mercurial/sshrepo.py
author mason@suse.com
Thu, 01 Sep 2005 07:34:53 -0700
changeset 1183 d9e85a75dbda
parent 1103 808a9f0e7af0
child 1251 84cf8834efb5
permissions -rw-r--r--
Optimize dirstate walking This generally cuts the time for hg status/diff in half, from 2s down to 1s. The main parts I'm trying to optimize are: 1) os.walk stats every file. dirstate.changes then stats every file again. 2) os.walk yields every file and subdir to dirstate.traverse who yields every file and everything in the dirstate map. dirstate.walk then filters this mass and yields every file to the caller. There should be fewer steps in here, and fewer duplicate strings yielded. 3) dirstate.walk runs util.unique on the results from dirstate.traverse, even though it is also passing things through dirstate.seen to look for duplicates. I've turned os.walk into something hg specific that takes all the dirstate ignore and matching rules into account. The new function also takes an function arg (statmatch()) the caller supplies to help filter out files it doesn't care about. dirstate.changes uses this to update state for each file, avoiding the second stat call. dirstate.walk is changed to turn the match function it is passed into a statmatch function. The only real difference is that a statmatch function takes the stat data as a second parameter. It now calls dirstate.walkhelper, who requires a statmatch function to be passed. This fails test-walk, but right now I think this is from a sorting error fixed by this patch. Index: crew/mercurial/dirstate.py ===================================================================

# sshrepo.py - ssh repository proxy class for mercurial
#
# Copyright 2005 Matt Mackall <mpm@selenic.com>
#
# This software may be used and distributed according to the terms
# of the GNU General Public License, incorporated herein by reference.

import os, re, select
from node import *
from remoterepo import *

class sshrepository(remoterepository):
    def __init__(self, ui, path):
        self.url = path
        self.ui = ui

        m = re.match(r'ssh://(([^@]+)@)?([^:/]+)(:(\d+))?(/(.*))?', path)
        if not m:
            raise RepoError("couldn't parse destination %s" % path)

        self.user = m.group(2)
        self.host = m.group(3)
        self.port = m.group(5)
        self.path = m.group(7) or "."

        args = self.user and ("%s@%s" % (self.user, self.host)) or self.host
        args = self.port and ("%s -p %s") % (args, self.port) or args

        sshcmd = self.ui.config("ui", "ssh", "ssh")
        remotecmd = self.ui.config("ui", "remotecmd", "hg")
        cmd = "%s %s '%s -R %s serve --stdio'"
        cmd = cmd % (sshcmd, args, remotecmd, self.path)

        self.pipeo, self.pipei, self.pipee = os.popen3(cmd)

    def readerr(self):
        while 1:
            r,w,x = select.select([self.pipee], [], [], 0)
            if not r: break
            l = self.pipee.readline()
            if not l: break
            self.ui.status("remote: ", l)

    def __del__(self):
        try:
            self.pipeo.close()
            self.pipei.close()
            for l in self.pipee:
                self.ui.status("remote: ", l)
            self.pipee.close()
        except:
            pass

    def dev(self):
        return -1

    def do_cmd(self, cmd, **args):
        self.ui.debug("sending %s command\n" % cmd)
        self.pipeo.write("%s\n" % cmd)
        for k, v in args.items():
            self.pipeo.write("%s %d\n" % (k, len(v)))
            self.pipeo.write(v)
        self.pipeo.flush()

        return self.pipei

    def call(self, cmd, **args):
        r = self.do_cmd(cmd, **args)
        l = r.readline()
        self.readerr()
        try:
            l = int(l)
        except:
            raise RepoError("unexpected response '%s'" % l)
        return r.read(l)

    def lock(self):
        self.call("lock")
        return remotelock(self)

    def unlock(self):
        self.call("unlock")

    def heads(self):
        d = self.call("heads")
        try:
            return map(bin, d[:-1].split(" "))
        except:
            raise RepoError("unexpected response '%s'" % (d[:400] + "..."))

    def branches(self, nodes):
        n = " ".join(map(hex, nodes))
        d = self.call("branches", nodes=n)
        try:
            br = [ tuple(map(bin, b.split(" "))) for b in d.splitlines() ]
            return br
        except:
            raise RepoError("unexpected response '%s'" % (d[:400] + "..."))

    def between(self, pairs):
        n = "\n".join(["-".join(map(hex, p)) for p in pairs])
        d = self.call("between", pairs=n)
        try:
            p = [ l and map(bin, l.split(" ")) or [] for l in d.splitlines() ]
            return p
        except:
            raise RepoError("unexpected response '%s'" % (d[:400] + "..."))

    def changegroup(self, nodes):
        n = " ".join(map(hex, nodes))
        f = self.do_cmd("changegroup", roots=n)
        return self.pipei

    def addchangegroup(self, cg):
        d = self.call("addchangegroup")
        if d:
            raise RepoError("push refused: %s", d)

        while 1:
            d = cg.read(4096)
            if not d: break
            self.pipeo.write(d)
            self.readerr()

        self.pipeo.flush()

        self.readerr()
        l = int(self.pipei.readline())
        return self.pipei.read(l) != ""