mirror of
https://github.com/facebook/sapling.git
synced 2024-10-08 15:57:43 +03:00
1538b87cfc
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.
136 lines
3.6 KiB
Python
136 lines
3.6 KiB
Python
# sshserver.py - ssh protocol server support for mercurial
|
|
#
|
|
# Copyright 2005-2007 Matt Mackall <mpm@selenic.com>
|
|
# Copyright 2006 Vadim Gelfer <vadim.gelfer@gmail.com>
|
|
#
|
|
# This software may be used and distributed according to the terms of the
|
|
# GNU General Public License version 2 or any later version.
|
|
|
|
from __future__ import absolute_import
|
|
|
|
import sys
|
|
|
|
from .i18n import _
|
|
from . import (
|
|
encoding,
|
|
error,
|
|
hook,
|
|
util,
|
|
wireproto,
|
|
)
|
|
|
|
class sshserver(wireproto.abstractserverproto):
|
|
def __init__(self, ui, repo):
|
|
self.ui = ui
|
|
self.repo = repo
|
|
self.lock = None
|
|
self.fin = ui.fin
|
|
self.fout = ui.fout
|
|
self.name = 'ssh'
|
|
|
|
hook.redirect(True)
|
|
ui.fout = repo.ui.fout = ui.ferr
|
|
|
|
# Prevent insertion/deletion of CRs
|
|
util.setbinary(self.fin)
|
|
util.setbinary(self.fout)
|
|
|
|
def getargs(self, args):
|
|
data = {}
|
|
keys = args.split()
|
|
for n in xrange(len(keys)):
|
|
argline = self.fin.readline()[:-1]
|
|
arg, l = argline.split()
|
|
if arg not in keys:
|
|
raise error.Abort(_("unexpected parameter %r") % arg)
|
|
if arg == '*':
|
|
star = {}
|
|
for k in xrange(int(l)):
|
|
argline = self.fin.readline()[:-1]
|
|
arg, l = argline.split()
|
|
val = self.fin.read(int(l))
|
|
star[arg] = val
|
|
data['*'] = star
|
|
else:
|
|
val = self.fin.read(int(l))
|
|
data[arg] = val
|
|
return [data[k] for k in keys]
|
|
|
|
def getarg(self, name):
|
|
return self.getargs(name)[0]
|
|
|
|
def getfile(self, fpout):
|
|
self.sendresponse('')
|
|
count = int(self.fin.readline())
|
|
while count:
|
|
fpout.write(self.fin.read(count))
|
|
count = int(self.fin.readline())
|
|
|
|
def redirect(self):
|
|
pass
|
|
|
|
def sendresponse(self, v):
|
|
self.fout.write("%d\n" % len(v))
|
|
self.fout.write(v)
|
|
self.fout.flush()
|
|
|
|
def sendstream(self, source):
|
|
write = self.fout.write
|
|
|
|
if source.reader:
|
|
gen = iter(lambda: source.reader.read(4096), '')
|
|
else:
|
|
gen = source.gen
|
|
|
|
for chunk in gen:
|
|
write(chunk)
|
|
self.fout.flush()
|
|
|
|
def sendpushresponse(self, rsp):
|
|
self.sendresponse('')
|
|
self.sendresponse(str(rsp.res))
|
|
|
|
def sendpusherror(self, rsp):
|
|
self.sendresponse(rsp.res)
|
|
|
|
def sendooberror(self, rsp):
|
|
self.ui.ferr.write('%s\n-\n' % rsp.message)
|
|
self.ui.ferr.flush()
|
|
self.fout.write('\n')
|
|
self.fout.flush()
|
|
|
|
def serve_forever(self):
|
|
try:
|
|
while self.serve_one():
|
|
pass
|
|
finally:
|
|
if self.lock is not None:
|
|
self.lock.release()
|
|
sys.exit(0)
|
|
|
|
handlers = {
|
|
str: sendresponse,
|
|
wireproto.streamres: sendstream,
|
|
wireproto.pushres: sendpushresponse,
|
|
wireproto.pusherr: sendpusherror,
|
|
wireproto.ooberror: sendooberror,
|
|
}
|
|
|
|
def serve_one(self):
|
|
cmd = self.fin.readline()[:-1]
|
|
if cmd and cmd in wireproto.commands:
|
|
rsp = wireproto.dispatch(self.repo, self, cmd)
|
|
self.handlers[rsp.__class__](self, rsp)
|
|
elif cmd:
|
|
impl = getattr(self, 'do_' + cmd, None)
|
|
if impl:
|
|
r = impl()
|
|
if r is not None:
|
|
self.sendresponse(r)
|
|
else: self.sendresponse("")
|
|
return cmd != ''
|
|
|
|
def _client(self):
|
|
client = encoding.environ.get('SSH_CLIENT', '').split(' ', 1)[0]
|
|
return 'remote:ssh:' + client
|