exchangev2: fetch manifest revisions
Now that the server has support for retrieving manifest data, we can
implement the client bits to call it.
We teach the changeset fetching code to capture the manifest revisions
that are encountered on incoming changesets. We then feed this into a
new function which filters out known manifests and then batches up
manifest data requests to the server.
This is different from the previous wire protocol in a few notable
ways.
First, the client fetches manifest data separately and explicitly.
Before, we'd ask the server for data pertaining to some changesets
(via a "getbundle" command) and manifests (and files) would be sent
automatically. Providing an API for looking up just manifest data
separately gives clients much more flexibility for manifest management.
For example, a client may choose to only fetch manifest data on demand
instead of prefetching it (i.e. partial clone).
Second, we send N commands to the server for manifest retrieval instead
of 1. This property has a few nice side-effects. One is that the
deterministic nature of the requests lends itself to server-side
caching. For example, say the remote has 50,000 manifests. If the
server is configured to cache responses, each time a new commit
arrives, you will have a cache miss and need to regenerate all outgoing
data. But if you makes N requests requesting 10,000 manifests each,
a new commit will still yield cache hits on the initial, unchanged
manifest batches/requests.
A derived benefit from these properties is that resumable clone is
conceptually simpler to implement. When making a monolithic request
for all of the repository data, recovering from an interrupted clone
is hard because the server was in the driver's seat and was maintaining
state about all the data that needed transferred. With the client
driving fetching, the client can persist the set of unfetched entities
and retry/resume a fetch if something goes wrong. Or we can fetch all
data N changesets at a time and slowly build up a repository. This
approach is drastically easier to implement when we have server APIs
exposing low-level repository primitives (such as manifests and files).
We don't yet support tree manifests. But it should be possible to
implement that with the existing wire protocol command.
Differential Revision: https://phab.mercurial-scm.org/D4489
# Infinite push
#
# Copyright 2016 Facebook, Inc.
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
"""
[infinitepush]
# Server-side option. Used only if indextype=disk.
# Filesystem path to the index store
indexpath = PATH
"""
from __future__ import absolute_import
import os
from mercurial import util
from mercurial.utils import stringutil
from . import indexapi
class fileindexapi(indexapi.indexapi):
def __init__(self, repo):
super(fileindexapi, self).__init__()
self._repo = repo
root = repo.ui.config('infinitepush', 'indexpath')
if not root:
root = os.path.join('scratchbranches', 'index')
self._nodemap = os.path.join(root, 'nodemap')
self._bookmarkmap = os.path.join(root, 'bookmarkmap')
self._metadatamap = os.path.join(root, 'nodemetadatamap')
self._lock = None
def __enter__(self):
self._lock = self._repo.wlock()
return self
def __exit__(self, exc_type, exc_val, exc_tb):
if self._lock:
self._lock.__exit__(exc_type, exc_val, exc_tb)
def addbundle(self, bundleid, nodesctx):
for node in nodesctx:
nodepath = os.path.join(self._nodemap, node.hex())
self._write(nodepath, bundleid)
def addbookmark(self, bookmark, node):
bookmarkpath = os.path.join(self._bookmarkmap, bookmark)
self._write(bookmarkpath, node)
def addmanybookmarks(self, bookmarks):
for bookmark, node in bookmarks.items():
self.addbookmark(bookmark, node)
def deletebookmarks(self, patterns):
for pattern in patterns:
for bookmark, _ in self._listbookmarks(pattern):
bookmarkpath = os.path.join(self._bookmarkmap, bookmark)
self._delete(bookmarkpath)
def getbundle(self, node):
nodepath = os.path.join(self._nodemap, node)
return self._read(nodepath)
def getnode(self, bookmark):
bookmarkpath = os.path.join(self._bookmarkmap, bookmark)
return self._read(bookmarkpath)
def getbookmarks(self, query):
return dict(self._listbookmarks(query))
def saveoptionaljsonmetadata(self, node, jsonmetadata):
vfs = self._repo.vfs
vfs.write(os.path.join(self._metadatamap, node), jsonmetadata)
def _listbookmarks(self, pattern):
if pattern.endswith('*'):
pattern = 're:^' + pattern[:-1] + '.*'
kind, pat, matcher = stringutil.stringmatcher(pattern)
prefixlen = len(self._bookmarkmap) + 1
for dirpath, _, books in self._repo.vfs.walk(self._bookmarkmap):
for book in books:
bookmark = os.path.join(dirpath, book)[prefixlen:]
bookmark = util.pconvert(bookmark)
if not matcher(bookmark):
continue
yield bookmark, self._read(os.path.join(dirpath, book))
def _write(self, path, value):
vfs = self._repo.vfs
dirname = vfs.dirname(path)
if not vfs.exists(dirname):
vfs.makedirs(dirname)
vfs.write(path, value)
def _read(self, path):
vfs = self._repo.vfs
if not vfs.exists(path):
return None
return vfs.read(path)
def _delete(self, path):
vfs = self._repo.vfs
if not vfs.exists(path):
return
return vfs.unlink(path)