i18n: cache the result of every gettext call
In looking at profiler output for 'hg log' on mozilla-central, I
noticed we spent a _huge_ amount of time in gettext relative to what
it's doing. Caching provides a roughly 15% performance improvement
even on repositories as small as hg.
== hg repo on linux ==
Before:
% cumulative self
time seconds seconds name
5.05 0.19 0.19 i18n.py:62:gettext
4.84 0.18 0.18 revlog.py:88:decompress
2.95 0.17 0.11 changelog.py:201:node
2.32 0.09 0.09 ui.py:577:write
2.11 0.08 0.08 i18n.py:72:gettext
2.11 0.08 0.08 obsolete.py:196:_fm0readmarkers
1.89 0.07 0.07 obsolete.py:569:_load
1.68 0.63 0.06 localrepo.py:29:__get__
real 0m4.026s
user 0m3.993s
sys 0m0.034s
After:
% cumulative self
time seconds seconds name
8.05 0.26 0.26 revlog.py:88:decompress
2.68 0.22 0.09 color.py:395:write
2.20 0.07 0.07 obsolete.py:196:_fm0readmarkers
1.95 0.06 0.06 obsolete.py:174:_fm0readmarkers
1.95 0.06 0.06 ui.py:577:write
1.95 0.06 0.06 util.py:1228:datestr
1.71 0.06 0.06 utf_8.py:16:decode
1.71 0.06 0.06 revlog.py:273:__len__
real 0m3.519s
user 0m3.447s
sys 0m0.073s
== mozilla-central repo on linux ==
Before:
% cumulative self
time seconds seconds name
7.72 2.35 2.35 revlog.py:88:decompress
4.46 1.36 1.36 i18n.py:62:gettext
2.22 0.67 0.67 i18n.py:72:gettext
2.19 1.14 0.67 changelog.py:201:node
2.16 0.66 0.66 ui.py:577:write
1.96 0.60 0.60 utf_8.py:16:decode
1.93 1.97 0.59 color.py:395:write
1.85 0.81 0.56 changelog.py:136:tip
real 0m30.822s
user 0m30.660s
sys 0m0.149s
After:
% cumulative self
time seconds seconds name
9.82 2.49 2.49 revlog.py:88:decompress
2.67 1.31 0.68 localrepo.py:29:__get__
2.57 0.65 0.65 utf_8.py:16:decode
2.48 1.01 0.63 changelog.py:201:node
2.10 0.82 0.53 changelog.py:136:tip
2.01 0.51 0.51 ui.py:577:write
1.91 0.49 0.49 util.py:1232:datestr
1.85 1.65 0.47 color.py:395:write
real 0m25.619s
user 0m25.446s
sys 0m0.166s
== cpython repo on os x =
Before:
% cumulative self
time seconds seconds name
5.05 1.35 1.35 cmdutil.py:982:_show
4.59 1.22 1.22 revlog.py:274:__len__
3.98 1.06 1.06 i18n.py:62:gettext
3.91 1.04 1.04 revlog.py:1016:revision
3.68 0.98 0.98 revlog.py:337:parents
3.45 0.92 0.92 revlog.py:88:decompress
2.91 0.78 0.78 revlog.py:309:rev
2.62 0.70 0.70 revlog.py:1033:revision
real 0m30.414s
user 0m28.145s
sys 0m0.541s
After:
% cumulative self
time seconds seconds name
7.98 1.66 1.66 cmdutil.py:982:_show
6.83 1.42 1.42 changelog.py:46:decodeextra
5.18 1.08 1.08 revlog.py:274:__len__
3.94 0.82 0.82 revlog.py:1016:revision
3.41 0.71 0.71 revlog.py:309:rev
3.32 0.69 0.69 revlog.py:88:decompress
2.99 0.63 0.62 revlog.py:1033:revision
2.69 0.56 0.56 revlog.py:341:start
real 0m22.811s
user 0m21.883s
sys 0m0.397s
# cvs.py: CVS conversion code inspired by hg-cvs-import and git-cvsimport
#
# Copyright 2005-2009 Matt Mackall <mpm@selenic.com> and others
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
import os, re, socket, errno
from cStringIO import StringIO
from mercurial import encoding, util
from mercurial.i18n import _
from common import NoRepo, commit, converter_source, checktool
from common import makedatetimestamp
import cvsps
class convert_cvs(converter_source):
def __init__(self, ui, path, rev=None):
super(convert_cvs, self).__init__(ui, path, rev=rev)
cvs = os.path.join(path, "CVS")
if not os.path.exists(cvs):
raise NoRepo(_("%s does not look like a CVS checkout") % path)
checktool('cvs')
self.changeset = None
self.files = {}
self.tags = {}
self.lastbranch = {}
self.socket = None
self.cvsroot = open(os.path.join(cvs, "Root")).read()[:-1]
self.cvsrepo = open(os.path.join(cvs, "Repository")).read()[:-1]
self.encoding = encoding.encoding
self._connect()
def _parse(self):
if self.changeset is not None:
return
self.changeset = {}
maxrev = 0
if self.rev:
# TODO: handle tags
try:
# patchset number?
maxrev = int(self.rev)
except ValueError:
raise util.Abort(_('revision %s is not a patchset number')
% self.rev)
d = os.getcwd()
try:
os.chdir(self.path)
id = None
cache = 'update'
if not self.ui.configbool('convert', 'cvsps.cache', True):
cache = None
db = cvsps.createlog(self.ui, cache=cache)
db = cvsps.createchangeset(self.ui, db,
fuzz=int(self.ui.config('convert', 'cvsps.fuzz', 60)),
mergeto=self.ui.config('convert', 'cvsps.mergeto', None),
mergefrom=self.ui.config('convert', 'cvsps.mergefrom', None))
for cs in db:
if maxrev and cs.id > maxrev:
break
id = str(cs.id)
cs.author = self.recode(cs.author)
self.lastbranch[cs.branch] = id
cs.comment = self.recode(cs.comment)
if self.ui.configbool('convert', 'localtimezone'):
cs.date = makedatetimestamp(cs.date[0])
date = util.datestr(cs.date, '%Y-%m-%d %H:%M:%S %1%2')
self.tags.update(dict.fromkeys(cs.tags, id))
files = {}
for f in cs.entries:
files[f.file] = "%s%s" % ('.'.join([str(x)
for x in f.revision]),
['', '(DEAD)'][f.dead])
# add current commit to set
c = commit(author=cs.author, date=date,
parents=[str(p.id) for p in cs.parents],
desc=cs.comment, branch=cs.branch or '')
self.changeset[id] = c
self.files[id] = files
self.heads = self.lastbranch.values()
finally:
os.chdir(d)
def _connect(self):
root = self.cvsroot
conntype = None
user, host = None, None
cmd = ['cvs', 'server']
self.ui.status(_("connecting to %s\n") % root)
if root.startswith(":pserver:"):
root = root[9:]
m = re.match(r'(?:(.*?)(?::(.*?))?@)?([^:\/]*)(?::(\d*))?(.*)',
root)
if m:
conntype = "pserver"
user, passw, serv, port, root = m.groups()
if not user:
user = "anonymous"
if not port:
port = 2401
else:
port = int(port)
format0 = ":pserver:%s@%s:%s" % (user, serv, root)
format1 = ":pserver:%s@%s:%d%s" % (user, serv, port, root)
if not passw:
passw = "A"
cvspass = os.path.expanduser("~/.cvspass")
try:
pf = open(cvspass)
for line in pf.read().splitlines():
part1, part2 = line.split(' ', 1)
# /1 :pserver:user@example.com:2401/cvsroot/foo
# Ah<Z
if part1 == '/1':
part1, part2 = part2.split(' ', 1)
format = format1
# :pserver:user@example.com:/cvsroot/foo Ah<Z
else:
format = format0
if part1 == format:
passw = part2
break
pf.close()
except IOError, inst:
if inst.errno != errno.ENOENT:
if not getattr(inst, 'filename', None):
inst.filename = cvspass
raise
sck = socket.socket()
sck.connect((serv, port))
sck.send("\n".join(["BEGIN AUTH REQUEST", root, user, passw,
"END AUTH REQUEST", ""]))
if sck.recv(128) != "I LOVE YOU\n":
raise util.Abort(_("CVS pserver authentication failed"))
self.writep = self.readp = sck.makefile('r+')
if not conntype and root.startswith(":local:"):
conntype = "local"
root = root[7:]
if not conntype:
# :ext:user@host/home/user/path/to/cvsroot
if root.startswith(":ext:"):
root = root[5:]
m = re.match(r'(?:([^@:/]+)@)?([^:/]+):?(.*)', root)
# Do not take Windows path "c:\foo\bar" for a connection strings
if os.path.isdir(root) or not m:
conntype = "local"
else:
conntype = "rsh"
user, host, root = m.group(1), m.group(2), m.group(3)
if conntype != "pserver":
if conntype == "rsh":
rsh = os.environ.get("CVS_RSH") or "ssh"
if user:
cmd = [rsh, '-l', user, host] + cmd
else:
cmd = [rsh, host] + cmd
# popen2 does not support argument lists under Windows
cmd = [util.shellquote(arg) for arg in cmd]
cmd = util.quotecommand(' '.join(cmd))
self.writep, self.readp = util.popen2(cmd)
self.realroot = root
self.writep.write("Root %s\n" % root)
self.writep.write("Valid-responses ok error Valid-requests Mode"
" M Mbinary E Checked-in Created Updated"
" Merged Removed\n")
self.writep.write("valid-requests\n")
self.writep.flush()
r = self.readp.readline()
if not r.startswith("Valid-requests"):
raise util.Abort(_('unexpected response from CVS server '
'(expected "Valid-requests", but got %r)')
% r)
if "UseUnchanged" in r:
self.writep.write("UseUnchanged\n")
self.writep.flush()
r = self.readp.readline()
def getheads(self):
self._parse()
return self.heads
def getfile(self, name, rev):
def chunkedread(fp, count):
# file-objects returned by socket.makefile() do not handle
# large read() requests very well.
chunksize = 65536
output = StringIO()
while count > 0:
data = fp.read(min(count, chunksize))
if not data:
raise util.Abort(_("%d bytes missing from remote file")
% count)
count -= len(data)
output.write(data)
return output.getvalue()
self._parse()
if rev.endswith("(DEAD)"):
return None, None
args = ("-N -P -kk -r %s --" % rev).split()
args.append(self.cvsrepo + '/' + name)
for x in args:
self.writep.write("Argument %s\n" % x)
self.writep.write("Directory .\n%s\nco\n" % self.realroot)
self.writep.flush()
data = ""
mode = None
while True:
line = self.readp.readline()
if line.startswith("Created ") or line.startswith("Updated "):
self.readp.readline() # path
self.readp.readline() # entries
mode = self.readp.readline()[:-1]
count = int(self.readp.readline()[:-1])
data = chunkedread(self.readp, count)
elif line.startswith(" "):
data += line[1:]
elif line.startswith("M "):
pass
elif line.startswith("Mbinary "):
count = int(self.readp.readline()[:-1])
data = chunkedread(self.readp, count)
else:
if line == "ok\n":
if mode is None:
raise util.Abort(_('malformed response from CVS'))
return (data, "x" in mode and "x" or "")
elif line.startswith("E "):
self.ui.warn(_("cvs server: %s\n") % line[2:])
elif line.startswith("Remove"):
self.readp.readline()
else:
raise util.Abort(_("unknown CVS response: %s") % line)
def getchanges(self, rev, full):
if full:
raise util.Abort(_("convert from cvs do not support --full"))
self._parse()
return sorted(self.files[rev].iteritems()), {}
def getcommit(self, rev):
self._parse()
return self.changeset[rev]
def gettags(self):
self._parse()
return self.tags
def getchangedfiles(self, rev, i):
self._parse()
return sorted(self.files[rev])