wireproto: compress data from a generator
Currently, the "getbundle" wire protocol command obtains a generator of
data, converts it to a util.chunkbuffer, then converts it back to a
generator via the protocol's groupchunks() implementation. For the SSH
protocol, groupchunks() simply reads 4kb chunks then write()s the
data to a file descriptor. For the HTTP protocol, groupchunks() reads
32kb chunks, feeds those into a zlib compressor, emits compressed data
as it is available, and that is sent to the WSGI layer, where it is
likely turned into HTTP chunked transfer chunks as is or further
buffered and turned into a larger chunk.
For both the SSH and HTTP protocols, there is inefficiency from using
util.chunkbuffer.
For SSH, emitting consistent 4kb chunks sounds nice. However, the file
descriptor it is writing to is almost certainly buffered. That means
that a Python .write() probably doesn't translate into exactly what is
written to the I/O layer.
For HTTP, we're going through an intermediate layer to zlib compress
data. So all util.chunkbuffer is doing is ensuring that the chunks we
feed into the zlib compressor are of uniform size. This means more CPU
time in Python buffering and emitting chunks in util.chunkbuffer but
fewer function calls to zlib.
This patch introduces and implements a new wire protocol abstract
method: compresschunks(). It is like groupchunks() except it operates
on a generator instead of something with a .read(). The SSH
implementation simply proxies chunks. The HTTP implementation uses
zlib compression.
To avoid duplicate code, the HTTP groupchunks() has been reimplemented
in terms of compresschunks().
To prove this all works, the "getbundle" wire protocol command has been
switched to compresschunks(). This removes the util.chunkbuffer from
that command. Now, data essentially streams straight from the
changegroup emitter to the wire, possibly through a zlib compressor.
Generators all the way, baby.
There were slim to no performance changes on the server as measured
with the mozilla-central repository. This is likely because CPU
time is dominated by reading revlogs, producing the changegroup, and
zlib compressing the output stream. Still, this brings us a little
closer to our ideal of using generators everywhere.
# ASCII graph log extension for Mercurial
#
# Copyright 2007 Joel Rosdahl <joel@rosdahl.net>
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
'''command to view revision graphs from a shell (DEPRECATED)
The functionality of this extension has been include in core Mercurial
since version 2.3. Please use :hg:`log -G ...` instead.
This extension adds a --graph option to the incoming, outgoing and log
commands. When this options is given, an ASCII representation of the
revision graph is also shown.
'''
from __future__ import absolute_import
from mercurial.i18n import _
from mercurial import (
cmdutil,
commands,
)
cmdtable = {}
command = cmdutil.command(cmdtable)
# Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
# extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
# be specifying the version(s) of Mercurial they are tested with, or
# leave the attribute unspecified.
testedwith = 'ships-with-hg-core'
@command('glog',
[('f', 'follow', None,
_('follow changeset history, or file history across copies and renames')),
('', 'follow-first', None,
_('only follow the first parent of merge changesets (DEPRECATED)')),
('d', 'date', '', _('show revisions matching date spec'), _('DATE')),
('C', 'copies', None, _('show copied files')),
('k', 'keyword', [],
_('do case-insensitive search for a given text'), _('TEXT')),
('r', 'rev', [], _('show the specified revision or revset'), _('REV')),
('', 'removed', None, _('include revisions where files were removed')),
('m', 'only-merges', None, _('show only merges (DEPRECATED)')),
('u', 'user', [], _('revisions committed by user'), _('USER')),
('', 'only-branch', [],
_('show only changesets within the given named branch (DEPRECATED)'),
_('BRANCH')),
('b', 'branch', [],
_('show changesets within the given named branch'), _('BRANCH')),
('P', 'prune', [],
_('do not display revision or any of its ancestors'), _('REV')),
] + commands.logopts + commands.walkopts,
_('[OPTION]... [FILE]'),
inferrepo=True)
def glog(ui, repo, *pats, **opts):
"""show revision history alongside an ASCII revision graph
Print a revision history alongside a revision graph drawn with
ASCII characters.
Nodes printed as an @ character are parents of the working
directory.
This is an alias to :hg:`log -G`.
"""
opts['graph'] = True
return commands.log(ui, repo, *pats, **opts)