that is stored on another host,
as the network round-trip time starts to dominate the operation cost.
RevWalk, ObjectWalk, UploadPack and PackWriter are the first major
users of this new bulk interface, with the goal being to support an
efficient way to pack a repository for a fetch/clone client when the
source repository is stored in a high-latency storage system.
Processing the want/have lists is now done in bulk, to remove
the high costs associated with common ancestor negotiation.
PackWriter already performs object reuse selection in bulk, but it
now can also do the object size lookup and object counting phases
with higher efficiency. Actual object reuse, deltification, and
final output are still doing sequential lookups, making them a bit
more expensive to perform.
Change-Id: I4c966f84917482598012074c370b9831451404ee
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
14 years ago Added read/write support for pack bitmap index.
A pack bitmap index is an additional index of compressed
bitmaps of the object graph. Furthermore, a logical API of the index
functionality is included, as it is expected to be used by the
PackWriter.
Compressed bitmaps are created using the javaewah library, which is a
word-aligned compressed variant of the Java bitset class based on
run-length encoding. The library only works with positive integer
values. Thus, the maximum number of ObjectIds in a pack file that
this index can currently support is limited to Integer.MAX_VALUE.
Every ObjectId is given an integer mapping. The integer is the
position of the ObjectId in the complete ObjectId list, sorted
by offset, for the pack file. That integer is what the bitmaps
use to reference the ObjectId. Currently, the new index format can
only be used with pack files that contain a complete closure of the
object graph e.g. the result of a garbage collection.
The index file includes four bitmaps for the Git object types i.e.
commits, trees, blobs, and tags. In addition, a collection of
bitmaps keyed by an ObjectId is also included. The bitmap for each entry
in the collection represents the full closure of ObjectIds reachable
from the keyed ObjectId (including the keyed ObjectId itself). The
bitmaps are further compressed by XORing the current bitmaps against
prior bitmaps in the index, and selecting the smallest representation.
The XOR'd bitmap and offset from the current entry to the position
of the bitmap to XOR against is the actual representation of the entry
in the index file. Each entry contains one byte, which is currently
used to note whether the bitmap should be blindly reused.
Change-Id: Id328724bf6b4c8366a088233098c18643edcf40f
11 years ago Added read/write support for pack bitmap index.
A pack bitmap index is an additional index of compressed
bitmaps of the object graph. Furthermore, a logical API of the index
functionality is included, as it is expected to be used by the
PackWriter.
Compressed bitmaps are created using the javaewah library, which is a
word-aligned compressed variant of the Java bitset class based on
run-length encoding. The library only works with positive integer
values. Thus, the maximum number of ObjectIds in a pack file that
this index can currently support is limited to Integer.MAX_VALUE.
Every ObjectId is given an integer mapping. The integer is the
position of the ObjectId in the complete ObjectId list, sorted
by offset, for the pack file. That integer is what the bitmaps
use to reference the ObjectId. Currently, the new index format can
only be used with pack files that contain a complete closure of the
object graph e.g. the result of a garbage collection.
The index file includes four bitmaps for the Git object types i.e.
commits, trees, blobs, and tags. In addition, a collection of
bitmaps keyed by an ObjectId is also included. The bitmap for each entry
in the collection represents the full closure of ObjectIds reachable
from the keyed ObjectId (including the keyed ObjectId itself). The
bitmaps are further compressed by XORing the current bitmaps against
prior bitmaps in the index, and selecting the smallest representation.
The XOR'd bitmap and offset from the current entry to the position
of the bitmap to XOR against is the actual representation of the entry
in the index file. Each entry contains one byte, which is currently
used to note whether the bitmap should be blindly reused.
Change-Id: Id328724bf6b4c8366a088233098c18643edcf40f
11 years ago |
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580 |
- /*
- * Copyright (C) 2010, Google Inc.
- * and other copyright owners as documented in the project's IP log.
- *
- * This program and the accompanying materials are made available
- * under the terms of the Eclipse Distribution License v1.0 which
- * accompanies this distribution, is reproduced below, and is
- * available at http://www.eclipse.org/org/documents/edl-v10.php
- *
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or
- * without modification, are permitted provided that the following
- * conditions are met:
- *
- * - Redistributions of source code must retain the above copyright
- * notice, this list of conditions and the following disclaimer.
- *
- * - Redistributions in binary form must reproduce the above
- * copyright notice, this list of conditions and the following
- * disclaimer in the documentation and/or other materials provided
- * with the distribution.
- *
- * - Neither the name of the Eclipse Foundation, Inc. nor the
- * names of its contributors may be used to endorse or promote
- * products derived from this software without specific prior
- * written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
- * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
- * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
- * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
- * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
- * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
- * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
- * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
- * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
- * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- */
-
- package org.eclipse.jgit.lib;
-
- import java.io.IOException;
- import java.util.ArrayList;
- import java.util.Collection;
- import java.util.Iterator;
- import java.util.List;
- import java.util.Set;
-
- import org.eclipse.jgit.annotations.Nullable;
- import org.eclipse.jgit.errors.IncorrectObjectTypeException;
- import org.eclipse.jgit.errors.MissingObjectException;
- import org.eclipse.jgit.internal.storage.pack.ObjectReuseAsIs;
-
- /**
- * Reads an {@link ObjectDatabase} for a single thread.
- * <p>
- * Readers that can support efficient reuse of pack encoded objects should also
- * implement the companion interface {@link ObjectReuseAsIs}.
- */
- public abstract class ObjectReader implements AutoCloseable {
- /** Type hint indicating the caller doesn't know the type. */
- public static final int OBJ_ANY = -1;
-
- /**
- * The threshold at which a file will be streamed rather than loaded
- * entirely into memory.
- * @since 4.6
- */
- protected int streamFileThreshold;
-
- /**
- * Construct a new reader from the same data.
- * <p>
- * Applications can use this method to build a new reader from the same data
- * source, but for an different thread.
- *
- * @return a brand new reader, using the same data source.
- */
- public abstract ObjectReader newReader();
-
- /**
- * Obtain a unique abbreviation (prefix) of an object SHA-1.
- *
- * This method uses a reasonable default for the minimum length. Callers who
- * don't care about the minimum length should prefer this method.
- *
- * The returned abbreviation would expand back to the argument ObjectId when
- * passed to {@link #resolve(AbbreviatedObjectId)}, assuming no new objects
- * are added to this repository between calls.
- *
- * @param objectId
- * object identity that needs to be abbreviated.
- * @return SHA-1 abbreviation.
- * @throws IOException
- * the object store cannot be read.
- */
- public AbbreviatedObjectId abbreviate(AnyObjectId objectId)
- throws IOException {
- return abbreviate(objectId, 7);
- }
-
- /**
- * Obtain a unique abbreviation (prefix) of an object SHA-1.
- *
- * The returned abbreviation would expand back to the argument ObjectId when
- * passed to {@link #resolve(AbbreviatedObjectId)}, assuming no new objects
- * are added to this repository between calls.
- *
- * The default implementation of this method abbreviates the id to the
- * minimum length, then resolves it to see if there are multiple results.
- * When multiple results are found, the length is extended by 1 and resolve
- * is tried again.
- *
- * @param objectId
- * object identity that needs to be abbreviated.
- * @param len
- * minimum length of the abbreviated string. Must be in the range
- * [2, {@value Constants#OBJECT_ID_STRING_LENGTH}].
- * @return SHA-1 abbreviation. If no matching objects exist in the
- * repository, the abbreviation will match the minimum length.
- * @throws IOException
- * the object store cannot be read.
- */
- public AbbreviatedObjectId abbreviate(AnyObjectId objectId, int len)
- throws IOException {
- if (len == Constants.OBJECT_ID_STRING_LENGTH)
- return AbbreviatedObjectId.fromObjectId(objectId);
-
- AbbreviatedObjectId abbrev = objectId.abbreviate(len);
- Collection<ObjectId> matches = resolve(abbrev);
- while (1 < matches.size() && len < Constants.OBJECT_ID_STRING_LENGTH) {
- abbrev = objectId.abbreviate(++len);
- List<ObjectId> n = new ArrayList<ObjectId>(8);
- for (ObjectId candidate : matches) {
- if (abbrev.prefixCompare(candidate) == 0)
- n.add(candidate);
- }
- if (1 < n.size())
- matches = n;
- else
- matches = resolve(abbrev);
- }
- return abbrev;
- }
-
- /**
- * Resolve an abbreviated ObjectId to its full form.
- *
- * This method searches for an ObjectId that begins with the abbreviation,
- * and returns at least some matching candidates.
- *
- * If the returned collection is empty, no objects start with this
- * abbreviation. The abbreviation doesn't belong to this repository, or the
- * repository lacks the necessary objects to complete it.
- *
- * If the collection contains exactly one member, the abbreviation is
- * (currently) unique within this database. There is a reasonably high
- * probability that the returned id is what was previously abbreviated.
- *
- * If the collection contains 2 or more members, the abbreviation is not
- * unique. In this case the implementation is only required to return at
- * least 2 candidates to signal the abbreviation has conflicts. User
- * friendly implementations should return as many candidates as reasonably
- * possible, as the caller may be able to disambiguate further based on
- * context. However since databases can be very large (e.g. 10 million
- * objects) returning 625,000 candidates for the abbreviation "0" is simply
- * unreasonable, so implementors should draw the line at around 256 matches.
- *
- * @param id
- * abbreviated id to resolve to a complete identity. The
- * abbreviation must have a length of at least 2.
- * @return candidates that begin with the abbreviated identity.
- * @throws IOException
- * the object store cannot be read.
- */
- public abstract Collection<ObjectId> resolve(AbbreviatedObjectId id)
- throws IOException;
-
- /**
- * Does the requested object exist in this database?
- *
- * @param objectId
- * identity of the object to test for existence of.
- * @return true if the specified object is stored in this database.
- * @throws IOException
- * the object store cannot be accessed.
- */
- public boolean has(AnyObjectId objectId) throws IOException {
- return has(objectId, OBJ_ANY);
- }
-
- /**
- * Does the requested object exist in this database?
- *
- * @param objectId
- * identity of the object to test for existence of.
- * @param typeHint
- * hint about the type of object being requested, e.g.
- * {@link Constants#OBJ_BLOB}; {@link #OBJ_ANY} if the object
- * type is not known, or does not matter to the caller.
- * @return true if the specified object is stored in this database.
- * @throws IncorrectObjectTypeException
- * typeHint was not OBJ_ANY, and the object's actual type does
- * not match typeHint.
- * @throws IOException
- * the object store cannot be accessed.
- */
- public boolean has(AnyObjectId objectId, int typeHint) throws IOException {
- try {
- open(objectId, typeHint);
- return true;
- } catch (MissingObjectException notFound) {
- return false;
- }
- }
-
- /**
- * Open an object from this database.
- *
- * @param objectId
- * identity of the object to open.
- * @return a {@link ObjectLoader} for accessing the object.
- * @throws MissingObjectException
- * the object does not exist.
- * @throws IOException
- * the object store cannot be accessed.
- */
- public ObjectLoader open(AnyObjectId objectId)
- throws MissingObjectException, IOException {
- return open(objectId, OBJ_ANY);
- }
-
- /**
- * Open an object from this database.
- *
- * @param objectId
- * identity of the object to open.
- * @param typeHint
- * hint about the type of object being requested, e.g.
- * {@link Constants#OBJ_BLOB}; {@link #OBJ_ANY} if the object
- * type is not known, or does not matter to the caller.
- * @return a {@link ObjectLoader} for accessing the object.
- * @throws MissingObjectException
- * the object does not exist.
- * @throws IncorrectObjectTypeException
- * typeHint was not OBJ_ANY, and the object's actual type does
- * not match typeHint.
- * @throws IOException
- * the object store cannot be accessed.
- */
- public abstract ObjectLoader open(AnyObjectId objectId, int typeHint)
- throws MissingObjectException, IncorrectObjectTypeException,
- IOException;
-
- /**
- * Returns IDs for those commits which should be considered as shallow.
- *
- * @return IDs of shallow commits
- * @throws IOException
- */
- public abstract Set<ObjectId> getShallowCommits() throws IOException;
-
- /**
- * Asynchronous object opening.
- *
- * @param <T>
- * type of identifier being supplied.
- * @param objectIds
- * objects to open from the object store. The supplied collection
- * must not be modified until the queue has finished.
- * @param reportMissing
- * if true missing objects are reported by calling failure with a
- * MissingObjectException. This may be more expensive for the
- * implementation to guarantee. If false the implementation may
- * choose to report MissingObjectException, or silently skip over
- * the object with no warning.
- * @return queue to read the objects from.
- */
- public <T extends ObjectId> AsyncObjectLoaderQueue<T> open(
- Iterable<T> objectIds, final boolean reportMissing) {
- final Iterator<T> idItr = objectIds.iterator();
- return new AsyncObjectLoaderQueue<T>() {
- private T cur;
-
- public boolean next() throws MissingObjectException, IOException {
- if (idItr.hasNext()) {
- cur = idItr.next();
- return true;
- } else {
- return false;
- }
- }
-
- public T getCurrent() {
- return cur;
- }
-
- public ObjectId getObjectId() {
- return cur;
- }
-
- public ObjectLoader open() throws IOException {
- return ObjectReader.this.open(cur, OBJ_ANY);
- }
-
- public boolean cancel(boolean mayInterruptIfRunning) {
- return true;
- }
-
- public void release() {
- // Since we are sequential by default, we don't
- // have any state to clean up if we terminate early.
- }
- };
- }
-
- /**
- * Get only the size of an object.
- * <p>
- * The default implementation of this method opens an ObjectLoader.
- * Databases are encouraged to override this if a faster access method is
- * available to them.
- *
- * @param objectId
- * identity of the object to open.
- * @param typeHint
- * hint about the type of object being requested, e.g.
- * {@link Constants#OBJ_BLOB}; {@link #OBJ_ANY} if the object
- * type is not known, or does not matter to the caller.
- * @return size of object in bytes.
- * @throws MissingObjectException
- * the object does not exist.
- * @throws IncorrectObjectTypeException
- * typeHint was not OBJ_ANY, and the object's actual type does
- * not match typeHint.
- * @throws IOException
- * the object store cannot be accessed.
- */
- public long getObjectSize(AnyObjectId objectId, int typeHint)
- throws MissingObjectException, IncorrectObjectTypeException,
- IOException {
- return open(objectId, typeHint).getSize();
- }
-
- /**
- * Asynchronous object size lookup.
- *
- * @param <T>
- * type of identifier being supplied.
- * @param objectIds
- * objects to get the size of from the object store. The supplied
- * collection must not be modified until the queue has finished.
- * @param reportMissing
- * if true missing objects are reported by calling failure with a
- * MissingObjectException. This may be more expensive for the
- * implementation to guarantee. If false the implementation may
- * choose to report MissingObjectException, or silently skip over
- * the object with no warning.
- * @return queue to read object sizes from.
- */
- public <T extends ObjectId> AsyncObjectSizeQueue<T> getObjectSize(
- Iterable<T> objectIds, final boolean reportMissing) {
- final Iterator<T> idItr = objectIds.iterator();
- return new AsyncObjectSizeQueue<T>() {
- private T cur;
-
- private long sz;
-
- public boolean next() throws MissingObjectException, IOException {
- if (idItr.hasNext()) {
- cur = idItr.next();
- sz = getObjectSize(cur, OBJ_ANY);
- return true;
- } else {
- return false;
- }
- }
-
- public T getCurrent() {
- return cur;
- }
-
- public ObjectId getObjectId() {
- return cur;
- }
-
- public long getSize() {
- return sz;
- }
-
- public boolean cancel(boolean mayInterruptIfRunning) {
- return true;
- }
-
- public void release() {
- // Since we are sequential by default, we don't
- // have any state to clean up if we terminate early.
- }
- };
- }
-
- /**
- * Advise the reader to avoid unreachable objects.
- * <p>
- * While enabled the reader will skip over anything previously proven to be
- * unreachable. This may be dangerous in the face of concurrent writes.
- *
- * @param avoid
- * true to avoid unreachable objects.
- * @since 3.0
- */
- public void setAvoidUnreachableObjects(boolean avoid) {
- // Do nothing by default.
- }
-
- /**
- * An index that can be used to speed up ObjectWalks.
- *
- * @return the index or null if one does not exist.
- * @throws IOException
- * when the index fails to load
- * @since 3.0
- */
- public BitmapIndex getBitmapIndex() throws IOException {
- return null;
- }
-
- /**
- * @return the {@link ObjectInserter} from which this reader was created
- * using {@code inserter.newReader()}, or null if this reader was not
- * created from an inserter.
- * @since 4.4
- */
- @Nullable
- public ObjectInserter getCreatedFromInserter() {
- return null;
- }
-
- /**
- * Release any resources used by this reader.
- * <p>
- * A reader that has been released can be used again, but may need to be
- * released after the subsequent usage.
- *
- * @since 4.0
- */
- @Override
- public abstract void close();
-
- /**
- * Sets the threshold at which a file will be streamed rather than loaded
- * entirely into memory
- *
- * @param threshold
- * the new threshold
- * @since 4.6
- */
- public void setStreamFileThreshold(int threshold) {
- streamFileThreshold = threshold;
- }
-
- /**
- * Returns the threshold at which a file will be streamed rather than loaded
- * entirely into memory
- *
- * @return the threshold in bytes
- * @since 4.6
- */
- public int getStreamFileThreshold() {
- return streamFileThreshold;
- }
-
- /**
- * Wraps a delegate ObjectReader.
- *
- * @since 4.4
- */
- public static abstract class Filter extends ObjectReader {
- /**
- * @return delegate ObjectReader to handle all processing.
- * @since 4.4
- */
- protected abstract ObjectReader delegate();
-
- @Override
- public ObjectReader newReader() {
- return delegate().newReader();
- }
-
- @Override
- public AbbreviatedObjectId abbreviate(AnyObjectId objectId)
- throws IOException {
- return delegate().abbreviate(objectId);
- }
-
- @Override
- public AbbreviatedObjectId abbreviate(AnyObjectId objectId, int len)
- throws IOException {
- return delegate().abbreviate(objectId, len);
- }
-
- @Override
- public Collection<ObjectId> resolve(AbbreviatedObjectId id)
- throws IOException {
- return delegate().resolve(id);
- }
-
- @Override
- public boolean has(AnyObjectId objectId) throws IOException {
- return delegate().has(objectId);
- }
-
- @Override
- public boolean has(AnyObjectId objectId, int typeHint) throws IOException {
- return delegate().has(objectId, typeHint);
- }
-
- @Override
- public ObjectLoader open(AnyObjectId objectId)
- throws MissingObjectException, IOException {
- return delegate().open(objectId);
- }
-
- @Override
- public ObjectLoader open(AnyObjectId objectId, int typeHint)
- throws MissingObjectException, IncorrectObjectTypeException,
- IOException {
- return delegate().open(objectId, typeHint);
- }
-
- @Override
- public Set<ObjectId> getShallowCommits() throws IOException {
- return delegate().getShallowCommits();
- }
-
- @Override
- public <T extends ObjectId> AsyncObjectLoaderQueue<T> open(
- Iterable<T> objectIds, boolean reportMissing) {
- return delegate().open(objectIds, reportMissing);
- }
-
- @Override
- public long getObjectSize(AnyObjectId objectId, int typeHint)
- throws MissingObjectException, IncorrectObjectTypeException,
- IOException {
- return delegate().getObjectSize(objectId, typeHint);
- }
-
- @Override
- public <T extends ObjectId> AsyncObjectSizeQueue<T> getObjectSize(
- Iterable<T> objectIds, boolean reportMissing) {
- return delegate().getObjectSize(objectIds, reportMissing);
- }
-
- @Override
- public void setAvoidUnreachableObjects(boolean avoid) {
- delegate().setAvoidUnreachableObjects(avoid);
- }
-
- @Override
- public BitmapIndex getBitmapIndex() throws IOException {
- return delegate().getBitmapIndex();
- }
-
- @Override
- @Nullable
- public ObjectInserter getCreatedFromInserter() {
- return delegate().getCreatedFromInserter();
- }
-
- @Override
- public void close() {
- delegate().close();
- }
- }
- }
|