2005-08-27 12:13:28 +04:00
|
|
|
"""
|
|
|
|
util.py - Mercurial utility functions and platform specfic implementations
|
|
|
|
|
|
|
|
Copyright 2005 K. Thananchayan <thananck@yahoo.com>
|
2007-06-19 10:51:34 +04:00
|
|
|
Copyright 2005-2007 Matt Mackall <mpm@selenic.com>
|
2006-08-12 23:30:02 +04:00
|
|
|
Copyright 2006 Vadim Gelfer <vadim.gelfer@gmail.com>
|
2005-08-27 12:13:28 +04:00
|
|
|
|
|
|
|
This software may be used and distributed according to the terms
|
|
|
|
of the GNU General Public License, incorporated herein by reference.
|
|
|
|
|
|
|
|
This contains helper routines that are independent of the SCM core and hide
|
|
|
|
platform-specific details from the core.
|
|
|
|
"""
|
2005-06-22 07:31:13 +04:00
|
|
|
|
2006-12-15 05:25:19 +03:00
|
|
|
from i18n import _
|
2008-03-07 00:23:41 +03:00
|
|
|
import cStringIO, errno, getpass, re, shutil, sys, tempfile
|
2007-10-06 02:01:06 +04:00
|
|
|
import os, stat, threading, time, calendar, ConfigParser, locale, glob, osutil
|
2008-01-22 00:00:44 +03:00
|
|
|
import urlparse
|
2006-12-04 01:16:33 +03:00
|
|
|
|
2007-06-19 19:37:41 +04:00
|
|
|
try:
|
|
|
|
set = set
|
|
|
|
frozenset = frozenset
|
|
|
|
except NameError:
|
|
|
|
from sets import Set as set, ImmutableSet as frozenset
|
|
|
|
|
2007-01-30 23:32:23 +03:00
|
|
|
try:
|
2007-06-11 23:14:31 +04:00
|
|
|
_encoding = os.environ.get("HGENCODING")
|
|
|
|
if sys.platform == 'darwin' and not _encoding:
|
|
|
|
# On darwin, getpreferredencoding ignores the locale environment and
|
|
|
|
# always returns mac-roman. We override this if the environment is
|
|
|
|
# not C (has been customized by the user).
|
|
|
|
locale.setlocale(locale.LC_CTYPE, '')
|
|
|
|
_encoding = locale.getlocale()[1]
|
|
|
|
if not _encoding:
|
|
|
|
_encoding = locale.getpreferredencoding() or 'ascii'
|
2007-01-30 23:32:23 +03:00
|
|
|
except locale.Error:
|
|
|
|
_encoding = 'ascii'
|
2006-12-04 01:16:33 +03:00
|
|
|
_encodingmode = os.environ.get("HGENCODINGMODE", "strict")
|
2006-12-09 03:01:05 +03:00
|
|
|
_fallbackencoding = 'ISO-8859-1'
|
2006-12-04 01:16:33 +03:00
|
|
|
|
|
|
|
def tolocal(s):
|
|
|
|
"""
|
|
|
|
Convert a string from internal UTF-8 to local encoding
|
|
|
|
|
|
|
|
All internal strings should be UTF-8 but some repos before the
|
|
|
|
implementation of locale support may contain latin1 or possibly
|
|
|
|
other character sets. We attempt to decode everything strictly
|
|
|
|
using UTF-8, then Latin-1, and failing that, we use UTF-8 and
|
|
|
|
replace unknown characters.
|
|
|
|
"""
|
2006-12-09 03:01:05 +03:00
|
|
|
for e in ('UTF-8', _fallbackencoding):
|
2006-12-04 01:16:33 +03:00
|
|
|
try:
|
|
|
|
u = s.decode(e) # attempt strict decoding
|
|
|
|
return u.encode(_encoding, "replace")
|
2006-12-09 21:46:01 +03:00
|
|
|
except LookupError, k:
|
|
|
|
raise Abort(_("%s, please check your locale settings") % k)
|
2006-12-04 01:16:33 +03:00
|
|
|
except UnicodeDecodeError:
|
|
|
|
pass
|
|
|
|
u = s.decode("utf-8", "replace") # last ditch
|
|
|
|
return u.encode(_encoding, "replace")
|
|
|
|
|
|
|
|
def fromlocal(s):
|
|
|
|
"""
|
|
|
|
Convert a string from the local character encoding to UTF-8
|
|
|
|
|
|
|
|
We attempt to decode strings using the encoding mode set by
|
2007-07-12 18:55:49 +04:00
|
|
|
HGENCODINGMODE, which defaults to 'strict'. In this mode, unknown
|
2006-12-04 01:16:33 +03:00
|
|
|
characters will cause an error message. Other modes include
|
|
|
|
'replace', which replaces unknown characters with a special
|
|
|
|
Unicode character, and 'ignore', which drops the character.
|
|
|
|
"""
|
|
|
|
try:
|
|
|
|
return s.decode(_encoding, _encodingmode).encode("utf-8")
|
|
|
|
except UnicodeDecodeError, inst:
|
|
|
|
sub = s[max(0, inst.start-10):inst.start+10]
|
2006-12-09 21:46:01 +03:00
|
|
|
raise Abort("decoding near '%s': %s!" % (sub, inst))
|
|
|
|
except LookupError, k:
|
|
|
|
raise Abort(_("%s, please check your locale settings") % k)
|
2006-12-04 01:16:33 +03:00
|
|
|
|
|
|
|
def locallen(s):
|
|
|
|
"""Find the length in characters of a local string"""
|
|
|
|
return len(s.decode(_encoding, "replace"))
|
|
|
|
|
2006-07-13 20:40:01 +04:00
|
|
|
# used by parsedate
|
2006-12-06 22:13:31 +03:00
|
|
|
defaultdateformats = (
|
|
|
|
'%Y-%m-%d %H:%M:%S',
|
|
|
|
'%Y-%m-%d %I:%M:%S%p',
|
|
|
|
'%Y-%m-%d %H:%M',
|
|
|
|
'%Y-%m-%d %I:%M%p',
|
|
|
|
'%Y-%m-%d',
|
|
|
|
'%m-%d',
|
|
|
|
'%m/%d',
|
|
|
|
'%m/%d/%y',
|
|
|
|
'%m/%d/%Y',
|
|
|
|
'%a %b %d %H:%M:%S %Y',
|
|
|
|
'%a %b %d %I:%M:%S%p %Y',
|
2007-06-25 16:46:20 +04:00
|
|
|
'%a, %d %b %Y %H:%M:%S', # GNU coreutils "/bin/date --rfc-2822"
|
2006-12-06 22:13:31 +03:00
|
|
|
'%b %d %H:%M:%S %Y',
|
2006-12-07 00:11:44 +03:00
|
|
|
'%b %d %I:%M:%S%p %Y',
|
|
|
|
'%b %d %H:%M:%S',
|
2006-12-06 22:13:31 +03:00
|
|
|
'%b %d %I:%M:%S%p',
|
|
|
|
'%b %d %H:%M',
|
|
|
|
'%b %d %I:%M%p',
|
|
|
|
'%b %d %Y',
|
|
|
|
'%b %d',
|
|
|
|
'%H:%M:%S',
|
|
|
|
'%I:%M:%SP',
|
|
|
|
'%H:%M',
|
|
|
|
'%I:%M%p',
|
|
|
|
)
|
2006-07-13 20:40:01 +04:00
|
|
|
|
2006-12-07 00:11:44 +03:00
|
|
|
extendeddateformats = defaultdateformats + (
|
|
|
|
"%Y",
|
|
|
|
"%Y-%m",
|
|
|
|
"%b",
|
|
|
|
"%b %Y",
|
|
|
|
)
|
|
|
|
|
2006-04-29 01:50:23 +04:00
|
|
|
class SignalInterrupt(Exception):
|
|
|
|
"""Exception raised on SIGTERM and SIGHUP."""
|
|
|
|
|
2007-02-09 08:48:26 +03:00
|
|
|
# differences from SafeConfigParser:
|
|
|
|
# - case-sensitive keys
|
|
|
|
# - allows values that are not strings (this means that you may not
|
|
|
|
# be able to save the configuration to a file)
|
2006-10-16 22:38:53 +04:00
|
|
|
class configparser(ConfigParser.SafeConfigParser):
|
|
|
|
def optionxform(self, optionstr):
|
|
|
|
return optionstr
|
|
|
|
|
2007-02-09 08:48:26 +03:00
|
|
|
def set(self, section, option, value):
|
|
|
|
return ConfigParser.ConfigParser.set(self, section, option, value)
|
|
|
|
|
|
|
|
def _interpolate(self, section, option, rawval, vars):
|
|
|
|
if not isinstance(rawval, basestring):
|
|
|
|
return rawval
|
|
|
|
return ConfigParser.SafeConfigParser._interpolate(self, section,
|
|
|
|
option, rawval, vars)
|
|
|
|
|
2006-09-22 19:19:25 +04:00
|
|
|
def cachefunc(func):
|
|
|
|
'''cache the result of function calls'''
|
2006-09-22 19:58:22 +04:00
|
|
|
# XXX doesn't handle keywords args
|
2006-09-22 19:19:25 +04:00
|
|
|
cache = {}
|
|
|
|
if func.func_code.co_argcount == 1:
|
2006-09-22 19:58:22 +04:00
|
|
|
# we gain a small amount of time because
|
|
|
|
# we don't need to pack/unpack the list
|
2006-09-22 19:19:25 +04:00
|
|
|
def f(arg):
|
|
|
|
if arg not in cache:
|
|
|
|
cache[arg] = func(arg)
|
|
|
|
return cache[arg]
|
|
|
|
else:
|
|
|
|
def f(*args):
|
|
|
|
if args not in cache:
|
|
|
|
cache[args] = func(*args)
|
|
|
|
return cache[args]
|
|
|
|
|
|
|
|
return f
|
|
|
|
|
2005-09-21 22:44:08 +04:00
|
|
|
def pipefilter(s, cmd):
|
|
|
|
'''filter string S through command CMD, returning its output'''
|
2007-06-19 01:52:31 +04:00
|
|
|
(pin, pout) = os.popen2(cmd, 'b')
|
2005-09-15 11:59:16 +04:00
|
|
|
def writer():
|
2006-04-19 22:41:25 +04:00
|
|
|
try:
|
|
|
|
pin.write(s)
|
|
|
|
pin.close()
|
|
|
|
except IOError, inst:
|
|
|
|
if inst.errno != errno.EPIPE:
|
|
|
|
raise
|
2005-09-15 11:59:16 +04:00
|
|
|
|
|
|
|
# we should use select instead on UNIX, but this will work on most
|
|
|
|
# systems, including Windows
|
|
|
|
w = threading.Thread(target=writer)
|
|
|
|
w.start()
|
|
|
|
f = pout.read()
|
|
|
|
pout.close()
|
|
|
|
w.join()
|
|
|
|
return f
|
2005-06-22 07:31:13 +04:00
|
|
|
|
2005-09-21 22:44:08 +04:00
|
|
|
def tempfilter(s, cmd):
|
|
|
|
'''filter string S through a pair of temporary files with CMD.
|
|
|
|
CMD is used as a template to create the real command to be run,
|
|
|
|
with the strings INFILE and OUTFILE replaced by the real names of
|
|
|
|
the temporary files generated.'''
|
|
|
|
inname, outname = None, None
|
|
|
|
try:
|
2006-04-30 23:11:22 +04:00
|
|
|
infd, inname = tempfile.mkstemp(prefix='hg-filter-in-')
|
2005-09-21 22:44:08 +04:00
|
|
|
fp = os.fdopen(infd, 'wb')
|
|
|
|
fp.write(s)
|
|
|
|
fp.close()
|
2006-04-30 23:11:22 +04:00
|
|
|
outfd, outname = tempfile.mkstemp(prefix='hg-filter-out-')
|
2005-09-21 22:44:08 +04:00
|
|
|
os.close(outfd)
|
|
|
|
cmd = cmd.replace('INFILE', inname)
|
|
|
|
cmd = cmd.replace('OUTFILE', outname)
|
|
|
|
code = os.system(cmd)
|
2007-06-08 18:24:43 +04:00
|
|
|
if sys.platform == 'OpenVMS' and code & 1:
|
|
|
|
code = 0
|
2005-10-19 05:38:39 +04:00
|
|
|
if code: raise Abort(_("command '%s' failed: %s") %
|
2005-09-21 22:44:08 +04:00
|
|
|
(cmd, explain_exit(code)))
|
|
|
|
return open(outname, 'rb').read()
|
|
|
|
finally:
|
|
|
|
try:
|
|
|
|
if inname: os.unlink(inname)
|
|
|
|
except: pass
|
|
|
|
try:
|
|
|
|
if outname: os.unlink(outname)
|
|
|
|
except: pass
|
|
|
|
|
|
|
|
filtertable = {
|
|
|
|
'tempfile:': tempfilter,
|
|
|
|
'pipe:': pipefilter,
|
|
|
|
}
|
|
|
|
|
|
|
|
def filter(s, cmd):
|
|
|
|
"filter a string through a command that transforms its input to its output"
|
|
|
|
for name, fn in filtertable.iteritems():
|
|
|
|
if cmd.startswith(name):
|
|
|
|
return fn(s, cmd[len(name):].lstrip())
|
|
|
|
return pipefilter(s, cmd)
|
|
|
|
|
2005-08-24 06:58:46 +04:00
|
|
|
def binary(s):
|
2005-08-27 12:13:28 +04:00
|
|
|
"""return true if a string is binary data using diff's heuristic"""
|
2005-08-24 06:58:46 +04:00
|
|
|
if s and '\0' in s[:4096]:
|
|
|
|
return True
|
|
|
|
return False
|
|
|
|
|
2005-07-01 12:34:50 +04:00
|
|
|
def unique(g):
|
2005-08-27 12:13:28 +04:00
|
|
|
"""return the uniq elements of iterable g"""
|
2007-12-28 08:55:40 +03:00
|
|
|
return dict.fromkeys(g).keys()
|
2005-07-01 12:34:50 +04:00
|
|
|
|
2005-08-08 00:43:11 +04:00
|
|
|
class Abort(Exception):
|
|
|
|
"""Raised if a command needs to print an error and exit."""
|
2005-06-28 14:38:33 +04:00
|
|
|
|
2006-10-27 20:17:12 +04:00
|
|
|
class UnexpectedOutput(Abort):
|
|
|
|
"""Raised to print an error with part of output and exit."""
|
|
|
|
|
2005-07-18 18:54:21 +04:00
|
|
|
def always(fn): return True
|
|
|
|
def never(fn): return False
|
|
|
|
|
2007-01-30 23:32:18 +03:00
|
|
|
def expand_glob(pats):
|
|
|
|
'''On Windows, expand the implicit globs in a list of patterns'''
|
|
|
|
if os.name != 'nt':
|
|
|
|
return list(pats)
|
|
|
|
ret = []
|
|
|
|
for p in pats:
|
|
|
|
kind, name = patkind(p, None)
|
|
|
|
if kind is None:
|
|
|
|
globbed = glob.glob(name)
|
|
|
|
if globbed:
|
|
|
|
ret.extend(globbed)
|
|
|
|
continue
|
|
|
|
# if we couldn't expand the glob, just keep it around
|
|
|
|
ret.append(p)
|
|
|
|
return ret
|
|
|
|
|
2005-12-01 19:48:22 +03:00
|
|
|
def patkind(name, dflt_pat='glob'):
|
|
|
|
"""Split a string into an optional pattern kind prefix and the
|
|
|
|
actual pattern."""
|
|
|
|
for prefix in 're', 'glob', 'path', 'relglob', 'relpath', 'relre':
|
|
|
|
if name.startswith(prefix + ':'): return name.split(':', 1)
|
|
|
|
return dflt_pat, name
|
|
|
|
|
2005-08-26 15:06:58 +04:00
|
|
|
def globre(pat, head='^', tail='$'):
|
2005-07-18 18:54:21 +04:00
|
|
|
"convert a glob pattern into a regexp"
|
|
|
|
i, n = 0, len(pat)
|
|
|
|
res = ''
|
2008-01-25 23:54:25 +03:00
|
|
|
group = 0
|
2005-07-18 18:54:21 +04:00
|
|
|
def peek(): return i < n and pat[i]
|
|
|
|
while i < n:
|
|
|
|
c = pat[i]
|
|
|
|
i = i+1
|
|
|
|
if c == '*':
|
|
|
|
if peek() == '*':
|
|
|
|
i += 1
|
|
|
|
res += '.*'
|
|
|
|
else:
|
|
|
|
res += '[^/]*'
|
|
|
|
elif c == '?':
|
|
|
|
res += '.'
|
|
|
|
elif c == '[':
|
|
|
|
j = i
|
|
|
|
if j < n and pat[j] in '!]':
|
|
|
|
j += 1
|
|
|
|
while j < n and pat[j] != ']':
|
|
|
|
j += 1
|
|
|
|
if j >= n:
|
|
|
|
res += '\\['
|
|
|
|
else:
|
|
|
|
stuff = pat[i:j].replace('\\','\\\\')
|
|
|
|
i = j + 1
|
|
|
|
if stuff[0] == '!':
|
|
|
|
stuff = '^' + stuff[1:]
|
|
|
|
elif stuff[0] == '^':
|
|
|
|
stuff = '\\' + stuff
|
|
|
|
res = '%s[%s]' % (res, stuff)
|
|
|
|
elif c == '{':
|
2008-01-25 23:54:25 +03:00
|
|
|
group += 1
|
2005-07-18 18:54:21 +04:00
|
|
|
res += '(?:'
|
|
|
|
elif c == '}' and group:
|
|
|
|
res += ')'
|
2008-01-25 23:54:25 +03:00
|
|
|
group -= 1
|
2005-07-18 18:54:21 +04:00
|
|
|
elif c == ',' and group:
|
|
|
|
res += '|'
|
2006-03-24 01:39:53 +03:00
|
|
|
elif c == '\\':
|
|
|
|
p = peek()
|
|
|
|
if p:
|
|
|
|
i += 1
|
|
|
|
res += re.escape(p)
|
|
|
|
else:
|
|
|
|
res += re.escape(c)
|
2005-07-18 18:54:21 +04:00
|
|
|
else:
|
|
|
|
res += re.escape(c)
|
|
|
|
return head + res + tail
|
|
|
|
|
2005-07-30 00:30:12 +04:00
|
|
|
_globchars = {'[': 1, '{': 1, '*': 1, '?': 1}
|
|
|
|
|
2007-03-16 06:22:57 +03:00
|
|
|
def pathto(root, n1, n2):
|
2005-08-13 03:06:52 +04:00
|
|
|
'''return the relative path from one place to another.
|
2007-03-16 06:22:57 +03:00
|
|
|
root should use os.sep to separate directories
|
2006-11-15 23:56:47 +03:00
|
|
|
n1 should use os.sep to separate directories
|
|
|
|
n2 should use "/" to separate directories
|
|
|
|
returns an os.sep-separated path.
|
2007-03-16 06:22:57 +03:00
|
|
|
|
|
|
|
If n1 is a relative path, it's assumed it's
|
|
|
|
relative to root.
|
|
|
|
n2 should always be relative to root.
|
2006-11-15 23:56:47 +03:00
|
|
|
'''
|
2005-08-13 03:06:52 +04:00
|
|
|
if not n1: return localpath(n2)
|
2007-03-16 06:22:58 +03:00
|
|
|
if os.path.isabs(n1):
|
|
|
|
if os.path.splitdrive(root)[0] != os.path.splitdrive(n1)[0]:
|
|
|
|
return os.path.join(root, localpath(n2))
|
|
|
|
n2 = '/'.join((pconvert(root), n2))
|
2008-01-09 15:30:36 +03:00
|
|
|
a, b = splitpath(n1), n2.split('/')
|
2005-11-14 04:59:35 +03:00
|
|
|
a.reverse()
|
|
|
|
b.reverse()
|
2005-08-12 23:16:58 +04:00
|
|
|
while a and b and a[-1] == b[-1]:
|
2005-11-14 04:59:35 +03:00
|
|
|
a.pop()
|
|
|
|
b.pop()
|
2005-08-12 23:16:58 +04:00
|
|
|
b.reverse()
|
2008-02-15 15:38:37 +03:00
|
|
|
return os.sep.join((['..'] * len(a)) + b) or '.'
|
2005-08-12 23:16:58 +04:00
|
|
|
|
2005-08-27 10:29:10 +04:00
|
|
|
def canonpath(root, cwd, myname):
|
2005-08-27 12:13:28 +04:00
|
|
|
"""return the canonical path of myname, given cwd and root"""
|
2005-12-01 19:48:42 +03:00
|
|
|
if root == os.sep:
|
|
|
|
rootsep = os.sep
|
2008-01-09 15:30:35 +03:00
|
|
|
elif endswithsep(root):
|
2006-05-12 08:55:15 +04:00
|
|
|
rootsep = root
|
2005-12-01 19:48:42 +03:00
|
|
|
else:
|
2006-03-01 01:25:34 +03:00
|
|
|
rootsep = root + os.sep
|
2005-08-08 00:43:11 +04:00
|
|
|
name = myname
|
2006-04-19 20:08:45 +04:00
|
|
|
if not os.path.isabs(name):
|
2005-08-27 10:29:10 +04:00
|
|
|
name = os.path.join(root, cwd, name)
|
2005-08-08 00:43:11 +04:00
|
|
|
name = os.path.normpath(name)
|
2007-08-10 21:46:03 +04:00
|
|
|
audit_path = path_auditor(root)
|
2006-05-14 10:00:05 +04:00
|
|
|
if name != rootsep and name.startswith(rootsep):
|
2006-03-19 16:53:58 +03:00
|
|
|
name = name[len(rootsep):]
|
|
|
|
audit_path(name)
|
|
|
|
return pconvert(name)
|
2005-08-27 10:29:10 +04:00
|
|
|
elif name == root:
|
2005-08-08 00:43:11 +04:00
|
|
|
return ''
|
|
|
|
else:
|
2006-04-22 03:09:43 +04:00
|
|
|
# Determine whether `name' is in the hierarchy at or beneath `root',
|
|
|
|
# by iterating name=dirname(name) until that causes no change (can't
|
|
|
|
# check name == '/', because that doesn't work on windows). For each
|
|
|
|
# `name', compare dev/inode numbers. If they match, the list `rel'
|
|
|
|
# holds the reversed list of components making up the relative file
|
|
|
|
# name we want.
|
|
|
|
root_st = os.stat(root)
|
|
|
|
rel = []
|
|
|
|
while True:
|
|
|
|
try:
|
|
|
|
name_st = os.stat(name)
|
|
|
|
except OSError:
|
|
|
|
break
|
2006-05-04 09:47:08 +04:00
|
|
|
if samestat(name_st, root_st):
|
2007-02-15 10:18:23 +03:00
|
|
|
if not rel:
|
|
|
|
# name was actually the same as root (maybe a symlink)
|
|
|
|
return ''
|
2006-04-22 03:09:43 +04:00
|
|
|
rel.reverse()
|
|
|
|
name = os.path.join(*rel)
|
|
|
|
audit_path(name)
|
|
|
|
return pconvert(name)
|
|
|
|
dirname, basename = os.path.split(name)
|
|
|
|
rel.append(basename)
|
|
|
|
if dirname == name:
|
|
|
|
break
|
|
|
|
name = dirname
|
|
|
|
|
2005-08-27 10:29:10 +04:00
|
|
|
raise Abort('%s not under root' % myname)
|
2005-08-15 00:23:36 +04:00
|
|
|
|
2007-03-11 05:00:59 +03:00
|
|
|
def matcher(canonroot, cwd='', names=[], inc=[], exc=[], src=None):
|
|
|
|
return _matcher(canonroot, cwd, names, inc, exc, 'glob', src)
|
2005-10-19 11:02:41 +04:00
|
|
|
|
2007-03-11 05:00:59 +03:00
|
|
|
def cmdmatcher(canonroot, cwd='', names=[], inc=[], exc=[], src=None,
|
|
|
|
globbed=False, default=None):
|
2007-03-11 05:00:57 +03:00
|
|
|
default = default or 'relpath'
|
|
|
|
if default == 'relpath' and not globbed:
|
2007-01-30 23:32:20 +03:00
|
|
|
names = expand_glob(names)
|
2007-03-11 05:00:59 +03:00
|
|
|
return _matcher(canonroot, cwd, names, inc, exc, default, src)
|
2005-10-19 11:02:41 +04:00
|
|
|
|
2007-03-11 05:00:59 +03:00
|
|
|
def _matcher(canonroot, cwd, names, inc, exc, dflt_pat, src):
|
2005-08-27 12:13:28 +04:00
|
|
|
"""build a function to match a set of file patterns
|
|
|
|
|
|
|
|
arguments:
|
|
|
|
canonroot - the canonical root of the tree you're matching against
|
|
|
|
cwd - the current working directory, if relevant
|
|
|
|
names - patterns to find
|
|
|
|
inc - patterns to include
|
|
|
|
exc - patterns to exclude
|
2007-03-11 05:00:45 +03:00
|
|
|
dflt_pat - if a pattern in names has no explicit type, assume this one
|
|
|
|
src - where these patterns came from (e.g. .hgignore)
|
2005-08-27 12:13:28 +04:00
|
|
|
|
|
|
|
a pattern is one of:
|
2007-03-11 05:00:45 +03:00
|
|
|
'glob:<glob>' - a glob relative to cwd
|
|
|
|
're:<regexp>' - a regular expression
|
|
|
|
'path:<path>' - a path relative to canonroot
|
|
|
|
'relglob:<glob>' - an unrooted glob (*.c matches C files in all dirs)
|
|
|
|
'relpath:<path>' - a path relative to cwd
|
|
|
|
'relre:<regexp>' - a regexp that doesn't have to match the start of a name
|
|
|
|
'<something>' - one of the cases above, selected by the dflt_pat argument
|
2005-08-27 12:13:28 +04:00
|
|
|
|
|
|
|
returns:
|
|
|
|
a 3-tuple containing
|
2007-03-11 05:00:45 +03:00
|
|
|
- list of roots (places where one should start a recursive walk of the fs);
|
|
|
|
this often matches the explicit non-pattern names passed in, but also
|
|
|
|
includes the initial part of glob: patterns that has no glob characters
|
2005-08-27 12:13:28 +04:00
|
|
|
- a bool match(filename) function
|
|
|
|
- a bool indicating if any patterns were passed in
|
|
|
|
"""
|
|
|
|
|
2007-03-11 05:01:00 +03:00
|
|
|
# a common case: no patterns at all
|
|
|
|
if not names and not inc and not exc:
|
|
|
|
return [], always, False
|
|
|
|
|
2005-10-19 11:02:41 +04:00
|
|
|
def contains_glob(name):
|
2005-07-30 00:30:12 +04:00
|
|
|
for c in name:
|
2005-10-19 11:02:41 +04:00
|
|
|
if c in _globchars: return True
|
|
|
|
return False
|
Clean up walk and changes code to use normalised names properly.
New function: commands.pathto returns the relative path from one path
to another. For example, given foo/bar and baz/quux, it will return
../../baz/quux. This new function is used by the walk and status code
to print relative paths correctly.
New command: debugwalk exercises the walk code without doing anything
more.
hg.dirstate.walk now yields normalised names. For example, if you're
in the baz directory and you ask it to walk ../foo/bar/.., it will yield
names starting with foo/.
As a result of this change, all of the other walk and changes methods
in this module also return normalised names.
The util.matcher function now normalises globs and path names, so that
it will match normalised names properly.
Finally, util.matcher uses the non-glob prefix of a glob to tell walk
which directories to scan. Perviously, a glob like foo/* would scan
everything, but only return matches for foo/*. Now, foo/* only scans
under foo (using the globprefix function), which is much faster.
2005-08-01 05:42:46 +04:00
|
|
|
|
2007-04-04 11:22:05 +04:00
|
|
|
def regex(kind, name, tail):
|
2005-07-21 21:21:33 +04:00
|
|
|
'''convert a pattern into a regular expression'''
|
2007-03-11 05:00:51 +03:00
|
|
|
if not name:
|
|
|
|
return ''
|
Clean up walk and changes code to use normalised names properly.
New function: commands.pathto returns the relative path from one path
to another. For example, given foo/bar and baz/quux, it will return
../../baz/quux. This new function is used by the walk and status code
to print relative paths correctly.
New command: debugwalk exercises the walk code without doing anything
more.
hg.dirstate.walk now yields normalised names. For example, if you're
in the baz directory and you ask it to walk ../foo/bar/.., it will yield
names starting with foo/.
As a result of this change, all of the other walk and changes methods
in this module also return normalised names.
The util.matcher function now normalises globs and path names, so that
it will match normalised names properly.
Finally, util.matcher uses the non-glob prefix of a glob to tell walk
which directories to scan. Perviously, a glob like foo/* would scan
everything, but only return matches for foo/*. Now, foo/* only scans
under foo (using the globprefix function), which is much faster.
2005-08-01 05:42:46 +04:00
|
|
|
if kind == 're':
|
|
|
|
return name
|
|
|
|
elif kind == 'path':
|
2005-08-13 04:09:17 +04:00
|
|
|
return '^' + re.escape(name) + '(?:/|$)'
|
2005-09-17 11:27:27 +04:00
|
|
|
elif kind == 'relglob':
|
2007-04-04 11:22:06 +04:00
|
|
|
return globre(name, '(?:|.*/)', tail)
|
2005-08-13 04:09:17 +04:00
|
|
|
elif kind == 'relpath':
|
2007-03-11 05:00:59 +03:00
|
|
|
return re.escape(name) + '(?:/|$)'
|
2005-09-17 11:27:27 +04:00
|
|
|
elif kind == 'relre':
|
|
|
|
if name.startswith('^'):
|
|
|
|
return name
|
|
|
|
return '.*' + name
|
2007-04-04 11:22:05 +04:00
|
|
|
return globre(name, '', tail)
|
2005-07-21 21:21:33 +04:00
|
|
|
|
2007-04-04 11:22:05 +04:00
|
|
|
def matchfn(pats, tail):
|
2005-07-21 21:21:33 +04:00
|
|
|
"""build a matching function from a set of patterns"""
|
2005-10-27 03:10:56 +04:00
|
|
|
if not pats:
|
|
|
|
return
|
2007-04-24 21:53:25 +04:00
|
|
|
try:
|
|
|
|
pat = '(?:%s)' % '|'.join([regex(k, p, tail) for (k, p) in pats])
|
2008-02-12 01:13:43 +03:00
|
|
|
if len(pat) > 20000:
|
|
|
|
raise OverflowError()
|
2007-04-24 21:53:25 +04:00
|
|
|
return re.compile(pat).match
|
2007-08-19 23:03:56 +04:00
|
|
|
except OverflowError:
|
|
|
|
# We're using a Python with a tiny regex engine and we
|
|
|
|
# made it explode, so we'll divide the pattern list in two
|
|
|
|
# until it works
|
|
|
|
l = len(pats)
|
|
|
|
if l < 2:
|
|
|
|
raise
|
2007-10-11 18:19:12 +04:00
|
|
|
a, b = matchfn(pats[:l//2], tail), matchfn(pats[l//2:], tail)
|
2007-08-19 23:03:56 +04:00
|
|
|
return lambda s: a(s) or b(s)
|
2007-04-24 21:53:25 +04:00
|
|
|
except re.error:
|
|
|
|
for k, p in pats:
|
|
|
|
try:
|
|
|
|
re.compile('(?:%s)' % regex(k, p, tail))
|
|
|
|
except re.error:
|
|
|
|
if src:
|
|
|
|
raise Abort("%s: invalid pattern (%s): %s" %
|
|
|
|
(src, k, p))
|
|
|
|
else:
|
|
|
|
raise Abort("invalid pattern (%s): %s" % (k, p))
|
|
|
|
raise Abort("invalid pattern")
|
2005-07-21 21:21:33 +04:00
|
|
|
|
Clean up walk and changes code to use normalised names properly.
New function: commands.pathto returns the relative path from one path
to another. For example, given foo/bar and baz/quux, it will return
../../baz/quux. This new function is used by the walk and status code
to print relative paths correctly.
New command: debugwalk exercises the walk code without doing anything
more.
hg.dirstate.walk now yields normalised names. For example, if you're
in the baz directory and you ask it to walk ../foo/bar/.., it will yield
names starting with foo/.
As a result of this change, all of the other walk and changes methods
in this module also return normalised names.
The util.matcher function now normalises globs and path names, so that
it will match normalised names properly.
Finally, util.matcher uses the non-glob prefix of a glob to tell walk
which directories to scan. Perviously, a glob like foo/* would scan
everything, but only return matches for foo/*. Now, foo/* only scans
under foo (using the globprefix function), which is much faster.
2005-08-01 05:42:46 +04:00
|
|
|
def globprefix(pat):
|
|
|
|
'''return the non-glob prefix of a path, e.g. foo/* -> foo'''
|
|
|
|
root = []
|
2007-03-11 05:00:42 +03:00
|
|
|
for p in pat.split('/'):
|
2005-10-19 11:02:41 +04:00
|
|
|
if contains_glob(p): break
|
Clean up walk and changes code to use normalised names properly.
New function: commands.pathto returns the relative path from one path
to another. For example, given foo/bar and baz/quux, it will return
../../baz/quux. This new function is used by the walk and status code
to print relative paths correctly.
New command: debugwalk exercises the walk code without doing anything
more.
hg.dirstate.walk now yields normalised names. For example, if you're
in the baz directory and you ask it to walk ../foo/bar/.., it will yield
names starting with foo/.
As a result of this change, all of the other walk and changes methods
in this module also return normalised names.
The util.matcher function now normalises globs and path names, so that
it will match normalised names properly.
Finally, util.matcher uses the non-glob prefix of a glob to tell walk
which directories to scan. Perviously, a glob like foo/* would scan
everything, but only return matches for foo/*. Now, foo/* only scans
under foo (using the globprefix function), which is much faster.
2005-08-01 05:42:46 +04:00
|
|
|
root.append(p)
|
2007-03-11 05:00:47 +03:00
|
|
|
return '/'.join(root) or '.'
|
Clean up walk and changes code to use normalised names properly.
New function: commands.pathto returns the relative path from one path
to another. For example, given foo/bar and baz/quux, it will return
../../baz/quux. This new function is used by the walk and status code
to print relative paths correctly.
New command: debugwalk exercises the walk code without doing anything
more.
hg.dirstate.walk now yields normalised names. For example, if you're
in the baz directory and you ask it to walk ../foo/bar/.., it will yield
names starting with foo/.
As a result of this change, all of the other walk and changes methods
in this module also return normalised names.
The util.matcher function now normalises globs and path names, so that
it will match normalised names properly.
Finally, util.matcher uses the non-glob prefix of a glob to tell walk
which directories to scan. Perviously, a glob like foo/* would scan
everything, but only return matches for foo/*. Now, foo/* only scans
under foo (using the globprefix function), which is much faster.
2005-08-01 05:42:46 +04:00
|
|
|
|
2007-03-11 05:00:53 +03:00
|
|
|
def normalizepats(names, default):
|
|
|
|
pats = []
|
|
|
|
roots = []
|
|
|
|
anypats = False
|
|
|
|
for kind, name in [patkind(p, default) for p in names]:
|
|
|
|
if kind in ('glob', 'relpath'):
|
|
|
|
name = canonpath(canonroot, cwd, name)
|
|
|
|
elif kind in ('relglob', 'path'):
|
|
|
|
name = normpath(name)
|
2007-03-17 04:48:23 +03:00
|
|
|
|
|
|
|
pats.append((kind, name))
|
|
|
|
|
2007-03-11 05:00:53 +03:00
|
|
|
if kind in ('glob', 're', 'relglob', 'relre'):
|
|
|
|
anypats = True
|
2007-03-17 04:48:23 +03:00
|
|
|
|
2007-03-11 05:00:53 +03:00
|
|
|
if kind == 'glob':
|
|
|
|
root = globprefix(name)
|
|
|
|
roots.append(root)
|
|
|
|
elif kind in ('relpath', 'path'):
|
2007-03-17 04:48:20 +03:00
|
|
|
roots.append(name or '.')
|
2007-03-11 05:00:53 +03:00
|
|
|
elif kind == 'relglob':
|
|
|
|
roots.append('.')
|
2007-03-17 04:48:23 +03:00
|
|
|
return roots, pats, anypats
|
2007-03-11 05:00:53 +03:00
|
|
|
|
|
|
|
roots, pats, anypats = normalizepats(names, dflt_pat)
|
2005-08-15 00:23:36 +04:00
|
|
|
|
2007-04-04 11:22:05 +04:00
|
|
|
patmatch = matchfn(pats, '$') or always
|
2005-08-15 00:23:36 +04:00
|
|
|
incmatch = always
|
|
|
|
if inc:
|
2007-03-11 05:00:53 +03:00
|
|
|
dummy, inckinds, dummy = normalizepats(inc, 'glob')
|
2007-04-04 11:22:05 +04:00
|
|
|
incmatch = matchfn(inckinds, '(?:/|$)')
|
2005-08-15 00:23:36 +04:00
|
|
|
excmatch = lambda fn: False
|
|
|
|
if exc:
|
2007-03-11 05:00:53 +03:00
|
|
|
dummy, exckinds, dummy = normalizepats(exc, 'glob')
|
2007-04-04 11:22:05 +04:00
|
|
|
excmatch = matchfn(exckinds, '(?:/|$)')
|
2005-07-21 21:21:33 +04:00
|
|
|
|
2007-03-11 05:01:02 +03:00
|
|
|
if not names and inc and not exc:
|
|
|
|
# common case: hgignore patterns
|
|
|
|
match = incmatch
|
|
|
|
else:
|
|
|
|
match = lambda fn: incmatch(fn) and not excmatch(fn) and patmatch(fn)
|
|
|
|
|
|
|
|
return (roots, match, (inc or exc or anypats) and True)
|
2005-07-21 21:21:33 +04:00
|
|
|
|
2007-08-05 00:25:12 +04:00
|
|
|
_hgexecutable = None
|
2007-06-23 22:21:10 +04:00
|
|
|
|
2007-08-05 00:25:12 +04:00
|
|
|
def hgexecutable():
|
|
|
|
"""return location of the 'hg' executable.
|
2007-06-23 22:21:10 +04:00
|
|
|
|
2007-08-05 00:25:12 +04:00
|
|
|
Defaults to $HG or 'hg' in the search path.
|
2007-06-23 22:21:10 +04:00
|
|
|
"""
|
2007-08-05 00:25:12 +04:00
|
|
|
if _hgexecutable is None:
|
|
|
|
set_hgexecutable(os.environ.get('HG') or find_exe('hg', 'hg'))
|
|
|
|
return _hgexecutable
|
|
|
|
|
|
|
|
def set_hgexecutable(path):
|
|
|
|
"""set location of the 'hg' executable"""
|
2007-06-23 22:21:10 +04:00
|
|
|
global _hgexecutable
|
2007-08-05 00:25:12 +04:00
|
|
|
_hgexecutable = path
|
2007-06-23 22:21:10 +04:00
|
|
|
|
2006-03-11 09:42:59 +03:00
|
|
|
def system(cmd, environ={}, cwd=None, onerr=None, errprefix=None):
|
2006-03-11 09:24:19 +03:00
|
|
|
'''enhanced shell command execution.
|
2006-03-11 09:42:59 +03:00
|
|
|
run with environment maybe modified, maybe in different dir.
|
|
|
|
|
|
|
|
if command fails and onerr is None, return status. if ui object,
|
|
|
|
print error message and return status, else raise onerr object as
|
|
|
|
exception.'''
|
2006-07-12 19:59:20 +04:00
|
|
|
def py2shell(val):
|
|
|
|
'convert python object into string that is useful to shell'
|
|
|
|
if val in (None, False):
|
|
|
|
return '0'
|
|
|
|
if val == True:
|
|
|
|
return '1'
|
|
|
|
return str(val)
|
2006-03-11 09:24:19 +03:00
|
|
|
oldenv = {}
|
|
|
|
for k in environ:
|
|
|
|
oldenv[k] = os.environ.get(k)
|
|
|
|
if cwd is not None:
|
|
|
|
oldcwd = os.getcwd()
|
2006-12-16 07:38:18 +03:00
|
|
|
origcmd = cmd
|
|
|
|
if os.name == 'nt':
|
|
|
|
cmd = '"%s"' % cmd
|
2006-03-11 09:24:19 +03:00
|
|
|
try:
|
|
|
|
for k, v in environ.iteritems():
|
2006-07-12 19:59:20 +04:00
|
|
|
os.environ[k] = py2shell(v)
|
2007-08-05 00:25:12 +04:00
|
|
|
os.environ['HG'] = hgexecutable()
|
2006-03-11 09:24:19 +03:00
|
|
|
if cwd is not None and oldcwd != cwd:
|
|
|
|
os.chdir(cwd)
|
2006-03-11 09:42:59 +03:00
|
|
|
rc = os.system(cmd)
|
2007-06-08 18:24:43 +04:00
|
|
|
if sys.platform == 'OpenVMS' and rc & 1:
|
|
|
|
rc = 0
|
2006-03-11 09:42:59 +03:00
|
|
|
if rc and onerr:
|
2006-12-16 07:38:18 +03:00
|
|
|
errmsg = '%s %s' % (os.path.basename(origcmd.split(None, 1)[0]),
|
2006-03-11 09:42:59 +03:00
|
|
|
explain_exit(rc)[0])
|
|
|
|
if errprefix:
|
|
|
|
errmsg = '%s: %s' % (errprefix, errmsg)
|
|
|
|
try:
|
|
|
|
onerr.warn(errmsg + '\n')
|
|
|
|
except AttributeError:
|
|
|
|
raise onerr(errmsg)
|
|
|
|
return rc
|
2006-03-11 09:24:19 +03:00
|
|
|
finally:
|
|
|
|
for k, v in oldenv.iteritems():
|
|
|
|
if v is None:
|
|
|
|
del os.environ[k]
|
|
|
|
else:
|
|
|
|
os.environ[k] = v
|
|
|
|
if cwd is not None and oldcwd != cwd:
|
|
|
|
os.chdir(oldcwd)
|
2005-06-28 14:38:33 +04:00
|
|
|
|
2007-03-27 04:36:26 +04:00
|
|
|
# os.path.lexists is not available on python2.3
|
|
|
|
def lexists(filename):
|
|
|
|
"test whether a file with this name exists. does not follow symlinks"
|
|
|
|
try:
|
|
|
|
os.lstat(filename)
|
|
|
|
except:
|
|
|
|
return False
|
|
|
|
return True
|
|
|
|
|
2005-06-22 07:36:35 +04:00
|
|
|
def rename(src, dst):
|
2005-08-27 12:13:28 +04:00
|
|
|
"""forcibly rename a file"""
|
2005-06-22 07:36:35 +04:00
|
|
|
try:
|
|
|
|
os.rename(src, dst)
|
2007-07-20 11:31:32 +04:00
|
|
|
except OSError, err: # FIXME: check err (EEXIST ?)
|
2006-05-03 01:30:00 +04:00
|
|
|
# on windows, rename to existing file is not allowed, so we
|
|
|
|
# must delete destination first. but if file is open, unlink
|
|
|
|
# schedules it for delete but does not delete it. rename
|
|
|
|
# happens immediately even for open files, so we create
|
|
|
|
# temporary file, delete it, rename destination to that name,
|
|
|
|
# then delete that. then rename is safe to do.
|
|
|
|
fd, temp = tempfile.mkstemp(dir=os.path.dirname(dst) or '.')
|
|
|
|
os.close(fd)
|
|
|
|
os.unlink(temp)
|
|
|
|
os.rename(dst, temp)
|
|
|
|
os.unlink(temp)
|
2005-06-22 07:36:35 +04:00
|
|
|
os.rename(src, dst)
|
|
|
|
|
2005-10-19 11:10:52 +04:00
|
|
|
def unlink(f):
|
|
|
|
"""unlink and remove the directory if it is empty"""
|
|
|
|
os.unlink(f)
|
|
|
|
# try removing directories that might now be empty
|
2006-04-14 00:46:56 +04:00
|
|
|
try:
|
|
|
|
os.removedirs(os.path.dirname(f))
|
|
|
|
except OSError:
|
|
|
|
pass
|
2005-10-19 11:10:52 +04:00
|
|
|
|
2006-11-13 22:26:57 +03:00
|
|
|
def copyfile(src, dest):
|
|
|
|
"copy a file, preserving mode"
|
2007-03-22 06:20:56 +03:00
|
|
|
if os.path.islink(src):
|
|
|
|
try:
|
|
|
|
os.unlink(dest)
|
|
|
|
except:
|
|
|
|
pass
|
|
|
|
os.symlink(os.readlink(src), dest)
|
|
|
|
else:
|
|
|
|
try:
|
|
|
|
shutil.copyfile(src, dest)
|
|
|
|
shutil.copymode(src, dest)
|
|
|
|
except shutil.Error, inst:
|
|
|
|
raise Abort(str(inst))
|
2006-11-13 22:26:57 +03:00
|
|
|
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
def copyfiles(src, dst, hardlink=None):
|
|
|
|
"""Copy a directory tree using hardlinks if possible"""
|
|
|
|
|
|
|
|
if hardlink is None:
|
|
|
|
hardlink = (os.stat(src).st_dev ==
|
|
|
|
os.stat(os.path.dirname(dst)).st_dev)
|
2005-09-08 06:21:38 +04:00
|
|
|
|
|
|
|
if os.path.isdir(src):
|
|
|
|
os.mkdir(dst)
|
2007-10-06 02:01:06 +04:00
|
|
|
for name, kind in osutil.listdir(src):
|
2005-09-08 06:21:38 +04:00
|
|
|
srcname = os.path.join(src, name)
|
|
|
|
dstname = os.path.join(dst, name)
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
copyfiles(srcname, dstname, hardlink)
|
2005-09-08 06:21:38 +04:00
|
|
|
else:
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
if hardlink:
|
|
|
|
try:
|
|
|
|
os_link(src, dst)
|
2006-04-06 04:17:07 +04:00
|
|
|
except (IOError, OSError):
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
hardlink = False
|
2005-12-16 10:32:44 +03:00
|
|
|
shutil.copy(src, dst)
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
else:
|
2005-12-16 10:32:44 +03:00
|
|
|
shutil.copy(src, dst)
|
2005-07-14 18:30:49 +04:00
|
|
|
|
2007-08-10 21:46:03 +04:00
|
|
|
class path_auditor(object):
|
|
|
|
'''ensure that a filesystem path contains no banned components.
|
|
|
|
the following properties of a path are checked:
|
|
|
|
|
|
|
|
- under top-level .hg
|
|
|
|
- starts at the root of a windows drive
|
|
|
|
- contains ".."
|
|
|
|
- traverses a symlink (e.g. a/symlink_here/b)
|
|
|
|
- inside a nested repository'''
|
|
|
|
|
|
|
|
def __init__(self, root):
|
2007-08-19 04:36:10 +04:00
|
|
|
self.audited = set()
|
|
|
|
self.auditeddir = set()
|
2007-08-10 21:46:03 +04:00
|
|
|
self.root = root
|
|
|
|
|
|
|
|
def __call__(self, path):
|
|
|
|
if path in self.audited:
|
|
|
|
return
|
2007-08-10 21:51:47 +04:00
|
|
|
normpath = os.path.normcase(path)
|
2008-01-09 15:30:36 +03:00
|
|
|
parts = splitpath(normpath)
|
2007-08-10 21:46:03 +04:00
|
|
|
if (os.path.splitdrive(path)[0] or parts[0] in ('.hg', '')
|
|
|
|
or os.pardir in parts):
|
|
|
|
raise Abort(_("path contains illegal component: %s") % path)
|
|
|
|
def check(prefix):
|
|
|
|
curpath = os.path.join(self.root, prefix)
|
|
|
|
try:
|
|
|
|
st = os.lstat(curpath)
|
|
|
|
except OSError, err:
|
2007-08-13 21:42:50 +04:00
|
|
|
# EINVAL can be raised as invalid path syntax under win32.
|
|
|
|
# They must be ignored for patterns can be checked too.
|
2007-11-05 20:05:44 +03:00
|
|
|
if err.errno not in (errno.ENOENT, errno.ENOTDIR, errno.EINVAL):
|
2007-08-10 21:46:03 +04:00
|
|
|
raise
|
|
|
|
else:
|
|
|
|
if stat.S_ISLNK(st.st_mode):
|
|
|
|
raise Abort(_('path %r traverses symbolic link %r') %
|
|
|
|
(path, prefix))
|
2007-08-10 21:51:47 +04:00
|
|
|
elif (stat.S_ISDIR(st.st_mode) and
|
|
|
|
os.path.isdir(os.path.join(curpath, '.hg'))):
|
2007-08-10 21:46:03 +04:00
|
|
|
raise Abort(_('path %r is inside repo %r') %
|
|
|
|
(path, prefix))
|
2008-01-09 15:30:37 +03:00
|
|
|
parts.pop()
|
2007-08-19 04:36:10 +04:00
|
|
|
prefixes = []
|
2008-01-09 15:30:37 +03:00
|
|
|
for n in range(len(parts)):
|
|
|
|
prefix = os.sep.join(parts)
|
2007-08-19 04:36:10 +04:00
|
|
|
if prefix in self.auditeddir:
|
|
|
|
break
|
|
|
|
check(prefix)
|
|
|
|
prefixes.append(prefix)
|
2008-01-09 15:30:37 +03:00
|
|
|
parts.pop()
|
2007-08-19 04:36:10 +04:00
|
|
|
|
|
|
|
self.audited.add(path)
|
|
|
|
# only add prefixes to the cache after checking everything: we don't
|
|
|
|
# want to add "foo/bar/baz" before checking if there's a "foo/.hg"
|
|
|
|
self.auditeddir.update(prefixes)
|
2006-03-04 21:01:45 +03:00
|
|
|
|
2005-07-15 01:51:47 +04:00
|
|
|
def _makelock_file(info, pathname):
|
|
|
|
ld = os.open(pathname, os.O_CREAT | os.O_WRONLY | os.O_EXCL)
|
|
|
|
os.write(ld, info)
|
|
|
|
os.close(ld)
|
|
|
|
|
|
|
|
def _readlock_file(pathname):
|
2006-05-03 01:30:00 +04:00
|
|
|
return posixfile(pathname).read()
|
2005-07-15 01:51:47 +04:00
|
|
|
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
def nlinks(pathname):
|
|
|
|
"""Return number of hardlinks for the given file."""
|
2006-06-16 23:58:24 +04:00
|
|
|
return os.lstat(pathname).st_nlink
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
|
|
|
|
if hasattr(os, 'link'):
|
|
|
|
os_link = os.link
|
|
|
|
else:
|
|
|
|
def os_link(src, dst):
|
2005-10-19 05:38:39 +04:00
|
|
|
raise OSError(0, _("Hardlinks not supported"))
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
|
2006-05-03 01:30:00 +04:00
|
|
|
def fstat(fp):
|
|
|
|
'''stat file object that may not have fileno method.'''
|
|
|
|
try:
|
|
|
|
return os.fstat(fp.fileno())
|
|
|
|
except AttributeError:
|
|
|
|
return os.stat(fp.name)
|
|
|
|
|
|
|
|
posixfile = file
|
|
|
|
|
2007-12-18 23:01:42 +03:00
|
|
|
def openhardlinks():
|
|
|
|
'''return true if it is safe to hold open file handles to hardlinks'''
|
|
|
|
return True
|
2006-05-10 22:10:18 +04:00
|
|
|
|
2006-11-28 23:16:05 +03:00
|
|
|
getuser_fallback = None
|
|
|
|
|
|
|
|
def getuser():
|
|
|
|
'''return name of current user'''
|
|
|
|
try:
|
|
|
|
return getpass.getuser()
|
|
|
|
except ImportError:
|
|
|
|
# import of pwd will fail on windows - try fallback
|
|
|
|
if getuser_fallback:
|
|
|
|
return getuser_fallback()
|
|
|
|
# raised if win32api not available
|
|
|
|
raise Abort(_('user name not available - set USERNAME '
|
|
|
|
'environment variable'))
|
|
|
|
|
2006-10-26 21:25:44 +04:00
|
|
|
def username(uid=None):
|
|
|
|
"""Return the name of the user with the given uid.
|
|
|
|
|
|
|
|
If uid is None, return the name of the current user."""
|
|
|
|
try:
|
|
|
|
import pwd
|
|
|
|
if uid is None:
|
|
|
|
uid = os.getuid()
|
|
|
|
try:
|
|
|
|
return pwd.getpwuid(uid)[0]
|
|
|
|
except KeyError:
|
|
|
|
return str(uid)
|
|
|
|
except ImportError:
|
|
|
|
return None
|
|
|
|
|
|
|
|
def groupname(gid=None):
|
|
|
|
"""Return the name of the group with the given gid.
|
|
|
|
|
|
|
|
If gid is None, return the name of the current group."""
|
|
|
|
try:
|
|
|
|
import grp
|
|
|
|
if gid is None:
|
|
|
|
gid = os.getgid()
|
|
|
|
try:
|
|
|
|
return grp.getgrgid(gid)[0]
|
|
|
|
except KeyError:
|
|
|
|
return str(gid)
|
|
|
|
except ImportError:
|
|
|
|
return None
|
|
|
|
|
2006-12-05 02:10:29 +03:00
|
|
|
# File system features
|
|
|
|
|
|
|
|
def checkfolding(path):
|
|
|
|
"""
|
|
|
|
Check whether the given path is on a case-sensitive filesystem
|
|
|
|
|
|
|
|
Requires a path (like /foo/.hg) ending with a foldable final
|
|
|
|
directory component.
|
|
|
|
"""
|
|
|
|
s1 = os.stat(path)
|
|
|
|
d, b = os.path.split(path)
|
|
|
|
p2 = os.path.join(d, b.upper())
|
|
|
|
if path == p2:
|
|
|
|
p2 = os.path.join(d, b.lower())
|
|
|
|
try:
|
|
|
|
s2 = os.stat(p2)
|
|
|
|
if s2 == s1:
|
|
|
|
return False
|
|
|
|
return True
|
|
|
|
except:
|
|
|
|
return True
|
|
|
|
|
2006-12-30 05:04:30 +03:00
|
|
|
def checkexec(path):
|
|
|
|
"""
|
|
|
|
Check whether the given path is on a filesystem with UNIX-like exec flags
|
|
|
|
|
|
|
|
Requires a directory (like /foo/.hg)
|
|
|
|
"""
|
2007-12-28 08:55:40 +03:00
|
|
|
|
|
|
|
# VFAT on some Linux versions can flip mode but it doesn't persist
|
|
|
|
# a FS remount. Frequently we can detect it if files are created
|
|
|
|
# with exec bit on.
|
|
|
|
|
2007-08-24 02:38:08 +04:00
|
|
|
try:
|
2007-10-05 03:52:53 +04:00
|
|
|
EXECFLAGS = stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH
|
2007-08-24 02:38:08 +04:00
|
|
|
fh, fn = tempfile.mkstemp("", "", path)
|
2007-12-28 08:55:40 +03:00
|
|
|
try:
|
|
|
|
os.close(fh)
|
|
|
|
m = os.stat(fn).st_mode & 0777
|
|
|
|
new_file_has_exec = m & EXECFLAGS
|
|
|
|
os.chmod(fn, m ^ EXECFLAGS)
|
2007-12-29 03:14:45 +03:00
|
|
|
exec_flags_cannot_flip = ((os.stat(fn).st_mode & 0777) == m)
|
2007-12-28 08:55:40 +03:00
|
|
|
finally:
|
|
|
|
os.unlink(fn)
|
|
|
|
except (IOError, OSError):
|
2007-08-24 02:38:08 +04:00
|
|
|
# we don't care, the user probably won't be able to commit anyway
|
|
|
|
return False
|
2007-10-05 03:52:53 +04:00
|
|
|
return not (new_file_has_exec or exec_flags_cannot_flip)
|
2006-12-30 05:04:30 +03:00
|
|
|
|
2006-12-30 05:04:30 +03:00
|
|
|
def execfunc(path, fallback):
|
|
|
|
'''return an is_exec() function with default to fallback'''
|
|
|
|
if checkexec(path):
|
2006-12-30 05:04:31 +03:00
|
|
|
return lambda x: is_exec(os.path.join(path, x))
|
2006-12-30 05:04:30 +03:00
|
|
|
return fallback
|
|
|
|
|
2006-12-30 05:04:31 +03:00
|
|
|
def checklink(path):
|
2006-12-30 05:04:31 +03:00
|
|
|
"""check whether the given path is on a symlink-capable filesystem"""
|
|
|
|
# mktemp is not racy because symlink creation will fail if the
|
|
|
|
# file already exists
|
|
|
|
name = tempfile.mktemp(dir=path)
|
|
|
|
try:
|
|
|
|
os.symlink(".", name)
|
|
|
|
os.unlink(name)
|
|
|
|
return True
|
2007-01-05 05:01:51 +03:00
|
|
|
except (OSError, AttributeError):
|
2006-12-30 05:04:31 +03:00
|
|
|
return False
|
|
|
|
|
2006-12-30 05:04:31 +03:00
|
|
|
def linkfunc(path, fallback):
|
|
|
|
'''return an is_link() function with default to fallback'''
|
|
|
|
if checklink(path):
|
2007-03-24 05:40:25 +03:00
|
|
|
return lambda x: os.path.islink(os.path.join(path, x))
|
2006-12-30 05:04:31 +03:00
|
|
|
return fallback
|
|
|
|
|
2007-04-09 11:24:17 +04:00
|
|
|
_umask = os.umask(0)
|
|
|
|
os.umask(_umask)
|
|
|
|
|
2007-05-12 23:09:29 +04:00
|
|
|
def needbinarypatch():
|
|
|
|
"""return True if patches should be applied in binary mode by default."""
|
|
|
|
return os.name == 'nt'
|
|
|
|
|
2008-01-09 15:30:35 +03:00
|
|
|
def endswithsep(path):
|
|
|
|
'''Check path ends with os.sep or os.altsep.'''
|
|
|
|
return path.endswith(os.sep) or os.altsep and path.endswith(os.altsep)
|
|
|
|
|
2008-01-09 15:30:36 +03:00
|
|
|
def splitpath(path):
|
|
|
|
'''Split path by os.sep.
|
|
|
|
Note that this function does not use os.altsep because this is
|
|
|
|
an alternative of simple "xxx.split(os.sep)".
|
|
|
|
It is recommended to use os.path.normpath() before using this
|
|
|
|
function if need.'''
|
|
|
|
return path.split(os.sep)
|
|
|
|
|
2008-02-04 04:29:05 +03:00
|
|
|
def gui():
|
|
|
|
'''Are we running in a GUI?'''
|
|
|
|
return os.name == "nt" or os.name == "mac" or os.environ.get("DISPLAY")
|
|
|
|
|
2008-02-14 01:09:28 +03:00
|
|
|
def lookup_reg(key, name=None, scope=None):
|
|
|
|
return None
|
|
|
|
|
2005-08-27 12:13:28 +04:00
|
|
|
# Platform specific variants
|
2005-06-22 07:31:13 +04:00
|
|
|
if os.name == 'nt':
|
2006-12-13 22:27:09 +03:00
|
|
|
import msvcrt
|
2005-06-25 10:58:14 +04:00
|
|
|
nulldev = 'NUL:'
|
2006-01-03 02:25:33 +03:00
|
|
|
|
|
|
|
class winstdout:
|
|
|
|
'''stdout on windows misbehaves if sent through a pipe'''
|
|
|
|
|
|
|
|
def __init__(self, fp):
|
|
|
|
self.fp = fp
|
|
|
|
|
|
|
|
def __getattr__(self, key):
|
|
|
|
return getattr(self.fp, key)
|
|
|
|
|
|
|
|
def close(self):
|
|
|
|
try:
|
|
|
|
self.fp.close()
|
|
|
|
except: pass
|
|
|
|
|
|
|
|
def write(self, s):
|
|
|
|
try:
|
2007-12-14 18:47:41 +03:00
|
|
|
# This is workaround for "Not enough space" error on
|
|
|
|
# writing large size of data to console.
|
|
|
|
limit = 16000
|
|
|
|
l = len(s)
|
|
|
|
start = 0
|
|
|
|
while start < l:
|
|
|
|
end = start + limit
|
|
|
|
self.fp.write(s[start:end])
|
|
|
|
start = end
|
2006-01-03 02:25:33 +03:00
|
|
|
except IOError, inst:
|
|
|
|
if inst.errno != 0: raise
|
|
|
|
self.close()
|
|
|
|
raise IOError(errno.EPIPE, 'Broken pipe')
|
2007-06-06 22:22:52 +04:00
|
|
|
|
2007-02-19 12:32:46 +03:00
|
|
|
def flush(self):
|
|
|
|
try:
|
|
|
|
return self.fp.flush()
|
|
|
|
except IOError, inst:
|
|
|
|
if inst.errno != errno.EINVAL: raise
|
|
|
|
self.close()
|
|
|
|
raise IOError(errno.EPIPE, 'Broken pipe')
|
2006-01-03 02:25:33 +03:00
|
|
|
|
|
|
|
sys.stdout = winstdout(sys.stdout)
|
|
|
|
|
2007-12-18 23:01:42 +03:00
|
|
|
def _is_win_9x():
|
|
|
|
'''return true if run on windows 95, 98 or me.'''
|
|
|
|
try:
|
|
|
|
return sys.getwindowsversion()[3] == 1
|
|
|
|
except AttributeError:
|
|
|
|
return 'command' in os.environ.get('comspec', '')
|
|
|
|
|
|
|
|
def openhardlinks():
|
|
|
|
return not _is_win_9x and "win32api" in locals()
|
|
|
|
|
2006-04-09 01:12:43 +04:00
|
|
|
def system_rcpath():
|
2006-04-25 19:37:05 +04:00
|
|
|
try:
|
|
|
|
return system_rcpath_win32()
|
|
|
|
except:
|
|
|
|
return [r'c:\mercurial\mercurial.ini']
|
2006-04-09 01:12:43 +04:00
|
|
|
|
2006-05-15 10:44:50 +04:00
|
|
|
def user_rcpath():
|
2006-11-17 10:06:54 +03:00
|
|
|
'''return os-specific hgrc search path to the user dir'''
|
2007-02-16 10:53:03 +03:00
|
|
|
try:
|
2008-02-20 23:31:42 +03:00
|
|
|
path = user_rcpath_win32()
|
2007-02-16 10:53:03 +03:00
|
|
|
except:
|
2008-02-20 23:31:42 +03:00
|
|
|
home = os.path.expanduser('~')
|
|
|
|
path = [os.path.join(home, 'mercurial.ini'),
|
|
|
|
os.path.join(home, '.hgrc')]
|
2007-02-15 01:20:06 +03:00
|
|
|
userprofile = os.environ.get('USERPROFILE')
|
|
|
|
if userprofile:
|
|
|
|
path.append(os.path.join(userprofile, 'mercurial.ini'))
|
2008-02-20 23:31:42 +03:00
|
|
|
path.append(os.path.join(userprofile, '.hgrc'))
|
2007-02-15 01:20:06 +03:00
|
|
|
return path
|
2006-05-15 10:44:50 +04:00
|
|
|
|
2005-09-21 08:35:51 +04:00
|
|
|
def parse_patch_output(output_line):
|
|
|
|
"""parses the output produced by patch and returns the file name"""
|
|
|
|
pf = output_line[14:]
|
|
|
|
if pf[0] == '`':
|
|
|
|
pf = pf[1:-1] # Remove the quotes
|
|
|
|
return pf
|
|
|
|
|
2007-12-13 01:44:26 +03:00
|
|
|
def sshargs(sshcmd, host, user, port):
|
|
|
|
'''Build argument list for ssh or Plink'''
|
|
|
|
pflag = 'plink' in sshcmd.lower() and '-P' or '-p'
|
|
|
|
args = user and ("%s@%s" % (user, host)) or host
|
2007-12-13 23:25:51 +03:00
|
|
|
return port and ("%s %s %s" % (args, pflag, port)) or args
|
2007-12-13 01:44:26 +03:00
|
|
|
|
2006-04-09 01:12:43 +04:00
|
|
|
def testpid(pid):
|
|
|
|
'''return False if pid dead, True if running or not known'''
|
|
|
|
return True
|
Add support for cloning with hardlinks on windows.
In order to use hardlinks, the win32file module is needed, and this is
present in ActivePython. If it isn't present, or hardlinks are not supported
on the underlying filesystem, a regular copy is used.
When using hardlinks the biggest benefit is probably the saving in space,
but cloning can be much quicker. For example cloning the Xen tree
(non trivial) without an update goes from about 95s to 15s.
Unix-like platforms should be unaffected, although should be more tolerant on
filesystems that don't support hard links.
(tweaked by mpm to deal with new copyfiles function)
--- hg.orig/mercurial/commands.py 2005-09-13 19:32:53.000000000 -0500
+++ hg/mercurial/commands.py 2005-09-14 12:11:34.000000000 -0500
@@ -620,10 +620,6 @@ def clone(ui, source, dest=None, **opts)
if other.dev() != -1:
abspath = os.path.abspath(source)
- copyfile = (os.stat(dest).st_dev == other.dev()
- and getattr(os, 'link', None) or shutil.copy2)
- if copyfile is not shutil.copy2:
- ui.note("cloning by hardlink\n")
# we use a lock here because if we race with commit, we can
# end up with extra data in the cloned revlogs that's not
@@ -638,7 +634,7 @@ def clone(ui, source, dest=None, **opts)
for f in files.split():
src = os.path.join(source, ".hg", f)
dst = os.path.join(dest, ".hg", f)
- util.copyfiles(src, dst, copyfile)
+ util.copyfiles(src, dst)
repo = hg.repository(ui, dest)
Index: hg/mercurial/util.py
===================================================================
--- hg.orig/mercurial/util.py 2005-09-08 00:15:25.000000000 -0500
+++ hg/mercurial/util.py 2005-09-14 12:16:49.000000000 -0500
@@ -12,7 +12,7 @@ platform-specific details from the core.
import os, errno
from demandload import *
-demandload(globals(), "re cStringIO")
+demandload(globals(), "re cStringIO shutil")
def binary(s):
"""return true if a string is binary data using diff's heuristic"""
@@ -217,17 +217,28 @@ def rename(src, dst):
os.unlink(dst)
os.rename(src, dst)
-def copyfiles(src, dst, copyfile):
- """Copy a directory tree, files are copied using 'copyfile'."""
+def copyfiles(src, dst, hardlink=None):
+ """Copy a directory tree using hardlinks if possible"""
+
+ if hardlink is None:
+ hardlink = (os.stat(src).st_dev ==
+ os.stat(os.path.dirname(dst)).st_dev)
if os.path.isdir(src):
os.mkdir(dst)
for name in os.listdir(src):
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
- copyfiles(srcname, dstname, copyfile)
+ copyfiles(srcname, dstname, hardlink)
else:
- copyfile(src, dst)
+ if hardlink:
+ try:
+ os_link(src, dst)
+ except:
+ hardlink = False
+ shutil.copy2(src, dst)
+ else:
+ shutil.copy2(src, dst)
def opener(base):
"""
@@ -244,13 +255,13 @@ def opener(base):
if mode[0] != "r":
try:
- s = os.stat(f)
+ nlink = nlinks(f)
except OSError:
d = os.path.dirname(f)
if not os.path.isdir(d):
os.makedirs(d)
else:
- if s.st_nlink > 1:
+ if nlink > 1:
file(f + ".tmp", "wb").write(file(f, "rb").read())
rename(f+".tmp", f)
@@ -266,10 +277,41 @@ def _makelock_file(info, pathname):
def _readlock_file(pathname):
return file(pathname).read()
+def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ return os.stat(pathname).st_nlink
+
+if hasattr(os, 'link'):
+ os_link = os.link
+else:
+ def os_link(src, dst):
+ raise OSError(0, "Hardlinks not supported")
+
# Platform specific variants
if os.name == 'nt':
nulldev = 'NUL:'
+ try: # ActivePython can create hard links using win32file module
+ import win32file
+
+ def os_link(src, dst): # NB will only succeed on NTFS
+ win32file.CreateHardLink(dst, src)
+
+ def nlinks(pathname):
+ """Return number of hardlinks for the given file."""
+ try:
+ fh = win32file.CreateFile(pathname,
+ win32file.GENERIC_READ, win32file.FILE_SHARE_READ,
+ None, win32file.OPEN_EXISTING, 0, None)
+ res = win32file.GetFileInformationByHandle(fh)
+ fh.Close()
+ return res[7]
+ except:
+ return os.stat(pathname).st_nlink
+
+ except ImportError:
+ pass
+
def is_exec(f, last):
return last
2005-09-14 21:22:20 +04:00
|
|
|
|
2007-12-28 07:27:45 +03:00
|
|
|
def set_flags(f, flags):
|
|
|
|
pass
|
|
|
|
|
2005-10-25 01:34:48 +04:00
|
|
|
def set_binary(fd):
|
|
|
|
msvcrt.setmode(fd.fileno(), os.O_BINARY)
|
|
|
|
|
2005-06-22 07:31:13 +04:00
|
|
|
def pconvert(path):
|
2008-01-09 15:30:36 +03:00
|
|
|
return '/'.join(splitpath(path))
|
2005-06-22 07:43:40 +04:00
|
|
|
|
2005-08-13 03:06:52 +04:00
|
|
|
def localpath(path):
|
|
|
|
return path.replace('/', '\\')
|
|
|
|
|
|
|
|
def normpath(path):
|
|
|
|
return pconvert(os.path.normpath(path))
|
|
|
|
|
2005-07-15 01:51:47 +04:00
|
|
|
makelock = _makelock_file
|
|
|
|
readlock = _readlock_file
|
2005-06-25 10:58:14 +04:00
|
|
|
|
2006-05-04 09:47:08 +04:00
|
|
|
def samestat(s1, s2):
|
|
|
|
return False
|
|
|
|
|
2007-02-15 10:38:00 +03:00
|
|
|
# A sequence of backslashes is special iff it precedes a double quote:
|
|
|
|
# - if there's an even number of backslashes, the double quote is not
|
|
|
|
# quoted (i.e. it ends the quoted region)
|
|
|
|
# - if there's an odd number of backslashes, the double quote is quoted
|
|
|
|
# - in both cases, every pair of backslashes is unquoted into a single
|
|
|
|
# backslash
|
|
|
|
# (See http://msdn2.microsoft.com/en-us/library/a1y7w461.aspx )
|
|
|
|
# So, to quote a string, we must surround it in double quotes, double
|
|
|
|
# the number of backslashes that preceed double quotes and add another
|
|
|
|
# backslash before every double quote (being careful with the double
|
|
|
|
# quote we've appended to the end)
|
|
|
|
_quotere = None
|
2006-08-06 02:23:26 +04:00
|
|
|
def shellquote(s):
|
2007-02-15 10:38:00 +03:00
|
|
|
global _quotere
|
|
|
|
if _quotere is None:
|
|
|
|
_quotere = re.compile(r'(\\*)("|\\$)')
|
|
|
|
return '"%s"' % _quotere.sub(r'\1\1\\\2', s)
|
2006-08-06 02:23:26 +04:00
|
|
|
|
2007-09-11 01:34:33 +04:00
|
|
|
def quotecommand(cmd):
|
|
|
|
"""Build a command string suitable for os.popen* calls."""
|
|
|
|
# The extra quotes are needed because popen* runs the command
|
|
|
|
# through the current COMSPEC. cmd.exe suppress enclosing quotes.
|
|
|
|
return '"' + cmd + '"'
|
|
|
|
|
2007-11-01 14:05:14 +03:00
|
|
|
def popen(command):
|
|
|
|
# Work around "popen spawned process may not write to stdout
|
|
|
|
# under windows"
|
|
|
|
# http://bugs.python.org/issue1366
|
|
|
|
command += " 2> %s" % nulldev
|
|
|
|
return os.popen(quotecommand(command))
|
|
|
|
|
2005-07-27 22:18:22 +04:00
|
|
|
def explain_exit(code):
|
2005-10-19 05:38:39 +04:00
|
|
|
return _("exited with status %d") % code, code
|
2005-07-27 22:18:22 +04:00
|
|
|
|
2006-11-19 04:51:14 +03:00
|
|
|
# if you change this stub into a real check, please try to implement the
|
|
|
|
# username and groupname functions above, too.
|
|
|
|
def isowner(fp, st=None):
|
|
|
|
return True
|
2007-06-06 22:22:52 +04:00
|
|
|
|
2007-05-06 18:40:53 +04:00
|
|
|
def find_in_path(name, path, default=None):
|
|
|
|
'''find name in search path. path can be string (will be split
|
|
|
|
with os.pathsep), or iterable thing that returns strings. if name
|
|
|
|
found, return path to name. else return default. name is looked up
|
|
|
|
using cmd.exe rules, using PATHEXT.'''
|
|
|
|
if isinstance(path, str):
|
|
|
|
path = path.split(os.pathsep)
|
2007-06-06 22:22:52 +04:00
|
|
|
|
2007-05-06 18:40:53 +04:00
|
|
|
pathext = os.environ.get('PATHEXT', '.COM;.EXE;.BAT;.CMD')
|
|
|
|
pathext = pathext.lower().split(os.pathsep)
|
|
|
|
isexec = os.path.splitext(name)[1].lower() in pathext
|
2007-06-06 22:22:52 +04:00
|
|
|
|
2007-05-06 18:40:53 +04:00
|
|
|
for p in path:
|
|
|
|
p_name = os.path.join(p, name)
|
2007-06-06 22:22:52 +04:00
|
|
|
|
2007-05-06 18:40:53 +04:00
|
|
|
if isexec and os.path.exists(p_name):
|
|
|
|
return p_name
|
2007-06-06 22:22:52 +04:00
|
|
|
|
2007-05-06 18:40:53 +04:00
|
|
|
for ext in pathext:
|
|
|
|
p_name_ext = p_name + ext
|
|
|
|
if os.path.exists(p_name_ext):
|
|
|
|
return p_name_ext
|
|
|
|
return default
|
2006-11-19 04:51:14 +03:00
|
|
|
|
2007-06-26 18:48:58 +04:00
|
|
|
def set_signal_handler():
|
|
|
|
try:
|
|
|
|
set_signal_handler_win32()
|
|
|
|
except NameError:
|
|
|
|
pass
|
|
|
|
|
2006-04-09 01:12:43 +04:00
|
|
|
try:
|
|
|
|
# override functions with win32 versions if possible
|
|
|
|
from util_win32 import *
|
2007-12-18 23:01:42 +03:00
|
|
|
if not _is_win_9x():
|
2006-05-10 22:10:18 +04:00
|
|
|
posixfile = posixfile_nt
|
2006-04-09 01:12:43 +04:00
|
|
|
except ImportError:
|
|
|
|
pass
|
|
|
|
|
2005-06-22 07:31:13 +04:00
|
|
|
else:
|
2005-06-25 10:58:14 +04:00
|
|
|
nulldev = '/dev/null'
|
|
|
|
|
2005-11-04 22:51:01 +03:00
|
|
|
def rcfiles(path):
|
|
|
|
rcs = [os.path.join(path, 'hgrc')]
|
|
|
|
rcdir = os.path.join(path, 'hgrc.d')
|
|
|
|
try:
|
2007-10-06 02:01:06 +04:00
|
|
|
rcs.extend([os.path.join(rcdir, f)
|
|
|
|
for f, kind in osutil.listdir(rcdir)
|
2005-11-04 22:51:01 +03:00
|
|
|
if f.endswith(".rc")])
|
2006-09-18 19:43:31 +04:00
|
|
|
except OSError:
|
|
|
|
pass
|
2005-11-04 22:51:01 +03:00
|
|
|
return rcs
|
2006-03-15 08:40:46 +03:00
|
|
|
|
2007-02-15 01:20:06 +03:00
|
|
|
def system_rcpath():
|
2006-03-15 08:40:46 +03:00
|
|
|
path = []
|
2006-05-11 20:01:32 +04:00
|
|
|
# old mod_python does not set sys.argv
|
2006-05-11 19:48:36 +04:00
|
|
|
if len(getattr(sys, 'argv', [])) > 0:
|
2006-03-15 08:40:46 +03:00
|
|
|
path.extend(rcfiles(os.path.dirname(sys.argv[0]) +
|
|
|
|
'/../etc/mercurial'))
|
|
|
|
path.extend(rcfiles('/etc/mercurial'))
|
|
|
|
return path
|
2005-09-21 20:56:30 +04:00
|
|
|
|
2007-02-15 01:20:06 +03:00
|
|
|
def user_rcpath():
|
|
|
|
return [os.path.expanduser('~/.hgrc')]
|
|
|
|
|
2005-09-21 08:35:51 +04:00
|
|
|
def parse_patch_output(output_line):
|
|
|
|
"""parses the output produced by patch and returns the file name"""
|
2005-12-15 19:50:10 +03:00
|
|
|
pf = output_line[14:]
|
2007-06-08 18:24:43 +04:00
|
|
|
if os.sys.platform == 'OpenVMS':
|
|
|
|
if pf[0] == '`':
|
|
|
|
pf = pf[1:-1] # Remove the quotes
|
|
|
|
else:
|
|
|
|
if pf.startswith("'") and pf.endswith("'") and " " in pf:
|
|
|
|
pf = pf[1:-1] # Remove the quotes
|
2005-12-15 19:50:10 +03:00
|
|
|
return pf
|
2005-09-21 08:35:51 +04:00
|
|
|
|
2007-12-13 01:44:26 +03:00
|
|
|
def sshargs(sshcmd, host, user, port):
|
|
|
|
'''Build argument list for ssh'''
|
|
|
|
args = user and ("%s@%s" % (user, host)) or host
|
2007-12-13 23:25:51 +03:00
|
|
|
return port and ("%s -p %s" % (args, port)) or args
|
2007-12-13 01:44:26 +03:00
|
|
|
|
2006-12-30 05:04:31 +03:00
|
|
|
def is_exec(f):
|
2005-08-27 12:13:28 +04:00
|
|
|
"""check whether a file is executable"""
|
2006-06-16 23:58:24 +04:00
|
|
|
return (os.lstat(f).st_mode & 0100 != 0)
|
2005-06-23 21:33:18 +04:00
|
|
|
|
2007-12-28 07:27:45 +03:00
|
|
|
def set_flags(f, flags):
|
|
|
|
s = os.lstat(f).st_mode
|
|
|
|
x = "x" in flags
|
|
|
|
l = "l" in flags
|
|
|
|
if l:
|
|
|
|
if not stat.S_ISLNK(s):
|
|
|
|
# switch file to link
|
|
|
|
data = file(f).read()
|
|
|
|
os.unlink(f)
|
|
|
|
os.symlink(data, f)
|
|
|
|
# no chmod needed at this point
|
|
|
|
return
|
|
|
|
if stat.S_ISLNK(s):
|
|
|
|
# switch link to file
|
|
|
|
data = os.readlink(f)
|
|
|
|
os.unlink(f)
|
|
|
|
file(f, "w").write(data)
|
|
|
|
s = 0666 & ~_umask # avoid restatting for chmod
|
|
|
|
|
|
|
|
sx = s & 0100
|
|
|
|
if x and not sx:
|
|
|
|
# Turn on +x for every +r bit when making a file executable
|
|
|
|
# and obey umask.
|
|
|
|
os.chmod(f, s | (s & 0444) >> 2 & ~_umask)
|
|
|
|
elif not x and sx:
|
|
|
|
# Turn off all +x bits
|
|
|
|
os.chmod(f, s & 0666)
|
|
|
|
|
2005-10-25 01:34:48 +04:00
|
|
|
def set_binary(fd):
|
|
|
|
pass
|
|
|
|
|
2005-06-22 07:31:13 +04:00
|
|
|
def pconvert(path):
|
|
|
|
return path
|
|
|
|
|
2005-08-13 03:06:52 +04:00
|
|
|
def localpath(path):
|
|
|
|
return path
|
|
|
|
|
|
|
|
normpath = os.path.normpath
|
2006-05-04 09:47:08 +04:00
|
|
|
samestat = os.path.samestat
|
2005-08-13 03:06:52 +04:00
|
|
|
|
2005-06-22 07:43:40 +04:00
|
|
|
def makelock(info, pathname):
|
2005-07-15 01:51:47 +04:00
|
|
|
try:
|
|
|
|
os.symlink(info, pathname)
|
|
|
|
except OSError, why:
|
|
|
|
if why.errno == errno.EEXIST:
|
|
|
|
raise
|
|
|
|
else:
|
|
|
|
_makelock_file(info, pathname)
|
2005-06-22 07:43:40 +04:00
|
|
|
|
|
|
|
def readlock(pathname):
|
2005-07-15 01:51:47 +04:00
|
|
|
try:
|
|
|
|
return os.readlink(pathname)
|
|
|
|
except OSError, why:
|
2007-06-08 18:24:43 +04:00
|
|
|
if why.errno in (errno.EINVAL, errno.ENOSYS):
|
2005-07-15 01:51:47 +04:00
|
|
|
return _readlock_file(pathname)
|
|
|
|
else:
|
|
|
|
raise
|
2005-07-27 22:18:22 +04:00
|
|
|
|
2006-08-06 02:23:26 +04:00
|
|
|
def shellquote(s):
|
2007-06-08 18:24:43 +04:00
|
|
|
if os.sys.platform == 'OpenVMS':
|
|
|
|
return '"%s"' % s
|
|
|
|
else:
|
|
|
|
return "'%s'" % s.replace("'", "'\\''")
|
2006-08-06 02:23:26 +04:00
|
|
|
|
2007-09-11 01:34:33 +04:00
|
|
|
def quotecommand(cmd):
|
|
|
|
return cmd
|
|
|
|
|
2007-11-01 14:05:14 +03:00
|
|
|
def popen(command):
|
|
|
|
return os.popen(command)
|
|
|
|
|
2006-03-10 19:31:31 +03:00
|
|
|
def testpid(pid):
|
|
|
|
'''return False if pid dead, True if running or not sure'''
|
2007-06-08 18:24:43 +04:00
|
|
|
if os.sys.platform == 'OpenVMS':
|
|
|
|
return True
|
2006-03-10 19:31:31 +03:00
|
|
|
try:
|
|
|
|
os.kill(pid, 0)
|
|
|
|
return True
|
|
|
|
except OSError, inst:
|
|
|
|
return inst.errno != errno.ESRCH
|
|
|
|
|
2005-07-27 22:18:22 +04:00
|
|
|
def explain_exit(code):
|
|
|
|
"""return a 2-tuple (desc, code) describing a process's status"""
|
|
|
|
if os.WIFEXITED(code):
|
|
|
|
val = os.WEXITSTATUS(code)
|
2005-10-19 05:38:39 +04:00
|
|
|
return _("exited with status %d") % val, val
|
2005-07-27 22:18:22 +04:00
|
|
|
elif os.WIFSIGNALED(code):
|
|
|
|
val = os.WTERMSIG(code)
|
2005-10-19 05:38:39 +04:00
|
|
|
return _("killed by signal %d") % val, val
|
2005-07-27 22:18:22 +04:00
|
|
|
elif os.WIFSTOPPED(code):
|
2005-08-15 23:46:41 +04:00
|
|
|
val = os.WSTOPSIG(code)
|
2005-10-19 05:38:39 +04:00
|
|
|
return _("stopped by signal %d") % val, val
|
|
|
|
raise ValueError(_("invalid exit code"))
|
2005-09-05 01:11:51 +04:00
|
|
|
|
2006-11-19 04:51:14 +03:00
|
|
|
def isowner(fp, st=None):
|
|
|
|
"""Return True if the file object f belongs to the current user.
|
|
|
|
|
|
|
|
The return value of a util.fstat(f) may be passed as the st argument.
|
|
|
|
"""
|
|
|
|
if st is None:
|
2006-12-10 21:41:57 +03:00
|
|
|
st = fstat(fp)
|
2006-11-19 04:51:14 +03:00
|
|
|
return st.st_uid == os.getuid()
|
2007-06-06 22:22:52 +04:00
|
|
|
|
2007-05-06 18:40:53 +04:00
|
|
|
def find_in_path(name, path, default=None):
|
|
|
|
'''find name in search path. path can be string (will be split
|
|
|
|
with os.pathsep), or iterable thing that returns strings. if name
|
|
|
|
found, return path to name. else return default.'''
|
|
|
|
if isinstance(path, str):
|
|
|
|
path = path.split(os.pathsep)
|
|
|
|
for p in path:
|
|
|
|
p_name = os.path.join(p, name)
|
|
|
|
if os.path.exists(p_name):
|
|
|
|
return p_name
|
|
|
|
return default
|
2006-11-19 04:51:14 +03:00
|
|
|
|
2007-06-21 23:58:18 +04:00
|
|
|
def set_signal_handler():
|
|
|
|
pass
|
|
|
|
|
2007-05-28 01:26:54 +04:00
|
|
|
def find_exe(name, default=None):
|
|
|
|
'''find path of an executable.
|
|
|
|
if name contains a path component, return it as is. otherwise,
|
|
|
|
use normal executable search path.'''
|
|
|
|
|
2007-06-08 18:24:43 +04:00
|
|
|
if os.sep in name or sys.platform == 'OpenVMS':
|
2007-05-28 01:26:54 +04:00
|
|
|
# don't check the executable bit. if the file isn't
|
|
|
|
# executable, whoever tries to actually run it will give a
|
|
|
|
# much more useful error message.
|
|
|
|
return name
|
|
|
|
return find_in_path(name, os.environ.get('PATH', ''), default=default)
|
|
|
|
|
2006-12-10 02:07:02 +03:00
|
|
|
def _buildencodefun():
|
|
|
|
e = '_'
|
2006-12-10 23:53:01 +03:00
|
|
|
win_reserved = [ord(x) for x in '\\:*?"<>|']
|
2006-12-10 02:07:02 +03:00
|
|
|
cmap = dict([ (chr(x), chr(x)) for x in xrange(127) ])
|
|
|
|
for x in (range(32) + range(126, 256) + win_reserved):
|
|
|
|
cmap[chr(x)] = "~%02x" % x
|
|
|
|
for x in range(ord("A"), ord("Z")+1) + [ord(e)]:
|
|
|
|
cmap[chr(x)] = e + chr(x).lower()
|
|
|
|
dmap = {}
|
|
|
|
for k, v in cmap.iteritems():
|
|
|
|
dmap[v] = k
|
|
|
|
def decode(s):
|
|
|
|
i = 0
|
|
|
|
while i < len(s):
|
|
|
|
for l in xrange(1, 4):
|
|
|
|
try:
|
|
|
|
yield dmap[s[i:i+l]]
|
|
|
|
i += l
|
|
|
|
break
|
|
|
|
except KeyError:
|
|
|
|
pass
|
|
|
|
else:
|
|
|
|
raise KeyError
|
|
|
|
return (lambda s: "".join([cmap[c] for c in s]),
|
|
|
|
lambda s: "".join(list(decode(s))))
|
|
|
|
|
|
|
|
encodefilename, decodefilename = _buildencodefun()
|
|
|
|
|
2006-12-01 15:34:09 +03:00
|
|
|
def encodedopener(openerfn, fn):
|
|
|
|
def o(path, *args, **kw):
|
|
|
|
return openerfn(fn(path), *args, **kw)
|
|
|
|
return o
|
2006-11-19 04:51:14 +03:00
|
|
|
|
2008-02-09 23:38:54 +03:00
|
|
|
def mktempcopy(name, emptyok=False, createmode=None):
|
2007-07-12 00:40:41 +04:00
|
|
|
"""Create a temporary file with the same contents from name
|
2006-05-03 01:30:00 +04:00
|
|
|
|
2007-07-12 00:40:41 +04:00
|
|
|
The permission bits are copied from the original file.
|
|
|
|
|
|
|
|
If the temporary file is going to be truncated immediately, you
|
|
|
|
can use emptyok=True as an optimization.
|
|
|
|
|
|
|
|
Returns the name of the temporary file.
|
2006-05-03 01:30:00 +04:00
|
|
|
"""
|
2007-07-12 00:40:41 +04:00
|
|
|
d, fn = os.path.split(name)
|
|
|
|
fd, temp = tempfile.mkstemp(prefix='.%s-' % fn, dir=d)
|
|
|
|
os.close(fd)
|
|
|
|
# Temporary files are created with mode 0600, which is usually not
|
|
|
|
# what we want. If the original file already exists, just copy
|
|
|
|
# its mode. Otherwise, manually obey umask.
|
|
|
|
try:
|
2007-12-28 08:55:40 +03:00
|
|
|
st_mode = os.lstat(name).st_mode & 0777
|
2007-07-12 00:40:41 +04:00
|
|
|
except OSError, inst:
|
|
|
|
if inst.errno != errno.ENOENT:
|
|
|
|
raise
|
2008-02-09 23:38:54 +03:00
|
|
|
st_mode = createmode
|
|
|
|
if st_mode is None:
|
|
|
|
st_mode = ~_umask
|
|
|
|
st_mode &= 0666
|
2007-07-12 00:40:41 +04:00
|
|
|
os.chmod(temp, st_mode)
|
|
|
|
if emptyok:
|
|
|
|
return temp
|
|
|
|
try:
|
2006-05-03 01:30:00 +04:00
|
|
|
try:
|
2007-07-12 00:40:41 +04:00
|
|
|
ifp = posixfile(name, "rb")
|
|
|
|
except IOError, inst:
|
|
|
|
if inst.errno == errno.ENOENT:
|
|
|
|
return temp
|
|
|
|
if not getattr(inst, 'filename', None):
|
|
|
|
inst.filename = name
|
|
|
|
raise
|
|
|
|
ofp = posixfile(temp, "wb")
|
|
|
|
for chunk in filechunkiter(ifp):
|
|
|
|
ofp.write(chunk)
|
|
|
|
ifp.close()
|
|
|
|
ofp.close()
|
|
|
|
except:
|
|
|
|
try: os.unlink(temp)
|
|
|
|
except: pass
|
|
|
|
raise
|
|
|
|
return temp
|
|
|
|
|
|
|
|
class atomictempfile(posixfile):
|
|
|
|
"""file-like object that atomically updates a file
|
|
|
|
|
|
|
|
All writes will be redirected to a temporary copy of the original
|
|
|
|
file. When rename is called, the copy is renamed to the original
|
|
|
|
name, making the changes visible.
|
|
|
|
"""
|
2008-02-09 23:38:54 +03:00
|
|
|
def __init__(self, name, mode, createmode):
|
2007-07-12 00:40:41 +04:00
|
|
|
self.__name = name
|
2008-02-09 23:38:54 +03:00
|
|
|
self.temp = mktempcopy(name, emptyok=('w' in mode),
|
|
|
|
createmode=createmode)
|
2007-07-12 00:40:41 +04:00
|
|
|
posixfile.__init__(self, self.temp, mode)
|
|
|
|
|
|
|
|
def rename(self):
|
|
|
|
if not self.closed:
|
|
|
|
posixfile.close(self)
|
|
|
|
rename(self.temp, localpath(self.__name))
|
|
|
|
|
|
|
|
def __del__(self):
|
|
|
|
if not self.closed:
|
2006-05-08 19:20:56 +04:00
|
|
|
try:
|
2007-07-12 00:40:41 +04:00
|
|
|
os.unlink(self.temp)
|
2006-05-03 01:30:00 +04:00
|
|
|
except: pass
|
2007-07-12 00:40:41 +04:00
|
|
|
posixfile.close(self)
|
2006-05-03 01:30:00 +04:00
|
|
|
|
2008-02-09 23:38:54 +03:00
|
|
|
def makedirs(name, mode=None):
|
|
|
|
"""recursive directory creation with parent mode inheritance"""
|
|
|
|
try:
|
|
|
|
os.mkdir(name)
|
|
|
|
if mode is not None:
|
|
|
|
os.chmod(name, mode)
|
|
|
|
return
|
|
|
|
except OSError, err:
|
|
|
|
if err.errno == errno.EEXIST:
|
|
|
|
return
|
|
|
|
if err.errno != errno.ENOENT:
|
|
|
|
raise
|
|
|
|
parent = os.path.abspath(os.path.dirname(name))
|
|
|
|
makedirs(parent, mode)
|
|
|
|
makedirs(name, mode)
|
|
|
|
|
2007-07-12 00:40:41 +04:00
|
|
|
class opener(object):
|
|
|
|
"""Open files relative to a base directory
|
2006-05-03 01:30:00 +04:00
|
|
|
|
2007-07-12 00:40:41 +04:00
|
|
|
This class is used to hide the details of COW semantics and
|
|
|
|
remote file access from higher level code.
|
|
|
|
"""
|
|
|
|
def __init__(self, base, audit=True):
|
|
|
|
self.base = base
|
2007-08-10 21:46:03 +04:00
|
|
|
if audit:
|
|
|
|
self.audit_path = path_auditor(base)
|
|
|
|
else:
|
|
|
|
self.audit_path = always
|
2008-02-09 23:38:54 +03:00
|
|
|
self.createmode = None
|
2007-07-12 00:40:41 +04:00
|
|
|
|
2007-07-12 00:40:41 +04:00
|
|
|
def __getattr__(self, name):
|
|
|
|
if name == '_can_symlink':
|
|
|
|
self._can_symlink = checklink(self.base)
|
|
|
|
return self._can_symlink
|
|
|
|
raise AttributeError(name)
|
|
|
|
|
2008-02-09 23:38:54 +03:00
|
|
|
def _fixfilemode(self, name):
|
|
|
|
if self.createmode is None:
|
|
|
|
return
|
|
|
|
os.chmod(name, self.createmode & 0666)
|
|
|
|
|
2007-07-12 00:40:41 +04:00
|
|
|
def __call__(self, path, mode="r", text=False, atomictemp=False):
|
2007-08-10 21:46:03 +04:00
|
|
|
self.audit_path(path)
|
2007-07-12 00:40:41 +04:00
|
|
|
f = os.path.join(self.base, path)
|
2006-05-03 01:30:00 +04:00
|
|
|
|
2007-06-08 18:24:43 +04:00
|
|
|
if not text and "b" not in mode:
|
2006-05-03 01:30:00 +04:00
|
|
|
mode += "b" # for that other OS
|
|
|
|
|
2008-02-09 23:38:54 +03:00
|
|
|
nlink = -1
|
2006-05-03 01:30:00 +04:00
|
|
|
if mode[0] != "r":
|
|
|
|
try:
|
|
|
|
nlink = nlinks(f)
|
|
|
|
except OSError:
|
2007-04-09 11:24:17 +04:00
|
|
|
nlink = 0
|
2006-05-03 01:30:00 +04:00
|
|
|
d = os.path.dirname(f)
|
|
|
|
if not os.path.isdir(d):
|
2008-02-09 23:38:54 +03:00
|
|
|
makedirs(d, self.createmode)
|
2007-06-06 02:55:27 +04:00
|
|
|
if atomictemp:
|
2008-02-09 23:38:54 +03:00
|
|
|
return atomictempfile(f, mode, self.createmode)
|
2007-04-09 11:24:17 +04:00
|
|
|
if nlink > 1:
|
|
|
|
rename(mktempcopy(f), f)
|
2008-02-09 23:38:54 +03:00
|
|
|
fp = posixfile(f, mode)
|
|
|
|
if nlink == 0:
|
|
|
|
self._fixfilemode(f)
|
|
|
|
return fp
|
2006-05-03 01:30:00 +04:00
|
|
|
|
2007-07-12 00:40:41 +04:00
|
|
|
def symlink(self, src, dst):
|
2007-08-10 21:46:03 +04:00
|
|
|
self.audit_path(dst)
|
2007-07-12 00:40:41 +04:00
|
|
|
linkname = os.path.join(self.base, dst)
|
|
|
|
try:
|
|
|
|
os.unlink(linkname)
|
|
|
|
except OSError:
|
|
|
|
pass
|
|
|
|
|
|
|
|
dirname = os.path.dirname(linkname)
|
|
|
|
if not os.path.exists(dirname):
|
2008-02-09 23:38:54 +03:00
|
|
|
makedirs(dirname, self.createmode)
|
2007-07-12 00:40:41 +04:00
|
|
|
|
|
|
|
if self._can_symlink:
|
2007-07-20 02:29:33 +04:00
|
|
|
try:
|
|
|
|
os.symlink(src, linkname)
|
|
|
|
except OSError, err:
|
|
|
|
raise OSError(err.errno, _('could not symlink to %r: %s') %
|
|
|
|
(src, err.strerror), linkname)
|
2007-07-12 00:40:41 +04:00
|
|
|
else:
|
2007-08-06 12:08:28 +04:00
|
|
|
f = self(dst, "w")
|
2007-07-12 00:40:41 +04:00
|
|
|
f.write(src)
|
|
|
|
f.close()
|
2008-02-09 23:38:54 +03:00
|
|
|
self._fixfilemode(dst)
|
2007-07-12 00:40:41 +04:00
|
|
|
|
2005-09-05 01:11:51 +04:00
|
|
|
class chunkbuffer(object):
|
|
|
|
"""Allow arbitrary sized chunks of data to be efficiently read from an
|
|
|
|
iterator over chunks of arbitrary size."""
|
2005-09-05 01:21:53 +04:00
|
|
|
|
2007-10-11 09:46:48 +04:00
|
|
|
def __init__(self, in_iter):
|
2005-09-05 01:11:51 +04:00
|
|
|
"""in_iter is the iterator that's iterating over the input chunks.
|
|
|
|
targetsize is how big a buffer to try to maintain."""
|
2007-10-11 09:46:49 +04:00
|
|
|
self.iter = iter(in_iter)
|
2005-09-05 01:11:51 +04:00
|
|
|
self.buf = ''
|
2007-10-11 09:46:48 +04:00
|
|
|
self.targetsize = 2**16
|
2005-09-05 01:11:51 +04:00
|
|
|
|
|
|
|
def read(self, l):
|
2005-09-05 01:21:53 +04:00
|
|
|
"""Read L bytes of data from the iterator of chunks of data.
|
2005-09-21 09:56:19 +04:00
|
|
|
Returns less than L bytes if the iterator runs dry."""
|
2007-10-11 09:46:49 +04:00
|
|
|
if l > len(self.buf) and self.iter:
|
2005-09-05 01:11:51 +04:00
|
|
|
# Clamp to a multiple of self.targetsize
|
2007-10-11 09:46:52 +04:00
|
|
|
targetsize = max(l, self.targetsize)
|
2005-09-05 01:11:51 +04:00
|
|
|
collector = cStringIO.StringIO()
|
|
|
|
collector.write(self.buf)
|
|
|
|
collected = len(self.buf)
|
2007-10-11 09:46:49 +04:00
|
|
|
for chunk in self.iter:
|
2005-09-05 01:11:51 +04:00
|
|
|
collector.write(chunk)
|
|
|
|
collected += len(chunk)
|
|
|
|
if collected >= targetsize:
|
|
|
|
break
|
|
|
|
if collected < targetsize:
|
2007-10-11 09:46:49 +04:00
|
|
|
self.iter = False
|
2005-09-05 01:11:51 +04:00
|
|
|
self.buf = collector.getvalue()
|
2007-10-11 09:46:52 +04:00
|
|
|
if len(self.buf) == l:
|
2007-10-11 09:46:53 +04:00
|
|
|
s, self.buf = str(self.buf), ''
|
2007-10-11 09:46:52 +04:00
|
|
|
else:
|
|
|
|
s, self.buf = self.buf[:l], buffer(self.buf, l)
|
2005-09-05 01:11:51 +04:00
|
|
|
return s
|
|
|
|
|
2006-06-21 02:13:17 +04:00
|
|
|
def filechunkiter(f, size=65536, limit=None):
|
|
|
|
"""Create a generator that produces the data in the file size
|
|
|
|
(default 65536) bytes at a time, up to optional limit (default is
|
|
|
|
to read all data). Chunks may be less than size bytes if the
|
|
|
|
chunk is the last chunk in the file, or the file is a socket or
|
|
|
|
some other type of file that sometimes reads less data than is
|
|
|
|
requested."""
|
|
|
|
assert size >= 0
|
|
|
|
assert limit is None or limit >= 0
|
|
|
|
while True:
|
|
|
|
if limit is None: nbytes = size
|
|
|
|
else: nbytes = min(limit, size)
|
|
|
|
s = nbytes and f.read(nbytes)
|
|
|
|
if not s: break
|
|
|
|
if limit: limit -= len(s)
|
2005-09-05 01:11:51 +04:00
|
|
|
yield s
|
2005-09-23 09:46:50 +04:00
|
|
|
|
2005-09-23 10:19:47 +04:00
|
|
|
def makedate():
|
2005-10-31 20:56:42 +03:00
|
|
|
lt = time.localtime()
|
|
|
|
if lt[8] == 1 and time.daylight:
|
|
|
|
tz = time.altzone
|
|
|
|
else:
|
|
|
|
tz = time.timezone
|
|
|
|
return time.mktime(lt), tz
|
2005-09-23 21:28:55 +04:00
|
|
|
|
2008-03-12 01:42:41 +03:00
|
|
|
def datestr(date=None, format='%a %b %d %H:%M:%S %Y %1%2'):
|
2005-09-23 10:19:47 +04:00
|
|
|
"""represent a (unixtime, offset) tuple as a localized time.
|
|
|
|
unixtime is seconds since the epoch, and offset is the time zone's
|
2006-03-22 10:29:21 +03:00
|
|
|
number of seconds away from UTC. if timezone is false, do not
|
|
|
|
append time zone to string."""
|
2005-09-23 10:19:47 +04:00
|
|
|
t, tz = date or makedate()
|
2008-03-12 01:42:41 +03:00
|
|
|
if "%1" in format or "%2" in format:
|
|
|
|
sign = (tz > 0) and "-" or "+"
|
|
|
|
minutes = abs(tz) / 60
|
|
|
|
format = format.replace("%1", "%c%02d" % (sign, minutes / 60))
|
|
|
|
format = format.replace("%2", "%02d" % (minutes % 60))
|
2006-03-22 10:29:21 +03:00
|
|
|
s = time.strftime(format, time.gmtime(float(t) - tz))
|
|
|
|
return s
|
2006-02-27 22:32:10 +03:00
|
|
|
|
2008-02-16 15:33:38 +03:00
|
|
|
def shortdate(date=None):
|
|
|
|
"""turn (timestamp, tzoff) tuple into iso 8631 date."""
|
2008-03-12 01:42:41 +03:00
|
|
|
return datestr(date, format='%Y-%m-%d')
|
2008-02-16 15:33:38 +03:00
|
|
|
|
2007-10-03 00:46:59 +04:00
|
|
|
def strdate(string, format, defaults=[]):
|
2006-06-30 20:47:35 +04:00
|
|
|
"""parse a localized time string and return a (unixtime, offset) tuple.
|
|
|
|
if the string cannot be parsed, ValueError is raised."""
|
2006-12-06 22:13:42 +03:00
|
|
|
def timezone(string):
|
|
|
|
tz = string.split()[-1]
|
|
|
|
if tz[0] in "+-" and len(tz) == 5 and tz[1:].isdigit():
|
2008-03-12 01:42:41 +03:00
|
|
|
sign = (tz[0] == "+") and 1 or -1
|
|
|
|
hours = int(tz[1:3])
|
|
|
|
minutes = int(tz[3:5])
|
|
|
|
return -sign * (hours * 60 + minutes) * 60
|
2006-12-06 22:13:42 +03:00
|
|
|
if tz == "GMT" or tz == "UTC":
|
|
|
|
return 0
|
|
|
|
return None
|
2006-06-30 20:47:35 +04:00
|
|
|
|
2006-10-03 14:33:14 +04:00
|
|
|
# NOTE: unixtime = localunixtime + offset
|
2006-12-06 22:13:42 +03:00
|
|
|
offset, date = timezone(string), string
|
|
|
|
if offset != None:
|
|
|
|
date = " ".join(string.split()[:-1])
|
2006-12-06 22:13:31 +03:00
|
|
|
|
2006-12-07 00:11:44 +03:00
|
|
|
# add missing elements from defaults
|
|
|
|
for part in defaults:
|
|
|
|
found = [True for p in part if ("%"+p) in format]
|
|
|
|
if not found:
|
|
|
|
date += "@" + defaults[part]
|
|
|
|
format += "@%" + part[0]
|
2006-12-06 22:13:31 +03:00
|
|
|
|
2006-10-03 14:33:18 +04:00
|
|
|
timetuple = time.strptime(date, format)
|
|
|
|
localunixtime = int(calendar.timegm(timetuple))
|
|
|
|
if offset is None:
|
|
|
|
# local timezone
|
|
|
|
unixtime = int(time.mktime(timetuple))
|
|
|
|
offset = unixtime - localunixtime
|
|
|
|
else:
|
|
|
|
unixtime = localunixtime + offset
|
2006-10-03 14:33:14 +04:00
|
|
|
return unixtime, offset
|
2006-06-30 20:47:35 +04:00
|
|
|
|
2008-02-17 23:34:28 +03:00
|
|
|
def parsedate(date, formats=None, defaults=None):
|
|
|
|
"""parse a localized date/time string and return a (unixtime, offset) tuple.
|
|
|
|
|
2006-06-30 20:47:35 +04:00
|
|
|
The date may be a "unixtime offset" string or in one of the specified
|
2008-02-17 23:34:28 +03:00
|
|
|
formats. If the date already is a (unixtime, offset) tuple, it is returned.
|
|
|
|
"""
|
|
|
|
if not date:
|
2006-12-06 22:13:27 +03:00
|
|
|
return 0, 0
|
2008-03-12 01:42:51 +03:00
|
|
|
if isinstance(date, tuple) and len(date) == 2:
|
2008-02-17 23:34:28 +03:00
|
|
|
return date
|
2006-07-13 20:40:01 +04:00
|
|
|
if not formats:
|
|
|
|
formats = defaultdateformats
|
2008-02-17 23:34:28 +03:00
|
|
|
date = date.strip()
|
2006-06-30 20:47:35 +04:00
|
|
|
try:
|
2008-02-17 23:34:28 +03:00
|
|
|
when, offset = map(int, date.split(' '))
|
2006-06-30 20:48:06 +04:00
|
|
|
except ValueError:
|
2006-12-07 00:11:44 +03:00
|
|
|
# fill out defaults
|
|
|
|
if not defaults:
|
|
|
|
defaults = {}
|
|
|
|
now = makedate()
|
|
|
|
for part in "d mb yY HI M S".split():
|
|
|
|
if part not in defaults:
|
|
|
|
if part[0] in "HMS":
|
|
|
|
defaults[part] = "00"
|
|
|
|
else:
|
2008-03-12 01:42:41 +03:00
|
|
|
defaults[part] = datestr(now, "%" + part[0])
|
2006-12-07 00:11:44 +03:00
|
|
|
|
2006-06-30 20:48:06 +04:00
|
|
|
for format in formats:
|
|
|
|
try:
|
2008-02-17 23:34:28 +03:00
|
|
|
when, offset = strdate(date, format, defaults)
|
2008-02-13 18:46:43 +03:00
|
|
|
except (ValueError, OverflowError):
|
2006-06-30 20:48:06 +04:00
|
|
|
pass
|
|
|
|
else:
|
|
|
|
break
|
|
|
|
else:
|
2008-02-17 23:34:28 +03:00
|
|
|
raise Abort(_('invalid date: %r ') % date)
|
2006-06-30 20:48:06 +04:00
|
|
|
# validate explicit (probably user-specified) date and
|
|
|
|
# time zone offset. values must fit in signed 32 bits for
|
|
|
|
# current 32-bit linux runtimes. timezones go from UTC-12
|
|
|
|
# to UTC+14
|
|
|
|
if abs(when) > 0x7fffffff:
|
2006-12-06 22:13:26 +03:00
|
|
|
raise Abort(_('date exceeds 32 bits: %d') % when)
|
2006-06-30 20:48:06 +04:00
|
|
|
if offset < -50400 or offset > 43200:
|
2006-12-06 22:13:26 +03:00
|
|
|
raise Abort(_('impossible time zone offset: %d') % offset)
|
2006-06-30 20:48:06 +04:00
|
|
|
return when, offset
|
2006-06-30 20:47:35 +04:00
|
|
|
|
2006-12-07 00:11:44 +03:00
|
|
|
def matchdate(date):
|
|
|
|
"""Return a function that matches a given date match specifier
|
|
|
|
|
|
|
|
Formats include:
|
|
|
|
|
|
|
|
'{date}' match a given date to the accuracy provided
|
|
|
|
|
|
|
|
'<{date}' on or before a given date
|
|
|
|
|
|
|
|
'>{date}' on or after a given date
|
|
|
|
|
|
|
|
"""
|
|
|
|
|
|
|
|
def lower(date):
|
2008-03-12 01:42:51 +03:00
|
|
|
d = dict(mb="1", d="1")
|
|
|
|
return parsedate(date, extendeddateformats, d)[0]
|
2006-12-07 00:11:44 +03:00
|
|
|
|
|
|
|
def upper(date):
|
|
|
|
d = dict(mb="12", HI="23", M="59", S="59")
|
|
|
|
for days in "31 30 29".split():
|
|
|
|
try:
|
|
|
|
d["d"] = days
|
|
|
|
return parsedate(date, extendeddateformats, d)[0]
|
|
|
|
except:
|
|
|
|
pass
|
|
|
|
d["d"] = "28"
|
|
|
|
return parsedate(date, extendeddateformats, d)[0]
|
|
|
|
|
|
|
|
if date[0] == "<":
|
|
|
|
when = upper(date[1:])
|
|
|
|
return lambda x: x <= when
|
|
|
|
elif date[0] == ">":
|
|
|
|
when = lower(date[1:])
|
|
|
|
return lambda x: x >= when
|
|
|
|
elif date[0] == "-":
|
|
|
|
try:
|
|
|
|
days = int(date[1:])
|
|
|
|
except ValueError:
|
|
|
|
raise Abort(_("invalid day spec: %s") % date[1:])
|
|
|
|
when = makedate()[0] - days * 3600 * 24
|
2006-12-07 00:29:17 +03:00
|
|
|
return lambda x: x >= when
|
2006-12-07 00:11:44 +03:00
|
|
|
elif " to " in date:
|
|
|
|
a, b = date.split(" to ")
|
|
|
|
start, stop = lower(a), upper(b)
|
|
|
|
return lambda x: x >= start and x <= stop
|
|
|
|
else:
|
|
|
|
start, stop = lower(date), upper(date)
|
|
|
|
return lambda x: x >= start and x <= stop
|
|
|
|
|
2006-02-27 22:32:10 +03:00
|
|
|
def shortuser(user):
|
|
|
|
"""Return a short representation of a user name or email address."""
|
|
|
|
f = user.find('@')
|
|
|
|
if f >= 0:
|
|
|
|
user = user[:f]
|
|
|
|
f = user.find('<')
|
|
|
|
if f >= 0:
|
|
|
|
user = user[f+1:]
|
2006-09-27 22:34:52 +04:00
|
|
|
f = user.find(' ')
|
|
|
|
if f >= 0:
|
|
|
|
user = user[:f]
|
2006-10-27 20:30:20 +04:00
|
|
|
f = user.find('.')
|
|
|
|
if f >= 0:
|
|
|
|
user = user[:f]
|
2006-02-27 22:32:10 +03:00
|
|
|
return user
|
2006-03-13 03:21:59 +03:00
|
|
|
|
2008-01-31 23:44:19 +03:00
|
|
|
def email(author):
|
|
|
|
'''get email of author.'''
|
|
|
|
r = author.find('>')
|
|
|
|
if r == -1: r = None
|
|
|
|
return author[author.find('<')+1:r]
|
|
|
|
|
2006-12-03 00:35:17 +03:00
|
|
|
def ellipsis(text, maxlength=400):
|
|
|
|
"""Trim string to at most maxlength (default: 400) characters."""
|
|
|
|
if len(text) <= maxlength:
|
|
|
|
return text
|
|
|
|
else:
|
|
|
|
return "%s..." % (text[:maxlength-3])
|
|
|
|
|
2008-03-15 22:42:34 +03:00
|
|
|
def walkrepos(path, followsym=False, seen_dirs=None):
|
2006-02-10 22:25:07 +03:00
|
|
|
'''yield every hg repository under path, recursively.'''
|
|
|
|
def errhandler(err):
|
|
|
|
if err.filename == path:
|
|
|
|
raise err
|
2008-03-15 22:42:34 +03:00
|
|
|
if followsym and hasattr(os.path, 'samestat'):
|
|
|
|
def _add_dir_if_not_there(dirlst, dirname):
|
|
|
|
match = False
|
|
|
|
samestat = os.path.samestat
|
|
|
|
dirstat = os.stat(dirname)
|
|
|
|
for lstdirstat in dirlst:
|
|
|
|
if samestat(dirstat, lstdirstat):
|
|
|
|
match = True
|
|
|
|
break
|
|
|
|
if not match:
|
|
|
|
dirlst.append(dirstat)
|
|
|
|
return not match
|
|
|
|
else:
|
|
|
|
followsym = false
|
2006-02-10 22:25:07 +03:00
|
|
|
|
2008-03-15 22:42:34 +03:00
|
|
|
if (seen_dirs is None) and followsym:
|
|
|
|
seen_dirs = []
|
|
|
|
_add_dir_if_not_there(seen_dirs, path)
|
|
|
|
for root, dirs, files in os.walk(path, topdown=True, onerror=errhandler):
|
2007-08-28 20:00:07 +04:00
|
|
|
if '.hg' in dirs:
|
|
|
|
dirs[:] = [] # don't descend further
|
|
|
|
yield root # found a repository
|
2008-02-21 22:56:06 +03:00
|
|
|
qroot = os.path.join(root, '.hg', 'patches')
|
2008-03-02 19:51:02 +03:00
|
|
|
if os.path.isdir(os.path.join(qroot, '.hg')):
|
2008-02-21 22:56:06 +03:00
|
|
|
yield qroot # we have a patch queue repo here
|
2008-03-15 22:42:34 +03:00
|
|
|
elif followsym:
|
|
|
|
newdirs = []
|
|
|
|
for d in dirs:
|
|
|
|
fname = os.path.join(root, d)
|
|
|
|
if _add_dir_if_not_there(seen_dirs, fname):
|
|
|
|
if os.path.islink(fname):
|
|
|
|
for hgname in walkrepos(fname, True, seen_dirs):
|
|
|
|
yield hgname
|
|
|
|
else:
|
|
|
|
newdirs.append(d)
|
|
|
|
dirs[:] = newdirs
|
2006-03-15 08:40:46 +03:00
|
|
|
|
|
|
|
_rcpath = None
|
|
|
|
|
2007-02-15 19:51:57 +03:00
|
|
|
def os_rcpath():
|
|
|
|
'''return default os-specific hgrc search path'''
|
|
|
|
path = system_rcpath()
|
|
|
|
path.extend(user_rcpath())
|
|
|
|
path = [os.path.normpath(f) for f in path]
|
|
|
|
return path
|
|
|
|
|
2006-03-15 08:40:46 +03:00
|
|
|
def rcpath():
|
|
|
|
'''return hgrc search path. if env var HGRCPATH is set, use it.
|
|
|
|
for each item in path, if directory, use files ending in .rc,
|
|
|
|
else use item.
|
|
|
|
make HGRCPATH empty to only look in .hg/hgrc of current repo.
|
|
|
|
if no HGRCPATH, use default os-specific path.'''
|
|
|
|
global _rcpath
|
|
|
|
if _rcpath is None:
|
|
|
|
if 'HGRCPATH' in os.environ:
|
|
|
|
_rcpath = []
|
|
|
|
for p in os.environ['HGRCPATH'].split(os.pathsep):
|
|
|
|
if not p: continue
|
2006-03-15 09:09:42 +03:00
|
|
|
if os.path.isdir(p):
|
2007-10-06 02:01:06 +04:00
|
|
|
for f, kind in osutil.listdir(p):
|
2006-03-15 08:40:46 +03:00
|
|
|
if f.endswith('.rc'):
|
|
|
|
_rcpath.append(os.path.join(p, f))
|
2006-03-15 09:09:42 +03:00
|
|
|
else:
|
|
|
|
_rcpath.append(p)
|
2006-03-15 08:40:46 +03:00
|
|
|
else:
|
|
|
|
_rcpath = os_rcpath()
|
|
|
|
return _rcpath
|
2006-07-14 22:17:22 +04:00
|
|
|
|
|
|
|
def bytecount(nbytes):
|
|
|
|
'''return byte count formatted as readable string, with units'''
|
|
|
|
|
|
|
|
units = (
|
|
|
|
(100, 1<<30, _('%.0f GB')),
|
|
|
|
(10, 1<<30, _('%.1f GB')),
|
|
|
|
(1, 1<<30, _('%.2f GB')),
|
|
|
|
(100, 1<<20, _('%.0f MB')),
|
|
|
|
(10, 1<<20, _('%.1f MB')),
|
|
|
|
(1, 1<<20, _('%.2f MB')),
|
|
|
|
(100, 1<<10, _('%.0f KB')),
|
|
|
|
(10, 1<<10, _('%.1f KB')),
|
|
|
|
(1, 1<<10, _('%.2f KB')),
|
|
|
|
(1, 1, _('%.0f bytes')),
|
|
|
|
)
|
|
|
|
|
|
|
|
for multiplier, divisor, format in units:
|
|
|
|
if nbytes >= divisor * multiplier:
|
|
|
|
return format % (nbytes / float(divisor))
|
|
|
|
return units[-1][2] % nbytes
|
2006-07-31 18:11:12 +04:00
|
|
|
|
|
|
|
def drop_scheme(scheme, path):
|
|
|
|
sc = scheme + ':'
|
|
|
|
if path.startswith(sc):
|
|
|
|
path = path[len(sc):]
|
|
|
|
if path.startswith('//'):
|
|
|
|
path = path[2:]
|
|
|
|
return path
|
2007-09-11 01:36:01 +04:00
|
|
|
|
|
|
|
def uirepr(s):
|
|
|
|
# Avoid double backslash in Windows path repr()
|
|
|
|
return repr(s).replace('\\\\', '\\')
|
2007-11-05 22:29:32 +03:00
|
|
|
|
2008-01-05 00:56:31 +03:00
|
|
|
def hidepassword(url):
|
2008-01-04 22:58:27 +03:00
|
|
|
'''hide user credential in a url string'''
|
|
|
|
scheme, netloc, path, params, query, fragment = urlparse.urlparse(url)
|
|
|
|
netloc = re.sub('([^:]*):([^@]*)@(.*)', r'\1:***@\3', netloc)
|
|
|
|
return urlparse.urlunparse((scheme, netloc, path, params, query, fragment))
|
|
|
|
|
|
|
|
def removeauth(url):
|
|
|
|
'''remove all authentication information from a url string'''
|
|
|
|
scheme, netloc, path, params, query, fragment = urlparse.urlparse(url)
|
|
|
|
netloc = netloc[netloc.find('@')+1:]
|
|
|
|
return urlparse.urlunparse((scheme, netloc, path, params, query, fragment))
|