PackWriter: Support reuse of entire packs
The most expensive part of packing a repository for transport to
another system is enumerating all of the objects in the repository.
Once this gets to the size of the linux-2.6 repository (1.8 million
objects), enumeration can take several CPU minutes and costs a lot
of temporary working set memory.
Teach PackWriter to efficiently reuse an existing "cached pack"
by answering a clone request with a thin pack followed by a larger
cached pack appended to the end. This requires the repository
owner to first construct the cached pack by hand, and record the
tip commits inside of $GIT_DIR/objects/info/cached-packs:
cd $GIT_DIR
root=$(git rev-parse master)
tmp=objects/.tmp-$$
names=$(echo $root | git pack-objects --keep-true-parents --revs $tmp)
for n in $names; do
chmod a-w $tmp-$n.pack $tmp-$n.idx
touch objects/pack/pack-$n.keep
mv $tmp-$n.pack objects/pack/pack-$n.pack
mv $tmp-$n.idx objects/pack/pack-$n.idx
done
(echo "+ $root";
for n in $names; do echo "P $n"; done;
echo) >>objects/info/cached-packs
git repack -a -d
When a clone request needs to include $root, the corresponding
cached pack will be copied as-is, rather than enumerating all of
the objects that are reachable from $root.
For a linux-2.6 kernel repository that should be about 376 MiB,
the above process creates two packs of 368 MiB and 38 MiB[1].
This is a local disk usage increase of ~26 MiB, due to reduced
delta compression between the large cached pack and the smaller
recent activity pack. The overhead is similar to 1 full copy of
the compressed project sources.
With this cached pack in hand, JGit daemon completes a clone request
in 1m17s less time, but a slightly larger data transfer (+2.39 MiB):
Before:
remote: Counting objects: 1861830, done
remote: Finding sources: 100% (1861830/1861830)
remote: Getting sizes: 100% (88243/88243)
remote: Compressing objects: 100% (88184/88184)
Receiving objects: 100% (1861830/1861830), 376.01 MiB | 19.01 MiB/s, done.
remote: Total 1861830 (delta 4706), reused 1851053 (delta 1553844)
Resolving deltas: 100% (1564621/1564621), done.
real 3m19.005s
After:
remote: Counting objects: 1601, done
remote: Counting objects: 1828460, done
remote: Finding sources: 100% (50475/50475)
remote: Getting sizes: 100% (18843/18843)
remote: Compressing objects: 100% (7585/7585)
remote: Total 1861830 (delta 2407), reused 1856197 (delta 37510)
Receiving objects: 100% (1861830/1861830), 378.40 MiB | 31.31 MiB/s, done.
Resolving deltas: 100% (1559477/1559477), done.
real 2m2.938s
Repository owners can periodically refresh their cached packs by
repacking their repository, folding all newer objects into a larger
cached pack. Since repacking is already considered to be a normal
Git maintenance activity, this isn't a very big burden.
[1] In this test $root was set back about two weeks.
Change-Id: Ib87131d5c4b5e8c5cacb0f4fe16ff4ece554734b
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
13 jaren geleden Support creating pack bitmap indexes in PackWriter.
Update the PackWriter to support writing out pack bitmap indexes,
a parallel ".bitmap" file to the ".pack" file.
Bitmaps are selected at commits every 1 to 5,000 commits for
each unique path from the start. The most recent 100 commits are
all bitmapped. The next 19,000 commits have a bitmaps every 100
commits. The remaining commits have a bitmap every 5,000 commits.
Commits with more than 1 parent are prefered over ones
with 1 or less. Furthermore, previously computed bitmaps are reused,
if the previous entry had the reuse flag set, which is set when the
bitmap was placed at the max allowed distance.
Bitmaps are used to speed up the counting phase when packing, for
requests that are not shallow. The PackWriterBitmapWalker uses
a RevFilter to proactively mark commits with RevFlag.SEEN, when
they appear in a bitmap. The walker produces the full closure
of reachable ObjectIds, given the collection of starting ObjectIds.
For fetch request, two ObjectWalks are executed to compute the
ObjectIds reachable from the haves and from the wants. The
ObjectIds needed to be written are determined by taking all the
resulting wants AND NOT the haves.
For clone requests, we get cached pack support for "free" since
it is possible to determine if all of the ObjectIds in a pack file
are included in the resulting list of ObjectIds to write.
On my machine, the best times for clones and fetches of the linux
kernel repository (with about 2.6M objects and 300K commits) are
tabulated below:
Operation Index V2 Index VE003
Clone 37530ms (524.06 MiB) 82ms (524.06 MiB)
Fetch (1 commit back) 75ms 107ms
Fetch (10 commits back) 456ms (269.51 KiB) 341ms (265.19 KiB)
Fetch (100 commits back) 449ms (269.91 KiB) 337ms (267.28 KiB)
Fetch (1000 commits back) 2229ms ( 14.75 MiB) 189ms ( 14.42 MiB)
Fetch (10000 commits back) 2177ms ( 16.30 MiB) 254ms ( 15.88 MiB)
Fetch (100000 commits back) 14340ms (185.83 MiB) 1655ms (189.39 MiB)
Change-Id: Icdb0cdd66ff168917fb9ef17b96093990cc6a98d
11 jaren geleden PackWriter: Support reuse of entire packs
The most expensive part of packing a repository for transport to
another system is enumerating all of the objects in the repository.
Once this gets to the size of the linux-2.6 repository (1.8 million
objects), enumeration can take several CPU minutes and costs a lot
of temporary working set memory.
Teach PackWriter to efficiently reuse an existing "cached pack"
by answering a clone request with a thin pack followed by a larger
cached pack appended to the end. This requires the repository
owner to first construct the cached pack by hand, and record the
tip commits inside of $GIT_DIR/objects/info/cached-packs:
cd $GIT_DIR
root=$(git rev-parse master)
tmp=objects/.tmp-$$
names=$(echo $root | git pack-objects --keep-true-parents --revs $tmp)
for n in $names; do
chmod a-w $tmp-$n.pack $tmp-$n.idx
touch objects/pack/pack-$n.keep
mv $tmp-$n.pack objects/pack/pack-$n.pack
mv $tmp-$n.idx objects/pack/pack-$n.idx
done
(echo "+ $root";
for n in $names; do echo "P $n"; done;
echo) >>objects/info/cached-packs
git repack -a -d
When a clone request needs to include $root, the corresponding
cached pack will be copied as-is, rather than enumerating all of
the objects that are reachable from $root.
For a linux-2.6 kernel repository that should be about 376 MiB,
the above process creates two packs of 368 MiB and 38 MiB[1].
This is a local disk usage increase of ~26 MiB, due to reduced
delta compression between the large cached pack and the smaller
recent activity pack. The overhead is similar to 1 full copy of
the compressed project sources.
With this cached pack in hand, JGit daemon completes a clone request
in 1m17s less time, but a slightly larger data transfer (+2.39 MiB):
Before:
remote: Counting objects: 1861830, done
remote: Finding sources: 100% (1861830/1861830)
remote: Getting sizes: 100% (88243/88243)
remote: Compressing objects: 100% (88184/88184)
Receiving objects: 100% (1861830/1861830), 376.01 MiB | 19.01 MiB/s, done.
remote: Total 1861830 (delta 4706), reused 1851053 (delta 1553844)
Resolving deltas: 100% (1564621/1564621), done.
real 3m19.005s
After:
remote: Counting objects: 1601, done
remote: Counting objects: 1828460, done
remote: Finding sources: 100% (50475/50475)
remote: Getting sizes: 100% (18843/18843)
remote: Compressing objects: 100% (7585/7585)
remote: Total 1861830 (delta 2407), reused 1856197 (delta 37510)
Receiving objects: 100% (1861830/1861830), 378.40 MiB | 31.31 MiB/s, done.
Resolving deltas: 100% (1559477/1559477), done.
real 2m2.938s
Repository owners can periodically refresh their cached packs by
repacking their repository, folding all newer objects into a larger
cached pack. Since repacking is already considered to be a normal
Git maintenance activity, this isn't a very big burden.
[1] In this test $root was set back about two weeks.
Change-Id: Ib87131d5c4b5e8c5cacb0f4fe16ff4ece554734b
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
13 jaren geleden PackWriter: Support reuse of entire packs
The most expensive part of packing a repository for transport to
another system is enumerating all of the objects in the repository.
Once this gets to the size of the linux-2.6 repository (1.8 million
objects), enumeration can take several CPU minutes and costs a lot
of temporary working set memory.
Teach PackWriter to efficiently reuse an existing "cached pack"
by answering a clone request with a thin pack followed by a larger
cached pack appended to the end. This requires the repository
owner to first construct the cached pack by hand, and record the
tip commits inside of $GIT_DIR/objects/info/cached-packs:
cd $GIT_DIR
root=$(git rev-parse master)
tmp=objects/.tmp-$$
names=$(echo $root | git pack-objects --keep-true-parents --revs $tmp)
for n in $names; do
chmod a-w $tmp-$n.pack $tmp-$n.idx
touch objects/pack/pack-$n.keep
mv $tmp-$n.pack objects/pack/pack-$n.pack
mv $tmp-$n.idx objects/pack/pack-$n.idx
done
(echo "+ $root";
for n in $names; do echo "P $n"; done;
echo) >>objects/info/cached-packs
git repack -a -d
When a clone request needs to include $root, the corresponding
cached pack will be copied as-is, rather than enumerating all of
the objects that are reachable from $root.
For a linux-2.6 kernel repository that should be about 376 MiB,
the above process creates two packs of 368 MiB and 38 MiB[1].
This is a local disk usage increase of ~26 MiB, due to reduced
delta compression between the large cached pack and the smaller
recent activity pack. The overhead is similar to 1 full copy of
the compressed project sources.
With this cached pack in hand, JGit daemon completes a clone request
in 1m17s less time, but a slightly larger data transfer (+2.39 MiB):
Before:
remote: Counting objects: 1861830, done
remote: Finding sources: 100% (1861830/1861830)
remote: Getting sizes: 100% (88243/88243)
remote: Compressing objects: 100% (88184/88184)
Receiving objects: 100% (1861830/1861830), 376.01 MiB | 19.01 MiB/s, done.
remote: Total 1861830 (delta 4706), reused 1851053 (delta 1553844)
Resolving deltas: 100% (1564621/1564621), done.
real 3m19.005s
After:
remote: Counting objects: 1601, done
remote: Counting objects: 1828460, done
remote: Finding sources: 100% (50475/50475)
remote: Getting sizes: 100% (18843/18843)
remote: Compressing objects: 100% (7585/7585)
remote: Total 1861830 (delta 2407), reused 1856197 (delta 37510)
Receiving objects: 100% (1861830/1861830), 378.40 MiB | 31.31 MiB/s, done.
Resolving deltas: 100% (1559477/1559477), done.
real 2m2.938s
Repository owners can periodically refresh their cached packs by
repacking their repository, folding all newer objects into a larger
cached pack. Since repacking is already considered to be a normal
Git maintenance activity, this isn't a very big burden.
[1] In this test $root was set back about two weeks.
Change-Id: Ib87131d5c4b5e8c5cacb0f4fe16ff4ece554734b
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
13 jaren geleden Support creating pack bitmap indexes in PackWriter.
Update the PackWriter to support writing out pack bitmap indexes,
a parallel ".bitmap" file to the ".pack" file.
Bitmaps are selected at commits every 1 to 5,000 commits for
each unique path from the start. The most recent 100 commits are
all bitmapped. The next 19,000 commits have a bitmaps every 100
commits. The remaining commits have a bitmap every 5,000 commits.
Commits with more than 1 parent are prefered over ones
with 1 or less. Furthermore, previously computed bitmaps are reused,
if the previous entry had the reuse flag set, which is set when the
bitmap was placed at the max allowed distance.
Bitmaps are used to speed up the counting phase when packing, for
requests that are not shallow. The PackWriterBitmapWalker uses
a RevFilter to proactively mark commits with RevFlag.SEEN, when
they appear in a bitmap. The walker produces the full closure
of reachable ObjectIds, given the collection of starting ObjectIds.
For fetch request, two ObjectWalks are executed to compute the
ObjectIds reachable from the haves and from the wants. The
ObjectIds needed to be written are determined by taking all the
resulting wants AND NOT the haves.
For clone requests, we get cached pack support for "free" since
it is possible to determine if all of the ObjectIds in a pack file
are included in the resulting list of ObjectIds to write.
On my machine, the best times for clones and fetches of the linux
kernel repository (with about 2.6M objects and 300K commits) are
tabulated below:
Operation Index V2 Index VE003
Clone 37530ms (524.06 MiB) 82ms (524.06 MiB)
Fetch (1 commit back) 75ms 107ms
Fetch (10 commits back) 456ms (269.51 KiB) 341ms (265.19 KiB)
Fetch (100 commits back) 449ms (269.91 KiB) 337ms (267.28 KiB)
Fetch (1000 commits back) 2229ms ( 14.75 MiB) 189ms ( 14.42 MiB)
Fetch (10000 commits back) 2177ms ( 16.30 MiB) 254ms ( 15.88 MiB)
Fetch (100000 commits back) 14340ms (185.83 MiB) 1655ms (189.39 MiB)
Change-Id: Icdb0cdd66ff168917fb9ef17b96093990cc6a98d
11 jaren geleden Support creating pack bitmap indexes in PackWriter.
Update the PackWriter to support writing out pack bitmap indexes,
a parallel ".bitmap" file to the ".pack" file.
Bitmaps are selected at commits every 1 to 5,000 commits for
each unique path from the start. The most recent 100 commits are
all bitmapped. The next 19,000 commits have a bitmaps every 100
commits. The remaining commits have a bitmap every 5,000 commits.
Commits with more than 1 parent are prefered over ones
with 1 or less. Furthermore, previously computed bitmaps are reused,
if the previous entry had the reuse flag set, which is set when the
bitmap was placed at the max allowed distance.
Bitmaps are used to speed up the counting phase when packing, for
requests that are not shallow. The PackWriterBitmapWalker uses
a RevFilter to proactively mark commits with RevFlag.SEEN, when
they appear in a bitmap. The walker produces the full closure
of reachable ObjectIds, given the collection of starting ObjectIds.
For fetch request, two ObjectWalks are executed to compute the
ObjectIds reachable from the haves and from the wants. The
ObjectIds needed to be written are determined by taking all the
resulting wants AND NOT the haves.
For clone requests, we get cached pack support for "free" since
it is possible to determine if all of the ObjectIds in a pack file
are included in the resulting list of ObjectIds to write.
On my machine, the best times for clones and fetches of the linux
kernel repository (with about 2.6M objects and 300K commits) are
tabulated below:
Operation Index V2 Index VE003
Clone 37530ms (524.06 MiB) 82ms (524.06 MiB)
Fetch (1 commit back) 75ms 107ms
Fetch (10 commits back) 456ms (269.51 KiB) 341ms (265.19 KiB)
Fetch (100 commits back) 449ms (269.91 KiB) 337ms (267.28 KiB)
Fetch (1000 commits back) 2229ms ( 14.75 MiB) 189ms ( 14.42 MiB)
Fetch (10000 commits back) 2177ms ( 16.30 MiB) 254ms ( 15.88 MiB)
Fetch (100000 commits back) 14340ms (185.83 MiB) 1655ms (189.39 MiB)
Change-Id: Icdb0cdd66ff168917fb9ef17b96093990cc6a98d
11 jaren geleden Support creating pack bitmap indexes in PackWriter.
Update the PackWriter to support writing out pack bitmap indexes,
a parallel ".bitmap" file to the ".pack" file.
Bitmaps are selected at commits every 1 to 5,000 commits for
each unique path from the start. The most recent 100 commits are
all bitmapped. The next 19,000 commits have a bitmaps every 100
commits. The remaining commits have a bitmap every 5,000 commits.
Commits with more than 1 parent are prefered over ones
with 1 or less. Furthermore, previously computed bitmaps are reused,
if the previous entry had the reuse flag set, which is set when the
bitmap was placed at the max allowed distance.
Bitmaps are used to speed up the counting phase when packing, for
requests that are not shallow. The PackWriterBitmapWalker uses
a RevFilter to proactively mark commits with RevFlag.SEEN, when
they appear in a bitmap. The walker produces the full closure
of reachable ObjectIds, given the collection of starting ObjectIds.
For fetch request, two ObjectWalks are executed to compute the
ObjectIds reachable from the haves and from the wants. The
ObjectIds needed to be written are determined by taking all the
resulting wants AND NOT the haves.
For clone requests, we get cached pack support for "free" since
it is possible to determine if all of the ObjectIds in a pack file
are included in the resulting list of ObjectIds to write.
On my machine, the best times for clones and fetches of the linux
kernel repository (with about 2.6M objects and 300K commits) are
tabulated below:
Operation Index V2 Index VE003
Clone 37530ms (524.06 MiB) 82ms (524.06 MiB)
Fetch (1 commit back) 75ms 107ms
Fetch (10 commits back) 456ms (269.51 KiB) 341ms (265.19 KiB)
Fetch (100 commits back) 449ms (269.91 KiB) 337ms (267.28 KiB)
Fetch (1000 commits back) 2229ms ( 14.75 MiB) 189ms ( 14.42 MiB)
Fetch (10000 commits back) 2177ms ( 16.30 MiB) 254ms ( 15.88 MiB)
Fetch (100000 commits back) 14340ms (185.83 MiB) 1655ms (189.39 MiB)
Change-Id: Icdb0cdd66ff168917fb9ef17b96093990cc6a98d
11 jaren geleden |
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207 |
- /*
- * Copyright (C) 2010, Google Inc. and others
- *
- * This program and the accompanying materials are made available under the
- * terms of the Eclipse Distribution License v. 1.0 which is available at
- * https://www.eclipse.org/org/documents/edl-v10.php.
- *
- * SPDX-License-Identifier: BSD-3-Clause
- */
-
- package org.eclipse.jgit.internal.storage.pack;
-
- import java.io.IOException;
- import java.util.Collection;
- import java.util.List;
-
- import org.eclipse.jgit.errors.MissingObjectException;
- import org.eclipse.jgit.errors.StoredObjectRepresentationNotAvailableException;
- import org.eclipse.jgit.errors.StoredPackRepresentationNotAvailableException;
- import org.eclipse.jgit.lib.AnyObjectId;
- import org.eclipse.jgit.lib.BitmapIndex.BitmapBuilder;
- import org.eclipse.jgit.lib.ProgressMonitor;
-
- /**
- * Extension of {@link org.eclipse.jgit.lib.ObjectReader} that supports reusing
- * objects in packs.
- * <p>
- * {@code ObjectReader} implementations may also optionally implement this
- * interface to support
- * {@link org.eclipse.jgit.internal.storage.pack.PackWriter} with a means of
- * copying an object that is already in pack encoding format directly into the
- * output stream, without incurring decompression and recompression overheads.
- */
- public interface ObjectReuseAsIs {
- /**
- * Allocate a new {@code PackWriter} state structure for an object.
- * <p>
- * {@link org.eclipse.jgit.internal.storage.pack.PackWriter} allocates these
- * objects to keep track of the per-object state, and how to load the
- * objects efficiently into the generated stream. Implementers may subclass
- * this type with additional object state, such as to remember what file and
- * offset contains the object's pack encoded data.
- *
- * @param objectId
- * the id of the object that will be packed.
- * @param type
- * the Git type of the object that will be packed.
- * @return a new instance for this object.
- */
- ObjectToPack newObjectToPack(AnyObjectId objectId, int type);
-
- /**
- * Select the best object representation for a packer.
- * <p>
- * Implementations should iterate through all available representations of
- * an object, and pass them in turn to the PackWriter though
- * {@link org.eclipse.jgit.internal.storage.pack.PackWriter#select(ObjectToPack, StoredObjectRepresentation)}
- * so the writer can select the most suitable representation to reuse into
- * the output stream.
- * <p>
- * If the implementation returns CachedPack from
- * {@link #getCachedPacksAndUpdate(BitmapBuilder)} it must consider the
- * representation of any object that is stored in any of the offered
- * CachedPacks. PackWriter relies on this behavior to prune duplicate
- * objects out of the pack stream when it selects a CachedPack and the
- * object was also reached through the thin-pack enumeration.
- * <p>
- * The implementation may choose to consider multiple objects at once on
- * concurrent threads, but must evaluate all representations of an object
- * within the same thread.
- *
- * @param packer
- * the packer that will write the object in the near future.
- * @param monitor
- * progress monitor, implementation should update the monitor
- * once for each item in the iteration when selection is done.
- * @param objects
- * the objects that are being packed.
- * @throws org.eclipse.jgit.errors.MissingObjectException
- * there is no representation available for the object, as it is
- * no longer in the repository. Packing will abort.
- * @throws java.io.IOException
- * the repository cannot be accessed. Packing will abort.
- */
- void selectObjectRepresentation(PackWriter packer,
- ProgressMonitor monitor, Iterable<ObjectToPack> objects)
- throws IOException, MissingObjectException;
-
- /**
- * Write objects to the pack stream in roughly the order given.
- *
- * {@code PackWriter} invokes this method to write out one or more objects,
- * in approximately the order specified by the iteration over the list. A
- * simple implementation of this method would just iterate the list and
- * output each object:
- *
- * <pre>
- * for (ObjectToPack obj : list)
- * out.writeObject(obj)
- * </pre>
- *
- * However more sophisticated implementors may try to perform some (small)
- * reordering to access objects that are stored close to each other at
- * roughly the same time. Implementations may choose to write objects out of
- * order, but this may increase pack file size due to using a larger header
- * format to reach a delta base that is later in the stream. It may also
- * reduce data locality for the reader, slowing down data access.
- *
- * Invoking
- * {@link org.eclipse.jgit.internal.storage.pack.PackOutputStream#writeObject(ObjectToPack)}
- * will cause
- * {@link #copyObjectAsIs(PackOutputStream, ObjectToPack, boolean)} to be
- * invoked recursively on {@code this} if the current object is scheduled
- * for reuse.
- *
- * @param out
- * the stream to write each object to.
- * @param list
- * the list of objects to write. Objects should be written in
- * approximately this order. Implementors may resort the list
- * elements in-place during writing if desired.
- * @throws java.io.IOException
- * the stream cannot be written to, or one or more required
- * objects cannot be accessed from the object database.
- */
- void writeObjects(PackOutputStream out, List<ObjectToPack> list)
- throws IOException;
-
- /**
- * Output a previously selected representation.
- * <p>
- * {@code PackWriter} invokes this method only if a representation
- * previously given to it by {@code selectObjectRepresentation} was chosen
- * for reuse into the output stream. The {@code otp} argument is an instance
- * created by this reader's own {@code newObjectToPack}, and the
- * representation data saved within it also originated from this reader.
- * <p>
- * Implementors must write the object header before copying the raw data to
- * the output stream. The typical implementation is like:
- *
- * <pre>
- * MyToPack mtp = (MyToPack) otp;
- * byte[] raw;
- * if (validate)
- * raw = validate(mtp); // throw SORNAE here, if at all
- * else
- * raw = readFast(mtp);
- * out.writeHeader(mtp, mtp.inflatedSize);
- * out.write(raw);
- * </pre>
- *
- * @param out
- * stream the object should be written to.
- * @param otp
- * the object's saved representation information.
- * @param validate
- * if true the representation must be validated and not be
- * corrupt before being reused. If false, validation may be
- * skipped as it will be performed elsewhere in the processing
- * pipeline.
- * @throws org.eclipse.jgit.errors.StoredObjectRepresentationNotAvailableException
- * the previously selected representation is no longer
- * available. If thrown before {@code out.writeHeader} the pack
- * writer will try to find another representation, and write
- * that one instead. If throw after {@code out.writeHeader},
- * packing will abort.
- * @throws java.io.IOException
- * the stream's write method threw an exception. Packing will
- * abort.
- */
- void copyObjectAsIs(PackOutputStream out, ObjectToPack otp,
- boolean validate) throws IOException,
- StoredObjectRepresentationNotAvailableException;
-
- /**
- * Append an entire pack's contents onto the output stream.
- * <p>
- * The entire pack, excluding its header and trailing footer is sent.
- *
- * @param out
- * stream to append the pack onto.
- * @param pack
- * the cached pack to send.
- * @throws java.io.IOException
- * the pack cannot be read, or stream did not accept a write.
- */
- void copyPackAsIs(PackOutputStream out, CachedPack pack)
- throws IOException, StoredPackRepresentationNotAvailableException;
-
- /**
- * Obtain the available cached packs that match the bitmap and update
- * the bitmap by removing the items that are in the CachedPack.
- * <p>
- * A cached pack has known starting points and may be sent entirely as-is,
- * with almost no effort on the sender's part.
- *
- * @param needBitmap
- * the bitmap that contains all of the objects the client wants.
- * @return the available cached packs.
- * @throws java.io.IOException
- * the cached packs cannot be listed from the repository.
- * Callers may choose to ignore this and continue as-if there
- * were no cached packs.
- */
- Collection<CachedPack> getCachedPacksAndUpdate(
- BitmapBuilder needBitmap) throws IOException;
- }
|