osutil: switch to policy importer

"make clean" is recommended to test this change, though C API compatibility
should be preserved.

contrib/check-py3-compat.py

@@ -1,107 +1,106 b''
 #!/usr/bin/env python
 #
 # check-py3-compat - check Python 3 compatibility of Mercurial files
 #
 # Copyright 2015 Gregory Szorc <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import, print_function
 
 import ast
 import importlib
 import os
 import sys
 import traceback
 
 # Modules that have both Python and C implementations.
 _dualmodules = (
     'base85.py',
     'bdiff.py',
     'diffhelpers.py',
     'mpatch.py',
-    'osutil.py',
     'parsers.py',
 )
 
 def check_compat_py2(f):
     """Check Python 3 compatibility for a file with Python 2"""
     with open(f, 'rb') as fh:
         content = fh.read()
     root = ast.parse(content)
 
     # Ignore empty files.
     if not root.body:
         return
 
     futures = set()
     haveprint = False
     for node in ast.walk(root):
         if isinstance(node, ast.ImportFrom):
             if node.module == '__future__':
                 futures |= set(n.name for n in node.names)
         elif isinstance(node, ast.Print):
             haveprint = True
 
     if 'absolute_import' not in futures:
         print('%s not using absolute_import' % f)
     if haveprint and 'print_function' not in futures:
         print('%s requires print_function' % f)
 
 def check_compat_py3(f):
     """Check Python 3 compatibility of a file with Python 3."""
     with open(f, 'rb') as fh:
         content = fh.read()
 
     try:
         ast.parse(content)
     except SyntaxError as e:
         print('%s: invalid syntax: %s' % (f, e))
         return
 
     # Try to import the module.
     # For now we only support mercurial.* and hgext.* modules because figuring
     # out module paths for things not in a package can be confusing.
     if f.startswith(('hgext/', 'mercurial/')) and not f.endswith('__init__.py'):
         assert f.endswith('.py')
         name = f.replace('/', '.')[:-3]
         if f.endswith(_dualmodules):
             name = name.replace('.pure.', '.')
         try:
             importlib.import_module(name)
         except Exception as e:
             exc_type, exc_value, tb = sys.exc_info()
             # We walk the stack and ignore frames from our custom importer,
             # import mechanisms, and stdlib modules. This kinda/sorta
             # emulates CPython behavior in import.c while also attempting
             # to pin blame on a Mercurial file.
             for frame in reversed(traceback.extract_tb(tb)):
                 if frame.name == '_call_with_frames_removed':
                     continue
                 if 'importlib' in frame.filename:
                     continue
                 if 'mercurial/__init__.py' in frame.filename:
                     continue
                 if frame.filename.startswith(sys.prefix):
                     continue
                 break
 
             if frame.filename:
                 filename = os.path.basename(frame.filename)
                 print('%s: error importing: <%s> %s (error at %s:%d)' % (
                       f, type(e).__name__, e, filename, frame.lineno))
             else:
                 print('%s: error importing module: <%s> %s (line %d)' % (
                       f, type(e).__name__, e, frame.lineno))
 
 if __name__ == '__main__':
     if sys.version_info[0] == 2:
         fn = check_compat_py2
     else:
         fn = check_compat_py3
 
     for f in sys.argv[1:]:
         fn(f)
 
     sys.exit(0)

contrib/import-checker.py

@@ -1,736 +1,735 b''
 #!/usr/bin/env python
 
 from __future__ import absolute_import, print_function
 
 import ast
 import collections
 import os
 import re
 import sys
 
 # Import a minimal set of stdlib modules needed for list_stdlib_modules()
 # to work when run from a virtualenv.  The modules were chosen empirically
 # so that the return value matches the return value without virtualenv.
 if True: # disable lexical sorting checks
     import BaseHTTPServer
     import zlib
 
 # Whitelist of modules that symbols can be directly imported from.
 allowsymbolimports = (
     '__future__',
     'mercurial.hgweb.common',
     'mercurial.hgweb.request',
     'mercurial.i18n',
     'mercurial.node',
 )
 
 # Modules that have both Python and C implementations.
 _dualmodules = (
     'base85.py',
     'bdiff.py',
     'diffhelpers.py',
     'mpatch.py',
-    'osutil.py',
     'parsers.py',
 )
 
 # Modules that must be aliased because they are commonly confused with
 # common variables and can create aliasing and readability issues.
 requirealias = {
     'ui': 'uimod',
 }
 
 def usingabsolute(root):
     """Whether absolute imports are being used."""
     if sys.version_info[0] >= 3:
         return True
 
     for node in ast.walk(root):
         if isinstance(node, ast.ImportFrom):
             if node.module == '__future__':
                 for n in node.names:
                     if n.name == 'absolute_import':
                         return True
 
     return False
 
 def walklocal(root):
     """Recursively yield all descendant nodes but not in a different scope"""
     todo = collections.deque(ast.iter_child_nodes(root))
     yield root, False
     while todo:
         node = todo.popleft()
         newscope = isinstance(node, ast.FunctionDef)
         if not newscope:
             todo.extend(ast.iter_child_nodes(node))
         yield node, newscope
 
 def dotted_name_of_path(path, trimpure=False):
     """Given a relative path to a source file, return its dotted module name.
 
     >>> dotted_name_of_path('mercurial/error.py')
     'mercurial.error'
     >>> dotted_name_of_path('mercurial/pure/parsers.py', trimpure=True)
     'mercurial.parsers'
     >>> dotted_name_of_path('zlibmodule.so')
     'zlib'
     """
     parts = path.replace(os.sep, '/').split('/')
     parts[-1] = parts[-1].split('.', 1)[0] # remove .py and .so and .ARCH.so
     if parts[-1].endswith('module'):
         parts[-1] = parts[-1][:-6]
     if trimpure:
         return '.'.join(p for p in parts if p != 'pure')
     return '.'.join(parts)
 
 def fromlocalfunc(modulename, localmods):
     """Get a function to examine which locally defined module the
     target source imports via a specified name.
 
     `modulename` is an `dotted_name_of_path()`-ed source file path,
     which may have `.__init__` at the end of it, of the target source.
 
     `localmods` is a dict (or set), of which key is an absolute
     `dotted_name_of_path()`-ed source file path of locally defined (=
     Mercurial specific) modules.
 
     This function assumes that module names not existing in
     `localmods` are from the Python standard library.
 
     This function returns the function, which takes `name` argument,
     and returns `(absname, dottedpath, hassubmod)` tuple if `name`
     matches against locally defined module. Otherwise, it returns
     False.
 
     It is assumed that `name` doesn't have `.__init__`.
 
     `absname` is an absolute module name of specified `name`
     (e.g. "hgext.convert"). This can be used to compose prefix for sub
     modules or so.
 
     `dottedpath` is a `dotted_name_of_path()`-ed source file path
     (e.g. "hgext.convert.__init__") of `name`. This is used to look
     module up in `localmods` again.
 
     `hassubmod` is whether it may have sub modules under it (for
     convenient, even though this is also equivalent to "absname !=
     dottednpath")
 
     >>> localmods = {'foo.__init__': True, 'foo.foo1': True,
     ...              'foo.bar.__init__': True, 'foo.bar.bar1': True,
     ...              'baz.__init__': True, 'baz.baz1': True }
     >>> fromlocal = fromlocalfunc('foo.xxx', localmods)
     >>> # relative
     >>> fromlocal('foo1')
     ('foo.foo1', 'foo.foo1', False)
     >>> fromlocal('bar')
     ('foo.bar', 'foo.bar.__init__', True)
     >>> fromlocal('bar.bar1')
     ('foo.bar.bar1', 'foo.bar.bar1', False)
     >>> # absolute
     >>> fromlocal('baz')
     ('baz', 'baz.__init__', True)
     >>> fromlocal('baz.baz1')
     ('baz.baz1', 'baz.baz1', False)
     >>> # unknown = maybe standard library
     >>> fromlocal('os')
     False
     >>> fromlocal(None, 1)
     ('foo', 'foo.__init__', True)
     >>> fromlocal('foo1', 1)
     ('foo.foo1', 'foo.foo1', False)
     >>> fromlocal2 = fromlocalfunc('foo.xxx.yyy', localmods)
     >>> fromlocal2(None, 2)
     ('foo', 'foo.__init__', True)
     >>> fromlocal2('bar2', 1)
     False
     >>> fromlocal2('bar', 2)
     ('foo.bar', 'foo.bar.__init__', True)
     """
     prefix = '.'.join(modulename.split('.')[:-1])
     if prefix:
         prefix += '.'
     def fromlocal(name, level=0):
         # name is false value when relative imports are used.
         if not name:
             # If relative imports are used, level must not be absolute.
             assert level > 0
             candidates = ['.'.join(modulename.split('.')[:-level])]
         else:
             if not level:
                 # Check relative name first.
                 candidates = [prefix + name, name]
             else:
                 candidates = ['.'.join(modulename.split('.')[:-level]) +
                               '.' + name]
 
         for n in candidates:
             if n in localmods:
                 return (n, n, False)
             dottedpath = n + '.__init__'
             if dottedpath in localmods:
                 return (n, dottedpath, True)
         return False
     return fromlocal
 
 def list_stdlib_modules():
     """List the modules present in the stdlib.
 
     >>> mods = set(list_stdlib_modules())
     >>> 'BaseHTTPServer' in mods
     True
 
     os.path isn't really a module, so it's missing:
 
     >>> 'os.path' in mods
     False
 
     sys requires special treatment, because it's baked into the
     interpreter, but it should still appear:
 
     >>> 'sys' in mods
     True
 
     >>> 'collections' in mods
     True
 
     >>> 'cStringIO' in mods
     True
 
     >>> 'cffi' in mods
     True
     """
     for m in sys.builtin_module_names:
         yield m
     # These modules only exist on windows, but we should always
     # consider them stdlib.
     for m in ['msvcrt', '_winreg']:
         yield m
     yield 'builtins' # python3 only
     for m in 'fcntl', 'grp', 'pwd', 'termios':  # Unix only
         yield m
     for m in 'cPickle', 'datetime': # in Python (not C) on PyPy
         yield m
     for m in ['cffi']:
         yield m
     stdlib_prefixes = {sys.prefix, sys.exec_prefix}
     # We need to supplement the list of prefixes for the search to work
     # when run from within a virtualenv.
     for mod in (BaseHTTPServer, zlib):
         try:
             # Not all module objects have a __file__ attribute.
             filename = mod.__file__
         except AttributeError:
             continue
         dirname = os.path.dirname(filename)
         for prefix in stdlib_prefixes:
             if dirname.startswith(prefix):
                 # Then this directory is redundant.
                 break
         else:
             stdlib_prefixes.add(dirname)
     for libpath in sys.path:
         # We want to walk everything in sys.path that starts with
         # something in stdlib_prefixes.
         if not any(libpath.startswith(p) for p in stdlib_prefixes):
             continue
         for top, dirs, files in os.walk(libpath):
             for i, d in reversed(list(enumerate(dirs))):
                 if (not os.path.exists(os.path.join(top, d, '__init__.py'))
                     or top == libpath and d in ('hgext', 'mercurial')):
                     del dirs[i]
             for name in files:
                 if not name.endswith(('.py', '.so', '.pyc', '.pyo', '.pyd')):
                     continue
                 if name.startswith('__init__.py'):
                     full_path = top
                 else:
                     full_path = os.path.join(top, name)
                 rel_path = full_path[len(libpath) + 1:]
                 mod = dotted_name_of_path(rel_path)
                 yield mod
 
 stdlib_modules = set(list_stdlib_modules())
 
 def imported_modules(source, modulename, f, localmods, ignore_nested=False):
     """Given the source of a file as a string, yield the names
     imported by that file.
 
     Args:
       source: The python source to examine as a string.
       modulename: of specified python source (may have `__init__`)
       localmods: dict of locally defined module names (may have `__init__`)
       ignore_nested: If true, import statements that do not start in
                      column zero will be ignored.
 
     Returns:
       A list of absolute module names imported by the given source.
 
     >>> f = 'foo/xxx.py'
     >>> modulename = 'foo.xxx'
     >>> localmods = {'foo.__init__': True,
     ...              'foo.foo1': True, 'foo.foo2': True,
     ...              'foo.bar.__init__': True, 'foo.bar.bar1': True,
     ...              'baz.__init__': True, 'baz.baz1': True }
     >>> # standard library (= not locally defined ones)
     >>> sorted(imported_modules(
     ...        'from stdlib1 import foo, bar; import stdlib2',
     ...        modulename, f, localmods))
     []
     >>> # relative importing
     >>> sorted(imported_modules(
     ...        'import foo1; from bar import bar1',
     ...        modulename, f, localmods))
     ['foo.bar.bar1', 'foo.foo1']
     >>> sorted(imported_modules(
     ...        'from bar.bar1 import name1, name2, name3',
     ...        modulename, f, localmods))
     ['foo.bar.bar1']
     >>> # absolute importing
     >>> sorted(imported_modules(
     ...        'from baz import baz1, name1',
     ...        modulename, f, localmods))
     ['baz.__init__', 'baz.baz1']
     >>> # mixed importing, even though it shouldn't be recommended
     >>> sorted(imported_modules(
     ...        'import stdlib, foo1, baz',
     ...        modulename, f, localmods))
     ['baz.__init__', 'foo.foo1']
     >>> # ignore_nested
     >>> sorted(imported_modules(
     ... '''import foo
     ... def wat():
     ...     import bar
     ... ''', modulename, f, localmods))
     ['foo.__init__', 'foo.bar.__init__']
     >>> sorted(imported_modules(
     ... '''import foo
     ... def wat():
     ...     import bar
     ... ''', modulename, f, localmods, ignore_nested=True))
     ['foo.__init__']
     """
     fromlocal = fromlocalfunc(modulename, localmods)
     for node in ast.walk(ast.parse(source, f)):
         if ignore_nested and getattr(node, 'col_offset', 0) > 0:
             continue
         if isinstance(node, ast.Import):
             for n in node.names:
                 found = fromlocal(n.name)
                 if not found:
                     # this should import standard library
                     continue
                 yield found[1]
         elif isinstance(node, ast.ImportFrom):
             found = fromlocal(node.module, node.level)
             if not found:
                 # this should import standard library
                 continue
 
             absname, dottedpath, hassubmod = found
             if not hassubmod:
                 # "dottedpath" is not a package; must be imported
                 yield dottedpath
                 # examination of "node.names" should be redundant
                 # e.g.: from mercurial.node import nullid, nullrev
                 continue
 
             modnotfound = False
             prefix = absname + '.'
             for n in node.names:
                 found = fromlocal(prefix + n.name)
                 if not found:
                     # this should be a function or a property of "node.module"
                     modnotfound = True
                     continue
                 yield found[1]
             if modnotfound:
                 # "dottedpath" is a package, but imported because of non-module
                 # lookup
                 yield dottedpath
 
 def verify_import_convention(module, source, localmods):
     """Verify imports match our established coding convention.
 
     We have 2 conventions: legacy and modern. The modern convention is in
     effect when using absolute imports.
 
     The legacy convention only looks for mixed imports. The modern convention
     is much more thorough.
     """
     root = ast.parse(source)
     absolute = usingabsolute(root)
 
     if absolute:
         return verify_modern_convention(module, root, localmods)
     else:
         return verify_stdlib_on_own_line(root)
 
 def verify_modern_convention(module, root, localmods, root_col_offset=0):
     """Verify a file conforms to the modern import convention rules.
 
     The rules of the modern convention are:
 
     * Ordering is stdlib followed by local imports. Each group is lexically
       sorted.
     * Importing multiple modules via "import X, Y" is not allowed: use
       separate import statements.
     * Importing multiple modules via "from X import ..." is allowed if using
       parenthesis and one entry per line.
     * Only 1 relative import statement per import level ("from .", "from ..")
       is allowed.
     * Relative imports from higher levels must occur before lower levels. e.g.
       "from .." must be before "from .".
     * Imports from peer packages should use relative import (e.g. do not
       "import mercurial.foo" from a "mercurial.*" module).
     * Symbols can only be imported from specific modules (see
       `allowsymbolimports`). For other modules, first import the module then
       assign the symbol to a module-level variable. In addition, these imports
       must be performed before other local imports. This rule only
       applies to import statements outside of any blocks.
     * Relative imports from the standard library are not allowed.
     * Certain modules must be aliased to alternate names to avoid aliasing
       and readability problems. See `requirealias`.
     """
     topmodule = module.split('.')[0]
     fromlocal = fromlocalfunc(module, localmods)
 
     # Whether a local/non-stdlib import has been performed.
     seenlocal = None
     # Whether a local/non-stdlib, non-symbol import has been seen.
     seennonsymbollocal = False
     # The last name to be imported (for sorting).
     lastname = None
     laststdlib = None
     # Relative import levels encountered so far.
     seenlevels = set()
 
     for node, newscope in walklocal(root):
         def msg(fmt, *args):
             return (fmt % args, node.lineno)
         if newscope:
             # Check for local imports in function
             for r in verify_modern_convention(module, node, localmods,
                                               node.col_offset + 4):
                 yield r
         elif isinstance(node, ast.Import):
             # Disallow "import foo, bar" and require separate imports
             # for each module.
             if len(node.names) > 1:
                 yield msg('multiple imported names: %s',
                           ', '.join(n.name for n in node.names))
 
             name = node.names[0].name
             asname = node.names[0].asname
 
             stdlib = name in stdlib_modules
 
             # Ignore sorting rules on imports inside blocks.
             if node.col_offset == root_col_offset:
                 if lastname and name < lastname and laststdlib == stdlib:
                     yield msg('imports not lexically sorted: %s < %s',
                               name, lastname)
 
             lastname = name
             laststdlib = stdlib
 
             # stdlib imports should be before local imports.
             if stdlib and seenlocal and node.col_offset == root_col_offset:
                 yield msg('stdlib import "%s" follows local import: %s',
                           name, seenlocal)
 
             if not stdlib:
                 seenlocal = name
 
             # Import of sibling modules should use relative imports.
             topname = name.split('.')[0]
             if topname == topmodule:
                 yield msg('import should be relative: %s', name)
 
             if name in requirealias and asname != requirealias[name]:
                 yield msg('%s module must be "as" aliased to %s',
                           name, requirealias[name])
 
         elif isinstance(node, ast.ImportFrom):
             # Resolve the full imported module name.
             if node.level > 0:
                 fullname = '.'.join(module.split('.')[:-node.level])
                 if node.module:
                     fullname += '.%s' % node.module
             else:
                 assert node.module
                 fullname = node.module
 
                 topname = fullname.split('.')[0]
                 if topname == topmodule:
                     yield msg('import should be relative: %s', fullname)
 
             # __future__ is special since it needs to come first and use
             # symbol import.
             if fullname != '__future__':
                 if not fullname or fullname in stdlib_modules:
                     yield msg('relative import of stdlib module')
                 else:
                     seenlocal = fullname
 
             # Direct symbol import is only allowed from certain modules and
             # must occur before non-symbol imports.
             found = fromlocal(node.module, node.level)
             if found and found[2]:  # node.module is a package
                 prefix = found[0] + '.'
                 symbols = [n.name for n in node.names
                            if not fromlocal(prefix + n.name)]
             else:
                 symbols = [n.name for n in node.names]
             if node.module and node.col_offset == root_col_offset:
                 if symbols and fullname not in allowsymbolimports:
                     yield msg('direct symbol import %s from %s',
                               ', '.join(symbols), fullname)
 
                 if symbols and seennonsymbollocal:
                     yield msg('symbol import follows non-symbol import: %s',
                               fullname)
             if not symbols and fullname not in stdlib_modules:
                 seennonsymbollocal = True
 
             if not node.module:
                 assert node.level
 
                 # Only allow 1 group per level.
                 if (node.level in seenlevels
                     and node.col_offset == root_col_offset):
                     yield msg('multiple "from %s import" statements',
                               '.' * node.level)
 
                 # Higher-level groups come before lower-level groups.
                 if any(node.level > l for l in seenlevels):
                     yield msg('higher-level import should come first: %s',
                               fullname)
 
                 seenlevels.add(node.level)
 
             # Entries in "from .X import ( ... )" lists must be lexically
             # sorted.
             lastentryname = None
 
             for n in node.names:
                 if lastentryname and n.name < lastentryname:
                     yield msg('imports from %s not lexically sorted: %s < %s',
                               fullname, n.name, lastentryname)
 
                 lastentryname = n.name
 
                 if n.name in requirealias and n.asname != requirealias[n.name]:
                     yield msg('%s from %s must be "as" aliased to %s',
                               n.name, fullname, requirealias[n.name])
 
 def verify_stdlib_on_own_line(root):
     """Given some python source, verify that stdlib imports are done
     in separate statements from relative local module imports.
 
     >>> list(verify_stdlib_on_own_line(ast.parse('import sys, foo')))
     [('mixed imports\\n   stdlib:    sys\\n   relative:  foo', 1)]
     >>> list(verify_stdlib_on_own_line(ast.parse('import sys, os')))
     []
     >>> list(verify_stdlib_on_own_line(ast.parse('import foo, bar')))
     []
     """
     for node in ast.walk(root):
         if isinstance(node, ast.Import):
             from_stdlib = {False: [], True: []}
             for n in node.names:
                 from_stdlib[n.name in stdlib_modules].append(n.name)
             if from_stdlib[True] and from_stdlib[False]:
                 yield ('mixed imports\n   stdlib:    %s\n   relative:  %s' %
                        (', '.join(sorted(from_stdlib[True])),
                         ', '.join(sorted(from_stdlib[False]))), node.lineno)
 
 class CircularImport(Exception):
     pass
 
 def checkmod(mod, imports):
     shortest = {}
     visit = [[mod]]
     while visit:
         path = visit.pop(0)
         for i in sorted(imports.get(path[-1], [])):
             if len(path) < shortest.get(i, 1000):
                 shortest[i] = len(path)
                 if i in path:
                     if i == path[0]:
                         raise CircularImport(path)
                     continue
                 visit.append(path + [i])
 
 def rotatecycle(cycle):
     """arrange a cycle so that the lexicographically first module listed first
 
     >>> rotatecycle(['foo', 'bar'])
     ['bar', 'foo', 'bar']
     """
     lowest = min(cycle)
     idx = cycle.index(lowest)
     return cycle[idx:] + cycle[:idx] + [lowest]
 
 def find_cycles(imports):
     """Find cycles in an already-loaded import graph.
 
     All module names recorded in `imports` should be absolute one.
 
     >>> from __future__ import print_function
     >>> imports = {'top.foo': ['top.bar', 'os.path', 'top.qux'],
     ...            'top.bar': ['top.baz', 'sys'],
     ...            'top.baz': ['top.foo'],
     ...            'top.qux': ['top.foo']}
     >>> print('\\n'.join(sorted(find_cycles(imports))))
     top.bar -> top.baz -> top.foo -> top.bar
     top.foo -> top.qux -> top.foo
     """
     cycles = set()
     for mod in sorted(imports.keys()):
         try:
             checkmod(mod, imports)
         except CircularImport as e:
             cycle = e.args[0]
             cycles.add(" -> ".join(rotatecycle(cycle)))
     return cycles
 
 def _cycle_sortkey(c):
     return len(c), c
 
 def embedded(f, modname, src):
     """Extract embedded python code
 
     >>> def test(fn, lines):
     ...     for s, m, f, l in embedded(fn, "example", lines):
     ...         print("%s %s %s" % (m, f, l))
     ...         print(repr(s))
     >>> lines = [
     ...   'comment',
     ...   '  >>> from __future__ import print_function',
     ...   "  >>> ' multiline",
     ...   "  ... string'",
     ...   '  ',
     ...   'comment',
     ...   '  $ cat > foo.py <<EOF',
     ...   '  > from __future__ import print_function',
     ...   '  > EOF',
     ... ]
     >>> test("example.t", lines)
     example[2] doctest.py 2
     "from __future__ import print_function\\n' multiline\\nstring'\\n"
     example[7] foo.py 7
     'from __future__ import print_function\\n'
     """
     inlinepython = 0
     shpython = 0
     script = []
     prefix = 6
     t = ''
     n = 0
     for l in src:
         n += 1
         if not l.endswith(b'\n'):
             l += b'\n'
         if l.startswith(b'  >>> '): # python inlines
             if shpython:
                 print("%s:%d: Parse Error" % (f, n))
             if not inlinepython:
                 # We've just entered a Python block.
                 inlinepython = n
                 t = 'doctest.py'
             script.append(l[prefix:])
             continue
         if l.startswith(b'  ... '): # python inlines
             script.append(l[prefix:])
             continue
         cat = re.search(r"\$ \s*cat\s*>\s*(\S+\.py)\s*<<\s*EOF", l)
         if cat:
             if inlinepython:
                 yield ''.join(script), ("%s[%d]" %
                        (modname, inlinepython)), t, inlinepython
                 script = []
                 inlinepython = 0
             shpython = n
             t = cat.group(1)
             continue
         if shpython and l.startswith(b'  > '): # sh continuation
             if l == b'  > EOF\n':
                 yield ''.join(script), ("%s[%d]" %
                        (modname, shpython)), t, shpython
                 script = []
                 shpython = 0
             else:
                 script.append(l[4:])
             continue
         if inlinepython and l == b'  \n':
             yield ''.join(script), ("%s[%d]" %
                    (modname, inlinepython)), t, inlinepython
             script = []
             inlinepython = 0
             continue
 
 def sources(f, modname):
     """Yields possibly multiple sources from a filepath
 
     input: filepath, modulename
     yields:  script(string), modulename, filepath, linenumber
 
     For embedded scripts, the modulename and filepath will be different
     from the function arguments. linenumber is an offset relative to
     the input file.
     """
     py = False
     if not f.endswith('.t'):
         with open(f) as src:
             yield src.read(), modname, f, 0
             py = True
     if py or f.endswith('.t'):
         with open(f) as src:
             for script, modname, t, line in embedded(f, modname, src):
                 yield script, modname, t, line
 
 def main(argv):
     if len(argv) < 2 or (argv[1] == '-' and len(argv) > 2):
         print('Usage: %s {-|file [file] [file] ...}')
         return 1
     if argv[1] == '-':
         argv = argv[:1]
         argv.extend(l.rstrip() for l in sys.stdin.readlines())
     localmods = {}
     used_imports = {}
     any_errors = False
     for source_path in argv[1:]:
         trimpure = source_path.endswith(_dualmodules)
         modname = dotted_name_of_path(source_path, trimpure=trimpure)
         localmods[modname] = source_path
     for localmodname, source_path in sorted(localmods.items()):
         for src, modname, name, line in sources(source_path, localmodname):
             try:
                 used_imports[modname] = sorted(
                     imported_modules(src, modname, name, localmods,
                                      ignore_nested=True))
                 for error, lineno in verify_import_convention(modname, src,
                                                               localmods):
                     any_errors = True
                     print('%s:%d: %s' % (source_path, lineno + line, error))
             except SyntaxError as e:
                 print('%s:%d: SyntaxError: %s' %
                       (source_path, e.lineno + line, e))
     cycles = find_cycles(used_imports)
     if cycles:
         firstmods = set()
         for c in sorted(cycles, key=_cycle_sortkey):
             first = c.split()[0]
             # As a rough cut, ignore any cycle that starts with the
             # same module as some other cycle. Otherwise we see lots
             # of cycles that are effectively duplicates.
             if first in firstmods:
                 continue
             print('Import cycle:', c)
             firstmods.add(first)
         any_errors = True
     return any_errors != 0
 
 if __name__ == '__main__':
     sys.exit(int(main(sys.argv)))

contrib/wix/dist.wxs

@@ -1,35 +1,35 b''
 <?xml version="1.0" encoding="utf-8"?>
 <Wix xmlns="http://schemas.microsoft.com/wix/2006/wi">
 
   <?include guids.wxi ?>
   <?include defines.wxi ?>
 
   <Fragment>
     <DirectoryRef Id="INSTALLDIR" FileSource="$(var.SourceDir)">
       <Component Id="distOutput" Guid="$(var.dist.guid)" Win64='$(var.IsX64)'>
         <File Name="python27.dll" KeyPath="yes" />
       </Component>
       <Directory Id="libdir" Name="lib" FileSource="$(var.SourceDir)/lib">
         <Component Id="libOutput" Guid="$(var.lib.guid)" Win64='$(var.IsX64)'>
           <File Name="library.zip" KeyPath="yes" />
           <File Name="mercurial.base85.pyd" />
           <File Name="mercurial.bdiff.pyd" />
           <File Name="mercurial.diffhelpers.pyd" />
           <File Name="mercurial.mpatch.pyd" />
-          <File Name="mercurial.osutil.pyd" />
+          <File Name="mercurial.cext.osutil.pyd" />
           <File Name="mercurial.parsers.pyd" />
           <File Name="pyexpat.pyd" />
           <File Name="bz2.pyd" />
           <File Name="select.pyd" />
           <File Name="unicodedata.pyd" />
           <File Name="_ctypes.pyd" />
           <File Name="_elementtree.pyd" />
           <File Name="_hashlib.pyd" />
           <File Name="_socket.pyd" />
           <File Name="_ssl.pyd" />
         </Component>
       </Directory>
     </DirectoryRef>
   </Fragment>
 
 </Wix>

mercurial/__init__.py

@@ -1,404 +1,403 b''
 # __init__.py - Startup and module loading logic for Mercurial.
 #
 # Copyright 2015 Gregory Szorc <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import imp
 import os
 import sys
 import zipimport
 
 from . import (
     policy
 )
 
 __all__ = []
 
 modulepolicy = policy.policy
 
 # Modules that have both Python and C implementations. See also the
 # set of .py files under mercurial/pure/.
 _dualmodules = {
     'mercurial.base85',
     'mercurial.bdiff',
     'mercurial.diffhelpers',
     'mercurial.mpatch',
-    'mercurial.osutil',
     'mercurial.parsers',
 }
 
 class hgimporter(object):
     """Object that conforms to import hook interface defined in PEP-302."""
     def find_module(self, name, path=None):
         # We only care about modules that have both C and pure implementations.
         if name in _dualmodules:
             return self
         return None
 
     def load_module(self, name):
         mod = sys.modules.get(name, None)
         if mod:
             return mod
 
         mercurial = sys.modules['mercurial']
 
         # The zip importer behaves sufficiently differently from the default
         # importer to warrant its own code path.
         loader = getattr(mercurial, '__loader__', None)
         if isinstance(loader, zipimport.zipimporter):
             def ziploader(*paths):
                 """Obtain a zipimporter for a directory under the main zip."""
                 path = os.path.join(loader.archive, *paths)
                 zl = sys.path_importer_cache.get(path)
                 if not zl:
                     zl = zipimport.zipimporter(path)
                 return zl
 
             try:
                 if modulepolicy in policy.policynoc:
                     raise ImportError()
 
                 zl = ziploader('mercurial')
                 mod = zl.load_module(name)
                 # Unlike imp, ziploader doesn't expose module metadata that
                 # indicates the type of module. So just assume what we found
                 # is OK (even though it could be a pure Python module).
             except ImportError:
                 if modulepolicy == b'c':
                     raise
                 zl = ziploader('mercurial', 'pure')
                 mod = zl.load_module(name)
 
             sys.modules[name] = mod
             return mod
 
         # Unlike the default importer which searches special locations and
         # sys.path, we only look in the directory where "mercurial" was
         # imported from.
 
         # imp.find_module doesn't support submodules (modules with ".").
         # Instead you have to pass the parent package's __path__ attribute
         # as the path argument.
         stem = name.split('.')[-1]
 
         try:
             if modulepolicy in policy.policynoc:
                 raise ImportError()
 
             modinfo = imp.find_module(stem, mercurial.__path__)
 
             # The Mercurial installer used to copy files from
             # mercurial/pure/*.py to mercurial/*.py. Therefore, it's possible
             # for some installations to have .py files under mercurial/*.
             # Loading Python modules when we expected C versions could result
             # in a) poor performance b) loading a version from a previous
             # Mercurial version, potentially leading to incompatibility. Either
             # scenario is bad. So we verify that modules loaded from
             # mercurial/* are C extensions. If the current policy allows the
             # loading of .py modules, the module will be re-imported from
             # mercurial/pure/* below.
             if modinfo[2][2] != imp.C_EXTENSION:
                 raise ImportError('.py version of %s found where C '
                                   'version should exist' % name)
 
         except ImportError:
             if modulepolicy == b'c':
                 raise
 
             # Could not load the C extension and pure Python is allowed. So
             # try to load them.
             from . import pure
             modinfo = imp.find_module(stem, pure.__path__)
             if not modinfo:
                 raise ImportError('could not find mercurial module %s' %
                                   name)
 
         mod = imp.load_module(name, *modinfo)
         sys.modules[name] = mod
         return mod
 
 # Python 3 uses a custom module loader that transforms source code between
 # source file reading and compilation. This is done by registering a custom
 # finder that changes the spec for Mercurial modules to use a custom loader.
 if sys.version_info[0] >= 3:
     from . import pure
     import importlib
     import io
     import token
     import tokenize
 
     class hgpathentryfinder(importlib.abc.MetaPathFinder):
         """A sys.meta_path finder that uses a custom module loader."""
         def find_spec(self, fullname, path, target=None):
             # Only handle Mercurial-related modules.
             if not fullname.startswith(('mercurial.', 'hgext.', 'hgext3rd.')):
                 return None
             # zstd is already dual-version clean, don't try and mangle it
             if fullname.startswith('mercurial.zstd'):
                 return None
 
             # This assumes Python 3 doesn't support loading C modules.
             if fullname in _dualmodules:
                 stem = fullname.split('.')[-1]
                 fullname = 'mercurial.pure.%s' % stem
                 target = pure
                 assert len(path) == 1
                 path = [os.path.join(path[0], 'pure')]
 
             # Try to find the module using other registered finders.
             spec = None
             for finder in sys.meta_path:
                 if finder == self:
                     continue
 
                 spec = finder.find_spec(fullname, path, target=target)
                 if spec:
                     break
 
             # This is a Mercurial-related module but we couldn't find it
             # using the previously-registered finders. This likely means
             # the module doesn't exist.
             if not spec:
                 return None
 
             if (fullname.startswith('mercurial.pure.')
                 and fullname.replace('.pure.', '.') in _dualmodules):
                 spec.name = spec.name.replace('.pure.', '.')
 
             # TODO need to support loaders from alternate specs, like zip
             # loaders.
             spec.loader = hgloader(spec.name, spec.origin)
             return spec
 
     def replacetokens(tokens, fullname):
         """Transform a stream of tokens from raw to Python 3.
 
         It is called by the custom module loading machinery to rewrite
         source/tokens between source decoding and compilation.
 
         Returns a generator of possibly rewritten tokens.
 
         The input token list may be mutated as part of processing. However,
         its changes do not necessarily match the output token stream.
 
         REMEMBER TO CHANGE ``BYTECODEHEADER`` WHEN CHANGING THIS FUNCTION
         OR CACHED FILES WON'T GET INVALIDATED PROPERLY.
         """
         futureimpline = False
 
         # The following utility functions access the tokens list and i index of
         # the for i, t enumerate(tokens) loop below
         def _isop(j, *o):
             """Assert that tokens[j] is an OP with one of the given values"""
             try:
                 return tokens[j].type == token.OP and tokens[j].string in o
             except IndexError:
                 return False
 
         def _findargnofcall(n):
             """Find arg n of a call expression (start at 0)
 
             Returns index of the first token of that argument, or None if
             there is not that many arguments.
 
             Assumes that token[i + 1] is '('.
 
             """
             nested = 0
             for j in range(i + 2, len(tokens)):
                 if _isop(j, ')', ']', '}'):
                     # end of call, tuple, subscription or dict / set
                     nested -= 1
                     if nested < 0:
                         return None
                 elif n == 0:
                     # this is the starting position of arg
                     return j
                 elif _isop(j, '(', '[', '{'):
                     nested += 1
                 elif _isop(j, ',') and nested == 0:
                     n -= 1
 
             return None
 
         def _ensureunicode(j):
             """Make sure the token at j is a unicode string
 
             This rewrites a string token to include the unicode literal prefix
             so the string transformer won't add the byte prefix.
 
             Ignores tokens that are not strings. Assumes bounds checking has
             already been done.
 
             """
             st = tokens[j]
             if st.type == token.STRING and st.string.startswith(("'", '"')):
                 tokens[j] = st._replace(string='u%s' % st.string)
 
         for i, t in enumerate(tokens):
             # Convert most string literals to byte literals. String literals
             # in Python 2 are bytes. String literals in Python 3 are unicode.
             # Most strings in Mercurial are bytes and unicode strings are rare.
             # Rather than rewrite all string literals to use ``b''`` to indicate
             # byte strings, we apply this token transformer to insert the ``b``
             # prefix nearly everywhere.
             if t.type == token.STRING:
                 s = t.string
 
                 # Preserve docstrings as string literals. This is inconsistent
                 # with regular unprefixed strings. However, the
                 # "from __future__" parsing (which allows a module docstring to
                 # exist before it) doesn't properly handle the docstring if it
                 # is b''' prefixed, leading to a SyntaxError. We leave all
                 # docstrings as unprefixed to avoid this. This means Mercurial
                 # components touching docstrings need to handle unicode,
                 # unfortunately.
                 if s[0:3] in ("'''", '"""'):
                     yield t
                     continue
 
                 # If the first character isn't a quote, it is likely a string
                 # prefixing character (such as 'b', 'u', or 'r'. Ignore.
                 if s[0] not in ("'", '"'):
                     yield t
                     continue
 
                 # String literal. Prefix to make a b'' string.
                 yield t._replace(string='b%s' % t.string)
                 continue
 
             # Insert compatibility imports at "from __future__ import" line.
             # No '\n' should be added to preserve line numbers.
             if (t.type == token.NAME and t.string == 'import' and
                 all(u.type == token.NAME for u in tokens[i - 2:i]) and
                 [u.string for u in tokens[i - 2:i]] == ['from', '__future__']):
                 futureimpline = True
             if t.type == token.NEWLINE and futureimpline:
                 futureimpline = False
                 if fullname == 'mercurial.pycompat':
                     yield t
                     continue
                 r, c = t.start
                 l = (b'; from mercurial.pycompat import '
                      b'delattr, getattr, hasattr, setattr, xrange, '
                      b'open, unicode\n')
                 for u in tokenize.tokenize(io.BytesIO(l).readline):
                     if u.type in (tokenize.ENCODING, token.ENDMARKER):
                         continue
                     yield u._replace(
                         start=(r, c + u.start[1]), end=(r, c + u.end[1]))
                 continue
 
             # This looks like a function call.
             if t.type == token.NAME and _isop(i + 1, '('):
                 fn = t.string
 
                 # *attr() builtins don't accept byte strings to 2nd argument.
                 if (fn in ('getattr', 'setattr', 'hasattr', 'safehasattr') and
                         not _isop(i - 1, '.')):
                     arg1idx = _findargnofcall(1)
                     if arg1idx is not None:
                         _ensureunicode(arg1idx)
 
                 # .encode() and .decode() on str/bytes/unicode don't accept
                 # byte strings on Python 3.
                 elif fn in ('encode', 'decode') and _isop(i - 1, '.'):
                     for argn in range(2):
                         argidx = _findargnofcall(argn)
                         if argidx is not None:
                             _ensureunicode(argidx)
 
                 # It changes iteritems/values to items/values as they are not
                 # present in Python 3 world.
                 elif fn in ('iteritems', 'itervalues'):
                     yield t._replace(string=fn[4:])
                     continue
 
             # Emit unmodified token.
             yield t
 
     # Header to add to bytecode files. This MUST be changed when
     # ``replacetoken`` or any mechanism that changes semantics of module
     # loading is changed. Otherwise cached bytecode may get loaded without
     # the new transformation mechanisms applied.
     BYTECODEHEADER = b'HG\x00\x0a'
 
     class hgloader(importlib.machinery.SourceFileLoader):
         """Custom module loader that transforms source code.
 
         When the source code is converted to a code object, we transform
         certain patterns to be Python 3 compatible. This allows us to write code
         that is natively Python 2 and compatible with Python 3 without
         making the code excessively ugly.
 
         We do this by transforming the token stream between parse and compile.
 
         Implementing transformations invalidates caching assumptions made
         by the built-in importer. The built-in importer stores a header on
         saved bytecode files indicating the Python/bytecode version. If the
         version changes, the cached bytecode is ignored. The Mercurial
         transformations could change at any time. This means we need to check
         that cached bytecode was generated with the current transformation
         code or there could be a mismatch between cached bytecode and what
         would be generated from this class.
 
         We supplement the bytecode caching layer by wrapping ``get_data``
         and ``set_data``. These functions are called when the
         ``SourceFileLoader`` retrieves and saves bytecode cache files,
         respectively. We simply add an additional header on the file. As
         long as the version in this file is changed when semantics change,
         cached bytecode should be invalidated when transformations change.
 
         The added header has the form ``HG<VERSION>``. That is a literal
         ``HG`` with 2 binary bytes indicating the transformation version.
         """
         def get_data(self, path):
             data = super(hgloader, self).get_data(path)
 
             if not path.endswith(tuple(importlib.machinery.BYTECODE_SUFFIXES)):
                 return data
 
             # There should be a header indicating the Mercurial transformation
             # version. If it doesn't exist or doesn't match the current version,
             # we raise an OSError because that is what
             # ``SourceFileLoader.get_code()`` expects when loading bytecode
             # paths to indicate the cached file is "bad."
             if data[0:2] != b'HG':
                 raise OSError('no hg header')
             if data[0:4] != BYTECODEHEADER:
                 raise OSError('hg header version mismatch')
 
             return data[4:]
 
         def set_data(self, path, data, *args, **kwargs):
             if path.endswith(tuple(importlib.machinery.BYTECODE_SUFFIXES)):
                 data = BYTECODEHEADER + data
 
             return super(hgloader, self).set_data(path, data, *args, **kwargs)
 
         def source_to_code(self, data, path):
             """Perform token transformation before compilation."""
             buf = io.BytesIO(data)
             tokens = tokenize.tokenize(buf.readline)
             data = tokenize.untokenize(replacetokens(list(tokens), self.name))
             # Python's built-in importer strips frames from exceptions raised
             # for this code. Unfortunately, that mechanism isn't extensible
             # and our frame will be blamed for the import failure. There
             # are extremely hacky ways to do frame stripping. We haven't
             # implemented them because they are very ugly.
             return super(hgloader, self).source_to_code(data, path)
 
 # We automagically register our custom importer as a side-effect of loading.
 # This is necessary to ensure that any entry points are able to import
 # mercurial.* modules without having to perform this registration themselves.
 if sys.version_info[0] >= 3:
     _importercls = hgpathentryfinder
 else:
     _importercls = hgimporter
 if not any(isinstance(x, _importercls) for x in sys.meta_path):
     # meta_path is used before any implicit finders and before sys.path.
     sys.meta_path.insert(0, _importercls())

mercurial/debugcommands.py

@@ -1,2204 +1,2206 b''
 # debugcommands.py - command processing for debug* commands
 #
 # Copyright 2005-2016 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import difflib
 import errno
 import operator
 import os
 import random
 import socket
 import string
 import sys
 import tempfile
 import time
 
 from .i18n import _
 from .node import (
     bin,
     hex,
     nullhex,
     nullid,
     nullrev,
     short,
 )
 from . import (
     bundle2,
     changegroup,
     cmdutil,
     color,
     commands,
     context,
     dagparser,
     dagutil,
     encoding,
     error,
     exchange,
     extensions,
     filemerge,
     fileset,
     formatter,
     hg,
     localrepo,
     lock as lockmod,
     merge as mergemod,
     obsolete,
     policy,
     pvec,
     pycompat,
     registrar,
     repair,
     revlog,
     revset,
     revsetlang,
     scmutil,
     setdiscovery,
     simplemerge,
     smartset,
     sslutil,
     streamclone,
     templater,
     treediscovery,
     upgrade,
     util,
     vfs as vfsmod,
 )
 
 release = lockmod.release
 
 # We reuse the command table from commands because it is easier than
 # teaching dispatch about multiple tables.
 command = registrar.command(commands.table)
 
 @command('debugancestor', [], _('[INDEX] REV1 REV2'), optionalrepo=True)
 def debugancestor(ui, repo, *args):
     """find the ancestor revision of two revisions in a given index"""
     if len(args) == 3:
         index, rev1, rev2 = args
         r = revlog.revlog(vfsmod.vfs(pycompat.getcwd(), audit=False), index)
         lookup = r.lookup
     elif len(args) == 2:
         if not repo:
             raise error.Abort(_('there is no Mercurial repository here '
                                 '(.hg not found)'))
         rev1, rev2 = args
         r = repo.changelog
         lookup = repo.lookup
     else:
         raise error.Abort(_('either two or three arguments required'))
     a = r.ancestor(lookup(rev1), lookup(rev2))
     ui.write('%d:%s\n' % (r.rev(a), hex(a)))
 
 @command('debugapplystreamclonebundle', [], 'FILE')
 def debugapplystreamclonebundle(ui, repo, fname):
     """apply a stream clone bundle file"""
     f = hg.openpath(ui, fname)
     gen = exchange.readbundle(ui, f, fname)
     gen.apply(repo)
 
 @command('debugbuilddag',
     [('m', 'mergeable-file', None, _('add single file mergeable changes')),
     ('o', 'overwritten-file', None, _('add single file all revs overwrite')),
     ('n', 'new-file', None, _('add new file at each rev'))],
     _('[OPTION]... [TEXT]'))
 def debugbuilddag(ui, repo, text=None,
                   mergeable_file=False,
                   overwritten_file=False,
                   new_file=False):
     """builds a repo with a given DAG from scratch in the current empty repo
 
     The description of the DAG is read from stdin if not given on the
     command line.
 
     Elements:
 
      - "+n" is a linear run of n nodes based on the current default parent
      - "." is a single node based on the current default parent
      - "$" resets the default parent to null (implied at the start);
            otherwise the default parent is always the last node created
      - "<p" sets the default parent to the backref p
      - "*p" is a fork at parent p, which is a backref
      - "*p1/p2" is a merge of parents p1 and p2, which are backrefs
      - "/p2" is a merge of the preceding node and p2
      - ":tag" defines a local tag for the preceding node
      - "@branch" sets the named branch for subsequent nodes
      - "#...\\n" is a comment up to the end of the line
 
     Whitespace between the above elements is ignored.
 
     A backref is either
 
      - a number n, which references the node curr-n, where curr is the current
        node, or
      - the name of a local tag you placed earlier using ":tag", or
      - empty to denote the default parent.
 
     All string valued-elements are either strictly alphanumeric, or must
     be enclosed in double quotes ("..."), with "\\" as escape character.
     """
 
     if text is None:
         ui.status(_("reading DAG from stdin\n"))
         text = ui.fin.read()
 
     cl = repo.changelog
     if len(cl) > 0:
         raise error.Abort(_('repository is not empty'))
 
     # determine number of revs in DAG
     total = 0
     for type, data in dagparser.parsedag(text):
         if type == 'n':
             total += 1
 
     if mergeable_file:
         linesperrev = 2
         # make a file with k lines per rev
         initialmergedlines = [str(i) for i in xrange(0, total * linesperrev)]
         initialmergedlines.append("")
 
     tags = []
 
     wlock = lock = tr = None
     try:
         wlock = repo.wlock()
         lock = repo.lock()
         tr = repo.transaction("builddag")
 
         at = -1
         atbranch = 'default'
         nodeids = []
         id = 0
         ui.progress(_('building'), id, unit=_('revisions'), total=total)
         for type, data in dagparser.parsedag(text):
             if type == 'n':
                 ui.note(('node %s\n' % str(data)))
                 id, ps = data
 
                 files = []
                 fctxs = {}
 
                 p2 = None
                 if mergeable_file:
                     fn = "mf"
                     p1 = repo[ps[0]]
                     if len(ps) > 1:
                         p2 = repo[ps[1]]
                         pa = p1.ancestor(p2)
                         base, local, other = [x[fn].data() for x in (pa, p1,
                                                                      p2)]
                         m3 = simplemerge.Merge3Text(base, local, other)
                         ml = [l.strip() for l in m3.merge_lines()]
                         ml.append("")
                     elif at > 0:
                         ml = p1[fn].data().split("\n")
                     else:
                         ml = initialmergedlines
                     ml[id * linesperrev] += " r%i" % id
                     mergedtext = "\n".join(ml)
                     files.append(fn)
                     fctxs[fn] = context.memfilectx(repo, fn, mergedtext)
 
                 if overwritten_file:
                     fn = "of"
                     files.append(fn)
                     fctxs[fn] = context.memfilectx(repo, fn, "r%i\n" % id)
 
                 if new_file:
                     fn = "nf%i" % id
                     files.append(fn)
                     fctxs[fn] = context.memfilectx(repo, fn, "r%i\n" % id)
                     if len(ps) > 1:
                         if not p2:
                             p2 = repo[ps[1]]
                         for fn in p2:
                             if fn.startswith("nf"):
                                 files.append(fn)
                                 fctxs[fn] = p2[fn]
 
                 def fctxfn(repo, cx, path):
                     return fctxs.get(path)
 
                 if len(ps) == 0 or ps[0] < 0:
                     pars = [None, None]
                 elif len(ps) == 1:
                     pars = [nodeids[ps[0]], None]
                 else:
                     pars = [nodeids[p] for p in ps]
                 cx = context.memctx(repo, pars, "r%i" % id, files, fctxfn,
                                     date=(id, 0),
                                     user="debugbuilddag",
                                     extra={'branch': atbranch})
                 nodeid = repo.commitctx(cx)
                 nodeids.append(nodeid)
                 at = id
             elif type == 'l':
                 id, name = data
                 ui.note(('tag %s\n' % name))
                 tags.append("%s %s\n" % (hex(repo.changelog.node(id)), name))
             elif type == 'a':
                 ui.note(('branch %s\n' % data))
                 atbranch = data
             ui.progress(_('building'), id, unit=_('revisions'), total=total)
         tr.close()
 
         if tags:
             repo.vfs.write("localtags", "".join(tags))
     finally:
         ui.progress(_('building'), None)
         release(tr, lock, wlock)
 
 def _debugchangegroup(ui, gen, all=None, indent=0, **opts):
     indent_string = ' ' * indent
     if all:
         ui.write(("%sformat: id, p1, p2, cset, delta base, len(delta)\n")
                  % indent_string)
 
         def showchunks(named):
             ui.write("\n%s%s\n" % (indent_string, named))
             chain = None
             for chunkdata in iter(lambda: gen.deltachunk(chain), {}):
                 node = chunkdata['node']
                 p1 = chunkdata['p1']
                 p2 = chunkdata['p2']
                 cs = chunkdata['cs']
                 deltabase = chunkdata['deltabase']
                 delta = chunkdata['delta']
                 ui.write("%s%s %s %s %s %s %s\n" %
                          (indent_string, hex(node), hex(p1), hex(p2),
                           hex(cs), hex(deltabase), len(delta)))
                 chain = node
 
         chunkdata = gen.changelogheader()
         showchunks("changelog")
         chunkdata = gen.manifestheader()
         showchunks("manifest")
         for chunkdata in iter(gen.filelogheader, {}):
             fname = chunkdata['filename']
             showchunks(fname)
     else:
         if isinstance(gen, bundle2.unbundle20):
             raise error.Abort(_('use debugbundle2 for this file'))
         chunkdata = gen.changelogheader()
         chain = None
         for chunkdata in iter(lambda: gen.deltachunk(chain), {}):
             node = chunkdata['node']
             ui.write("%s%s\n" % (indent_string, hex(node)))
             chain = node
 
 def _debugbundle2(ui, gen, all=None, **opts):
     """lists the contents of a bundle2"""
     if not isinstance(gen, bundle2.unbundle20):
         raise error.Abort(_('not a bundle2 file'))
     ui.write(('Stream params: %s\n' % repr(gen.params)))
     for part in gen.iterparts():
         ui.write('%s -- %r\n' % (part.type, repr(part.params)))
         if part.type == 'changegroup':
             version = part.params.get('version', '01')
             cg = changegroup.getunbundler(version, part, 'UN')
             _debugchangegroup(ui, cg, all=all, indent=4, **opts)
 
 @command('debugbundle',
         [('a', 'all', None, _('show all details')),
          ('', 'spec', None, _('print the bundlespec of the bundle'))],
         _('FILE'),
         norepo=True)
 def debugbundle(ui, bundlepath, all=None, spec=None, **opts):
     """lists the contents of a bundle"""
     with hg.openpath(ui, bundlepath) as f:
         if spec:
             spec = exchange.getbundlespec(ui, f)
             ui.write('%s\n' % spec)
             return
 
         gen = exchange.readbundle(ui, f, bundlepath)
         if isinstance(gen, bundle2.unbundle20):
             return _debugbundle2(ui, gen, all=all, **opts)
         _debugchangegroup(ui, gen, all=all, **opts)
 
 @command('debugcheckstate', [], '')
 def debugcheckstate(ui, repo):
     """validate the correctness of the current dirstate"""
     parent1, parent2 = repo.dirstate.parents()
     m1 = repo[parent1].manifest()
     m2 = repo[parent2].manifest()
     errors = 0
     for f in repo.dirstate:
         state = repo.dirstate[f]
         if state in "nr" and f not in m1:
             ui.warn(_("%s in state %s, but not in manifest1\n") % (f, state))
             errors += 1
         if state in "a" and f in m1:
             ui.warn(_("%s in state %s, but also in manifest1\n") % (f, state))
             errors += 1
         if state in "m" and f not in m1 and f not in m2:
             ui.warn(_("%s in state %s, but not in either manifest\n") %
                     (f, state))
             errors += 1
     for f in m1:
         state = repo.dirstate[f]
         if state not in "nrm":
             ui.warn(_("%s in manifest1, but listed as state %s") % (f, state))
             errors += 1
     if errors:
         error = _(".hg/dirstate inconsistent with current parent's manifest")
         raise error.Abort(error)
 
 @command('debugcolor',
         [('', 'style', None, _('show all configured styles'))],
         'hg debugcolor')
 def debugcolor(ui, repo, **opts):
     """show available color, effects or style"""
     ui.write(('color mode: %s\n') % ui._colormode)
     if opts.get('style'):
         return _debugdisplaystyle(ui)
     else:
         return _debugdisplaycolor(ui)
 
 def _debugdisplaycolor(ui):
     ui = ui.copy()
     ui._styles.clear()
     for effect in color._activeeffects(ui).keys():
         ui._styles[effect] = effect
     if ui._terminfoparams:
         for k, v in ui.configitems('color'):
             if k.startswith('color.'):
                 ui._styles[k] = k[6:]
             elif k.startswith('terminfo.'):
                 ui._styles[k] = k[9:]
     ui.write(_('available colors:\n'))
     # sort label with a '_' after the other to group '_background' entry.
     items = sorted(ui._styles.items(),
                    key=lambda i: ('_' in i[0], i[0], i[1]))
     for colorname, label in items:
         ui.write(('%s\n') % colorname, label=label)
 
 def _debugdisplaystyle(ui):
     ui.write(_('available style:\n'))
     width = max(len(s) for s in ui._styles)
     for label, effects in sorted(ui._styles.items()):
         ui.write('%s' % label, label=label)
         if effects:
             # 50
             ui.write(': ')
             ui.write(' ' * (max(0, width - len(label))))
             ui.write(', '.join(ui.label(e, e) for e in effects.split()))
         ui.write('\n')
 
 @command('debugcommands', [], _('[COMMAND]'), norepo=True)
 def debugcommands(ui, cmd='', *args):
     """list all available commands and options"""
     for cmd, vals in sorted(commands.table.iteritems()):
         cmd = cmd.split('|')[0].strip('^')
         opts = ', '.join([i[1] for i in vals[1]])
         ui.write('%s: %s\n' % (cmd, opts))
 
 @command('debugcomplete',
     [('o', 'options', None, _('show the command options'))],
     _('[-o] CMD'),
     norepo=True)
 def debugcomplete(ui, cmd='', **opts):
     """returns the completion list associated with the given command"""
 
     if opts.get('options'):
         options = []
         otables = [commands.globalopts]
         if cmd:
             aliases, entry = cmdutil.findcmd(cmd, commands.table, False)
             otables.append(entry[1])
         for t in otables:
             for o in t:
                 if "(DEPRECATED)" in o[3]:
                     continue
                 if o[0]:
                     options.append('-%s' % o[0])
                 options.append('--%s' % o[1])
         ui.write("%s\n" % "\n".join(options))
         return
 
     cmdlist, unused_allcmds = cmdutil.findpossible(cmd, commands.table)
     if ui.verbose:
         cmdlist = [' '.join(c[0]) for c in cmdlist.values()]
     ui.write("%s\n" % "\n".join(sorted(cmdlist)))
 
 @command('debugcreatestreamclonebundle', [], 'FILE')
 def debugcreatestreamclonebundle(ui, repo, fname):
     """create a stream clone bundle file
 
     Stream bundles are special bundles that are essentially archives of
     revlog files. They are commonly used for cloning very quickly.
     """
     requirements, gen = streamclone.generatebundlev1(repo)
     changegroup.writechunks(ui, gen, fname)
 
     ui.write(_('bundle requirements: %s\n') % ', '.join(sorted(requirements)))
 
 @command('debugdag',
     [('t', 'tags', None, _('use tags as labels')),
     ('b', 'branches', None, _('annotate with branch names')),
     ('', 'dots', None, _('use dots for runs')),
     ('s', 'spaces', None, _('separate elements by spaces'))],
     _('[OPTION]... [FILE [REV]...]'),
     optionalrepo=True)
 def debugdag(ui, repo, file_=None, *revs, **opts):
     """format the changelog or an index DAG as a concise textual description
 
     If you pass a revlog index, the revlog's DAG is emitted. If you list
     revision numbers, they get labeled in the output as rN.
 
     Otherwise, the changelog DAG of the current repo is emitted.
     """
     spaces = opts.get('spaces')
     dots = opts.get('dots')
     if file_:
         rlog = revlog.revlog(vfsmod.vfs(pycompat.getcwd(), audit=False),
                              file_)
         revs = set((int(r) for r in revs))
         def events():
             for r in rlog:
                 yield 'n', (r, list(p for p in rlog.parentrevs(r)
                                         if p != -1))
                 if r in revs:
                     yield 'l', (r, "r%i" % r)
     elif repo:
         cl = repo.changelog
         tags = opts.get('tags')
         branches = opts.get('branches')
         if tags:
             labels = {}
             for l, n in repo.tags().items():
                 labels.setdefault(cl.rev(n), []).append(l)
         def events():
             b = "default"
             for r in cl:
                 if branches:
                     newb = cl.read(cl.node(r))[5]['branch']
                     if newb != b:
                         yield 'a', newb
                         b = newb
                 yield 'n', (r, list(p for p in cl.parentrevs(r)
                                         if p != -1))
                 if tags:
                     ls = labels.get(r)
                     if ls:
                         for l in ls:
                             yield 'l', (r, l)
     else:
         raise error.Abort(_('need repo for changelog dag'))
 
     for line in dagparser.dagtextlines(events(),
                                        addspaces=spaces,
                                        wraplabels=True,
                                        wrapannotations=True,
                                        wrapnonlinear=dots,
                                        usedots=dots,
                                        maxlinewidth=70):
         ui.write(line)
         ui.write("\n")
 
 @command('debugdata', commands.debugrevlogopts, _('-c|-m|FILE REV'))
 def debugdata(ui, repo, file_, rev=None, **opts):
     """dump the contents of a data file revision"""
     if opts.get('changelog') or opts.get('manifest') or opts.get('dir'):
         if rev is not None:
             raise error.CommandError('debugdata', _('invalid arguments'))
         file_, rev = None, file_
     elif rev is None:
         raise error.CommandError('debugdata', _('invalid arguments'))
     r = cmdutil.openrevlog(repo, 'debugdata', file_, opts)
     try:
         ui.write(r.revision(r.lookup(rev), raw=True))
     except KeyError:
         raise error.Abort(_('invalid revision identifier %s') % rev)
 
 @command('debugdate',
     [('e', 'extended', None, _('try extended date formats'))],
     _('[-e] DATE [RANGE]'),
     norepo=True, optionalrepo=True)
 def debugdate(ui, date, range=None, **opts):
     """parse and display a date"""
     if opts["extended"]:
         d = util.parsedate(date, util.extendeddateformats)
     else:
         d = util.parsedate(date)
     ui.write(("internal: %s %s\n") % d)
     ui.write(("standard: %s\n") % util.datestr(d))
     if range:
         m = util.matchdate(range)
         ui.write(("match: %s\n") % m(d[0]))
 
 @command('debugdeltachain',
     commands.debugrevlogopts + commands.formatteropts,
     _('-c|-m|FILE'),
     optionalrepo=True)
 def debugdeltachain(ui, repo, file_=None, **opts):
     """dump information about delta chains in a revlog
 
     Output can be templatized. Available template keywords are:
 
     :``rev``:       revision number
     :``chainid``:   delta chain identifier (numbered by unique base)
     :``chainlen``:  delta chain length to this revision
     :``prevrev``:   previous revision in delta chain
     :``deltatype``: role of delta / how it was computed
     :``compsize``:  compressed size of revision
     :``uncompsize``: uncompressed size of revision
     :``chainsize``: total size of compressed revisions in chain
     :``chainratio``: total chain size divided by uncompressed revision size
                     (new delta chains typically start at ratio 2.00)
     :``lindist``:   linear distance from base revision in delta chain to end
                     of this revision
     :``extradist``: total size of revisions not part of this delta chain from
                     base of delta chain to end of this revision; a measurement
                     of how much extra data we need to read/seek across to read
                     the delta chain for this revision
     :``extraratio``: extradist divided by chainsize; another representation of
                     how much unrelated data is needed to load this delta chain
     """
     r = cmdutil.openrevlog(repo, 'debugdeltachain', file_, opts)
     index = r.index
     generaldelta = r.version & revlog.FLAG_GENERALDELTA
 
     def revinfo(rev):
         e = index[rev]
         compsize = e[1]
         uncompsize = e[2]
         chainsize = 0
 
         if generaldelta:
             if e[3] == e[5]:
                 deltatype = 'p1'
             elif e[3] == e[6]:
                 deltatype = 'p2'
             elif e[3] == rev - 1:
                 deltatype = 'prev'
             elif e[3] == rev:
                 deltatype = 'base'
             else:
                 deltatype = 'other'
         else:
             if e[3] == rev:
                 deltatype = 'base'
             else:
                 deltatype = 'prev'
 
         chain = r._deltachain(rev)[0]
         for iterrev in chain:
             e = index[iterrev]
             chainsize += e[1]
 
         return compsize, uncompsize, deltatype, chain, chainsize
 
     fm = ui.formatter('debugdeltachain', opts)
 
     fm.plain('    rev  chain# chainlen     prev   delta       '
              'size    rawsize  chainsize     ratio   lindist extradist '
              'extraratio\n')
 
     chainbases = {}
     for rev in r:
         comp, uncomp, deltatype, chain, chainsize = revinfo(rev)
         chainbase = chain[0]
         chainid = chainbases.setdefault(chainbase, len(chainbases) + 1)
         basestart = r.start(chainbase)
         revstart = r.start(rev)
         lineardist = revstart + comp - basestart
         extradist = lineardist - chainsize
         try:
             prevrev = chain[-2]
         except IndexError:
             prevrev = -1
 
         chainratio = float(chainsize) / float(uncomp)
         extraratio = float(extradist) / float(chainsize)
 
         fm.startitem()
         fm.write('rev chainid chainlen prevrev deltatype compsize '
                  'uncompsize chainsize chainratio lindist extradist '
                  'extraratio',
                  '%7d %7d %8d %8d %7s %10d %10d %10d %9.5f %9d %9d %10.5f\n',
                  rev, chainid, len(chain), prevrev, deltatype, comp,
                  uncomp, chainsize, chainratio, lineardist, extradist,
                  extraratio,
                  rev=rev, chainid=chainid, chainlen=len(chain),
                  prevrev=prevrev, deltatype=deltatype, compsize=comp,
                  uncompsize=uncomp, chainsize=chainsize,
                  chainratio=chainratio, lindist=lineardist,
                  extradist=extradist, extraratio=extraratio)
 
     fm.end()
 
 @command('debugdirstate|debugstate',
     [('', 'nodates', None, _('do not display the saved mtime')),
     ('', 'datesort', None, _('sort by saved mtime'))],
     _('[OPTION]...'))
 def debugstate(ui, repo, **opts):
     """show the contents of the current dirstate"""
 
     nodates = opts.get('nodates')
     datesort = opts.get('datesort')
 
     timestr = ""
     if datesort:
         keyfunc = lambda x: (x[1][3], x[0]) # sort by mtime, then by filename
     else:
         keyfunc = None # sort by filename
     for file_, ent in sorted(repo.dirstate._map.iteritems(), key=keyfunc):
         if ent[3] == -1:
             timestr = 'unset               '
         elif nodates:
             timestr = 'set                 '
         else:
             timestr = time.strftime("%Y-%m-%d %H:%M:%S ",
                                     time.localtime(ent[3]))
         if ent[1] & 0o20000:
             mode = 'lnk'
         else:
             mode = '%3o' % (ent[1] & 0o777 & ~util.umask)
         ui.write("%c %s %10d %s%s\n" % (ent[0], mode, ent[2], timestr, file_))
     for f in repo.dirstate.copies():
         ui.write(_("copy: %s -> %s\n") % (repo.dirstate.copied(f), f))
 
 @command('debugdiscovery',
     [('', 'old', None, _('use old-style discovery')),
     ('', 'nonheads', None,
      _('use old-style discovery with non-heads included')),
     ] + commands.remoteopts,
     _('[-l REV] [-r REV] [-b BRANCH]... [OTHER]'))
 def debugdiscovery(ui, repo, remoteurl="default", **opts):
     """runs the changeset discovery protocol in isolation"""
     remoteurl, branches = hg.parseurl(ui.expandpath(remoteurl),
                                       opts.get('branch'))
     remote = hg.peer(repo, opts, remoteurl)
     ui.status(_('comparing with %s\n') % util.hidepassword(remoteurl))
 
     # make sure tests are repeatable
     random.seed(12323)
 
     def doit(localheads, remoteheads, remote=remote):
         if opts.get('old'):
             if localheads:
                 raise error.Abort('cannot use localheads with old style '
                                  'discovery')
             if not util.safehasattr(remote, 'branches'):
                 # enable in-client legacy support
                 remote = localrepo.locallegacypeer(remote.local())
             common, _in, hds = treediscovery.findcommonincoming(repo, remote,
                                                                 force=True)
             common = set(common)
             if not opts.get('nonheads'):
                 ui.write(("unpruned common: %s\n") %
                          " ".join(sorted(short(n) for n in common)))
                 dag = dagutil.revlogdag(repo.changelog)
                 all = dag.ancestorset(dag.internalizeall(common))
                 common = dag.externalizeall(dag.headsetofconnecteds(all))
         else:
             common, any, hds = setdiscovery.findcommonheads(ui, repo, remote)
         common = set(common)
         rheads = set(hds)
         lheads = set(repo.heads())
         ui.write(("common heads: %s\n") %
                  " ".join(sorted(short(n) for n in common)))
         if lheads <= common:
             ui.write(("local is subset\n"))
         elif rheads <= common:
             ui.write(("remote is subset\n"))
 
     serverlogs = opts.get('serverlog')
     if serverlogs:
         for filename in serverlogs:
             with open(filename, 'r') as logfile:
                 line = logfile.readline()
                 while line:
                     parts = line.strip().split(';')
                     op = parts[1]
                     if op == 'cg':
                         pass
                     elif op == 'cgss':
                         doit(parts[2].split(' '), parts[3].split(' '))
                     elif op == 'unb':
                         doit(parts[3].split(' '), parts[2].split(' '))
                     line = logfile.readline()
     else:
         remoterevs, _checkout = hg.addbranchrevs(repo, remote, branches,
                                                  opts.get('remote_head'))
         localrevs = opts.get('local_head')
         doit(localrevs, remoterevs)
 
 @command('debugextensions', commands.formatteropts, [], norepo=True)
 def debugextensions(ui, **opts):
     '''show information about active extensions'''
     exts = extensions.extensions(ui)
     hgver = util.version()
     fm = ui.formatter('debugextensions', opts)
     for extname, extmod in sorted(exts, key=operator.itemgetter(0)):
         isinternal = extensions.ismoduleinternal(extmod)
         extsource = pycompat.fsencode(extmod.__file__)
         if isinternal:
             exttestedwith = []  # never expose magic string to users
         else:
             exttestedwith = getattr(extmod, 'testedwith', '').split()
         extbuglink = getattr(extmod, 'buglink', None)
 
         fm.startitem()
 
         if ui.quiet or ui.verbose:
             fm.write('name', '%s\n', extname)
         else:
             fm.write('name', '%s', extname)
             if isinternal or hgver in exttestedwith:
                 fm.plain('\n')
             elif not exttestedwith:
                 fm.plain(_(' (untested!)\n'))
             else:
                 lasttestedversion = exttestedwith[-1]
                 fm.plain(' (%s!)\n' % lasttestedversion)
 
         fm.condwrite(ui.verbose and extsource, 'source',
                  _('  location: %s\n'), extsource or "")
 
         if ui.verbose:
             fm.plain(_('  bundled: %s\n') % ['no', 'yes'][isinternal])
         fm.data(bundled=isinternal)
 
         fm.condwrite(ui.verbose and exttestedwith, 'testedwith',
                      _('  tested with: %s\n'),
                      fm.formatlist(exttestedwith, name='ver'))
 
         fm.condwrite(ui.verbose and extbuglink, 'buglink',
                  _('  bug reporting: %s\n'), extbuglink or "")
 
     fm.end()
 
 @command('debugfileset',
     [('r', 'rev', '', _('apply the filespec on this revision'), _('REV'))],
     _('[-r REV] FILESPEC'))
 def debugfileset(ui, repo, expr, **opts):
     '''parse and apply a fileset specification'''
     ctx = scmutil.revsingle(repo, opts.get('rev'), None)
     if ui.verbose:
         tree = fileset.parse(expr)
         ui.note(fileset.prettyformat(tree), "\n")
 
     for f in ctx.getfileset(expr):
         ui.write("%s\n" % f)
 
 @command('debugfsinfo', [], _('[PATH]'), norepo=True)
 def debugfsinfo(ui, path="."):
     """show information detected about current filesystem"""
     ui.write(('exec: %s\n') % (util.checkexec(path) and 'yes' or 'no'))
     ui.write(('fstype: %s\n') % (util.getfstype(path) or '(unknown)'))
     ui.write(('symlink: %s\n') % (util.checklink(path) and 'yes' or 'no'))
     ui.write(('hardlink: %s\n') % (util.checknlink(path) and 'yes' or 'no'))
     casesensitive = '(unknown)'
     try:
         with tempfile.NamedTemporaryFile(prefix='.debugfsinfo', dir=path) as f:
             casesensitive = util.fscasesensitive(f.name) and 'yes' or 'no'
     except OSError:
         pass
     ui.write(('case-sensitive: %s\n') % casesensitive)
 
 @command('debuggetbundle',
     [('H', 'head', [], _('id of head node'), _('ID')),
     ('C', 'common', [], _('id of common node'), _('ID')),
     ('t', 'type', 'bzip2', _('bundle compression type to use'), _('TYPE'))],
     _('REPO FILE [-H|-C ID]...'),
     norepo=True)
 def debuggetbundle(ui, repopath, bundlepath, head=None, common=None, **opts):
     """retrieves a bundle from a repo
 
     Every ID must be a full-length hex node id string. Saves the bundle to the
     given file.
     """
     repo = hg.peer(ui, opts, repopath)
     if not repo.capable('getbundle'):
         raise error.Abort("getbundle() not supported by target repository")
     args = {}
     if common:
         args['common'] = [bin(s) for s in common]
     if head:
         args['heads'] = [bin(s) for s in head]
     # TODO: get desired bundlecaps from command line.
     args['bundlecaps'] = None
     bundle = repo.getbundle('debug', **args)
 
     bundletype = opts.get('type', 'bzip2').lower()
     btypes = {'none': 'HG10UN',
               'bzip2': 'HG10BZ',
               'gzip': 'HG10GZ',
               'bundle2': 'HG20'}
     bundletype = btypes.get(bundletype)
     if bundletype not in bundle2.bundletypes:
         raise error.Abort(_('unknown bundle type specified with --type'))
     bundle2.writebundle(ui, bundle, bundlepath, bundletype)
 
 @command('debugignore', [], '[FILE]')
 def debugignore(ui, repo, *files, **opts):
     """display the combined ignore pattern and information about ignored files
 
     With no argument display the combined ignore pattern.
 
     Given space separated file names, shows if the given file is ignored and
     if so, show the ignore rule (file and line number) that matched it.
     """
     ignore = repo.dirstate._ignore
     if not files:
         # Show all the patterns
         includepat = getattr(ignore, 'includepat', None)
         if includepat is not None:
             ui.write("%s\n" % includepat)
         else:
             raise error.Abort(_("no ignore patterns found"))
     else:
         for f in files:
             nf = util.normpath(f)
             ignored = None
             ignoredata = None
             if nf != '.':
                 if ignore(nf):
                     ignored = nf
                     ignoredata = repo.dirstate._ignorefileandline(nf)
                 else:
                     for p in util.finddirs(nf):
                         if ignore(p):
                             ignored = p
                             ignoredata = repo.dirstate._ignorefileandline(p)
                             break
             if ignored:
                 if ignored == nf:
                     ui.write(_("%s is ignored\n") % f)
                 else:
                     ui.write(_("%s is ignored because of "
                                "containing folder %s\n")
                              % (f, ignored))
                 ignorefile, lineno, line = ignoredata
                 ui.write(_("(ignore rule in %s, line %d: '%s')\n")
                          % (ignorefile, lineno, line))
             else:
                 ui.write(_("%s is not ignored\n") % f)
 
 @command('debugindex', commands.debugrevlogopts +
     [('f', 'format', 0, _('revlog format'), _('FORMAT'))],
     _('[-f FORMAT] -c|-m|FILE'),
     optionalrepo=True)
 def debugindex(ui, repo, file_=None, **opts):
     """dump the contents of an index file"""
     r = cmdutil.openrevlog(repo, 'debugindex', file_, opts)
     format = opts.get('format', 0)
     if format not in (0, 1):
         raise error.Abort(_("unknown format %d") % format)
 
     generaldelta = r.version & revlog.FLAG_GENERALDELTA
     if generaldelta:
         basehdr = ' delta'
     else:
         basehdr = '  base'
 
     if ui.debugflag:
         shortfn = hex
     else:
         shortfn = short
 
     # There might not be anything in r, so have a sane default
     idlen = 12
     for i in r:
         idlen = len(shortfn(r.node(i)))
         break
 
     if format == 0:
         ui.write(("   rev    offset  length " + basehdr + " linkrev"
                  " %s %s p2\n") % ("nodeid".ljust(idlen), "p1".ljust(idlen)))
     elif format == 1:
         ui.write(("   rev flag   offset   length"
                  "     size " + basehdr + "   link     p1     p2"
                  " %s\n") % "nodeid".rjust(idlen))
 
     for i in r:
         node = r.node(i)
         if generaldelta:
             base = r.deltaparent(i)
         else:
             base = r.chainbase(i)
         if format == 0:
             try:
                 pp = r.parents(node)
             except Exception:
                 pp = [nullid, nullid]
             ui.write("% 6d % 9d % 7d % 6d % 7d %s %s %s\n" % (
                     i, r.start(i), r.length(i), base, r.linkrev(i),
                     shortfn(node), shortfn(pp[0]), shortfn(pp[1])))
         elif format == 1:
             pr = r.parentrevs(i)
             ui.write("% 6d %04x % 8d % 8d % 8d % 6d % 6d % 6d % 6d %s\n" % (
                     i, r.flags(i), r.start(i), r.length(i), r.rawsize(i),
                     base, r.linkrev(i), pr[0], pr[1], shortfn(node)))
 
 @command('debugindexdot', commands.debugrevlogopts,
     _('-c|-m|FILE'), optionalrepo=True)
 def debugindexdot(ui, repo, file_=None, **opts):
     """dump an index DAG as a graphviz dot file"""
     r = cmdutil.openrevlog(repo, 'debugindexdot', file_, opts)
     ui.write(("digraph G {\n"))
     for i in r:
         node = r.node(i)
         pp = r.parents(node)
         ui.write("\t%d -> %d\n" % (r.rev(pp[0]), i))
         if pp[1] != nullid:
             ui.write("\t%d -> %d\n" % (r.rev(pp[1]), i))
     ui.write("}\n")
 
 @command('debuginstall', [] + commands.formatteropts, '', norepo=True)
 def debuginstall(ui, **opts):
     '''test Mercurial installation
 
     Returns 0 on success.
     '''
 
     def writetemp(contents):
         (fd, name) = tempfile.mkstemp(prefix="hg-debuginstall-")
         f = os.fdopen(fd, pycompat.sysstr("wb"))
         f.write(contents)
         f.close()
         return name
 
     problems = 0
 
     fm = ui.formatter('debuginstall', opts)
     fm.startitem()
 
     # encoding
     fm.write('encoding', _("checking encoding (%s)...\n"), encoding.encoding)
     err = None
     try:
         encoding.fromlocal("test")
     except error.Abort as inst:
         err = inst
         problems += 1
     fm.condwrite(err, 'encodingerror', _(" %s\n"
                  " (check that your locale is properly set)\n"), err)
 
     # Python
     fm.write('pythonexe', _("checking Python executable (%s)\n"),
              pycompat.sysexecutable)
     fm.write('pythonver', _("checking Python version (%s)\n"),
              ("%d.%d.%d" % sys.version_info[:3]))
     fm.write('pythonlib', _("checking Python lib (%s)...\n"),
              os.path.dirname(pycompat.fsencode(os.__file__)))
 
     security = set(sslutil.supportedprotocols)
     if sslutil.hassni:
         security.add('sni')
 
     fm.write('pythonsecurity', _("checking Python security support (%s)\n"),
              fm.formatlist(sorted(security), name='protocol',
                            fmt='%s', sep=','))
 
     # These are warnings, not errors. So don't increment problem count. This
     # may change in the future.
     if 'tls1.2' not in security:
         fm.plain(_('  TLS 1.2 not supported by Python install; '
                    'network connections lack modern security\n'))
     if 'sni' not in security:
         fm.plain(_('  SNI not supported by Python install; may have '
                    'connectivity issues with some servers\n'))
 
     # TODO print CA cert info
 
     # hg version
     hgver = util.version()
     fm.write('hgver', _("checking Mercurial version (%s)\n"),
              hgver.split('+')[0])
     fm.write('hgverextra', _("checking Mercurial custom build (%s)\n"),
              '+'.join(hgver.split('+')[1:]))
 
     # compiled modules
     fm.write('hgmodulepolicy', _("checking module policy (%s)\n"),
              policy.policy)
     fm.write('hgmodules', _("checking installed modules (%s)...\n"),
              os.path.dirname(pycompat.fsencode(__file__)))
 
     if policy.policy in ('c', 'allow'):
         err = None
         try:
             from . import (
                 base85,
                 bdiff,
                 mpatch,
+            )
+            from .cext import (
                 osutil,
             )
             dir(bdiff), dir(mpatch), dir(base85), dir(osutil) # quiet pyflakes
         except Exception as inst:
             err = inst
             problems += 1
         fm.condwrite(err, 'extensionserror', " %s\n", err)
 
     compengines = util.compengines._engines.values()
     fm.write('compengines', _('checking registered compression engines (%s)\n'),
              fm.formatlist(sorted(e.name() for e in compengines),
                            name='compengine', fmt='%s', sep=', '))
     fm.write('compenginesavail', _('checking available compression engines '
                                    '(%s)\n'),
              fm.formatlist(sorted(e.name() for e in compengines
                                   if e.available()),
                            name='compengine', fmt='%s', sep=', '))
     wirecompengines = util.compengines.supportedwireengines(util.SERVERROLE)
     fm.write('compenginesserver', _('checking available compression engines '
                                     'for wire protocol (%s)\n'),
              fm.formatlist([e.name() for e in wirecompengines
                             if e.wireprotosupport()],
                            name='compengine', fmt='%s', sep=', '))
 
     # templates
     p = templater.templatepaths()
     fm.write('templatedirs', 'checking templates (%s)...\n', ' '.join(p))
     fm.condwrite(not p, '', _(" no template directories found\n"))
     if p:
         m = templater.templatepath("map-cmdline.default")
         if m:
             # template found, check if it is working
             err = None
             try:
                 templater.templater.frommapfile(m)
             except Exception as inst:
                 err = inst
                 p = None
             fm.condwrite(err, 'defaulttemplateerror', " %s\n", err)
         else:
             p = None
         fm.condwrite(p, 'defaulttemplate',
                      _("checking default template (%s)\n"), m)
         fm.condwrite(not m, 'defaulttemplatenotfound',
                      _(" template '%s' not found\n"), "default")
     if not p:
         problems += 1
     fm.condwrite(not p, '',
                  _(" (templates seem to have been installed incorrectly)\n"))
 
     # editor
     editor = ui.geteditor()
     editor = util.expandpath(editor)
     fm.write('editor', _("checking commit editor... (%s)\n"), editor)
     cmdpath = util.findexe(pycompat.shlexsplit(editor)[0])
     fm.condwrite(not cmdpath and editor == 'vi', 'vinotfound',
                  _(" No commit editor set and can't find %s in PATH\n"
                    " (specify a commit editor in your configuration"
                    " file)\n"), not cmdpath and editor == 'vi' and editor)
     fm.condwrite(not cmdpath and editor != 'vi', 'editornotfound',
                  _(" Can't find editor '%s' in PATH\n"
                    " (specify a commit editor in your configuration"
                    " file)\n"), not cmdpath and editor)
     if not cmdpath and editor != 'vi':
         problems += 1
 
     # check username
     username = None
     err = None
     try:
         username = ui.username()
     except error.Abort as e:
         err = e
         problems += 1
 
     fm.condwrite(username, 'username',  _("checking username (%s)\n"), username)
     fm.condwrite(err, 'usernameerror', _("checking username...\n %s\n"
         " (specify a username in your configuration file)\n"), err)
 
     fm.condwrite(not problems, '',
                  _("no problems detected\n"))
     if not problems:
         fm.data(problems=problems)
     fm.condwrite(problems, 'problems',
                  _("%d problems detected,"
                    " please check your install!\n"), problems)
     fm.end()
 
     return problems
 
 @command('debugknown', [], _('REPO ID...'), norepo=True)
 def debugknown(ui, repopath, *ids, **opts):
     """test whether node ids are known to a repo
 
     Every ID must be a full-length hex node id string. Returns a list of 0s
     and 1s indicating unknown/known.
     """
     repo = hg.peer(ui, opts, repopath)
     if not repo.capable('known'):
         raise error.Abort("known() not supported by target repository")
     flags = repo.known([bin(s) for s in ids])
     ui.write("%s\n" % ("".join([f and "1" or "0" for f in flags])))
 
 @command('debuglabelcomplete', [], _('LABEL...'))
 def debuglabelcomplete(ui, repo, *args):
     '''backwards compatibility with old bash completion scripts (DEPRECATED)'''
     debugnamecomplete(ui, repo, *args)
 
 @command('debuglocks',
          [('L', 'force-lock', None, _('free the store lock (DANGEROUS)')),
           ('W', 'force-wlock', None,
            _('free the working state lock (DANGEROUS)'))],
          _('[OPTION]...'))
 def debuglocks(ui, repo, **opts):
     """show or modify state of locks
 
     By default, this command will show which locks are held. This
     includes the user and process holding the lock, the amount of time
     the lock has been held, and the machine name where the process is
     running if it's not local.
 
     Locks protect the integrity of Mercurial's data, so should be
     treated with care. System crashes or other interruptions may cause
     locks to not be properly released, though Mercurial will usually
     detect and remove such stale locks automatically.
 
     However, detecting stale locks may not always be possible (for
     instance, on a shared filesystem). Removing locks may also be
     blocked by filesystem permissions.
 
     Returns 0 if no locks are held.
 
     """
 
     if opts.get('force_lock'):
         repo.svfs.unlink('lock')
     if opts.get('force_wlock'):
         repo.vfs.unlink('wlock')
     if opts.get('force_lock') or opts.get('force_lock'):
         return 0
 
     now = time.time()
     held = 0
 
     def report(vfs, name, method):
         # this causes stale locks to get reaped for more accurate reporting
         try:
             l = method(False)
         except error.LockHeld:
             l = None
 
         if l:
             l.release()
         else:
             try:
                 stat = vfs.lstat(name)
                 age = now - stat.st_mtime
                 user = util.username(stat.st_uid)
                 locker = vfs.readlock(name)
                 if ":" in locker:
                     host, pid = locker.split(':')
                     if host == socket.gethostname():
                         locker = 'user %s, process %s' % (user, pid)
                     else:
                         locker = 'user %s, process %s, host %s' \
                                  % (user, pid, host)
                 ui.write(("%-6s %s (%ds)\n") % (name + ":", locker, age))
                 return 1
             except OSError as e:
                 if e.errno != errno.ENOENT:
                     raise
 
         ui.write(("%-6s free\n") % (name + ":"))
         return 0
 
     held += report(repo.svfs, "lock", repo.lock)
     held += report(repo.vfs, "wlock", repo.wlock)
 
     return held
 
 @command('debugmergestate', [], '')
 def debugmergestate(ui, repo, *args):
     """print merge state
 
     Use --verbose to print out information about whether v1 or v2 merge state
     was chosen."""
     def _hashornull(h):
         if h == nullhex:
             return 'null'
         else:
             return h
 
     def printrecords(version):
         ui.write(('* version %s records\n') % version)
         if version == 1:
             records = v1records
         else:
             records = v2records
 
         for rtype, record in records:
             # pretty print some record types
             if rtype == 'L':
                 ui.write(('local: %s\n') % record)
             elif rtype == 'O':
                 ui.write(('other: %s\n') % record)
             elif rtype == 'm':
                 driver, mdstate = record.split('\0', 1)
                 ui.write(('merge driver: %s (state "%s")\n')
                          % (driver, mdstate))
             elif rtype in 'FDC':
                 r = record.split('\0')
                 f, state, hash, lfile, afile, anode, ofile = r[0:7]
                 if version == 1:
                     onode = 'not stored in v1 format'
                     flags = r[7]
                 else:
                     onode, flags = r[7:9]
                 ui.write(('file: %s (record type "%s", state "%s", hash %s)\n')
                          % (f, rtype, state, _hashornull(hash)))
                 ui.write(('  local path: %s (flags "%s")\n') % (lfile, flags))
                 ui.write(('  ancestor path: %s (node %s)\n')
                          % (afile, _hashornull(anode)))
                 ui.write(('  other path: %s (node %s)\n')
                          % (ofile, _hashornull(onode)))
             elif rtype == 'f':
                 filename, rawextras = record.split('\0', 1)
                 extras = rawextras.split('\0')
                 i = 0
                 extrastrings = []
                 while i < len(extras):
                     extrastrings.append('%s = %s' % (extras[i], extras[i + 1]))
                     i += 2
 
                 ui.write(('file extras: %s (%s)\n')
                          % (filename, ', '.join(extrastrings)))
             elif rtype == 'l':
                 labels = record.split('\0', 2)
                 labels = [l for l in labels if len(l) > 0]
                 ui.write(('labels:\n'))
                 ui.write(('  local: %s\n' % labels[0]))
                 ui.write(('  other: %s\n' % labels[1]))
                 if len(labels) > 2:
                     ui.write(('  base:  %s\n' % labels[2]))
             else:
                 ui.write(('unrecognized entry: %s\t%s\n')
                          % (rtype, record.replace('\0', '\t')))
 
     # Avoid mergestate.read() since it may raise an exception for unsupported
     # merge state records. We shouldn't be doing this, but this is OK since this
     # command is pretty low-level.
     ms = mergemod.mergestate(repo)
 
     # sort so that reasonable information is on top
     v1records = ms._readrecordsv1()
     v2records = ms._readrecordsv2()
     order = 'LOml'
     def key(r):
         idx = order.find(r[0])
         if idx == -1:
             return (1, r[1])
         else:
             return (0, idx)
     v1records.sort(key=key)
     v2records.sort(key=key)
 
     if not v1records and not v2records:
         ui.write(('no merge state found\n'))
     elif not v2records:
         ui.note(('no version 2 merge state\n'))
         printrecords(1)
     elif ms._v1v2match(v1records, v2records):
         ui.note(('v1 and v2 states match: using v2\n'))
         printrecords(2)
     else:
         ui.note(('v1 and v2 states mismatch: using v1\n'))
         printrecords(1)
         if ui.verbose:
             printrecords(2)
 
 @command('debugnamecomplete', [], _('NAME...'))
 def debugnamecomplete(ui, repo, *args):
     '''complete "names" - tags, open branch names, bookmark names'''
 
     names = set()
     # since we previously only listed open branches, we will handle that
     # specially (after this for loop)
     for name, ns in repo.names.iteritems():
         if name != 'branches':
             names.update(ns.listnames(repo))
     names.update(tag for (tag, heads, tip, closed)
                  in repo.branchmap().iterbranches() if not closed)
     completions = set()
     if not args:
         args = ['']
     for a in args:
         completions.update(n for n in names if n.startswith(a))
     ui.write('\n'.join(sorted(completions)))
     ui.write('\n')
 
 @command('debugobsolete',
         [('', 'flags', 0, _('markers flag')),
          ('', 'record-parents', False,
           _('record parent information for the precursor')),
          ('r', 'rev', [], _('display markers relevant to REV')),
          ('', 'index', False, _('display index of the marker')),
          ('', 'delete', [], _('delete markers specified by indices')),
         ] + commands.commitopts2 + commands.formatteropts,
          _('[OBSOLETED [REPLACEMENT ...]]'))
 def debugobsolete(ui, repo, precursor=None, *successors, **opts):
     """create arbitrary obsolete marker
 
     With no arguments, displays the list of obsolescence markers."""
 
     def parsenodeid(s):
         try:
             # We do not use revsingle/revrange functions here to accept
             # arbitrary node identifiers, possibly not present in the
             # local repository.
             n = bin(s)
             if len(n) != len(nullid):
                 raise TypeError()
             return n
         except TypeError:
             raise error.Abort('changeset references must be full hexadecimal '
                              'node identifiers')
 
     if opts.get('delete'):
         indices = []
         for v in opts.get('delete'):
             try:
                 indices.append(int(v))
             except ValueError:
                 raise error.Abort(_('invalid index value: %r') % v,
                                   hint=_('use integers for indices'))
 
         if repo.currenttransaction():
             raise error.Abort(_('cannot delete obsmarkers in the middle '
                                 'of transaction.'))
 
         with repo.lock():
             n = repair.deleteobsmarkers(repo.obsstore, indices)
             ui.write(_('deleted %i obsolescence markers\n') % n)
 
         return
 
     if precursor is not None:
         if opts['rev']:
             raise error.Abort('cannot select revision when creating marker')
         metadata = {}
         metadata['user'] = opts['user'] or ui.username()
         succs = tuple(parsenodeid(succ) for succ in successors)
         l = repo.lock()
         try:
             tr = repo.transaction('debugobsolete')
             try:
                 date = opts.get('date')
                 if date:
                     date = util.parsedate(date)
                 else:
                     date = None
                 prec = parsenodeid(precursor)
                 parents = None
                 if opts['record_parents']:
                     if prec not in repo.unfiltered():
                         raise error.Abort('cannot used --record-parents on '
                                          'unknown changesets')
                     parents = repo.unfiltered()[prec].parents()
                     parents = tuple(p.node() for p in parents)
                 repo.obsstore.create(tr, prec, succs, opts['flags'],
                                      parents=parents, date=date,
                                      metadata=metadata)
                 tr.close()
             except ValueError as exc:
                 raise error.Abort(_('bad obsmarker input: %s') % exc)
             finally:
                 tr.release()
         finally:
             l.release()
     else:
         if opts['rev']:
             revs = scmutil.revrange(repo, opts['rev'])
             nodes = [repo[r].node() for r in revs]
             markers = list(obsolete.getmarkers(repo, nodes=nodes))
             markers.sort(key=lambda x: x._data)
         else:
             markers = obsolete.getmarkers(repo)
 
         markerstoiter = markers
         isrelevant = lambda m: True
         if opts.get('rev') and opts.get('index'):
             markerstoiter = obsolete.getmarkers(repo)
             markerset = set(markers)
             isrelevant = lambda m: m in markerset
 
         fm = ui.formatter('debugobsolete', opts)
         for i, m in enumerate(markerstoiter):
             if not isrelevant(m):
                 # marker can be irrelevant when we're iterating over a set
                 # of markers (markerstoiter) which is bigger than the set
                 # of markers we want to display (markers)
                 # this can happen if both --index and --rev options are
                 # provided and thus we need to iterate over all of the markers
                 # to get the correct indices, but only display the ones that
                 # are relevant to --rev value
                 continue
             fm.startitem()
             ind = i if opts.get('index') else None
             cmdutil.showmarker(fm, m, index=ind)
         fm.end()
 
 @command('debugpathcomplete',
          [('f', 'full', None, _('complete an entire path')),
           ('n', 'normal', None, _('show only normal files')),
           ('a', 'added', None, _('show only added files')),
           ('r', 'removed', None, _('show only removed files'))],
          _('FILESPEC...'))
 def debugpathcomplete(ui, repo, *specs, **opts):
     '''complete part or all of a tracked path
 
     This command supports shells that offer path name completion. It
     currently completes only files already known to the dirstate.
 
     Completion extends only to the next path segment unless
     --full is specified, in which case entire paths are used.'''
 
     def complete(path, acceptable):
         dirstate = repo.dirstate
         spec = os.path.normpath(os.path.join(pycompat.getcwd(), path))
         rootdir = repo.root + pycompat.ossep
         if spec != repo.root and not spec.startswith(rootdir):
             return [], []
         if os.path.isdir(spec):
             spec += '/'
         spec = spec[len(rootdir):]
         fixpaths = pycompat.ossep != '/'
         if fixpaths:
             spec = spec.replace(pycompat.ossep, '/')
         speclen = len(spec)
         fullpaths = opts['full']
         files, dirs = set(), set()
         adddir, addfile = dirs.add, files.add
         for f, st in dirstate.iteritems():
             if f.startswith(spec) and st[0] in acceptable:
                 if fixpaths:
                     f = f.replace('/', pycompat.ossep)
                 if fullpaths:
                     addfile(f)
                     continue
                 s = f.find(pycompat.ossep, speclen)
                 if s >= 0:
                     adddir(f[:s])
                 else:
                     addfile(f)
         return files, dirs
 
     acceptable = ''
     if opts['normal']:
         acceptable += 'nm'
     if opts['added']:
         acceptable += 'a'
     if opts['removed']:
         acceptable += 'r'
     cwd = repo.getcwd()
     if not specs:
         specs = ['.']
 
     files, dirs = set(), set()
     for spec in specs:
         f, d = complete(spec, acceptable or 'nmar')
         files.update(f)
         dirs.update(d)
     files.update(dirs)
     ui.write('\n'.join(repo.pathto(p, cwd) for p in sorted(files)))
     ui.write('\n')
 
 @command('debugpickmergetool',
         [('r', 'rev', '', _('check for files in this revision'), _('REV')),
          ('', 'changedelete', None, _('emulate merging change and delete')),
         ] + commands.walkopts + commands.mergetoolopts,
         _('[PATTERN]...'),
         inferrepo=True)
 def debugpickmergetool(ui, repo, *pats, **opts):
     """examine which merge tool is chosen for specified file
 
     As described in :hg:`help merge-tools`, Mercurial examines
     configurations below in this order to decide which merge tool is
     chosen for specified file.
 
     1. ``--tool`` option
     2. ``HGMERGE`` environment variable
     3. configurations in ``merge-patterns`` section
     4. configuration of ``ui.merge``
     5. configurations in ``merge-tools`` section
     6. ``hgmerge`` tool (for historical reason only)
     7. default tool for fallback (``:merge`` or ``:prompt``)
 
     This command writes out examination result in the style below::
 
         FILE = MERGETOOL
 
     By default, all files known in the first parent context of the
     working directory are examined. Use file patterns and/or -I/-X
     options to limit target files. -r/--rev is also useful to examine
     files in another context without actual updating to it.
 
     With --debug, this command shows warning messages while matching
     against ``merge-patterns`` and so on, too. It is recommended to
     use this option with explicit file patterns and/or -I/-X options,
     because this option increases amount of output per file according
     to configurations in hgrc.
 
     With -v/--verbose, this command shows configurations below at
     first (only if specified).
 
     - ``--tool`` option
     - ``HGMERGE`` environment variable
     - configuration of ``ui.merge``
 
     If merge tool is chosen before matching against
     ``merge-patterns``, this command can't show any helpful
     information, even with --debug. In such case, information above is
     useful to know why a merge tool is chosen.
     """
     overrides = {}
     if opts['tool']:
         overrides[('ui', 'forcemerge')] = opts['tool']
         ui.note(('with --tool %r\n') % (opts['tool']))
 
     with ui.configoverride(overrides, 'debugmergepatterns'):
         hgmerge = encoding.environ.get("HGMERGE")
         if hgmerge is not None:
             ui.note(('with HGMERGE=%r\n') % (hgmerge))
         uimerge = ui.config("ui", "merge")
         if uimerge:
             ui.note(('with ui.merge=%r\n') % (uimerge))
 
         ctx = scmutil.revsingle(repo, opts.get('rev'))
         m = scmutil.match(ctx, pats, opts)
         changedelete = opts['changedelete']
         for path in ctx.walk(m):
             fctx = ctx[path]
             try:
                 if not ui.debugflag:
                     ui.pushbuffer(error=True)
                 tool, toolpath = filemerge._picktool(repo, ui, path,
                                                      fctx.isbinary(),
                                                      'l' in fctx.flags(),
                                                      changedelete)
             finally:
                 if not ui.debugflag:
                     ui.popbuffer()
             ui.write(('%s = %s\n') % (path, tool))
 
 @command('debugpushkey', [], _('REPO NAMESPACE [KEY OLD NEW]'), norepo=True)
 def debugpushkey(ui, repopath, namespace, *keyinfo, **opts):
     '''access the pushkey key/value protocol
 
     With two args, list the keys in the given namespace.
 
     With five args, set a key to new if it currently is set to old.
     Reports success or failure.
     '''
 
     target = hg.peer(ui, {}, repopath)
     if keyinfo:
         key, old, new = keyinfo
         r = target.pushkey(namespace, key, old, new)
         ui.status(str(r) + '\n')
         return not r
     else:
         for k, v in sorted(target.listkeys(namespace).iteritems()):
             ui.write("%s\t%s\n" % (util.escapestr(k),
                                    util.escapestr(v)))
 
 @command('debugpvec', [], _('A B'))
 def debugpvec(ui, repo, a, b=None):
     ca = scmutil.revsingle(repo, a)
     cb = scmutil.revsingle(repo, b)
     pa = pvec.ctxpvec(ca)
     pb = pvec.ctxpvec(cb)
     if pa == pb:
         rel = "="
     elif pa > pb:
         rel = ">"
     elif pa < pb:
         rel = "<"
     elif pa | pb:
         rel = "|"
     ui.write(_("a: %s\n") % pa)
     ui.write(_("b: %s\n") % pb)
     ui.write(_("depth(a): %d depth(b): %d\n") % (pa._depth, pb._depth))
     ui.write(_("delta: %d hdist: %d distance: %d relation: %s\n") %
              (abs(pa._depth - pb._depth), pvec._hamming(pa._vec, pb._vec),
               pa.distance(pb), rel))
 
 @command('debugrebuilddirstate|debugrebuildstate',
     [('r', 'rev', '', _('revision to rebuild to'), _('REV')),
      ('', 'minimal', None, _('only rebuild files that are inconsistent with '
                              'the working copy parent')),
     ],
     _('[-r REV]'))
 def debugrebuilddirstate(ui, repo, rev, **opts):
     """rebuild the dirstate as it would look like for the given revision
 
     If no revision is specified the first current parent will be used.
 
     The dirstate will be set to the files of the given revision.
     The actual working directory content or existing dirstate
     information such as adds or removes is not considered.
 
     ``minimal`` will only rebuild the dirstate status for files that claim to be
     tracked but are not in the parent manifest, or that exist in the parent
     manifest but are not in the dirstate. It will not change adds, removes, or
     modified files that are in the working copy parent.
 
     One use of this command is to make the next :hg:`status` invocation
     check the actual file content.
     """
     ctx = scmutil.revsingle(repo, rev)
     with repo.wlock():
         dirstate = repo.dirstate
         changedfiles = None
         # See command doc for what minimal does.
         if opts.get('minimal'):
             manifestfiles = set(ctx.manifest().keys())
             dirstatefiles = set(dirstate)
             manifestonly = manifestfiles - dirstatefiles
             dsonly = dirstatefiles - manifestfiles
             dsnotadded = set(f for f in dsonly if dirstate[f] != 'a')
             changedfiles = manifestonly | dsnotadded
 
         dirstate.rebuild(ctx.node(), ctx.manifest(), changedfiles)
 
 @command('debugrebuildfncache', [], '')
 def debugrebuildfncache(ui, repo):
     """rebuild the fncache file"""
     repair.rebuildfncache(ui, repo)
 
 @command('debugrename',
     [('r', 'rev', '', _('revision to debug'), _('REV'))],
     _('[-r REV] FILE'))
 def debugrename(ui, repo, file1, *pats, **opts):
     """dump rename information"""
 
     ctx = scmutil.revsingle(repo, opts.get('rev'))
     m = scmutil.match(ctx, (file1,) + pats, opts)
     for abs in ctx.walk(m):
         fctx = ctx[abs]
         o = fctx.filelog().renamed(fctx.filenode())
         rel = m.rel(abs)
         if o:
             ui.write(_("%s renamed from %s:%s\n") % (rel, o[0], hex(o[1])))
         else:
             ui.write(_("%s not renamed\n") % rel)
 
 @command('debugrevlog', commands.debugrevlogopts +
     [('d', 'dump', False, _('dump index data'))],
     _('-c|-m|FILE'),
     optionalrepo=True)
 def debugrevlog(ui, repo, file_=None, **opts):
     """show data and statistics about a revlog"""
     r = cmdutil.openrevlog(repo, 'debugrevlog', file_, opts)
 
     if opts.get("dump"):
         numrevs = len(r)
         ui.write(("# rev p1rev p2rev start   end deltastart base   p1   p2"
                  " rawsize totalsize compression heads chainlen\n"))
         ts = 0
         heads = set()
 
         for rev in xrange(numrevs):
             dbase = r.deltaparent(rev)
             if dbase == -1:
                 dbase = rev
             cbase = r.chainbase(rev)
             clen = r.chainlen(rev)
             p1, p2 = r.parentrevs(rev)
             rs = r.rawsize(rev)
             ts = ts + rs
             heads -= set(r.parentrevs(rev))
             heads.add(rev)
             try:
                 compression = ts / r.end(rev)
             except ZeroDivisionError:
                 compression = 0
             ui.write("%5d %5d %5d %5d %5d %10d %4d %4d %4d %7d %9d "
                      "%11d %5d %8d\n" %
                      (rev, p1, p2, r.start(rev), r.end(rev),
                       r.start(dbase), r.start(cbase),
                       r.start(p1), r.start(p2),
                       rs, ts, compression, len(heads), clen))
         return 0
 
     v = r.version
     format = v & 0xFFFF
     flags = []
     gdelta = False
     if v & revlog.FLAG_INLINE_DATA:
         flags.append('inline')
     if v & revlog.FLAG_GENERALDELTA:
         gdelta = True
         flags.append('generaldelta')
     if not flags:
         flags = ['(none)']
 
     nummerges = 0
     numfull = 0
     numprev = 0
     nump1 = 0
     nump2 = 0
     numother = 0
     nump1prev = 0
     nump2prev = 0
     chainlengths = []
 
     datasize = [None, 0, 0]
     fullsize = [None, 0, 0]
     deltasize = [None, 0, 0]
     chunktypecounts = {}
     chunktypesizes = {}
 
     def addsize(size, l):
         if l[0] is None or size < l[0]:
             l[0] = size
         if size > l[1]:
             l[1] = size
         l[2] += size
 
     numrevs = len(r)
     for rev in xrange(numrevs):
         p1, p2 = r.parentrevs(rev)
         delta = r.deltaparent(rev)
         if format > 0:
             addsize(r.rawsize(rev), datasize)
         if p2 != nullrev:
             nummerges += 1
         size = r.length(rev)
         if delta == nullrev:
             chainlengths.append(0)
             numfull += 1
             addsize(size, fullsize)
         else:
             chainlengths.append(chainlengths[delta] + 1)
             addsize(size, deltasize)
             if delta == rev - 1:
                 numprev += 1
                 if delta == p1:
                     nump1prev += 1
                 elif delta == p2:
                     nump2prev += 1
             elif delta == p1:
                 nump1 += 1
             elif delta == p2:
                 nump2 += 1
             elif delta != nullrev:
                 numother += 1
 
         # Obtain data on the raw chunks in the revlog.
         segment = r._getsegmentforrevs(rev, rev)[1]
         if segment:
             chunktype = segment[0]
         else:
             chunktype = 'empty'
 
         if chunktype not in chunktypecounts:
             chunktypecounts[chunktype] = 0
             chunktypesizes[chunktype] = 0
 
         chunktypecounts[chunktype] += 1
         chunktypesizes[chunktype] += size
 
     # Adjust size min value for empty cases
     for size in (datasize, fullsize, deltasize):
         if size[0] is None:
             size[0] = 0
 
     numdeltas = numrevs - numfull
     numoprev = numprev - nump1prev - nump2prev
     totalrawsize = datasize[2]
     datasize[2] /= numrevs
     fulltotal = fullsize[2]
     fullsize[2] /= numfull
     deltatotal = deltasize[2]
     if numrevs - numfull > 0:
         deltasize[2] /= numrevs - numfull
     totalsize = fulltotal + deltatotal
     avgchainlen = sum(chainlengths) / numrevs
     maxchainlen = max(chainlengths)
     compratio = 1
     if totalsize:
         compratio = totalrawsize / totalsize
 
     basedfmtstr = '%%%dd\n'
     basepcfmtstr = '%%%dd %s(%%5.2f%%%%)\n'
 
     def dfmtstr(max):
         return basedfmtstr % len(str(max))
     def pcfmtstr(max, padding=0):
         return basepcfmtstr % (len(str(max)), ' ' * padding)
 
     def pcfmt(value, total):
         if total:
             return (value, 100 * float(value) / total)
         else:
             return value, 100.0
 
     ui.write(('format : %d\n') % format)
     ui.write(('flags  : %s\n') % ', '.join(flags))
 
     ui.write('\n')
     fmt = pcfmtstr(totalsize)
     fmt2 = dfmtstr(totalsize)
     ui.write(('revisions     : ') + fmt2 % numrevs)
     ui.write(('    merges    : ') + fmt % pcfmt(nummerges, numrevs))
     ui.write(('    normal    : ') + fmt % pcfmt(numrevs - nummerges, numrevs))
     ui.write(('revisions     : ') + fmt2 % numrevs)
     ui.write(('    full      : ') + fmt % pcfmt(numfull, numrevs))
     ui.write(('    deltas    : ') + fmt % pcfmt(numdeltas, numrevs))
     ui.write(('revision size : ') + fmt2 % totalsize)
     ui.write(('    full      : ') + fmt % pcfmt(fulltotal, totalsize))
     ui.write(('    deltas    : ') + fmt % pcfmt(deltatotal, totalsize))
 
     def fmtchunktype(chunktype):
         if chunktype == 'empty':
             return '    %s     : ' % chunktype
         elif chunktype in string.ascii_letters:
             return '    0x%s (%s)  : ' % (hex(chunktype), chunktype)
         else:
             return '    0x%s      : ' % hex(chunktype)
 
     ui.write('\n')
     ui.write(('chunks        : ') + fmt2 % numrevs)
     for chunktype in sorted(chunktypecounts):
         ui.write(fmtchunktype(chunktype))
         ui.write(fmt % pcfmt(chunktypecounts[chunktype], numrevs))
     ui.write(('chunks size   : ') + fmt2 % totalsize)
     for chunktype in sorted(chunktypecounts):
         ui.write(fmtchunktype(chunktype))
         ui.write(fmt % pcfmt(chunktypesizes[chunktype], totalsize))
 
     ui.write('\n')
     fmt = dfmtstr(max(avgchainlen, compratio))
     ui.write(('avg chain length  : ') + fmt % avgchainlen)
     ui.write(('max chain length  : ') + fmt % maxchainlen)
     ui.write(('compression ratio : ') + fmt % compratio)
 
     if format > 0:
         ui.write('\n')
         ui.write(('uncompressed data size (min/max/avg) : %d / %d / %d\n')
                  % tuple(datasize))
     ui.write(('full revision size (min/max/avg)     : %d / %d / %d\n')
              % tuple(fullsize))
     ui.write(('delta size (min/max/avg)             : %d / %d / %d\n')
              % tuple(deltasize))
 
     if numdeltas > 0:
         ui.write('\n')
         fmt = pcfmtstr(numdeltas)
         fmt2 = pcfmtstr(numdeltas, 4)
         ui.write(('deltas against prev  : ') + fmt % pcfmt(numprev, numdeltas))
         if numprev > 0:
             ui.write(('    where prev = p1  : ') + fmt2 % pcfmt(nump1prev,
                                                               numprev))
             ui.write(('    where prev = p2  : ') + fmt2 % pcfmt(nump2prev,
                                                               numprev))
             ui.write(('    other            : ') + fmt2 % pcfmt(numoprev,
                                                               numprev))
         if gdelta:
             ui.write(('deltas against p1    : ')
                      + fmt % pcfmt(nump1, numdeltas))
             ui.write(('deltas against p2    : ')
                      + fmt % pcfmt(nump2, numdeltas))
             ui.write(('deltas against other : ') + fmt % pcfmt(numother,
                                                              numdeltas))
 
 @command('debugrevspec',
     [('', 'optimize', None,
       _('print parsed tree after optimizing (DEPRECATED)')),
      ('p', 'show-stage', [],
       _('print parsed tree at the given stage'), _('NAME')),
      ('', 'no-optimized', False, _('evaluate tree without optimization')),
      ('', 'verify-optimized', False, _('verify optimized result')),
      ],
     ('REVSPEC'))
 def debugrevspec(ui, repo, expr, **opts):
     """parse and apply a revision specification
 
     Use -p/--show-stage option to print the parsed tree at the given stages.
     Use -p all to print tree at every stage.
 
     Use --verify-optimized to compare the optimized result with the unoptimized
     one. Returns 1 if the optimized result differs.
     """
     stages = [
         ('parsed', lambda tree: tree),
         ('expanded', lambda tree: revsetlang.expandaliases(ui, tree)),
         ('concatenated', revsetlang.foldconcat),
         ('analyzed', revsetlang.analyze),
         ('optimized', revsetlang.optimize),
     ]
     if opts['no_optimized']:
         stages = stages[:-1]
     if opts['verify_optimized'] and opts['no_optimized']:
         raise error.Abort(_('cannot use --verify-optimized with '
                             '--no-optimized'))
     stagenames = set(n for n, f in stages)
 
     showalways = set()
     showchanged = set()
     if ui.verbose and not opts['show_stage']:
         # show parsed tree by --verbose (deprecated)
         showalways.add('parsed')
         showchanged.update(['expanded', 'concatenated'])
         if opts['optimize']:
             showalways.add('optimized')
     if opts['show_stage'] and opts['optimize']:
         raise error.Abort(_('cannot use --optimize with --show-stage'))
     if opts['show_stage'] == ['all']:
         showalways.update(stagenames)
     else:
         for n in opts['show_stage']:
             if n not in stagenames:
                 raise error.Abort(_('invalid stage name: %s') % n)
         showalways.update(opts['show_stage'])
 
     treebystage = {}
     printedtree = None
     tree = revsetlang.parse(expr, lookup=repo.__contains__)
     for n, f in stages:
         treebystage[n] = tree = f(tree)
         if n in showalways or (n in showchanged and tree != printedtree):
             if opts['show_stage'] or n != 'parsed':
                 ui.write(("* %s:\n") % n)
             ui.write(revsetlang.prettyformat(tree), "\n")
             printedtree = tree
 
     if opts['verify_optimized']:
         arevs = revset.makematcher(treebystage['analyzed'])(repo)
         brevs = revset.makematcher(treebystage['optimized'])(repo)
         if ui.verbose:
             ui.note(("* analyzed set:\n"), smartset.prettyformat(arevs), "\n")
             ui.note(("* optimized set:\n"), smartset.prettyformat(brevs), "\n")
         arevs = list(arevs)
         brevs = list(brevs)
         if arevs == brevs:
             return 0
         ui.write(('--- analyzed\n'), label='diff.file_a')
         ui.write(('+++ optimized\n'), label='diff.file_b')
         sm = difflib.SequenceMatcher(None, arevs, brevs)
         for tag, alo, ahi, blo, bhi in sm.get_opcodes():
             if tag in ('delete', 'replace'):
                 for c in arevs[alo:ahi]:
                     ui.write('-%s\n' % c, label='diff.deleted')
             if tag in ('insert', 'replace'):
                 for c in brevs[blo:bhi]:
                     ui.write('+%s\n' % c, label='diff.inserted')
             if tag == 'equal':
                 for c in arevs[alo:ahi]:
                     ui.write(' %s\n' % c)
         return 1
 
     func = revset.makematcher(tree)
     revs = func(repo)
     if ui.verbose:
         ui.note(("* set:\n"), smartset.prettyformat(revs), "\n")
     for c in revs:
         ui.write("%s\n" % c)
 
 @command('debugsetparents', [], _('REV1 [REV2]'))
 def debugsetparents(ui, repo, rev1, rev2=None):
     """manually set the parents of the current working directory
 
     This is useful for writing repository conversion tools, but should
     be used with care. For example, neither the working directory nor the
     dirstate is updated, so file status may be incorrect after running this
     command.
 
     Returns 0 on success.
     """
 
     r1 = scmutil.revsingle(repo, rev1).node()
     r2 = scmutil.revsingle(repo, rev2, 'null').node()
 
     with repo.wlock():
         repo.setparents(r1, r2)
 
 @command('debugsub',
     [('r', 'rev', '',
      _('revision to check'), _('REV'))],
     _('[-r REV] [REV]'))
 def debugsub(ui, repo, rev=None):
     ctx = scmutil.revsingle(repo, rev, None)
     for k, v in sorted(ctx.substate.items()):
         ui.write(('path %s\n') % k)
         ui.write((' source   %s\n') % v[0])
         ui.write((' revision %s\n') % v[1])
 
 @command('debugsuccessorssets',
     [],
     _('[REV]'))
 def debugsuccessorssets(ui, repo, *revs):
     """show set of successors for revision
 
     A successors set of changeset A is a consistent group of revisions that
     succeed A. It contains non-obsolete changesets only.
 
     In most cases a changeset A has a single successors set containing a single
     successor (changeset A replaced by A').
 
     A changeset that is made obsolete with no successors are called "pruned".
     Such changesets have no successors sets at all.
 
     A changeset that has been "split" will have a successors set containing
     more than one successor.
 
     A changeset that has been rewritten in multiple different ways is called
     "divergent". Such changesets have multiple successor sets (each of which
     may also be split, i.e. have multiple successors).
 
     Results are displayed as follows::
 
         <rev1>
             <successors-1A>
         <rev2>
             <successors-2A>
             <successors-2B1> <successors-2B2> <successors-2B3>
 
     Here rev2 has two possible (i.e. divergent) successors sets. The first
     holds one element, whereas the second holds three (i.e. the changeset has
     been split).
     """
     # passed to successorssets caching computation from one call to another
     cache = {}
     ctx2str = str
     node2str = short
     if ui.debug():
         def ctx2str(ctx):
             return ctx.hex()
         node2str = hex
     for rev in scmutil.revrange(repo, revs):
         ctx = repo[rev]
         ui.write('%s\n'% ctx2str(ctx))
         for succsset in obsolete.successorssets(repo, ctx.node(), cache):
             if succsset:
                 ui.write('    ')
                 ui.write(node2str(succsset[0]))
                 for node in succsset[1:]:
                     ui.write(' ')
                     ui.write(node2str(node))
             ui.write('\n')
 
 @command('debugtemplate',
     [('r', 'rev', [], _('apply template on changesets'), _('REV')),
      ('D', 'define', [], _('define template keyword'), _('KEY=VALUE'))],
     _('[-r REV]... [-D KEY=VALUE]... TEMPLATE'),
     optionalrepo=True)
 def debugtemplate(ui, repo, tmpl, **opts):
     """parse and apply a template
 
     If -r/--rev is given, the template is processed as a log template and
     applied to the given changesets. Otherwise, it is processed as a generic
     template.
 
     Use --verbose to print the parsed tree.
     """
     revs = None
     if opts['rev']:
         if repo is None:
             raise error.RepoError(_('there is no Mercurial repository here '
                                     '(.hg not found)'))
         revs = scmutil.revrange(repo, opts['rev'])
 
     props = {}
     for d in opts['define']:
         try:
             k, v = (e.strip() for e in d.split('=', 1))
             if not k or k == 'ui':
                 raise ValueError
             props[k] = v
         except ValueError:
             raise error.Abort(_('malformed keyword definition: %s') % d)
 
     if ui.verbose:
         aliases = ui.configitems('templatealias')
         tree = templater.parse(tmpl)
         ui.note(templater.prettyformat(tree), '\n')
         newtree = templater.expandaliases(tree, aliases)
         if newtree != tree:
             ui.note(("* expanded:\n"), templater.prettyformat(newtree), '\n')
 
     mapfile = None
     if revs is None:
         k = 'debugtemplate'
         t = formatter.maketemplater(ui, k, tmpl)
         ui.write(templater.stringify(t(k, ui=ui, **props)))
     else:
         displayer = cmdutil.changeset_templater(ui, repo, None, opts, tmpl,
                                                 mapfile, buffered=False)
         for r in revs:
             displayer.show(repo[r], **props)
         displayer.close()
 
 @command('debugupdatecaches', [])
 def debugupdatecaches(ui, repo, *pats, **opts):
     """warm all known caches in the repository"""
     with repo.wlock():
         with repo.lock():
             repo.updatecaches()
 
 @command('debugupgraderepo', [
     ('o', 'optimize', [], _('extra optimization to perform'), _('NAME')),
     ('', 'run', False, _('performs an upgrade')),
 ])
 def debugupgraderepo(ui, repo, run=False, optimize=None):
     """upgrade a repository to use different features
 
     If no arguments are specified, the repository is evaluated for upgrade
     and a list of problems and potential optimizations is printed.
 
     With ``--run``, a repository upgrade is performed. Behavior of the upgrade
     can be influenced via additional arguments. More details will be provided
     by the command output when run without ``--run``.
 
     During the upgrade, the repository will be locked and no writes will be
     allowed.
 
     At the end of the upgrade, the repository may not be readable while new
     repository data is swapped in. This window will be as long as it takes to
     rename some directories inside the ``.hg`` directory. On most machines, this
     should complete almost instantaneously and the chances of a consumer being
     unable to access the repository should be low.
     """
     return upgrade.upgraderepo(ui, repo, run=run, optimize=optimize)
 
 @command('debugwalk', commands.walkopts, _('[OPTION]... [FILE]...'),
          inferrepo=True)
 def debugwalk(ui, repo, *pats, **opts):
     """show how files match on given patterns"""
     m = scmutil.match(repo[None], pats, opts)
     items = list(repo[None].walk(m))
     if not items:
         return
     f = lambda fn: fn
     if ui.configbool('ui', 'slash') and pycompat.ossep != '/':
         f = lambda fn: util.normpath(fn)
     fmt = 'f  %%-%ds  %%-%ds  %%s' % (
         max([len(abs) for abs in items]),
         max([len(m.rel(abs)) for abs in items]))
     for abs in items:
         line = fmt % (abs, f(m.rel(abs)), m.exact(abs) and 'exact' or '')
         ui.write("%s\n" % line.rstrip())
 
 @command('debugwireargs',
     [('', 'three', '', 'three'),
     ('', 'four', '', 'four'),
     ('', 'five', '', 'five'),
     ] + commands.remoteopts,
     _('REPO [OPTIONS]... [ONE [TWO]]'),
     norepo=True)
 def debugwireargs(ui, repopath, *vals, **opts):
     repo = hg.peer(ui, opts, repopath)
     for opt in commands.remoteopts:
         del opts[opt[1]]
     args = {}
     for k, v in opts.iteritems():
         if v:
             args[k] = v
     # run twice to check that we don't mess up the stream for the next command
     res1 = repo.debugwireargs(*vals, **args)
     res2 = repo.debugwireargs(*vals, **args)
     ui.write("%s\n" % res1)
     if res1 != res2:
         ui.warn("%s\n" % res2)

mercurial/pure/osutil.py

@@ -1,365 +1,365 b''
 # osutil.py - pure Python version of osutil.c
 #
 #  Copyright 2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import ctypes
 import ctypes.util
 import os
 import socket
 import stat as statmod
 
-from . import (
+from .. import (
     policy,
     pycompat,
 )
 
 modulepolicy = policy.policy
 policynocffi = policy.policynocffi
 
 def _mode_to_kind(mode):
     if statmod.S_ISREG(mode):
         return statmod.S_IFREG
     if statmod.S_ISDIR(mode):
         return statmod.S_IFDIR
     if statmod.S_ISLNK(mode):
         return statmod.S_IFLNK
     if statmod.S_ISBLK(mode):
         return statmod.S_IFBLK
     if statmod.S_ISCHR(mode):
         return statmod.S_IFCHR
     if statmod.S_ISFIFO(mode):
         return statmod.S_IFIFO
     if statmod.S_ISSOCK(mode):
         return statmod.S_IFSOCK
     return mode
 
 def listdirpure(path, stat=False, skip=None):
     '''listdir(path, stat=False) -> list_of_tuples
 
     Return a sorted list containing information about the entries
     in the directory.
 
     If stat is True, each element is a 3-tuple:
 
       (name, type, stat object)
 
     Otherwise, each element is a 2-tuple:
 
       (name, type)
     '''
     result = []
     prefix = path
     if not prefix.endswith(pycompat.ossep):
         prefix += pycompat.ossep
     names = os.listdir(path)
     names.sort()
     for fn in names:
         st = os.lstat(prefix + fn)
         if fn == skip and statmod.S_ISDIR(st.st_mode):
             return []
         if stat:
             result.append((fn, _mode_to_kind(st.st_mode), st))
         else:
             result.append((fn, _mode_to_kind(st.st_mode)))
     return result
 
 ffi = None
 if modulepolicy not in policynocffi and pycompat.sysplatform == 'darwin':
     try:
         from _osutil_cffi import ffi, lib
     except ImportError:
         if modulepolicy == 'cffi': # strict cffi import
             raise
 
 if pycompat.sysplatform == 'darwin' and ffi is not None:
     listdir_batch_size = 4096
     # tweakable number, only affects performance, which chunks
     # of bytes do we get back from getattrlistbulk
 
     attrkinds = [None] * 20 # we need the max no for enum VXXX, 20 is plenty
 
     attrkinds[lib.VREG] = statmod.S_IFREG
     attrkinds[lib.VDIR] = statmod.S_IFDIR
     attrkinds[lib.VLNK] = statmod.S_IFLNK
     attrkinds[lib.VBLK] = statmod.S_IFBLK
     attrkinds[lib.VCHR] = statmod.S_IFCHR
     attrkinds[lib.VFIFO] = statmod.S_IFIFO
     attrkinds[lib.VSOCK] = statmod.S_IFSOCK
 
     class stat_res(object):
         def __init__(self, st_mode, st_mtime, st_size):
             self.st_mode = st_mode
             self.st_mtime = st_mtime
             self.st_size = st_size
 
     tv_sec_ofs = ffi.offsetof("struct timespec", "tv_sec")
     buf = ffi.new("char[]", listdir_batch_size)
 
     def listdirinternal(dfd, req, stat, skip):
         ret = []
         while True:
             r = lib.getattrlistbulk(dfd, req, buf, listdir_batch_size, 0)
             if r == 0:
                 break
             if r == -1:
                 raise OSError(ffi.errno, os.strerror(ffi.errno))
             cur = ffi.cast("val_attrs_t*", buf)
             for i in range(r):
                 lgt = cur.length
                 assert lgt == ffi.cast('uint32_t*', cur)[0]
                 ofs = cur.name_info.attr_dataoffset
                 str_lgt = cur.name_info.attr_length
                 base_ofs = ffi.offsetof('val_attrs_t', 'name_info')
                 name = str(ffi.buffer(ffi.cast("char*", cur) + base_ofs + ofs,
                            str_lgt - 1))
                 tp = attrkinds[cur.obj_type]
                 if name == "." or name == "..":
                     continue
                 if skip == name and tp == statmod.S_ISDIR:
                     return []
                 if stat:
                     mtime = cur.mtime.tv_sec
                     mode = (cur.accessmask & ~lib.S_IFMT)| tp
                     ret.append((name, tp, stat_res(st_mode=mode, st_mtime=mtime,
                                 st_size=cur.datalength)))
                 else:
                     ret.append((name, tp))
                 cur = ffi.cast("val_attrs_t*", int(ffi.cast("intptr_t", cur))
                     + lgt)
         return ret
 
     def listdir(path, stat=False, skip=None):
         req = ffi.new("struct attrlist*")
         req.bitmapcount = lib.ATTR_BIT_MAP_COUNT
         req.commonattr = (lib.ATTR_CMN_RETURNED_ATTRS |
                           lib.ATTR_CMN_NAME |
                           lib.ATTR_CMN_OBJTYPE |
                           lib.ATTR_CMN_ACCESSMASK |
                           lib.ATTR_CMN_MODTIME)
         req.fileattr = lib.ATTR_FILE_DATALENGTH
         dfd = lib.open(path, lib.O_RDONLY, 0)
         if dfd == -1:
             raise OSError(ffi.errno, os.strerror(ffi.errno))
 
         try:
             ret = listdirinternal(dfd, req, stat, skip)
         finally:
             try:
                 lib.close(dfd)
             except BaseException:
                 pass # we ignore all the errors from closing, not
                 # much we can do about that
         return ret
 else:
     listdir = listdirpure
 
 if pycompat.osname != 'nt':
     posixfile = open
 
     _SCM_RIGHTS = 0x01
     _socklen_t = ctypes.c_uint
 
     if pycompat.sysplatform.startswith('linux'):
         # socket.h says "the type should be socklen_t but the definition of
         # the kernel is incompatible with this."
         _cmsg_len_t = ctypes.c_size_t
         _msg_controllen_t = ctypes.c_size_t
         _msg_iovlen_t = ctypes.c_size_t
     else:
         _cmsg_len_t = _socklen_t
         _msg_controllen_t = _socklen_t
         _msg_iovlen_t = ctypes.c_int
 
     class _iovec(ctypes.Structure):
         _fields_ = [
             (u'iov_base', ctypes.c_void_p),
             (u'iov_len', ctypes.c_size_t),
         ]
 
     class _msghdr(ctypes.Structure):
         _fields_ = [
             (u'msg_name', ctypes.c_void_p),
             (u'msg_namelen', _socklen_t),
             (u'msg_iov', ctypes.POINTER(_iovec)),
             (u'msg_iovlen', _msg_iovlen_t),
             (u'msg_control', ctypes.c_void_p),
             (u'msg_controllen', _msg_controllen_t),
             (u'msg_flags', ctypes.c_int),
         ]
 
     class _cmsghdr(ctypes.Structure):
         _fields_ = [
             (u'cmsg_len', _cmsg_len_t),
             (u'cmsg_level', ctypes.c_int),
             (u'cmsg_type', ctypes.c_int),
             (u'cmsg_data', ctypes.c_ubyte * 0),
         ]
 
     _libc = ctypes.CDLL(ctypes.util.find_library(u'c'), use_errno=True)
     _recvmsg = getattr(_libc, 'recvmsg', None)
     if _recvmsg:
         _recvmsg.restype = getattr(ctypes, 'c_ssize_t', ctypes.c_long)
         _recvmsg.argtypes = (ctypes.c_int, ctypes.POINTER(_msghdr),
                              ctypes.c_int)
     else:
         # recvmsg isn't always provided by libc; such systems are unsupported
         def _recvmsg(sockfd, msg, flags):
             raise NotImplementedError('unsupported platform')
 
     def _CMSG_FIRSTHDR(msgh):
         if msgh.msg_controllen < ctypes.sizeof(_cmsghdr):
             return
         cmsgptr = ctypes.cast(msgh.msg_control, ctypes.POINTER(_cmsghdr))
         return cmsgptr.contents
 
     # The pure version is less portable than the native version because the
     # handling of socket ancillary data heavily depends on C preprocessor.
     # Also, some length fields are wrongly typed in Linux kernel.
     def recvfds(sockfd):
         """receive list of file descriptors via socket"""
         dummy = (ctypes.c_ubyte * 1)()
         iov = _iovec(ctypes.cast(dummy, ctypes.c_void_p), ctypes.sizeof(dummy))
         cbuf = ctypes.create_string_buffer(256)
         msgh = _msghdr(None, 0,
                        ctypes.pointer(iov), 1,
                        ctypes.cast(cbuf, ctypes.c_void_p), ctypes.sizeof(cbuf),
                        0)
         r = _recvmsg(sockfd, ctypes.byref(msgh), 0)
         if r < 0:
             e = ctypes.get_errno()
             raise OSError(e, os.strerror(e))
         # assumes that the first cmsg has fds because it isn't easy to write
         # portable CMSG_NXTHDR() with ctypes.
         cmsg = _CMSG_FIRSTHDR(msgh)
         if not cmsg:
             return []
         if (cmsg.cmsg_level != socket.SOL_SOCKET or
             cmsg.cmsg_type != _SCM_RIGHTS):
             return []
         rfds = ctypes.cast(cmsg.cmsg_data, ctypes.POINTER(ctypes.c_int))
         rfdscount = ((cmsg.cmsg_len - _cmsghdr.cmsg_data.offset) /
                      ctypes.sizeof(ctypes.c_int))
         return [rfds[i] for i in xrange(rfdscount)]
 
 else:
     import msvcrt
 
     _kernel32 = ctypes.windll.kernel32
 
     _DWORD = ctypes.c_ulong
     _LPCSTR = _LPSTR = ctypes.c_char_p
     _HANDLE = ctypes.c_void_p
 
     _INVALID_HANDLE_VALUE = _HANDLE(-1).value
 
     # CreateFile
     _FILE_SHARE_READ = 0x00000001
     _FILE_SHARE_WRITE = 0x00000002
     _FILE_SHARE_DELETE = 0x00000004
 
     _CREATE_ALWAYS = 2
     _OPEN_EXISTING = 3
     _OPEN_ALWAYS = 4
 
     _GENERIC_READ = 0x80000000
     _GENERIC_WRITE = 0x40000000
 
     _FILE_ATTRIBUTE_NORMAL = 0x80
 
     # open_osfhandle flags
     _O_RDONLY = 0x0000
     _O_RDWR = 0x0002
     _O_APPEND = 0x0008
 
     _O_TEXT = 0x4000
     _O_BINARY = 0x8000
 
     # types of parameters of C functions used (required by pypy)
 
     _kernel32.CreateFileA.argtypes = [_LPCSTR, _DWORD, _DWORD, ctypes.c_void_p,
         _DWORD, _DWORD, _HANDLE]
     _kernel32.CreateFileA.restype = _HANDLE
 
     def _raiseioerror(name):
         err = ctypes.WinError()
         raise IOError(err.errno, '%s: %s' % (name, err.strerror))
 
     class posixfile(object):
         '''a file object aiming for POSIX-like semantics
 
         CPython's open() returns a file that was opened *without* setting the
         _FILE_SHARE_DELETE flag, which causes rename and unlink to abort.
         This even happens if any hardlinked copy of the file is in open state.
         We set _FILE_SHARE_DELETE here, so files opened with posixfile can be
         renamed and deleted while they are held open.
         Note that if a file opened with posixfile is unlinked, the file
         remains but cannot be opened again or be recreated under the same name,
         until all reading processes have closed the file.'''
 
         def __init__(self, name, mode='r', bufsize=-1):
             if 'b' in mode:
                 flags = _O_BINARY
             else:
                 flags = _O_TEXT
 
             m0 = mode[0]
             if m0 == 'r' and '+' not in mode:
                 flags |= _O_RDONLY
                 access = _GENERIC_READ
             else:
                 # work around http://support.microsoft.com/kb/899149 and
                 # set _O_RDWR for 'w' and 'a', even if mode has no '+'
                 flags |= _O_RDWR
                 access = _GENERIC_READ | _GENERIC_WRITE
 
             if m0 == 'r':
                 creation = _OPEN_EXISTING
             elif m0 == 'w':
                 creation = _CREATE_ALWAYS
             elif m0 == 'a':
                 creation = _OPEN_ALWAYS
                 flags |= _O_APPEND
             else:
                 raise ValueError("invalid mode: %s" % mode)
 
             fh = _kernel32.CreateFileA(name, access,
                     _FILE_SHARE_READ | _FILE_SHARE_WRITE | _FILE_SHARE_DELETE,
                     None, creation, _FILE_ATTRIBUTE_NORMAL, None)
             if fh == _INVALID_HANDLE_VALUE:
                 _raiseioerror(name)
 
             fd = msvcrt.open_osfhandle(fh, flags)
             if fd == -1:
                 _kernel32.CloseHandle(fh)
                 _raiseioerror(name)
 
             f = os.fdopen(fd, pycompat.sysstr(mode), bufsize)
             # unfortunately, f.name is '<fdopen>' at this point -- so we store
             # the name on this wrapper. We cannot just assign to f.name,
             # because that attribute is read-only.
             object.__setattr__(self, r'name', name)
             object.__setattr__(self, r'_file', f)
 
         def __iter__(self):
             return self._file
 
         def __getattr__(self, name):
             return getattr(self._file, name)
 
         def __setattr__(self, name, value):
             '''mimics the read-only attributes of Python file objects
             by raising 'TypeError: readonly attribute' if someone tries:
               f = posixfile('foo.txt')
               f.name = 'bla'  '''
             return self._file.__setattr__(name, value)
 
         def __enter__(self):
             return self._file.__enter__()
 
         def __exit__(self, exc_type, exc_value, exc_tb):
             return self._file.__exit__(exc_type, exc_value, exc_tb)

mercurial/util.py

@@ -1,3729 +1,3731 b''
 # util.py - Mercurial utility functions and platform specific implementations
 #
 #  Copyright 2005 K. Thananchayan <[e-mail]>
 #  Copyright 2005-2007 Matt Mackall <[e-mail]>
 #  Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 """Mercurial utility functions and platform specific implementations.
 
 This contains helper routines that are independent of the SCM core and
 hide platform-specific details from the core.
 """
 
 from __future__ import absolute_import
 
 import bz2
 import calendar
 import codecs
 import collections
 import datetime
 import errno
 import gc
 import hashlib
 import imp
 import os
 import platform as pyplatform
 import re as remod
 import shutil
 import signal
 import socket
 import stat
 import string
 import subprocess
 import sys
 import tempfile
 import textwrap
 import time
 import traceback
 import warnings
 import zlib
 
 from . import (
     base85,
     encoding,
     error,
     i18n,
-    osutil,
     parsers,
+    policy,
     pycompat,
 )
 
+osutil = policy.importmod(r'osutil')
+
 b85decode = base85.b85decode
 b85encode = base85.b85encode
 
 cookielib = pycompat.cookielib
 empty = pycompat.empty
 httplib = pycompat.httplib
 httpserver = pycompat.httpserver
 pickle = pycompat.pickle
 queue = pycompat.queue
 socketserver = pycompat.socketserver
 stderr = pycompat.stderr
 stdin = pycompat.stdin
 stdout = pycompat.stdout
 stringio = pycompat.stringio
 urlerr = pycompat.urlerr
 urlreq = pycompat.urlreq
 xmlrpclib = pycompat.xmlrpclib
 
 def isatty(fp):
     try:
         return fp.isatty()
     except AttributeError:
         return False
 
 # glibc determines buffering on first write to stdout - if we replace a TTY
 # destined stdout with a pipe destined stdout (e.g. pager), we want line
 # buffering
 if isatty(stdout):
     stdout = os.fdopen(stdout.fileno(), pycompat.sysstr('wb'), 1)
 
 if pycompat.osname == 'nt':
     from . import windows as platform
     stdout = platform.winstdout(stdout)
 else:
     from . import posix as platform
 
 _ = i18n._
 
 bindunixsocket = platform.bindunixsocket
 cachestat = platform.cachestat
 checkexec = platform.checkexec
 checklink = platform.checklink
 copymode = platform.copymode
 executablepath = platform.executablepath
 expandglobs = platform.expandglobs
 explainexit = platform.explainexit
 findexe = platform.findexe
 gethgcmd = platform.gethgcmd
 getuser = platform.getuser
 getpid = os.getpid
 groupmembers = platform.groupmembers
 groupname = platform.groupname
 hidewindow = platform.hidewindow
 isexec = platform.isexec
 isowner = platform.isowner
 listdir = osutil.listdir
 localpath = platform.localpath
 lookupreg = platform.lookupreg
 makedir = platform.makedir
 nlinks = platform.nlinks
 normpath = platform.normpath
 normcase = platform.normcase
 normcasespec = platform.normcasespec
 normcasefallback = platform.normcasefallback
 openhardlinks = platform.openhardlinks
 oslink = platform.oslink
 parsepatchoutput = platform.parsepatchoutput
 pconvert = platform.pconvert
 poll = platform.poll
 popen = platform.popen
 posixfile = platform.posixfile
 quotecommand = platform.quotecommand
 readpipe = platform.readpipe
 rename = platform.rename
 removedirs = platform.removedirs
 samedevice = platform.samedevice
 samefile = platform.samefile
 samestat = platform.samestat
 setbinary = platform.setbinary
 setflags = platform.setflags
 setsignalhandler = platform.setsignalhandler
 shellquote = platform.shellquote
 spawndetached = platform.spawndetached
 split = platform.split
 sshargs = platform.sshargs
 statfiles = getattr(osutil, 'statfiles', platform.statfiles)
 statisexec = platform.statisexec
 statislink = platform.statislink
 testpid = platform.testpid
 umask = platform.umask
 unlink = platform.unlink
 username = platform.username
 
 try:
     recvfds = osutil.recvfds
 except AttributeError:
     pass
 try:
     setprocname = osutil.setprocname
 except AttributeError:
     pass
 
 # Python compatibility
 
 _notset = object()
 
 # disable Python's problematic floating point timestamps (issue4836)
 # (Python hypocritically says you shouldn't change this behavior in
 # libraries, and sure enough Mercurial is not a library.)
 os.stat_float_times(False)
 
 def safehasattr(thing, attr):
     return getattr(thing, attr, _notset) is not _notset
 
 def bitsfrom(container):
     bits = 0
     for bit in container:
         bits |= bit
     return bits
 
 # python 2.6 still have deprecation warning enabled by default. We do not want
 # to display anything to standard user so detect if we are running test and
 # only use python deprecation warning in this case.
 _dowarn = bool(encoding.environ.get('HGEMITWARNINGS'))
 if _dowarn:
     # explicitly unfilter our warning for python 2.7
     #
     # The option of setting PYTHONWARNINGS in the test runner was investigated.
     # However, module name set through PYTHONWARNINGS was exactly matched, so
     # we cannot set 'mercurial' and have it match eg: 'mercurial.scmutil'. This
     # makes the whole PYTHONWARNINGS thing useless for our usecase.
     warnings.filterwarnings(r'default', r'', DeprecationWarning, r'mercurial')
     warnings.filterwarnings(r'default', r'', DeprecationWarning, r'hgext')
     warnings.filterwarnings(r'default', r'', DeprecationWarning, r'hgext3rd')
 
 def nouideprecwarn(msg, version, stacklevel=1):
     """Issue an python native deprecation warning
 
     This is a noop outside of tests, use 'ui.deprecwarn' when possible.
     """
     if _dowarn:
         msg += ("\n(compatibility will be dropped after Mercurial-%s,"
                 " update your code.)") % version
         warnings.warn(msg, DeprecationWarning, stacklevel + 1)
 
 DIGESTS = {
     'md5': hashlib.md5,
     'sha1': hashlib.sha1,
     'sha512': hashlib.sha512,
 }
 # List of digest types from strongest to weakest
 DIGESTS_BY_STRENGTH = ['sha512', 'sha1', 'md5']
 
 for k in DIGESTS_BY_STRENGTH:
     assert k in DIGESTS
 
 class digester(object):
     """helper to compute digests.
 
     This helper can be used to compute one or more digests given their name.
 
     >>> d = digester(['md5', 'sha1'])
     >>> d.update('foo')
     >>> [k for k in sorted(d)]
     ['md5', 'sha1']
     >>> d['md5']
     'acbd18db4cc2f85cedef654fccc4a4d8'
     >>> d['sha1']
     '0beec7b5ea3f0fdbc95d0dd47f3c5bc275da8a33'
     >>> digester.preferred(['md5', 'sha1'])
     'sha1'
     """
 
     def __init__(self, digests, s=''):
         self._hashes = {}
         for k in digests:
             if k not in DIGESTS:
                 raise Abort(_('unknown digest type: %s') % k)
             self._hashes[k] = DIGESTS[k]()
         if s:
             self.update(s)
 
     def update(self, data):
         for h in self._hashes.values():
             h.update(data)
 
     def __getitem__(self, key):
         if key not in DIGESTS:
             raise Abort(_('unknown digest type: %s') % k)
         return self._hashes[key].hexdigest()
 
     def __iter__(self):
         return iter(self._hashes)
 
     @staticmethod
     def preferred(supported):
         """returns the strongest digest type in both supported and DIGESTS."""
 
         for k in DIGESTS_BY_STRENGTH:
             if k in supported:
                 return k
         return None
 
 class digestchecker(object):
     """file handle wrapper that additionally checks content against a given
     size and digests.
 
         d = digestchecker(fh, size, {'md5': '...'})
 
     When multiple digests are given, all of them are validated.
     """
 
     def __init__(self, fh, size, digests):
         self._fh = fh
         self._size = size
         self._got = 0
         self._digests = dict(digests)
         self._digester = digester(self._digests.keys())
 
     def read(self, length=-1):
         content = self._fh.read(length)
         self._digester.update(content)
         self._got += len(content)
         return content
 
     def validate(self):
         if self._size != self._got:
             raise Abort(_('size mismatch: expected %d, got %d') %
                 (self._size, self._got))
         for k, v in self._digests.items():
             if v != self._digester[k]:
                 # i18n: first parameter is a digest name
                 raise Abort(_('%s mismatch: expected %s, got %s') %
                     (k, v, self._digester[k]))
 
 try:
     buffer = buffer
 except NameError:
     if not pycompat.ispy3:
         def buffer(sliceable, offset=0, length=None):
             if length is not None:
                 return sliceable[offset:offset + length]
             return sliceable[offset:]
     else:
         def buffer(sliceable, offset=0, length=None):
             if length is not None:
                 return memoryview(sliceable)[offset:offset + length]
             return memoryview(sliceable)[offset:]
 
 closefds = pycompat.osname == 'posix'
 
 _chunksize = 4096
 
 class bufferedinputpipe(object):
     """a manually buffered input pipe
 
     Python will not let us use buffered IO and lazy reading with 'polling' at
     the same time. We cannot probe the buffer state and select will not detect
     that data are ready to read if they are already buffered.
 
     This class let us work around that by implementing its own buffering
     (allowing efficient readline) while offering a way to know if the buffer is
     empty from the output (allowing collaboration of the buffer with polling).
 
     This class lives in the 'util' module because it makes use of the 'os'
     module from the python stdlib.
     """
 
     def __init__(self, input):
         self._input = input
         self._buffer = []
         self._eof = False
         self._lenbuf = 0
 
     @property
     def hasbuffer(self):
         """True is any data is currently buffered
 
         This will be used externally a pre-step for polling IO. If there is
         already data then no polling should be set in place."""
         return bool(self._buffer)
 
     @property
     def closed(self):
         return self._input.closed
 
     def fileno(self):
         return self._input.fileno()
 
     def close(self):
         return self._input.close()
 
     def read(self, size):
         while (not self._eof) and (self._lenbuf < size):
             self._fillbuffer()
         return self._frombuffer(size)
 
     def readline(self, *args, **kwargs):
         if 1 < len(self._buffer):
             # this should not happen because both read and readline end with a
             # _frombuffer call that collapse it.
             self._buffer = [''.join(self._buffer)]
             self._lenbuf = len(self._buffer[0])
         lfi = -1
         if self._buffer:
             lfi = self._buffer[-1].find('\n')
         while (not self._eof) and lfi < 0:
             self._fillbuffer()
             if self._buffer:
                 lfi = self._buffer[-1].find('\n')
         size = lfi + 1
         if lfi < 0: # end of file
             size = self._lenbuf
         elif 1 < len(self._buffer):
             # we need to take previous chunks into account
             size += self._lenbuf - len(self._buffer[-1])
         return self._frombuffer(size)
 
     def _frombuffer(self, size):
         """return at most 'size' data from the buffer
 
         The data are removed from the buffer."""
         if size == 0 or not self._buffer:
             return ''
         buf = self._buffer[0]
         if 1 < len(self._buffer):
             buf = ''.join(self._buffer)
 
         data = buf[:size]
         buf = buf[len(data):]
         if buf:
             self._buffer = [buf]
             self._lenbuf = len(buf)
         else:
             self._buffer = []
             self._lenbuf = 0
         return data
 
     def _fillbuffer(self):
         """read data to the buffer"""
         data = os.read(self._input.fileno(), _chunksize)
         if not data:
             self._eof = True
         else:
             self._lenbuf += len(data)
             self._buffer.append(data)
 
 def popen2(cmd, env=None, newlines=False):
     # Setting bufsize to -1 lets the system decide the buffer size.
     # The default for bufsize is 0, meaning unbuffered. This leads to
     # poor performance on Mac OS X: http://bugs.python.org/issue4194
     p = subprocess.Popen(cmd, shell=True, bufsize=-1,
                          close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE,
                          universal_newlines=newlines,
                          env=env)
     return p.stdin, p.stdout
 
 def popen3(cmd, env=None, newlines=False):
     stdin, stdout, stderr, p = popen4(cmd, env, newlines)
     return stdin, stdout, stderr
 
 def popen4(cmd, env=None, newlines=False, bufsize=-1):
     p = subprocess.Popen(cmd, shell=True, bufsize=bufsize,
                          close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE,
                          stderr=subprocess.PIPE,
                          universal_newlines=newlines,
                          env=env)
     return p.stdin, p.stdout, p.stderr, p
 
 def version():
     """Return version information if available."""
     try:
         from . import __version__
         return __version__.version
     except ImportError:
         return 'unknown'
 
 def versiontuple(v=None, n=4):
     """Parses a Mercurial version string into an N-tuple.
 
     The version string to be parsed is specified with the ``v`` argument.
     If it isn't defined, the current Mercurial version string will be parsed.
 
     ``n`` can be 2, 3, or 4. Here is how some version strings map to
     returned values:
 
     >>> v = '3.6.1+190-df9b73d2d444'
     >>> versiontuple(v, 2)
     (3, 6)
     >>> versiontuple(v, 3)
     (3, 6, 1)
     >>> versiontuple(v, 4)
     (3, 6, 1, '190-df9b73d2d444')
 
     >>> versiontuple('3.6.1+190-df9b73d2d444+20151118')
     (3, 6, 1, '190-df9b73d2d444+20151118')
 
     >>> v = '3.6'
     >>> versiontuple(v, 2)
     (3, 6)
     >>> versiontuple(v, 3)
     (3, 6, None)
     >>> versiontuple(v, 4)
     (3, 6, None, None)
 
     >>> v = '3.9-rc'
     >>> versiontuple(v, 2)
     (3, 9)
     >>> versiontuple(v, 3)
     (3, 9, None)
     >>> versiontuple(v, 4)
     (3, 9, None, 'rc')
 
     >>> v = '3.9-rc+2-02a8fea4289b'
     >>> versiontuple(v, 2)
     (3, 9)
     >>> versiontuple(v, 3)
     (3, 9, None)
     >>> versiontuple(v, 4)
     (3, 9, None, 'rc+2-02a8fea4289b')
     """
     if not v:
         v = version()
     parts = remod.split('[\+-]', v, 1)
     if len(parts) == 1:
         vparts, extra = parts[0], None
     else:
         vparts, extra = parts
 
     vints = []
     for i in vparts.split('.'):
         try:
             vints.append(int(i))
         except ValueError:
             break
     # (3, 6) -> (3, 6, None)
     while len(vints) < 3:
         vints.append(None)
 
     if n == 2:
         return (vints[0], vints[1])
     if n == 3:
         return (vints[0], vints[1], vints[2])
     if n == 4:
         return (vints[0], vints[1], vints[2], extra)
 
 # used by parsedate
 defaultdateformats = (
     '%Y-%m-%dT%H:%M:%S', # the 'real' ISO8601
     '%Y-%m-%dT%H:%M',    #   without seconds
     '%Y-%m-%dT%H%M%S',   # another awful but legal variant without :
     '%Y-%m-%dT%H%M',     #   without seconds
     '%Y-%m-%d %H:%M:%S', # our common legal variant
     '%Y-%m-%d %H:%M',    #   without seconds
     '%Y-%m-%d %H%M%S',   # without :
     '%Y-%m-%d %H%M',     #   without seconds
     '%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',
     '%a, %d %b %Y %H:%M:%S',        #  GNU coreutils "/bin/date --rfc-2822"
     '%b %d %H:%M:%S %Y',
     '%b %d %I:%M:%S%p %Y',
     '%b %d %H:%M:%S',
     '%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:%S%p',
     '%H:%M',
     '%I:%M%p',
 )
 
 extendeddateformats = defaultdateformats + (
     "%Y",
     "%Y-%m",
     "%b",
     "%b %Y",
     )
 
 def cachefunc(func):
     '''cache the result of function calls'''
     # XXX doesn't handle keywords args
     if func.__code__.co_argcount == 0:
         cache = []
         def f():
             if len(cache) == 0:
                 cache.append(func())
             return cache[0]
         return f
     cache = {}
     if func.__code__.co_argcount == 1:
         # we gain a small amount of time because
         # we don't need to pack/unpack the list
         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
 
 class sortdict(collections.OrderedDict):
     '''a simple sorted dictionary
 
     >>> d1 = sortdict([('a', 0), ('b', 1)])
     >>> d2 = d1.copy()
     >>> d2
     sortdict([('a', 0), ('b', 1)])
     >>> d2.update([('a', 2)])
     >>> d2.keys() # should still be in last-set order
     ['b', 'a']
     '''
 
     def __setitem__(self, key, value):
         if key in self:
             del self[key]
         super(sortdict, self).__setitem__(key, value)
 
 class _lrucachenode(object):
     """A node in a doubly linked list.
 
     Holds a reference to nodes on either side as well as a key-value
     pair for the dictionary entry.
     """
     __slots__ = (u'next', u'prev', u'key', u'value')
 
     def __init__(self):
         self.next = None
         self.prev = None
 
         self.key = _notset
         self.value = None
 
     def markempty(self):
         """Mark the node as emptied."""
         self.key = _notset
 
 class lrucachedict(object):
     """Dict that caches most recent accesses and sets.
 
     The dict consists of an actual backing dict - indexed by original
     key - and a doubly linked circular list defining the order of entries in
     the cache.
 
     The head node is the newest entry in the cache. If the cache is full,
     we recycle head.prev and make it the new head. Cache accesses result in
     the node being moved to before the existing head and being marked as the
     new head node.
     """
     def __init__(self, max):
         self._cache = {}
 
         self._head = head = _lrucachenode()
         head.prev = head
         head.next = head
         self._size = 1
         self._capacity = max
 
     def __len__(self):
         return len(self._cache)
 
     def __contains__(self, k):
         return k in self._cache
 
     def __iter__(self):
         # We don't have to iterate in cache order, but why not.
         n = self._head
         for i in range(len(self._cache)):
             yield n.key
             n = n.next
 
     def __getitem__(self, k):
         node = self._cache[k]
         self._movetohead(node)
         return node.value
 
     def __setitem__(self, k, v):
         node = self._cache.get(k)
         # Replace existing value and mark as newest.
         if node is not None:
             node.value = v
             self._movetohead(node)
             return
 
         if self._size < self._capacity:
             node = self._addcapacity()
         else:
             # Grab the last/oldest item.
             node = self._head.prev
 
         # At capacity. Kill the old entry.
         if node.key is not _notset:
             del self._cache[node.key]
 
         node.key = k
         node.value = v
         self._cache[k] = node
         # And mark it as newest entry. No need to adjust order since it
         # is already self._head.prev.
         self._head = node
 
     def __delitem__(self, k):
         node = self._cache.pop(k)
         node.markempty()
 
         # Temporarily mark as newest item before re-adjusting head to make
         # this node the oldest item.
         self._movetohead(node)
         self._head = node.next
 
     # Additional dict methods.
 
     def get(self, k, default=None):
         try:
             return self._cache[k].value
         except KeyError:
             return default
 
     def clear(self):
         n = self._head
         while n.key is not _notset:
             n.markempty()
             n = n.next
 
         self._cache.clear()
 
     def copy(self):
         result = lrucachedict(self._capacity)
         n = self._head.prev
         # Iterate in oldest-to-newest order, so the copy has the right ordering
         for i in range(len(self._cache)):
             result[n.key] = n.value
             n = n.prev
         return result
 
     def _movetohead(self, node):
         """Mark a node as the newest, making it the new head.
 
         When a node is accessed, it becomes the freshest entry in the LRU
         list, which is denoted by self._head.
 
         Visually, let's make ``N`` the new head node (* denotes head):
 
             previous/oldest <-> head <-> next/next newest
 
             ----<->--- A* ---<->-----
             |                       |
             E <-> D <-> N <-> C <-> B
 
         To:
 
             ----<->--- N* ---<->-----
             |                       |
             E <-> D <-> C <-> B <-> A
 
         This requires the following moves:
 
            C.next = D  (node.prev.next = node.next)
            D.prev = C  (node.next.prev = node.prev)
            E.next = N  (head.prev.next = node)
            N.prev = E  (node.prev = head.prev)
            N.next = A  (node.next = head)
            A.prev = N  (head.prev = node)
         """
         head = self._head
         # C.next = D
         node.prev.next = node.next
         # D.prev = C
         node.next.prev = node.prev
         # N.prev = E
         node.prev = head.prev
         # N.next = A
         # It is tempting to do just "head" here, however if node is
         # adjacent to head, this will do bad things.
         node.next = head.prev.next
         # E.next = N
         node.next.prev = node
         # A.prev = N
         node.prev.next = node
 
         self._head = node
 
     def _addcapacity(self):
         """Add a node to the circular linked list.
 
         The new node is inserted before the head node.
         """
         head = self._head
         node = _lrucachenode()
         head.prev.next = node
         node.prev = head.prev
         node.next = head
         head.prev = node
         self._size += 1
         return node
 
 def lrucachefunc(func):
     '''cache most recent results of function calls'''
     cache = {}
     order = collections.deque()
     if func.__code__.co_argcount == 1:
         def f(arg):
             if arg not in cache:
                 if len(cache) > 20:
                     del cache[order.popleft()]
                 cache[arg] = func(arg)
             else:
                 order.remove(arg)
             order.append(arg)
             return cache[arg]
     else:
         def f(*args):
             if args not in cache:
                 if len(cache) > 20:
                     del cache[order.popleft()]
                 cache[args] = func(*args)
             else:
                 order.remove(args)
             order.append(args)
             return cache[args]
 
     return f
 
 class propertycache(object):
     def __init__(self, func):
         self.func = func
         self.name = func.__name__
     def __get__(self, obj, type=None):
         result = self.func(obj)
         self.cachevalue(obj, result)
         return result
 
     def cachevalue(self, obj, value):
         # __dict__ assignment required to bypass __setattr__ (eg: repoview)
         obj.__dict__[self.name] = value
 
 def pipefilter(s, cmd):
     '''filter string S through command CMD, returning its output'''
     p = subprocess.Popen(cmd, shell=True, close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE)
     pout, perr = p.communicate(s)
     return pout
 
 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:
         infd, inname = tempfile.mkstemp(prefix='hg-filter-in-')
         fp = os.fdopen(infd, pycompat.sysstr('wb'))
         fp.write(s)
         fp.close()
         outfd, outname = tempfile.mkstemp(prefix='hg-filter-out-')
         os.close(outfd)
         cmd = cmd.replace('INFILE', inname)
         cmd = cmd.replace('OUTFILE', outname)
         code = os.system(cmd)
         if pycompat.sysplatform == 'OpenVMS' and code & 1:
             code = 0
         if code:
             raise Abort(_("command '%s' failed: %s") %
                         (cmd, explainexit(code)))
         return readfile(outname)
     finally:
         try:
             if inname:
                 os.unlink(inname)
         except OSError:
             pass
         try:
             if outname:
                 os.unlink(outname)
         except OSError:
             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)
 
 def binary(s):
     """return true if a string is binary data"""
     return bool(s and '\0' in s)
 
 def increasingchunks(source, min=1024, max=65536):
     '''return no less than min bytes per chunk while data remains,
     doubling min after each chunk until it reaches max'''
     def log2(x):
         if not x:
             return 0
         i = 0
         while x:
             x >>= 1
             i += 1
         return i - 1
 
     buf = []
     blen = 0
     for chunk in source:
         buf.append(chunk)
         blen += len(chunk)
         if blen >= min:
             if min < max:
                 min = min << 1
                 nmin = 1 << log2(blen)
                 if nmin > min:
                     min = nmin
                 if min > max:
                     min = max
             yield ''.join(buf)
             blen = 0
             buf = []
     if buf:
         yield ''.join(buf)
 
 Abort = error.Abort
 
 def always(fn):
     return True
 
 def never(fn):
     return False
 
 def nogc(func):
     """disable garbage collector
 
     Python's garbage collector triggers a GC each time a certain number of
     container objects (the number being defined by gc.get_threshold()) are
     allocated even when marked not to be tracked by the collector. Tracking has
     no effect on when GCs are triggered, only on what objects the GC looks
     into. As a workaround, disable GC while building complex (huge)
     containers.
 
     This garbage collector issue have been fixed in 2.7.
     """
     if sys.version_info >= (2, 7):
         return func
     def wrapper(*args, **kwargs):
         gcenabled = gc.isenabled()
         gc.disable()
         try:
             return func(*args, **kwargs)
         finally:
             if gcenabled:
                 gc.enable()
     return wrapper
 
 def pathto(root, n1, n2):
     '''return the relative path from one place to another.
     root should use os.sep to separate directories
     n1 should use os.sep to separate directories
     n2 should use "/" to separate directories
     returns an os.sep-separated path.
 
     If n1 is a relative path, it's assumed it's
     relative to root.
     n2 should always be relative to root.
     '''
     if not n1:
         return localpath(n2)
     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))
     a, b = splitpath(n1), n2.split('/')
     a.reverse()
     b.reverse()
     while a and b and a[-1] == b[-1]:
         a.pop()
         b.pop()
     b.reverse()
     return pycompat.ossep.join((['..'] * len(a)) + b) or '.'
 
 def mainfrozen():
     """return True if we are a frozen executable.
 
     The code supports py2exe (most common, Windows only) and tools/freeze
     (portable, not much used).
     """
     return (safehasattr(sys, "frozen") or # new py2exe
             safehasattr(sys, "importers") or # old py2exe
             imp.is_frozen(u"__main__")) # tools/freeze
 
 # the location of data files matching the source code
 if mainfrozen() and getattr(sys, 'frozen', None) != 'macosx_app':
     # executable version (py2exe) doesn't support __file__
     datapath = os.path.dirname(pycompat.sysexecutable)
 else:
     datapath = os.path.dirname(pycompat.fsencode(__file__))
 
 i18n.setdatapath(datapath)
 
 _hgexecutable = None
 
 def hgexecutable():
     """return location of the 'hg' executable.
 
     Defaults to $HG or 'hg' in the search path.
     """
     if _hgexecutable is None:
         hg = encoding.environ.get('HG')
         mainmod = sys.modules[pycompat.sysstr('__main__')]
         if hg:
             _sethgexecutable(hg)
         elif mainfrozen():
             if getattr(sys, 'frozen', None) == 'macosx_app':
                 # Env variable set by py2app
                 _sethgexecutable(encoding.environ['EXECUTABLEPATH'])
             else:
                 _sethgexecutable(pycompat.sysexecutable)
         elif (os.path.basename(
             pycompat.fsencode(getattr(mainmod, '__file__', ''))) == 'hg'):
             _sethgexecutable(pycompat.fsencode(mainmod.__file__))
         else:
             exe = findexe('hg') or os.path.basename(sys.argv[0])
             _sethgexecutable(exe)
     return _hgexecutable
 
 def _sethgexecutable(path):
     """set location of the 'hg' executable"""
     global _hgexecutable
     _hgexecutable = path
 
 def _isstdout(f):
     fileno = getattr(f, 'fileno', None)
     return fileno and fileno() == sys.__stdout__.fileno()
 
 def shellenviron(environ=None):
     """return environ with optional override, useful for shelling out"""
     def py2shell(val):
         'convert python object into string that is useful to shell'
         if val is None or val is False:
             return '0'
         if val is True:
             return '1'
         return str(val)
     env = dict(encoding.environ)
     if environ:
         env.update((k, py2shell(v)) for k, v in environ.iteritems())
     env['HG'] = hgexecutable()
     return env
 
 def system(cmd, environ=None, cwd=None, out=None):
     '''enhanced shell command execution.
     run with environment maybe modified, maybe in different dir.
 
     if out is specified, it is assumed to be a file-like object that has a
     write() method. stdout and stderr will be redirected to out.'''
     try:
         stdout.flush()
     except Exception:
         pass
     cmd = quotecommand(cmd)
     if pycompat.sysplatform == 'plan9' and (sys.version_info[0] == 2
                                     and sys.version_info[1] < 7):
         # subprocess kludge to work around issues in half-baked Python
         # ports, notably bichued/python:
         if not cwd is None:
             os.chdir(cwd)
         rc = os.system(cmd)
     else:
         env = shellenviron(environ)
         if out is None or _isstdout(out):
             rc = subprocess.call(cmd, shell=True, close_fds=closefds,
                                  env=env, cwd=cwd)
         else:
             proc = subprocess.Popen(cmd, shell=True, close_fds=closefds,
                                     env=env, cwd=cwd, stdout=subprocess.PIPE,
                                     stderr=subprocess.STDOUT)
             for line in iter(proc.stdout.readline, ''):
                 out.write(line)
             proc.wait()
             rc = proc.returncode
         if pycompat.sysplatform == 'OpenVMS' and rc & 1:
             rc = 0
     return rc
 
 def checksignature(func):
     '''wrap a function with code to check for calling errors'''
     def check(*args, **kwargs):
         try:
             return func(*args, **kwargs)
         except TypeError:
             if len(traceback.extract_tb(sys.exc_info()[2])) == 1:
                 raise error.SignatureError
             raise
 
     return check
 
 # a whilelist of known filesystems where hardlink works reliably
 _hardlinkfswhitelist = {
     'btrfs',
     'ext2',
     'ext3',
     'ext4',
     'hfs',
     'jfs',
     'reiserfs',
     'tmpfs',
     'ufs',
     'xfs',
     'zfs',
 }
 
 def copyfile(src, dest, hardlink=False, copystat=False, checkambig=False):
     '''copy a file, preserving mode and optionally other stat info like
     atime/mtime
 
     checkambig argument is used with filestat, and is useful only if
     destination file is guarded by any lock (e.g. repo.lock or
     repo.wlock).
 
     copystat and checkambig should be exclusive.
     '''
     assert not (copystat and checkambig)
     oldstat = None
     if os.path.lexists(dest):
         if checkambig:
             oldstat = checkambig and filestat(dest)
         unlink(dest)
     if hardlink:
         # Hardlinks are problematic on CIFS (issue4546), do not allow hardlinks
         # unless we are confident that dest is on a whitelisted filesystem.
         try:
             fstype = getfstype(os.path.dirname(dest))
         except OSError:
             fstype = None
         if fstype not in _hardlinkfswhitelist:
             hardlink = False
     if hardlink:
         try:
             oslink(src, dest)
             return
         except (IOError, OSError):
             pass # fall back to normal copy
     if os.path.islink(src):
         os.symlink(os.readlink(src), dest)
         # copytime is ignored for symlinks, but in general copytime isn't needed
         # for them anyway
     else:
         try:
             shutil.copyfile(src, dest)
             if copystat:
                 # copystat also copies mode
                 shutil.copystat(src, dest)
             else:
                 shutil.copymode(src, dest)
                 if oldstat and oldstat.stat:
                     newstat = filestat(dest)
                     if newstat.isambig(oldstat):
                         # stat of copied file is ambiguous to original one
                         advanced = (oldstat.stat.st_mtime + 1) & 0x7fffffff
                         os.utime(dest, (advanced, advanced))
         except shutil.Error as inst:
             raise Abort(str(inst))
 
 def copyfiles(src, dst, hardlink=None, progress=lambda t, pos: None):
     """Copy a directory tree using hardlinks if possible."""
     num = 0
 
     gettopic = lambda: hardlink and _('linking') or _('copying')
 
     if os.path.isdir(src):
         if hardlink is None:
             hardlink = (os.stat(src).st_dev ==
                         os.stat(os.path.dirname(dst)).st_dev)
         topic = gettopic()
         os.mkdir(dst)
         for name, kind in listdir(src):
             srcname = os.path.join(src, name)
             dstname = os.path.join(dst, name)
             def nprog(t, pos):
                 if pos is not None:
                     return progress(t, pos + num)
             hardlink, n = copyfiles(srcname, dstname, hardlink, progress=nprog)
             num += n
     else:
         if hardlink is None:
             hardlink = (os.stat(os.path.dirname(src)).st_dev ==
                         os.stat(os.path.dirname(dst)).st_dev)
         topic = gettopic()
 
         if hardlink:
             try:
                 oslink(src, dst)
             except (IOError, OSError):
                 hardlink = False
                 shutil.copy(src, dst)
         else:
             shutil.copy(src, dst)
         num += 1
         progress(topic, num)
     progress(topic, None)
 
     return hardlink, num
 
 _winreservednames = '''con prn aux nul
     com1 com2 com3 com4 com5 com6 com7 com8 com9
     lpt1 lpt2 lpt3 lpt4 lpt5 lpt6 lpt7 lpt8 lpt9'''.split()
 _winreservedchars = ':*?"<>|'
 def checkwinfilename(path):
     r'''Check that the base-relative path is a valid filename on Windows.
     Returns None if the path is ok, or a UI string describing the problem.
 
     >>> checkwinfilename("just/a/normal/path")
     >>> checkwinfilename("foo/bar/con.xml")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename("foo/con.xml/bar")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/xml.con")
     >>> checkwinfilename("foo/bar/AUX/bla.txt")
     "filename contains 'AUX', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/bla:.txt")
     "filename contains ':', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/b\07la.txt")
     "filename contains '\\x07', which is invalid on Windows"
     >>> checkwinfilename("foo/bar/bla ")
     "filename ends with ' ', which is not allowed on Windows"
     >>> checkwinfilename("../bar")
     >>> checkwinfilename("foo\\")
     "filename ends with '\\', which is invalid on Windows"
     >>> checkwinfilename("foo\\/bar")
     "directory name ends with '\\', which is invalid on Windows"
     '''
     if path.endswith('\\'):
         return _("filename ends with '\\', which is invalid on Windows")
     if '\\/' in path:
         return _("directory name ends with '\\', which is invalid on Windows")
     for n in path.replace('\\', '/').split('/'):
         if not n:
             continue
         for c in pycompat.bytestr(n):
             if c in _winreservedchars:
                 return _("filename contains '%s', which is reserved "
                          "on Windows") % c
             if ord(c) <= 31:
                 return _("filename contains %r, which is invalid "
                          "on Windows") % c
         base = n.split('.')[0]
         if base and base.lower() in _winreservednames:
             return _("filename contains '%s', which is reserved "
                      "on Windows") % base
         t = n[-1]
         if t in '. ' and n not in '..':
             return _("filename ends with '%s', which is not allowed "
                      "on Windows") % t
 
 if pycompat.osname == 'nt':
     checkosfilename = checkwinfilename
     timer = time.clock
 else:
     checkosfilename = platform.checkosfilename
     timer = time.time
 
 if safehasattr(time, "perf_counter"):
     timer = time.perf_counter
 
 def makelock(info, pathname):
     try:
         return os.symlink(info, pathname)
     except OSError as why:
         if why.errno == errno.EEXIST:
             raise
     except AttributeError: # no symlink in os
         pass
 
     ld = os.open(pathname, os.O_CREAT | os.O_WRONLY | os.O_EXCL)
     os.write(ld, info)
     os.close(ld)
 
 def readlock(pathname):
     try:
         return os.readlink(pathname)
     except OSError as why:
         if why.errno not in (errno.EINVAL, errno.ENOSYS):
             raise
     except AttributeError: # no symlink in os
         pass
     fp = posixfile(pathname)
     r = fp.read()
     fp.close()
     return r
 
 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)
 
 # File system features
 
 def fscasesensitive(path):
     """
     Return true if the given path is on a case-sensitive filesystem
 
     Requires a path (like /foo/.hg) ending with a foldable final
     directory component.
     """
     s1 = os.lstat(path)
     d, b = os.path.split(path)
     b2 = b.upper()
     if b == b2:
         b2 = b.lower()
         if b == b2:
             return True # no evidence against case sensitivity
     p2 = os.path.join(d, b2)
     try:
         s2 = os.lstat(p2)
         if s2 == s1:
             return False
         return True
     except OSError:
         return True
 
 try:
     import re2
     _re2 = None
 except ImportError:
     _re2 = False
 
 class _re(object):
     def _checkre2(self):
         global _re2
         try:
             # check if match works, see issue3964
             _re2 = bool(re2.match(r'\[([^\[]+)\]', '[ui]'))
         except ImportError:
             _re2 = False
 
     def compile(self, pat, flags=0):
         '''Compile a regular expression, using re2 if possible
 
         For best performance, use only re2-compatible regexp features. The
         only flags from the re module that are re2-compatible are
         IGNORECASE and MULTILINE.'''
         if _re2 is None:
             self._checkre2()
         if _re2 and (flags & ~(remod.IGNORECASE | remod.MULTILINE)) == 0:
             if flags & remod.IGNORECASE:
                 pat = '(?i)' + pat
             if flags & remod.MULTILINE:
                 pat = '(?m)' + pat
             try:
                 return re2.compile(pat)
             except re2.error:
                 pass
         return remod.compile(pat, flags)
 
     @propertycache
     def escape(self):
         '''Return the version of escape corresponding to self.compile.
 
         This is imperfect because whether re2 or re is used for a particular
         function depends on the flags, etc, but it's the best we can do.
         '''
         global _re2
         if _re2 is None:
             self._checkre2()
         if _re2:
             return re2.escape
         else:
             return remod.escape
 
 re = _re()
 
 _fspathcache = {}
 def fspath(name, root):
     '''Get name in the case stored in the filesystem
 
     The name should be relative to root, and be normcase-ed for efficiency.
 
     Note that this function is unnecessary, and should not be
     called, for case-sensitive filesystems (simply because it's expensive).
 
     The root should be normcase-ed, too.
     '''
     def _makefspathcacheentry(dir):
         return dict((normcase(n), n) for n in os.listdir(dir))
 
     seps = pycompat.ossep
     if pycompat.osaltsep:
         seps = seps + pycompat.osaltsep
     # Protect backslashes. This gets silly very quickly.
     seps.replace('\\','\\\\')
     pattern = remod.compile(br'([^%s]+)|([%s]+)' % (seps, seps))
     dir = os.path.normpath(root)
     result = []
     for part, sep in pattern.findall(name):
         if sep:
             result.append(sep)
             continue
 
         if dir not in _fspathcache:
             _fspathcache[dir] = _makefspathcacheentry(dir)
         contents = _fspathcache[dir]
 
         found = contents.get(part)
         if not found:
             # retry "once per directory" per "dirstate.walk" which
             # may take place for each patches of "hg qpush", for example
             _fspathcache[dir] = contents = _makefspathcacheentry(dir)
             found = contents.get(part)
 
         result.append(found or part)
         dir = os.path.join(dir, part)
 
     return ''.join(result)
 
 def getfstype(dirpath):
     '''Get the filesystem type name from a directory (best-effort)
 
     Returns None if we are unsure. Raises OSError on ENOENT, EPERM, etc.
     '''
     return getattr(osutil, 'getfstype', lambda x: None)(dirpath)
 
 def checknlink(testfile):
     '''check whether hardlink count reporting works properly'''
 
     # testfile may be open, so we need a separate file for checking to
     # work around issue2543 (or testfile may get lost on Samba shares)
     f1 = testfile + ".hgtmp1"
     if os.path.lexists(f1):
         return False
     try:
         posixfile(f1, 'w').close()
     except IOError:
         try:
             os.unlink(f1)
         except OSError:
             pass
         return False
 
     f2 = testfile + ".hgtmp2"
     fd = None
     try:
         oslink(f1, f2)
         # nlinks() may behave differently for files on Windows shares if
         # the file is open.
         fd = posixfile(f2)
         return nlinks(f2) > 1
     except OSError:
         return False
     finally:
         if fd is not None:
             fd.close()
         for f in (f1, f2):
             try:
                 os.unlink(f)
             except OSError:
                 pass
 
 def endswithsep(path):
     '''Check path ends with os.sep or os.altsep.'''
     return (path.endswith(pycompat.ossep)
             or pycompat.osaltsep and path.endswith(pycompat.osaltsep))
 
 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(pycompat.ossep)
 
 def gui():
     '''Are we running in a GUI?'''
     if pycompat.sysplatform == 'darwin':
         if 'SSH_CONNECTION' in encoding.environ:
             # handle SSH access to a box where the user is logged in
             return False
         elif getattr(osutil, 'isgui', None):
             # check if a CoreGraphics session is available
             return osutil.isgui()
         else:
             # pure build; use a safe default
             return True
     else:
         return pycompat.osname == "nt" or encoding.environ.get("DISPLAY")
 
 def mktempcopy(name, emptyok=False, createmode=None):
     """Create a temporary file with the same contents from name
 
     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.
     """
     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.
     copymode(name, temp, createmode)
     if emptyok:
         return temp
     try:
         try:
             ifp = posixfile(name, "rb")
         except IOError as 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: # re-raises
         try: os.unlink(temp)
         except OSError: pass
         raise
     return temp
 
 class filestat(object):
     """help to exactly detect change of a file
 
     'stat' attribute is result of 'os.stat()' if specified 'path'
     exists. Otherwise, it is None. This can avoid preparative
     'exists()' examination on client side of this class.
     """
     def __init__(self, path):
         try:
             self.stat = os.stat(path)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             self.stat = None
 
     __hash__ = object.__hash__
 
     def __eq__(self, old):
         try:
             # if ambiguity between stat of new and old file is
             # avoided, comparison of size, ctime and mtime is enough
             # to exactly detect change of a file regardless of platform
             return (self.stat.st_size == old.stat.st_size and
                     self.stat.st_ctime == old.stat.st_ctime and
                     self.stat.st_mtime == old.stat.st_mtime)
         except AttributeError:
             return False
 
     def isambig(self, old):
         """Examine whether new (= self) stat is ambiguous against old one
 
         "S[N]" below means stat of a file at N-th change:
 
         - S[n-1].ctime  < S[n].ctime: can detect change of a file
         - S[n-1].ctime == S[n].ctime
           - S[n-1].ctime  < S[n].mtime: means natural advancing (*1)
           - S[n-1].ctime == S[n].mtime: is ambiguous (*2)
           - S[n-1].ctime  > S[n].mtime: never occurs naturally (don't care)
         - S[n-1].ctime  > S[n].ctime: never occurs naturally (don't care)
 
         Case (*2) above means that a file was changed twice or more at
         same time in sec (= S[n-1].ctime), and comparison of timestamp
         is ambiguous.
 
         Base idea to avoid such ambiguity is "advance mtime 1 sec, if
         timestamp is ambiguous".
 
         But advancing mtime only in case (*2) doesn't work as
         expected, because naturally advanced S[n].mtime in case (*1)
         might be equal to manually advanced S[n-1 or earlier].mtime.
 
         Therefore, all "S[n-1].ctime == S[n].ctime" cases should be
         treated as ambiguous regardless of mtime, to avoid overlooking
         by confliction between such mtime.
 
         Advancing mtime "if isambig(oldstat)" ensures "S[n-1].mtime !=
         S[n].mtime", even if size of a file isn't changed.
         """
         try:
             return (self.stat.st_ctime == old.stat.st_ctime)
         except AttributeError:
             return False
 
     def avoidambig(self, path, old):
         """Change file stat of specified path to avoid ambiguity
 
         'old' should be previous filestat of 'path'.
 
         This skips avoiding ambiguity, if a process doesn't have
         appropriate privileges for 'path'.
         """
         advanced = (old.stat.st_mtime + 1) & 0x7fffffff
         try:
             os.utime(path, (advanced, advanced))
         except OSError as inst:
             if inst.errno == errno.EPERM:
                 # utime() on the file created by another user causes EPERM,
                 # if a process doesn't have appropriate privileges
                 return
             raise
 
     def __ne__(self, other):
         return not self == other
 
 class atomictempfile(object):
     '''writable file object that atomically updates a file
 
     All writes will go to a temporary copy of the original file. Call
     close() when you are done writing, and atomictempfile will rename
     the temporary copy to the original name, making the changes
     visible. If the object is destroyed without being closed, all your
     writes are discarded.
 
     checkambig argument of constructor is used with filestat, and is
     useful only if target file is guarded by any lock (e.g. repo.lock
     or repo.wlock).
     '''
     def __init__(self, name, mode='w+b', createmode=None, checkambig=False):
         self.__name = name      # permanent name
         self._tempname = mktempcopy(name, emptyok=('w' in mode),
                                     createmode=createmode)
         self._fp = posixfile(self._tempname, mode)
         self._checkambig = checkambig
 
         # delegated methods
         self.read = self._fp.read
         self.write = self._fp.write
         self.seek = self._fp.seek
         self.tell = self._fp.tell
         self.fileno = self._fp.fileno
 
     def close(self):
         if not self._fp.closed:
             self._fp.close()
             filename = localpath(self.__name)
             oldstat = self._checkambig and filestat(filename)
             if oldstat and oldstat.stat:
                 rename(self._tempname, filename)
                 newstat = filestat(filename)
                 if newstat.isambig(oldstat):
                     # stat of changed file is ambiguous to original one
                     advanced = (oldstat.stat.st_mtime + 1) & 0x7fffffff
                     os.utime(filename, (advanced, advanced))
             else:
                 rename(self._tempname, filename)
 
     def discard(self):
         if not self._fp.closed:
             try:
                 os.unlink(self._tempname)
             except OSError:
                 pass
             self._fp.close()
 
     def __del__(self):
         if safehasattr(self, '_fp'): # constructor actually did something
             self.discard()
 
     def __enter__(self):
         return self
 
     def __exit__(self, exctype, excvalue, traceback):
         if exctype is not None:
             self.discard()
         else:
             self.close()
 
 def unlinkpath(f, ignoremissing=False):
     """unlink and remove the directory if it is empty"""
     if ignoremissing:
         tryunlink(f)
     else:
         unlink(f)
     # try removing directories that might now be empty
     try:
         removedirs(os.path.dirname(f))
     except OSError:
         pass
 
 def tryunlink(f):
     """Attempt to remove a file, ignoring ENOENT errors."""
     try:
         unlink(f)
     except OSError as e:
         if e.errno != errno.ENOENT:
             raise
 
 def makedirs(name, mode=None, notindexed=False):
     """recursive directory creation with parent mode inheritance
 
     Newly created directories are marked as "not to be indexed by
     the content indexing service", if ``notindexed`` is specified
     for "write" mode access.
     """
     try:
         makedir(name, notindexed)
     except OSError as err:
         if err.errno == errno.EEXIST:
             return
         if err.errno != errno.ENOENT or not name:
             raise
         parent = os.path.dirname(os.path.abspath(name))
         if parent == name:
             raise
         makedirs(parent, mode, notindexed)
         try:
             makedir(name, notindexed)
         except OSError as err:
             # Catch EEXIST to handle races
             if err.errno == errno.EEXIST:
                 return
             raise
     if mode is not None:
         os.chmod(name, mode)
 
 def readfile(path):
     with open(path, 'rb') as fp:
         return fp.read()
 
 def writefile(path, text):
     with open(path, 'wb') as fp:
         fp.write(text)
 
 def appendfile(path, text):
     with open(path, 'ab') as fp:
         fp.write(text)
 
 class chunkbuffer(object):
     """Allow arbitrary sized chunks of data to be efficiently read from an
     iterator over chunks of arbitrary size."""
 
     def __init__(self, in_iter):
         """in_iter is the iterator that's iterating over the input chunks."""
         def splitbig(chunks):
             for chunk in chunks:
                 if len(chunk) > 2**20:
                     pos = 0
                     while pos < len(chunk):
                         end = pos + 2 ** 18
                         yield chunk[pos:end]
                         pos = end
                 else:
                     yield chunk
         self.iter = splitbig(in_iter)
         self._queue = collections.deque()
         self._chunkoffset = 0
 
     def read(self, l=None):
         """Read L bytes of data from the iterator of chunks of data.
         Returns less than L bytes if the iterator runs dry.
 
         If size parameter is omitted, read everything"""
         if l is None:
             return ''.join(self.iter)
 
         left = l
         buf = []
         queue = self._queue
         while left > 0:
             # refill the queue
             if not queue:
                 target = 2**18
                 for chunk in self.iter:
                     queue.append(chunk)
                     target -= len(chunk)
                     if target <= 0:
                         break
                 if not queue:
                     break
 
             # The easy way to do this would be to queue.popleft(), modify the
             # chunk (if necessary), then queue.appendleft(). However, for cases
             # where we read partial chunk content, this incurs 2 dequeue
             # mutations and creates a new str for the remaining chunk in the
             # queue. Our code below avoids this overhead.
 
             chunk = queue[0]
             chunkl = len(chunk)
             offset = self._chunkoffset
 
             # Use full chunk.
             if offset == 0 and left >= chunkl:
                 left -= chunkl
                 queue.popleft()
                 buf.append(chunk)
                 # self._chunkoffset remains at 0.
                 continue
 
             chunkremaining = chunkl - offset
 
             # Use all of unconsumed part of chunk.
             if left >= chunkremaining:
                 left -= chunkremaining
                 queue.popleft()
                 # offset == 0 is enabled by block above, so this won't merely
                 # copy via ``chunk[0:]``.
                 buf.append(chunk[offset:])
                 self._chunkoffset = 0
 
             # Partial chunk needed.
             else:
                 buf.append(chunk[offset:offset + left])
                 self._chunkoffset += left
                 left -= chunkremaining
 
         return ''.join(buf)
 
 def filechunkiter(f, size=131072, limit=None):
     """Create a generator that produces the data in the file size
     (default 131072) 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)
         yield s
 
 def makedate(timestamp=None):
     '''Return a unix timestamp (or the current time) as a (unixtime,
     offset) tuple based off the local timezone.'''
     if timestamp is None:
         timestamp = time.time()
     if timestamp < 0:
         hint = _("check your clock")
         raise Abort(_("negative timestamp: %d") % timestamp, hint=hint)
     delta = (datetime.datetime.utcfromtimestamp(timestamp) -
              datetime.datetime.fromtimestamp(timestamp))
     tz = delta.days * 86400 + delta.seconds
     return timestamp, tz
 
 def datestr(date=None, format='%a %b %d %H:%M:%S %Y %1%2'):
     """represent a (unixtime, offset) tuple as a localized time.
     unixtime is seconds since the epoch, and offset is the time zone's
     number of seconds away from UTC.
 
     >>> datestr((0, 0))
     'Thu Jan 01 00:00:00 1970 +0000'
     >>> datestr((42, 0))
     'Thu Jan 01 00:00:42 1970 +0000'
     >>> datestr((-42, 0))
     'Wed Dec 31 23:59:18 1969 +0000'
     >>> datestr((0x7fffffff, 0))
     'Tue Jan 19 03:14:07 2038 +0000'
     >>> datestr((-0x80000000, 0))
     'Fri Dec 13 20:45:52 1901 +0000'
     """
     t, tz = date or makedate()
     if "%1" in format or "%2" in format or "%z" in format:
         sign = (tz > 0) and "-" or "+"
         minutes = abs(tz) // 60
         q, r = divmod(minutes, 60)
         format = format.replace("%z", "%1%2")
         format = format.replace("%1", "%c%02d" % (sign, q))
         format = format.replace("%2", "%02d" % r)
     d = t - tz
     if d > 0x7fffffff:
         d = 0x7fffffff
     elif d < -0x80000000:
         d = -0x80000000
     # Never use time.gmtime() and datetime.datetime.fromtimestamp()
     # because they use the gmtime() system call which is buggy on Windows
     # for negative values.
     t = datetime.datetime(1970, 1, 1) + datetime.timedelta(seconds=d)
     s = encoding.strtolocal(t.strftime(encoding.strfromlocal(format)))
     return s
 
 def shortdate(date=None):
     """turn (timestamp, tzoff) tuple into iso 8631 date."""
     return datestr(date, format='%Y-%m-%d')
 
 def parsetimezone(s):
     """find a trailing timezone, if any, in string, and return a
        (offset, remainder) pair"""
 
     if s.endswith("GMT") or s.endswith("UTC"):
         return 0, s[:-3].rstrip()
 
     # Unix-style timezones [+-]hhmm
     if len(s) >= 5 and s[-5] in "+-" and s[-4:].isdigit():
         sign = (s[-5] == "+") and 1 or -1
         hours = int(s[-4:-2])
         minutes = int(s[-2:])
         return -sign * (hours * 60 + minutes) * 60, s[:-5].rstrip()
 
     # ISO8601 trailing Z
     if s.endswith("Z") and s[-2:-1].isdigit():
         return 0, s[:-1]
 
     # ISO8601-style [+-]hh:mm
     if (len(s) >= 6 and s[-6] in "+-" and s[-3] == ":" and
         s[-5:-3].isdigit() and s[-2:].isdigit()):
         sign = (s[-6] == "+") and 1 or -1
         hours = int(s[-5:-3])
         minutes = int(s[-2:])
         return -sign * (hours * 60 + minutes) * 60, s[:-6]
 
     return None, s
 
 def strdate(string, format, defaults=None):
     """parse a localized time string and return a (unixtime, offset) tuple.
     if the string cannot be parsed, ValueError is raised."""
     if defaults is None:
         defaults = {}
 
     # NOTE: unixtime = localunixtime + offset
     offset, date = parsetimezone(string)
 
     # add missing elements from defaults
     usenow = False # default to using biased defaults
     for part in ("S", "M", "HI", "d", "mb", "yY"): # decreasing specificity
         part = pycompat.bytestr(part)
         found = [True for p in part if ("%"+p) in format]
         if not found:
             date += "@" + defaults[part][usenow]
             format += "@%" + part[0]
         else:
             # We've found a specific time element, less specific time
             # elements are relative to today
             usenow = True
 
     timetuple = time.strptime(encoding.strfromlocal(date),
                               encoding.strfromlocal(format))
     localunixtime = int(calendar.timegm(timetuple))
     if offset is None:
         # local timezone
         unixtime = int(time.mktime(timetuple))
         offset = unixtime - localunixtime
     else:
         unixtime = localunixtime + offset
     return unixtime, offset
 
 def parsedate(date, formats=None, bias=None):
     """parse a localized date/time and return a (unixtime, offset) tuple.
 
     The date may be a "unixtime offset" string or in one of the specified
     formats. If the date already is a (unixtime, offset) tuple, it is returned.
 
     >>> parsedate(' today ') == parsedate(\
                                   datetime.date.today().strftime('%b %d'))
     True
     >>> parsedate( 'yesterday ') == parsedate((datetime.date.today() -\
                                                datetime.timedelta(days=1)\
                                               ).strftime('%b %d'))
     True
     >>> now, tz = makedate()
     >>> strnow, strtz = parsedate('now')
     >>> (strnow - now) < 1
     True
     >>> tz == strtz
     True
     """
     if bias is None:
         bias = {}
     if not date:
         return 0, 0
     if isinstance(date, tuple) and len(date) == 2:
         return date
     if not formats:
         formats = defaultdateformats
     date = date.strip()
 
     if date == 'now' or date == _('now'):
         return makedate()
     if date == 'today' or date == _('today'):
         date = datetime.date.today().strftime('%b %d')
     elif date == 'yesterday' or date == _('yesterday'):
         date = (datetime.date.today() -
                 datetime.timedelta(days=1)).strftime('%b %d')
 
     try:
         when, offset = map(int, date.split(' '))
     except ValueError:
         # fill out defaults
         now = makedate()
         defaults = {}
         for part in ("d", "mb", "yY", "HI", "M", "S"):
             # this piece is for rounding the specific end of unknowns
             b = bias.get(part)
             if b is None:
                 if part[0:1] in "HMS":
                     b = "00"
                 else:
                     b = "0"
 
             # this piece is for matching the generic end to today's date
             n = datestr(now, "%" + part[0:1])
 
             defaults[part] = (b, n)
 
         for format in formats:
             try:
                 when, offset = strdate(date, format, defaults)
             except (ValueError, OverflowError):
                 pass
             else:
                 break
         else:
             raise Abort(_('invalid date: %r') % date)
     # 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 when < -0x80000000 or when > 0x7fffffff:
         raise Abort(_('date exceeds 32 bits: %d') % when)
     if offset < -50400 or offset > 43200:
         raise Abort(_('impossible time zone offset: %d') % offset)
     return when, offset
 
 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
 
     >>> p1 = parsedate("10:29:59")
     >>> p2 = parsedate("10:30:00")
     >>> p3 = parsedate("10:30:59")
     >>> p4 = parsedate("10:31:00")
     >>> p5 = parsedate("Sep 15 10:30:00 1999")
     >>> f = matchdate("10:30")
     >>> f(p1[0])
     False
     >>> f(p2[0])
     True
     >>> f(p3[0])
     True
     >>> f(p4[0])
     False
     >>> f(p5[0])
     False
     """
 
     def lower(date):
         d = {'mb': "1", 'd': "1"}
         return parsedate(date, extendeddateformats, d)[0]
 
     def upper(date):
         d = {'mb': "12", 'HI': "23", 'M': "59", 'S': "59"}
         for days in ("31", "30", "29"):
             try:
                 d["d"] = days
                 return parsedate(date, extendeddateformats, d)[0]
             except Abort:
                 pass
         d["d"] = "28"
         return parsedate(date, extendeddateformats, d)[0]
 
     date = date.strip()
 
     if not date:
         raise Abort(_("dates cannot consist entirely of whitespace"))
     elif date[0] == "<":
         if not date[1:]:
             raise Abort(_("invalid day spec, use '<DATE'"))
         when = upper(date[1:])
         return lambda x: x <= when
     elif date[0] == ">":
         if not date[1:]:
             raise Abort(_("invalid day spec, use '>DATE'"))
         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:])
         if days < 0:
             raise Abort(_("%s must be nonnegative (see 'hg help dates')")
                 % date[1:])
         when = makedate()[0] - days * 3600 * 24
         return lambda x: x >= when
     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
 
 def stringmatcher(pattern, casesensitive=True):
     """
     accepts a string, possibly starting with 're:' or 'literal:' prefix.
     returns the matcher name, pattern, and matcher function.
     missing or unknown prefixes are treated as literal matches.
 
     helper for tests:
     >>> def test(pattern, *tests):
     ...     kind, pattern, matcher = stringmatcher(pattern)
     ...     return (kind, pattern, [bool(matcher(t)) for t in tests])
     >>> def itest(pattern, *tests):
     ...     kind, pattern, matcher = stringmatcher(pattern, casesensitive=False)
     ...     return (kind, pattern, [bool(matcher(t)) for t in tests])
 
     exact matching (no prefix):
     >>> test('abcdefg', 'abc', 'def', 'abcdefg')
     ('literal', 'abcdefg', [False, False, True])
 
     regex matching ('re:' prefix)
     >>> test('re:a.+b', 'nomatch', 'fooadef', 'fooadefbar')
     ('re', 'a.+b', [False, False, True])
 
     force exact matches ('literal:' prefix)
     >>> test('literal:re:foobar', 'foobar', 're:foobar')
     ('literal', 're:foobar', [False, True])
 
     unknown prefixes are ignored and treated as literals
     >>> test('foo:bar', 'foo', 'bar', 'foo:bar')
     ('literal', 'foo:bar', [False, False, True])
 
     case insensitive regex matches
     >>> itest('re:A.+b', 'nomatch', 'fooadef', 'fooadefBar')
     ('re', 'A.+b', [False, False, True])
 
     case insensitive literal matches
     >>> itest('ABCDEFG', 'abc', 'def', 'abcdefg')
     ('literal', 'ABCDEFG', [False, False, True])
     """
     if pattern.startswith('re:'):
         pattern = pattern[3:]
         try:
             flags = 0
             if not casesensitive:
                 flags = remod.I
             regex = remod.compile(pattern, flags)
         except remod.error as e:
             raise error.ParseError(_('invalid regular expression: %s')
                                    % e)
         return 're', pattern, regex.search
     elif pattern.startswith('literal:'):
         pattern = pattern[8:]
 
     match = pattern.__eq__
 
     if not casesensitive:
         ipat = encoding.lower(pattern)
         match = lambda s: ipat == encoding.lower(s)
     return 'literal', pattern, match
 
 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:]
     f = user.find(' ')
     if f >= 0:
         user = user[:f]
     f = user.find('.')
     if f >= 0:
         user = user[:f]
     return user
 
 def emailuser(user):
     """Return the user portion of an email address."""
     f = user.find('@')
     if f >= 0:
         user = user[:f]
     f = user.find('<')
     if f >= 0:
         user = user[f + 1:]
     return user
 
 def email(author):
     '''get email of author.'''
     r = author.find('>')
     if r == -1:
         r = None
     return author[author.find('<') + 1:r]
 
 def ellipsis(text, maxlength=400):
     """Trim string to at most maxlength (default: 400) columns in display."""
     return encoding.trim(text, maxlength, ellipsis='...')
 
 def unitcountfn(*unittable):
     '''return a function that renders a readable count of some quantity'''
 
     def go(count):
         for multiplier, divisor, format in unittable:
             if abs(count) >= divisor * multiplier:
                 return format % (count / float(divisor))
         return unittable[-1][2] % count
 
     return go
 
 def processlinerange(fromline, toline):
     """Check that linerange <fromline>:<toline> makes sense and return a
     0-based range.
 
     >>> processlinerange(10, 20)
     (9, 20)
     >>> processlinerange(2, 1)
     Traceback (most recent call last):
         ...
     ParseError: line range must be positive
     >>> processlinerange(0, 5)
     Traceback (most recent call last):
         ...
     ParseError: fromline must be strictly positive
     """
     if toline - fromline < 0:
         raise error.ParseError(_("line range must be positive"))
     if fromline < 1:
         raise error.ParseError(_("fromline must be strictly positive"))
     return fromline - 1, toline
 
 bytecount = unitcountfn(
     (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')),
     )
 
 # Matches a single EOL which can either be a CRLF where repeated CR
 # are removed or a LF. We do not care about old Macintosh files, so a
 # stray CR is an error.
 _eolre = remod.compile(br'\r*\n')
 
 def tolf(s):
     return _eolre.sub('\n', s)
 
 def tocrlf(s):
     return _eolre.sub('\r\n', s)
 
 if pycompat.oslinesep == '\r\n':
     tonativeeol = tocrlf
     fromnativeeol = tolf
 else:
     tonativeeol = pycompat.identity
     fromnativeeol = pycompat.identity
 
 def escapestr(s):
     # call underlying function of s.encode('string_escape') directly for
     # Python 3 compatibility
     return codecs.escape_encode(s)[0]
 
 def unescapestr(s):
     return codecs.escape_decode(s)[0]
 
 def uirepr(s):
     # Avoid double backslash in Windows path repr()
     return repr(s).replace('\\\\', '\\')
 
 # delay import of textwrap
 def MBTextWrapper(**kwargs):
     class tw(textwrap.TextWrapper):
         """
         Extend TextWrapper for width-awareness.
 
         Neither number of 'bytes' in any encoding nor 'characters' is
         appropriate to calculate terminal columns for specified string.
 
         Original TextWrapper implementation uses built-in 'len()' directly,
         so overriding is needed to use width information of each characters.
 
         In addition, characters classified into 'ambiguous' width are
         treated as wide in East Asian area, but as narrow in other.
 
         This requires use decision to determine width of such characters.
         """
         def _cutdown(self, ucstr, space_left):
             l = 0
             colwidth = encoding.ucolwidth
             for i in xrange(len(ucstr)):
                 l += colwidth(ucstr[i])
                 if space_left < l:
                     return (ucstr[:i], ucstr[i:])
             return ucstr, ''
 
         # overriding of base class
         def _handle_long_word(self, reversed_chunks, cur_line, cur_len, width):
             space_left = max(width - cur_len, 1)
 
             if self.break_long_words:
                 cut, res = self._cutdown(reversed_chunks[-1], space_left)
                 cur_line.append(cut)
                 reversed_chunks[-1] = res
             elif not cur_line:
                 cur_line.append(reversed_chunks.pop())
 
         # this overriding code is imported from TextWrapper of Python 2.6
         # to calculate columns of string by 'encoding.ucolwidth()'
         def _wrap_chunks(self, chunks):
             colwidth = encoding.ucolwidth
 
             lines = []
             if self.width <= 0:
                 raise ValueError("invalid width %r (must be > 0)" % self.width)
 
             # Arrange in reverse order so items can be efficiently popped
             # from a stack of chucks.
             chunks.reverse()
 
             while chunks:
 
                 # Start the list of chunks that will make up the current line.
                 # cur_len is just the length of all the chunks in cur_line.
                 cur_line = []
                 cur_len = 0
 
                 # Figure out which static string will prefix this line.
                 if lines:
                     indent = self.subsequent_indent
                 else:
                     indent = self.initial_indent
 
                 # Maximum width for this line.
                 width = self.width - len(indent)
 
                 # First chunk on line is whitespace -- drop it, unless this
                 # is the very beginning of the text (i.e. no lines started yet).
                 if self.drop_whitespace and chunks[-1].strip() == '' and lines:
                     del chunks[-1]
 
                 while chunks:
                     l = colwidth(chunks[-1])
 
                     # Can at least squeeze this chunk onto the current line.
                     if cur_len + l <= width:
                         cur_line.append(chunks.pop())
                         cur_len += l
 
                     # Nope, this line is full.
                     else:
                         break
 
                 # The current line is full, and the next chunk is too big to
                 # fit on *any* line (not just this one).
                 if chunks and colwidth(chunks[-1]) > width:
                     self._handle_long_word(chunks, cur_line, cur_len, width)
 
                 # If the last chunk on this line is all whitespace, drop it.
                 if (self.drop_whitespace and
                     cur_line and cur_line[-1].strip() == ''):
                     del cur_line[-1]
 
                 # Convert current line back to a string and store it in list
                 # of all lines (return value).
                 if cur_line:
                     lines.append(indent + ''.join(cur_line))
 
             return lines
 
     global MBTextWrapper
     MBTextWrapper = tw
     return tw(**kwargs)
 
 def wrap(line, width, initindent='', hangindent=''):
     maxindent = max(len(hangindent), len(initindent))
     if width <= maxindent:
         # adjust for weird terminal size
         width = max(78, maxindent + 1)
     line = line.decode(pycompat.sysstr(encoding.encoding),
                     pycompat.sysstr(encoding.encodingmode))
     initindent = initindent.decode(pycompat.sysstr(encoding.encoding),
                     pycompat.sysstr(encoding.encodingmode))
     hangindent = hangindent.decode(pycompat.sysstr(encoding.encoding),
                     pycompat.sysstr(encoding.encodingmode))
     wrapper = MBTextWrapper(width=width,
                             initial_indent=initindent,
                             subsequent_indent=hangindent)
     return wrapper.fill(line).encode(pycompat.sysstr(encoding.encoding))
 
 if (pyplatform.python_implementation() == 'CPython' and
     sys.version_info < (3, 0)):
     # There is an issue in CPython that some IO methods do not handle EINTR
     # correctly. The following table shows what CPython version (and functions)
     # are affected (buggy: has the EINTR bug, okay: otherwise):
     #
     #                | < 2.7.4 | 2.7.4 to 2.7.12 | >= 3.0
     #   --------------------------------------------------
     #    fp.__iter__ | buggy   | buggy           | okay
     #    fp.read*    | buggy   | okay [1]        | okay
     #
     # [1]: fixed by changeset 67dc99a989cd in the cpython hg repo.
     #
     # Here we workaround the EINTR issue for fileobj.__iter__. Other methods
     # like "read*" are ignored for now, as Python < 2.7.4 is a minority.
     #
     # Although we can workaround the EINTR issue for fp.__iter__, it is slower:
     # "for x in fp" is 4x faster than "for x in iter(fp.readline, '')" in
     # CPython 2, because CPython 2 maintains an internal readahead buffer for
     # fp.__iter__ but not other fp.read* methods.
     #
     # On modern systems like Linux, the "read" syscall cannot be interrupted
     # when reading "fast" files like on-disk files. So the EINTR issue only
     # affects things like pipes, sockets, ttys etc. We treat "normal" (S_ISREG)
     # files approximately as "fast" files and use the fast (unsafe) code path,
     # to minimize the performance impact.
     if sys.version_info >= (2, 7, 4):
         # fp.readline deals with EINTR correctly, use it as a workaround.
         def _safeiterfile(fp):
             return iter(fp.readline, '')
     else:
         # fp.read* are broken too, manually deal with EINTR in a stupid way.
         # note: this may block longer than necessary because of bufsize.
         def _safeiterfile(fp, bufsize=4096):
             fd = fp.fileno()
             line = ''
             while True:
                 try:
                     buf = os.read(fd, bufsize)
                 except OSError as ex:
                     # os.read only raises EINTR before any data is read
                     if ex.errno == errno.EINTR:
                         continue
                     else:
                         raise
                 line += buf
                 if '\n' in buf:
                     splitted = line.splitlines(True)
                     line = ''
                     for l in splitted:
                         if l[-1] == '\n':
                             yield l
                         else:
                             line = l
                 if not buf:
                     break
             if line:
                 yield line
 
     def iterfile(fp):
         fastpath = True
         if type(fp) is file:
             fastpath = stat.S_ISREG(os.fstat(fp.fileno()).st_mode)
         if fastpath:
             return fp
         else:
             return _safeiterfile(fp)
 else:
     # PyPy and CPython 3 do not have the EINTR issue thus no workaround needed.
     def iterfile(fp):
         return fp
 
 def iterlines(iterator):
     for chunk in iterator:
         for line in chunk.splitlines():
             yield line
 
 def expandpath(path):
     return os.path.expanduser(os.path.expandvars(path))
 
 def hgcmd():
     """Return the command used to execute current hg
 
     This is different from hgexecutable() because on Windows we want
     to avoid things opening new shell windows like batch files, so we
     get either the python call or current executable.
     """
     if mainfrozen():
         if getattr(sys, 'frozen', None) == 'macosx_app':
             # Env variable set by py2app
             return [encoding.environ['EXECUTABLEPATH']]
         else:
             return [pycompat.sysexecutable]
     return gethgcmd()
 
 def rundetached(args, condfn):
     """Execute the argument list in a detached process.
 
     condfn is a callable which is called repeatedly and should return
     True once the child process is known to have started successfully.
     At this point, the child process PID is returned. If the child
     process fails to start or finishes before condfn() evaluates to
     True, return -1.
     """
     # Windows case is easier because the child process is either
     # successfully starting and validating the condition or exiting
     # on failure. We just poll on its PID. On Unix, if the child
     # process fails to start, it will be left in a zombie state until
     # the parent wait on it, which we cannot do since we expect a long
     # running process on success. Instead we listen for SIGCHLD telling
     # us our child process terminated.
     terminated = set()
     def handler(signum, frame):
         terminated.add(os.wait())
     prevhandler = None
     SIGCHLD = getattr(signal, 'SIGCHLD', None)
     if SIGCHLD is not None:
         prevhandler = signal.signal(SIGCHLD, handler)
     try:
         pid = spawndetached(args)
         while not condfn():
             if ((pid in terminated or not testpid(pid))
                 and not condfn()):
                 return -1
             time.sleep(0.1)
         return pid
     finally:
         if prevhandler is not None:
             signal.signal(signal.SIGCHLD, prevhandler)
 
 def interpolate(prefix, mapping, s, fn=None, escape_prefix=False):
     """Return the result of interpolating items in the mapping into string s.
 
     prefix is a single character string, or a two character string with
     a backslash as the first character if the prefix needs to be escaped in
     a regular expression.
 
     fn is an optional function that will be applied to the replacement text
     just before replacement.
 
     escape_prefix is an optional flag that allows using doubled prefix for
     its escaping.
     """
     fn = fn or (lambda s: s)
     patterns = '|'.join(mapping.keys())
     if escape_prefix:
         patterns += '|' + prefix
         if len(prefix) > 1:
             prefix_char = prefix[1:]
         else:
             prefix_char = prefix
         mapping[prefix_char] = prefix_char
     r = remod.compile(r'%s(%s)' % (prefix, patterns))
     return r.sub(lambda x: fn(mapping[x.group()[1:]]), s)
 
 def getport(port):
     """Return the port for a given network service.
 
     If port is an integer, it's returned as is. If it's a string, it's
     looked up using socket.getservbyname(). If there's no matching
     service, error.Abort is raised.
     """
     try:
         return int(port)
     except ValueError:
         pass
 
     try:
         return socket.getservbyname(port)
     except socket.error:
         raise Abort(_("no port number associated with service '%s'") % port)
 
 _booleans = {'1': True, 'yes': True, 'true': True, 'on': True, 'always': True,
              '0': False, 'no': False, 'false': False, 'off': False,
              'never': False}
 
 def parsebool(s):
     """Parse s into a boolean.
 
     If s is not a valid boolean, returns None.
     """
     return _booleans.get(s.lower(), None)
 
 _hextochr = dict((a + b, chr(int(a + b, 16)))
                  for a in string.hexdigits for b in string.hexdigits)
 
 class url(object):
     r"""Reliable URL parser.
 
     This parses URLs and provides attributes for the following
     components:
 
     <scheme>://<user>:<passwd>@<host>:<port>/<path>?<query>#<fragment>
 
     Missing components are set to None. The only exception is
     fragment, which is set to '' if present but empty.
 
     If parsefragment is False, fragment is included in query. If
     parsequery is False, query is included in path. If both are
     False, both fragment and query are included in path.
 
     See http://www.ietf.org/rfc/rfc2396.txt for more information.
 
     Note that for backward compatibility reasons, bundle URLs do not
     take host names. That means 'bundle://../' has a path of '../'.
 
     Examples:
 
     >>> url('http://www.ietf.org/rfc/rfc2396.txt')
     <url scheme: 'http', host: 'www.ietf.org', path: 'rfc/rfc2396.txt'>
     >>> url('ssh://[::1]:2200//home/joe/repo')
     <url scheme: 'ssh', host: '[::1]', port: '2200', path: '/home/joe/repo'>
     >>> url('file:///home/joe/repo')
     <url scheme: 'file', path: '/home/joe/repo'>
     >>> url('file:///c:/temp/foo/')
     <url scheme: 'file', path: 'c:/temp/foo/'>
     >>> url('bundle:foo')
     <url scheme: 'bundle', path: 'foo'>
     >>> url('bundle://../foo')
     <url scheme: 'bundle', path: '../foo'>
     >>> url(r'c:\foo\bar')
     <url path: 'c:\\foo\\bar'>
     >>> url(r'\\blah\blah\blah')
     <url path: '\\\\blah\\blah\\blah'>
     >>> url(r'\\blah\blah\blah#baz')
     <url path: '\\\\blah\\blah\\blah', fragment: 'baz'>
     >>> url(r'file:///C:\users\me')
     <url scheme: 'file', path: 'C:\\users\\me'>
 
     Authentication credentials:
 
     >>> url('ssh://joe:xyz@x/repo')
     <url scheme: 'ssh', user: 'joe', passwd: 'xyz', host: 'x', path: 'repo'>
     >>> url('ssh://joe@x/repo')
     <url scheme: 'ssh', user: 'joe', host: 'x', path: 'repo'>
 
     Query strings and fragments:
 
     >>> url('http://host/a?b#c')
     <url scheme: 'http', host: 'host', path: 'a', query: 'b', fragment: 'c'>
     >>> url('http://host/a?b#c', parsequery=False, parsefragment=False)
     <url scheme: 'http', host: 'host', path: 'a?b#c'>
 
     Empty path:
 
     >>> url('')
     <url path: ''>
     >>> url('#a')
     <url path: '', fragment: 'a'>
     >>> url('http://host/')
     <url scheme: 'http', host: 'host', path: ''>
     >>> url('http://host/#a')
     <url scheme: 'http', host: 'host', path: '', fragment: 'a'>
 
     Only scheme:
 
     >>> url('http:')
     <url scheme: 'http'>
     """
 
     _safechars = "!~*'()+"
     _safepchars = "/!~*'()+:\\"
     _matchscheme = remod.compile('^[a-zA-Z0-9+.\\-]+:').match
 
     def __init__(self, path, parsequery=True, parsefragment=True):
         # We slowly chomp away at path until we have only the path left
         self.scheme = self.user = self.passwd = self.host = None
         self.port = self.path = self.query = self.fragment = None
         self._localpath = True
         self._hostport = ''
         self._origpath = path
 
         if parsefragment and '#' in path:
             path, self.fragment = path.split('#', 1)
 
         # special case for Windows drive letters and UNC paths
         if hasdriveletter(path) or path.startswith('\\\\'):
             self.path = path
             return
 
         # For compatibility reasons, we can't handle bundle paths as
         # normal URLS
         if path.startswith('bundle:'):
             self.scheme = 'bundle'
             path = path[7:]
             if path.startswith('//'):
                 path = path[2:]
             self.path = path
             return
 
         if self._matchscheme(path):
             parts = path.split(':', 1)
             if parts[0]:
                 self.scheme, path = parts
                 self._localpath = False
 
         if not path:
             path = None
             if self._localpath:
                 self.path = ''
                 return
         else:
             if self._localpath:
                 self.path = path
                 return
 
             if parsequery and '?' in path:
                 path, self.query = path.split('?', 1)
                 if not path:
                     path = None
                 if not self.query:
                     self.query = None
 
             # // is required to specify a host/authority
             if path and path.startswith('//'):
                 parts = path[2:].split('/', 1)
                 if len(parts) > 1:
                     self.host, path = parts
                 else:
                     self.host = parts[0]
                     path = None
                 if not self.host:
                     self.host = None
                     # path of file:///d is /d
                     # path of file:///d:/ is d:/, not /d:/
                     if path and not hasdriveletter(path):
                         path = '/' + path
 
             if self.host and '@' in self.host:
                 self.user, self.host = self.host.rsplit('@', 1)
                 if ':' in self.user:
                     self.user, self.passwd = self.user.split(':', 1)
                 if not self.host:
                     self.host = None
 
             # Don't split on colons in IPv6 addresses without ports
             if (self.host and ':' in self.host and
                 not (self.host.startswith('[') and self.host.endswith(']'))):
                 self._hostport = self.host
                 self.host, self.port = self.host.rsplit(':', 1)
                 if not self.host:
                     self.host = None
 
             if (self.host and self.scheme == 'file' and
                 self.host not in ('localhost', '127.0.0.1', '[::1]')):
                 raise Abort(_('file:// URLs can only refer to localhost'))
 
         self.path = path
 
         # leave the query string escaped
         for a in ('user', 'passwd', 'host', 'port',
                   'path', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 setattr(self, a, urlreq.unquote(v))
 
     def __repr__(self):
         attrs = []
         for a in ('scheme', 'user', 'passwd', 'host', 'port', 'path',
                   'query', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 attrs.append('%s: %r' % (a, v))
         return '<url %s>' % ', '.join(attrs)
 
     def __str__(self):
         r"""Join the URL's components back into a URL string.
 
         Examples:
 
         >>> str(url('http://user:pw@host:80/c:/bob?fo:oo#ba:ar'))
         'http://user:pw@host:80/c:/bob?fo:oo#ba:ar'
         >>> str(url('http://user:pw@host:80/?foo=bar&baz=42'))
         'http://user:pw@host:80/?foo=bar&baz=42'
         >>> str(url('http://user:pw@host:80/?foo=bar%3dbaz'))
         'http://user:pw@host:80/?foo=bar%3dbaz'
         >>> str(url('ssh://user:pw@[::1]:2200//home/joe#'))
         'ssh://user:pw@[::1]:2200//home/joe#'
         >>> str(url('http://localhost:80//'))
         'http://localhost:80//'
         >>> str(url('http://localhost:80/'))
         'http://localhost:80/'
         >>> str(url('http://localhost:80'))
         'http://localhost:80/'
         >>> str(url('bundle:foo'))
         'bundle:foo'
         >>> str(url('bundle://../foo'))
         'bundle:../foo'
         >>> str(url('path'))
         'path'
         >>> str(url('file:///tmp/foo/bar'))
         'file:///tmp/foo/bar'
         >>> str(url('file:///c:/tmp/foo/bar'))
         'file:///c:/tmp/foo/bar'
         >>> print url(r'bundle:foo\bar')
         bundle:foo\bar
         >>> print url(r'file:///D:\data\hg')
         file:///D:\data\hg
         """
         return encoding.strfromlocal(self.__bytes__())
 
     def __bytes__(self):
         if self._localpath:
             s = self.path
             if self.scheme == 'bundle':
                 s = 'bundle:' + s
             if self.fragment:
                 s += '#' + self.fragment
             return s
 
         s = self.scheme + ':'
         if self.user or self.passwd or self.host:
             s += '//'
         elif self.scheme and (not self.path or self.path.startswith('/')
                               or hasdriveletter(self.path)):
             s += '//'
             if hasdriveletter(self.path):
                 s += '/'
         if self.user:
             s += urlreq.quote(self.user, safe=self._safechars)
         if self.passwd:
             s += ':' + urlreq.quote(self.passwd, safe=self._safechars)
         if self.user or self.passwd:
             s += '@'
         if self.host:
             if not (self.host.startswith('[') and self.host.endswith(']')):
                 s += urlreq.quote(self.host)
             else:
                 s += self.host
         if self.port:
             s += ':' + urlreq.quote(self.port)
         if self.host:
             s += '/'
         if self.path:
             # TODO: similar to the query string, we should not unescape the
             # path when we store it, the path might contain '%2f' = '/',
             # which we should *not* escape.
             s += urlreq.quote(self.path, safe=self._safepchars)
         if self.query:
             # we store the query in escaped form.
             s += '?' + self.query
         if self.fragment is not None:
             s += '#' + urlreq.quote(self.fragment, safe=self._safepchars)
         return s
 
     def authinfo(self):
         user, passwd = self.user, self.passwd
         try:
             self.user, self.passwd = None, None
             s = bytes(self)
         finally:
             self.user, self.passwd = user, passwd
         if not self.user:
             return (s, None)
         # authinfo[1] is passed to urllib2 password manager, and its
         # URIs must not contain credentials. The host is passed in the
         # URIs list because Python < 2.4.3 uses only that to search for
         # a password.
         return (s, (None, (s, self.host),
                     self.user, self.passwd or ''))
 
     def isabs(self):
         if self.scheme and self.scheme != 'file':
             return True # remote URL
         if hasdriveletter(self.path):
             return True # absolute for our purposes - can't be joined()
         if self.path.startswith(r'\\'):
             return True # Windows UNC path
         if self.path.startswith('/'):
             return True # POSIX-style
         return False
 
     def localpath(self):
         if self.scheme == 'file' or self.scheme == 'bundle':
             path = self.path or '/'
             # For Windows, we need to promote hosts containing drive
             # letters to paths with drive letters.
             if hasdriveletter(self._hostport):
                 path = self._hostport + '/' + self.path
             elif (self.host is not None and self.path
                   and not hasdriveletter(path)):
                 path = '/' + path
             return path
         return self._origpath
 
     def islocal(self):
         '''whether localpath will return something that posixfile can open'''
         return (not self.scheme or self.scheme == 'file'
                 or self.scheme == 'bundle')
 
 def hasscheme(path):
     return bool(url(path).scheme)
 
 def hasdriveletter(path):
     return path and path[1:2] == ':' and path[0:1].isalpha()
 
 def urllocalpath(path):
     return url(path, parsequery=False, parsefragment=False).localpath()
 
 def hidepassword(u):
     '''hide user credential in a url string'''
     u = url(u)
     if u.passwd:
         u.passwd = '***'
     return bytes(u)
 
 def removeauth(u):
     '''remove all authentication information from a url string'''
     u = url(u)
     u.user = u.passwd = None
     return str(u)
 
 timecount = unitcountfn(
     (1, 1e3, _('%.0f s')),
     (100, 1, _('%.1f s')),
     (10, 1, _('%.2f s')),
     (1, 1, _('%.3f s')),
     (100, 0.001, _('%.1f ms')),
     (10, 0.001, _('%.2f ms')),
     (1, 0.001, _('%.3f ms')),
     (100, 0.000001, _('%.1f us')),
     (10, 0.000001, _('%.2f us')),
     (1, 0.000001, _('%.3f us')),
     (100, 0.000000001, _('%.1f ns')),
     (10, 0.000000001, _('%.2f ns')),
     (1, 0.000000001, _('%.3f ns')),
     )
 
 _timenesting = [0]
 
 def timed(func):
     '''Report the execution time of a function call to stderr.
 
     During development, use as a decorator when you need to measure
     the cost of a function, e.g. as follows:
 
     @util.timed
     def foo(a, b, c):
         pass
     '''
 
     def wrapper(*args, **kwargs):
         start = timer()
         indent = 2
         _timenesting[0] += indent
         try:
             return func(*args, **kwargs)
         finally:
             elapsed = timer() - start
             _timenesting[0] -= indent
             stderr.write('%s%s: %s\n' %
                          (' ' * _timenesting[0], func.__name__,
                           timecount(elapsed)))
     return wrapper
 
 _sizeunits = (('m', 2**20), ('k', 2**10), ('g', 2**30),
               ('kb', 2**10), ('mb', 2**20), ('gb', 2**30), ('b', 1))
 
 def sizetoint(s):
     '''Convert a space specifier to a byte count.
 
     >>> sizetoint('30')
     30
     >>> sizetoint('2.2kb')
     2252
     >>> sizetoint('6M')
     6291456
     '''
     t = s.strip().lower()
     try:
         for k, u in _sizeunits:
             if t.endswith(k):
                 return int(float(t[:-len(k)]) * u)
         return int(t)
     except ValueError:
         raise error.ParseError(_("couldn't parse size: %s") % s)
 
 class hooks(object):
     '''A collection of hook functions that can be used to extend a
     function's behavior. Hooks are called in lexicographic order,
     based on the names of their sources.'''
 
     def __init__(self):
         self._hooks = []
 
     def add(self, source, hook):
         self._hooks.append((source, hook))
 
     def __call__(self, *args):
         self._hooks.sort(key=lambda x: x[0])
         results = []
         for source, hook in self._hooks:
             results.append(hook(*args))
         return results
 
 def getstackframes(skip=0, line=' %-*s in %s\n', fileline='%s:%s', depth=0):
     '''Yields lines for a nicely formatted stacktrace.
     Skips the 'skip' last entries, then return the last 'depth' entries.
     Each file+linenumber is formatted according to fileline.
     Each line is formatted according to line.
     If line is None, it yields:
       length of longest filepath+line number,
       filepath+linenumber,
       function
 
     Not be used in production code but very convenient while developing.
     '''
     entries = [(fileline % (fn, ln), func)
         for fn, ln, func, _text in traceback.extract_stack()[:-skip - 1]
         ][-depth:]
     if entries:
         fnmax = max(len(entry[0]) for entry in entries)
         for fnln, func in entries:
             if line is None:
                 yield (fnmax, fnln, func)
             else:
                 yield line % (fnmax, fnln, func)
 
 def debugstacktrace(msg='stacktrace', skip=0,
                     f=stderr, otherf=stdout, depth=0):
     '''Writes a message to f (stderr) with a nicely formatted stacktrace.
     Skips the 'skip' entries closest to the call, then show 'depth' entries.
     By default it will flush stdout first.
     It can be used everywhere and intentionally does not require an ui object.
     Not be used in production code but very convenient while developing.
     '''
     if otherf:
         otherf.flush()
     f.write('%s at:\n' % msg.rstrip())
     for line in getstackframes(skip + 1, depth=depth):
         f.write(line)
     f.flush()
 
 class dirs(object):
     '''a multiset of directory names from a dirstate or manifest'''
 
     def __init__(self, map, skip=None):
         self._dirs = {}
         addpath = self.addpath
         if safehasattr(map, 'iteritems') and skip is not None:
             for f, s in map.iteritems():
                 if s[0] != skip:
                     addpath(f)
         else:
             for f in map:
                 addpath(f)
 
     def addpath(self, path):
         dirs = self._dirs
         for base in finddirs(path):
             if base in dirs:
                 dirs[base] += 1
                 return
             dirs[base] = 1
 
     def delpath(self, path):
         dirs = self._dirs
         for base in finddirs(path):
             if dirs[base] > 1:
                 dirs[base] -= 1
                 return
             del dirs[base]
 
     def __iter__(self):
         return iter(self._dirs)
 
     def __contains__(self, d):
         return d in self._dirs
 
 if safehasattr(parsers, 'dirs'):
     dirs = parsers.dirs
 
 def finddirs(path):
     pos = path.rfind('/')
     while pos != -1:
         yield path[:pos]
         pos = path.rfind('/', 0, pos)
 
 class ctxmanager(object):
     '''A context manager for use in 'with' blocks to allow multiple
     contexts to be entered at once.  This is both safer and more
     flexible than contextlib.nested.
 
     Once Mercurial supports Python 2.7+, this will become mostly
     unnecessary.
     '''
 
     def __init__(self, *args):
         '''Accepts a list of no-argument functions that return context
         managers.  These will be invoked at __call__ time.'''
         self._pending = args
         self._atexit = []
 
     def __enter__(self):
         return self
 
     def enter(self):
         '''Create and enter context managers in the order in which they were
         passed to the constructor.'''
         values = []
         for func in self._pending:
             obj = func()
             values.append(obj.__enter__())
             self._atexit.append(obj.__exit__)
         del self._pending
         return values
 
     def atexit(self, func, *args, **kwargs):
         '''Add a function to call when this context manager exits.  The
         ordering of multiple atexit calls is unspecified, save that
         they will happen before any __exit__ functions.'''
         def wrapper(exc_type, exc_val, exc_tb):
             func(*args, **kwargs)
         self._atexit.append(wrapper)
         return func
 
     def __exit__(self, exc_type, exc_val, exc_tb):
         '''Context managers are exited in the reverse order from which
         they were created.'''
         received = exc_type is not None
         suppressed = False
         pending = None
         self._atexit.reverse()
         for exitfunc in self._atexit:
             try:
                 if exitfunc(exc_type, exc_val, exc_tb):
                     suppressed = True
                     exc_type = None
                     exc_val = None
                     exc_tb = None
             except BaseException:
                 pending = sys.exc_info()
                 exc_type, exc_val, exc_tb = pending = sys.exc_info()
         del self._atexit
         if pending:
             raise exc_val
         return received and suppressed
 
 # compression code
 
 SERVERROLE = 'server'
 CLIENTROLE = 'client'
 
 compewireprotosupport = collections.namedtuple(u'compenginewireprotosupport',
                                                (u'name', u'serverpriority',
                                                 u'clientpriority'))
 
 class compressormanager(object):
     """Holds registrations of various compression engines.
 
     This class essentially abstracts the differences between compression
     engines to allow new compression formats to be added easily, possibly from
     extensions.
 
     Compressors are registered against the global instance by calling its
     ``register()`` method.
     """
     def __init__(self):
         self._engines = {}
         # Bundle spec human name to engine name.
         self._bundlenames = {}
         # Internal bundle identifier to engine name.
         self._bundletypes = {}
         # Revlog header to engine name.
         self._revlogheaders = {}
         # Wire proto identifier to engine name.
         self._wiretypes = {}
 
     def __getitem__(self, key):
         return self._engines[key]
 
     def __contains__(self, key):
         return key in self._engines
 
     def __iter__(self):
         return iter(self._engines.keys())
 
     def register(self, engine):
         """Register a compression engine with the manager.
 
         The argument must be a ``compressionengine`` instance.
         """
         if not isinstance(engine, compressionengine):
             raise ValueError(_('argument must be a compressionengine'))
 
         name = engine.name()
 
         if name in self._engines:
             raise error.Abort(_('compression engine %s already registered') %
                               name)
 
         bundleinfo = engine.bundletype()
         if bundleinfo:
             bundlename, bundletype = bundleinfo
 
             if bundlename in self._bundlenames:
                 raise error.Abort(_('bundle name %s already registered') %
                                   bundlename)
             if bundletype in self._bundletypes:
                 raise error.Abort(_('bundle type %s already registered by %s') %
                                   (bundletype, self._bundletypes[bundletype]))
 
             # No external facing name declared.
             if bundlename:
                 self._bundlenames[bundlename] = name
 
             self._bundletypes[bundletype] = name
 
         wiresupport = engine.wireprotosupport()
         if wiresupport:
             wiretype = wiresupport.name
             if wiretype in self._wiretypes:
                 raise error.Abort(_('wire protocol compression %s already '
                                     'registered by %s') %
                                   (wiretype, self._wiretypes[wiretype]))
 
             self._wiretypes[wiretype] = name
 
         revlogheader = engine.revlogheader()
         if revlogheader and revlogheader in self._revlogheaders:
             raise error.Abort(_('revlog header %s already registered by %s') %
                               (revlogheader, self._revlogheaders[revlogheader]))
 
         if revlogheader:
             self._revlogheaders[revlogheader] = name
 
         self._engines[name] = engine
 
     @property
     def supportedbundlenames(self):
         return set(self._bundlenames.keys())
 
     @property
     def supportedbundletypes(self):
         return set(self._bundletypes.keys())
 
     def forbundlename(self, bundlename):
         """Obtain a compression engine registered to a bundle name.
 
         Will raise KeyError if the bundle type isn't registered.
 
         Will abort if the engine is known but not available.
         """
         engine = self._engines[self._bundlenames[bundlename]]
         if not engine.available():
             raise error.Abort(_('compression engine %s could not be loaded') %
                               engine.name())
         return engine
 
     def forbundletype(self, bundletype):
         """Obtain a compression engine registered to a bundle type.
 
         Will raise KeyError if the bundle type isn't registered.
 
         Will abort if the engine is known but not available.
         """
         engine = self._engines[self._bundletypes[bundletype]]
         if not engine.available():
             raise error.Abort(_('compression engine %s could not be loaded') %
                               engine.name())
         return engine
 
     def supportedwireengines(self, role, onlyavailable=True):
         """Obtain compression engines that support the wire protocol.
 
         Returns a list of engines in prioritized order, most desired first.
 
         If ``onlyavailable`` is set, filter out engines that can't be
         loaded.
         """
         assert role in (SERVERROLE, CLIENTROLE)
 
         attr = 'serverpriority' if role == SERVERROLE else 'clientpriority'
 
         engines = [self._engines[e] for e in self._wiretypes.values()]
         if onlyavailable:
             engines = [e for e in engines if e.available()]
 
         def getkey(e):
             # Sort first by priority, highest first. In case of tie, sort
             # alphabetically. This is arbitrary, but ensures output is
             # stable.
             w = e.wireprotosupport()
             return -1 * getattr(w, attr), w.name
 
         return list(sorted(engines, key=getkey))
 
     def forwiretype(self, wiretype):
         engine = self._engines[self._wiretypes[wiretype]]
         if not engine.available():
             raise error.Abort(_('compression engine %s could not be loaded') %
                               engine.name())
         return engine
 
     def forrevlogheader(self, header):
         """Obtain a compression engine registered to a revlog header.
 
         Will raise KeyError if the revlog header value isn't registered.
         """
         return self._engines[self._revlogheaders[header]]
 
 compengines = compressormanager()
 
 class compressionengine(object):
     """Base class for compression engines.
 
     Compression engines must implement the interface defined by this class.
     """
     def name(self):
         """Returns the name of the compression engine.
 
         This is the key the engine is registered under.
 
         This method must be implemented.
         """
         raise NotImplementedError()
 
     def available(self):
         """Whether the compression engine is available.
 
         The intent of this method is to allow optional compression engines
         that may not be available in all installations (such as engines relying
         on C extensions that may not be present).
         """
         return True
 
     def bundletype(self):
         """Describes bundle identifiers for this engine.
 
         If this compression engine isn't supported for bundles, returns None.
 
         If this engine can be used for bundles, returns a 2-tuple of strings of
         the user-facing "bundle spec" compression name and an internal
         identifier used to denote the compression format within bundles. To
         exclude the name from external usage, set the first element to ``None``.
 
         If bundle compression is supported, the class must also implement
         ``compressstream`` and `decompressorreader``.
 
         The docstring of this method is used in the help system to tell users
         about this engine.
         """
         return None
 
     def wireprotosupport(self):
         """Declare support for this compression format on the wire protocol.
 
         If this compression engine isn't supported for compressing wire
         protocol payloads, returns None.
 
         Otherwise, returns ``compenginewireprotosupport`` with the following
         fields:
 
         * String format identifier
         * Integer priority for the server
         * Integer priority for the client
 
         The integer priorities are used to order the advertisement of format
         support by server and client. The highest integer is advertised
         first. Integers with non-positive values aren't advertised.
 
         The priority values are somewhat arbitrary and only used for default
         ordering. The relative order can be changed via config options.
 
         If wire protocol compression is supported, the class must also implement
         ``compressstream`` and ``decompressorreader``.
         """
         return None
 
     def revlogheader(self):
         """Header added to revlog chunks that identifies this engine.
 
         If this engine can be used to compress revlogs, this method should
         return the bytes used to identify chunks compressed with this engine.
         Else, the method should return ``None`` to indicate it does not
         participate in revlog compression.
         """
         return None
 
     def compressstream(self, it, opts=None):
         """Compress an iterator of chunks.
 
         The method receives an iterator (ideally a generator) of chunks of
         bytes to be compressed. It returns an iterator (ideally a generator)
         of bytes of chunks representing the compressed output.
 
         Optionally accepts an argument defining how to perform compression.
         Each engine treats this argument differently.
         """
         raise NotImplementedError()
 
     def decompressorreader(self, fh):
         """Perform decompression on a file object.
 
         Argument is an object with a ``read(size)`` method that returns
         compressed data. Return value is an object with a ``read(size)`` that
         returns uncompressed data.
         """
         raise NotImplementedError()
 
     def revlogcompressor(self, opts=None):
         """Obtain an object that can be used to compress revlog entries.
 
         The object has a ``compress(data)`` method that compresses binary
         data. This method returns compressed binary data or ``None`` if
         the data could not be compressed (too small, not compressible, etc).
         The returned data should have a header uniquely identifying this
         compression format so decompression can be routed to this engine.
         This header should be identified by the ``revlogheader()`` return
         value.
 
         The object has a ``decompress(data)`` method that decompresses
         data. The method will only be called if ``data`` begins with
         ``revlogheader()``. The method should return the raw, uncompressed
         data or raise a ``RevlogError``.
 
         The object is reusable but is not thread safe.
         """
         raise NotImplementedError()
 
 class _zlibengine(compressionengine):
     def name(self):
         return 'zlib'
 
     def bundletype(self):
         """zlib compression using the DEFLATE algorithm.
 
         All Mercurial clients should support this format. The compression
         algorithm strikes a reasonable balance between compression ratio
         and size.
         """
         return 'gzip', 'GZ'
 
     def wireprotosupport(self):
         return compewireprotosupport('zlib', 20, 20)
 
     def revlogheader(self):
         return 'x'
 
     def compressstream(self, it, opts=None):
         opts = opts or {}
 
         z = zlib.compressobj(opts.get('level', -1))
         for chunk in it:
             data = z.compress(chunk)
             # Not all calls to compress emit data. It is cheaper to inspect
             # here than to feed empty chunks through generator.
             if data:
                 yield data
 
         yield z.flush()
 
     def decompressorreader(self, fh):
         def gen():
             d = zlib.decompressobj()
             for chunk in filechunkiter(fh):
                 while chunk:
                     # Limit output size to limit memory.
                     yield d.decompress(chunk, 2 ** 18)
                     chunk = d.unconsumed_tail
 
         return chunkbuffer(gen())
 
     class zlibrevlogcompressor(object):
         def compress(self, data):
             insize = len(data)
             # Caller handles empty input case.
             assert insize > 0
 
             if insize < 44:
                 return None
 
             elif insize <= 1000000:
                 compressed = zlib.compress(data)
                 if len(compressed) < insize:
                     return compressed
                 return None
 
             # zlib makes an internal copy of the input buffer, doubling
             # memory usage for large inputs. So do streaming compression
             # on large inputs.
             else:
                 z = zlib.compressobj()
                 parts = []
                 pos = 0
                 while pos < insize:
                     pos2 = pos + 2**20
                     parts.append(z.compress(data[pos:pos2]))
                     pos = pos2
                 parts.append(z.flush())
 
                 if sum(map(len, parts)) < insize:
                     return ''.join(parts)
                 return None
 
         def decompress(self, data):
             try:
                 return zlib.decompress(data)
             except zlib.error as e:
                 raise error.RevlogError(_('revlog decompress error: %s') %
                                         str(e))
 
     def revlogcompressor(self, opts=None):
         return self.zlibrevlogcompressor()
 
 compengines.register(_zlibengine())
 
 class _bz2engine(compressionengine):
     def name(self):
         return 'bz2'
 
     def bundletype(self):
         """An algorithm that produces smaller bundles than ``gzip``.
 
         All Mercurial clients should support this format.
 
         This engine will likely produce smaller bundles than ``gzip`` but
         will be significantly slower, both during compression and
         decompression.
 
         If available, the ``zstd`` engine can yield similar or better
         compression at much higher speeds.
         """
         return 'bzip2', 'BZ'
 
     # We declare a protocol name but don't advertise by default because
     # it is slow.
     def wireprotosupport(self):
         return compewireprotosupport('bzip2', 0, 0)
 
     def compressstream(self, it, opts=None):
         opts = opts or {}
         z = bz2.BZ2Compressor(opts.get('level', 9))
         for chunk in it:
             data = z.compress(chunk)
             if data:
                 yield data
 
         yield z.flush()
 
     def decompressorreader(self, fh):
         def gen():
             d = bz2.BZ2Decompressor()
             for chunk in filechunkiter(fh):
                 yield d.decompress(chunk)
 
         return chunkbuffer(gen())
 
 compengines.register(_bz2engine())
 
 class _truncatedbz2engine(compressionengine):
     def name(self):
         return 'bz2truncated'
 
     def bundletype(self):
         return None, '_truncatedBZ'
 
     # We don't implement compressstream because it is hackily handled elsewhere.
 
     def decompressorreader(self, fh):
         def gen():
             # The input stream doesn't have the 'BZ' header. So add it back.
             d = bz2.BZ2Decompressor()
             d.decompress('BZ')
             for chunk in filechunkiter(fh):
                 yield d.decompress(chunk)
 
         return chunkbuffer(gen())
 
 compengines.register(_truncatedbz2engine())
 
 class _noopengine(compressionengine):
     def name(self):
         return 'none'
 
     def bundletype(self):
         """No compression is performed.
 
         Use this compression engine to explicitly disable compression.
         """
         return 'none', 'UN'
 
     # Clients always support uncompressed payloads. Servers don't because
     # unless you are on a fast network, uncompressed payloads can easily
     # saturate your network pipe.
     def wireprotosupport(self):
         return compewireprotosupport('none', 0, 10)
 
     # We don't implement revlogheader because it is handled specially
     # in the revlog class.
 
     def compressstream(self, it, opts=None):
         return it
 
     def decompressorreader(self, fh):
         return fh
 
     class nooprevlogcompressor(object):
         def compress(self, data):
             return None
 
     def revlogcompressor(self, opts=None):
         return self.nooprevlogcompressor()
 
 compengines.register(_noopengine())
 
 class _zstdengine(compressionengine):
     def name(self):
         return 'zstd'
 
     @propertycache
     def _module(self):
         # Not all installs have the zstd module available. So defer importing
         # until first access.
         try:
             from . import zstd
             # Force delayed import.
             zstd.__version__
             return zstd
         except ImportError:
             return None
 
     def available(self):
         return bool(self._module)
 
     def bundletype(self):
         """A modern compression algorithm that is fast and highly flexible.
 
         Only supported by Mercurial 4.1 and newer clients.
 
         With the default settings, zstd compression is both faster and yields
         better compression than ``gzip``. It also frequently yields better
         compression than ``bzip2`` while operating at much higher speeds.
 
         If this engine is available and backwards compatibility is not a
         concern, it is likely the best available engine.
         """
         return 'zstd', 'ZS'
 
     def wireprotosupport(self):
         return compewireprotosupport('zstd', 50, 50)
 
     def revlogheader(self):
         return '\x28'
 
     def compressstream(self, it, opts=None):
         opts = opts or {}
         # zstd level 3 is almost always significantly faster than zlib
         # while providing no worse compression. It strikes a good balance
         # between speed and compression.
         level = opts.get('level', 3)
 
         zstd = self._module
         z = zstd.ZstdCompressor(level=level).compressobj()
         for chunk in it:
             data = z.compress(chunk)
             if data:
                 yield data
 
         yield z.flush()
 
     def decompressorreader(self, fh):
         zstd = self._module
         dctx = zstd.ZstdDecompressor()
         return chunkbuffer(dctx.read_from(fh))
 
     class zstdrevlogcompressor(object):
         def __init__(self, zstd, level=3):
             # Writing the content size adds a few bytes to the output. However,
             # it allows decompression to be more optimal since we can
             # pre-allocate a buffer to hold the result.
             self._cctx = zstd.ZstdCompressor(level=level,
                                              write_content_size=True)
             self._dctx = zstd.ZstdDecompressor()
             self._compinsize = zstd.COMPRESSION_RECOMMENDED_INPUT_SIZE
             self._decompinsize = zstd.DECOMPRESSION_RECOMMENDED_INPUT_SIZE
 
         def compress(self, data):
             insize = len(data)
             # Caller handles empty input case.
             assert insize > 0
 
             if insize < 50:
                 return None
 
             elif insize <= 1000000:
                 compressed = self._cctx.compress(data)
                 if len(compressed) < insize:
                     return compressed
                 return None
             else:
                 z = self._cctx.compressobj()
                 chunks = []
                 pos = 0
                 while pos < insize:
                     pos2 = pos + self._compinsize
                     chunk = z.compress(data[pos:pos2])
                     if chunk:
                         chunks.append(chunk)
                     pos = pos2
                 chunks.append(z.flush())
 
                 if sum(map(len, chunks)) < insize:
                     return ''.join(chunks)
                 return None
 
         def decompress(self, data):
             insize = len(data)
 
             try:
                 # This was measured to be faster than other streaming
                 # decompressors.
                 dobj = self._dctx.decompressobj()
                 chunks = []
                 pos = 0
                 while pos < insize:
                     pos2 = pos + self._decompinsize
                     chunk = dobj.decompress(data[pos:pos2])
                     if chunk:
                         chunks.append(chunk)
                     pos = pos2
                 # Frame should be exhausted, so no finish() API.
 
                 return ''.join(chunks)
             except Exception as e:
                 raise error.RevlogError(_('revlog decompress error: %s') %
                                         str(e))
 
     def revlogcompressor(self, opts=None):
         opts = opts or {}
         return self.zstdrevlogcompressor(self._module,
                                          level=opts.get('level', 3))
 
 compengines.register(_zstdengine())
 
 def bundlecompressiontopics():
     """Obtains a list of available bundle compressions for use in help."""
     # help.makeitemsdocs() expects a dict of names to items with a .__doc__.
     items = {}
 
     # We need to format the docstring. So use a dummy object/type to hold it
     # rather than mutating the original.
     class docobject(object):
         pass
 
     for name in compengines:
         engine = compengines[name]
 
         if not engine.available():
             continue
 
         bt = engine.bundletype()
         if not bt or not bt[0]:
             continue
 
         doc = pycompat.sysstr('``%s``\n    %s') % (
             bt[0], engine.bundletype.__doc__)
 
         value = docobject()
         value.__doc__ = doc
 
         items[bt[0]] = value
 
     return items
 
 # convenient shortcut
 dst = debugstacktrace

mercurial/windows.py

@@ -1,475 +1,477 b''
 # windows.py - Windows utility function implementations for Mercurial
 #
 #  Copyright 2005-2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import msvcrt
 import os
 import re
 import stat
 import sys
 
 from .i18n import _
 from . import (
     encoding,
-    osutil,
+    policy,
     pycompat,
     win32,
 )
 
 try:
     import _winreg as winreg
     winreg.CloseKey
 except ImportError:
     import winreg
 
+osutil = policy.importmod(r'osutil')
+
 executablepath = win32.executablepath
 getuser = win32.getuser
 hidewindow = win32.hidewindow
 makedir = win32.makedir
 nlinks = win32.nlinks
 oslink = win32.oslink
 samedevice = win32.samedevice
 samefile = win32.samefile
 setsignalhandler = win32.setsignalhandler
 spawndetached = win32.spawndetached
 split = os.path.split
 testpid = win32.testpid
 unlink = win32.unlink
 
 umask = 0o022
 
 class mixedfilemodewrapper(object):
     """Wraps a file handle when it is opened in read/write mode.
 
     fopen() and fdopen() on Windows have a specific-to-Windows requirement
     that files opened with mode r+, w+, or a+ make a call to a file positioning
     function when switching between reads and writes. Without this extra call,
     Python will raise a not very intuitive "IOError: [Errno 0] Error."
 
     This class wraps posixfile instances when the file is opened in read/write
     mode and automatically adds checks or inserts appropriate file positioning
     calls when necessary.
     """
     OPNONE = 0
     OPREAD = 1
     OPWRITE = 2
 
     def __init__(self, fp):
         object.__setattr__(self, r'_fp', fp)
         object.__setattr__(self, r'_lastop', 0)
 
     def __enter__(self):
         return self._fp.__enter__()
 
     def __exit__(self, exc_type, exc_val, exc_tb):
         self._fp.__exit__(exc_type, exc_val, exc_tb)
 
     def __getattr__(self, name):
         return getattr(self._fp, name)
 
     def __setattr__(self, name, value):
         return self._fp.__setattr__(name, value)
 
     def _noopseek(self):
         self._fp.seek(0, os.SEEK_CUR)
 
     def seek(self, *args, **kwargs):
         object.__setattr__(self, r'_lastop', self.OPNONE)
         return self._fp.seek(*args, **kwargs)
 
     def write(self, d):
         if self._lastop == self.OPREAD:
             self._noopseek()
 
         object.__setattr__(self, r'_lastop', self.OPWRITE)
         return self._fp.write(d)
 
     def writelines(self, *args, **kwargs):
         if self._lastop == self.OPREAD:
             self._noopeseek()
 
         object.__setattr__(self, r'_lastop', self.OPWRITE)
         return self._fp.writelines(*args, **kwargs)
 
     def read(self, *args, **kwargs):
         if self._lastop == self.OPWRITE:
             self._noopseek()
 
         object.__setattr__(self, r'_lastop', self.OPREAD)
         return self._fp.read(*args, **kwargs)
 
     def readline(self, *args, **kwargs):
         if self._lastop == self.OPWRITE:
             self._noopseek()
 
         object.__setattr__(self, r'_lastop', self.OPREAD)
         return self._fp.readline(*args, **kwargs)
 
     def readlines(self, *args, **kwargs):
         if self._lastop == self.OPWRITE:
             self._noopseek()
 
         object.__setattr__(self, r'_lastop', self.OPREAD)
         return self._fp.readlines(*args, **kwargs)
 
 def posixfile(name, mode='r', buffering=-1):
     '''Open a file with even more POSIX-like semantics'''
     try:
         fp = osutil.posixfile(name, mode, buffering) # may raise WindowsError
 
         # The position when opening in append mode is implementation defined, so
         # make it consistent with other platforms, which position at EOF.
         if 'a' in mode:
             fp.seek(0, os.SEEK_END)
 
         if '+' in mode:
             return mixedfilemodewrapper(fp)
 
         return fp
     except WindowsError as err:
         # convert to a friendlier exception
         raise IOError(err.errno, '%s: %s' % (name, err.strerror))
 
 # may be wrapped by win32mbcs extension
 listdir = osutil.listdir
 
 class winstdout(object):
     '''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 IOError:
             pass
 
     def write(self, s):
         try:
             # This is workaround for "Not enough space" error on
             # writing large size of data to console.
             limit = 16000
             l = len(s)
             start = 0
             self.softspace = 0
             while start < l:
                 end = start + limit
                 self.fp.write(s[start:end])
                 start = end
         except IOError as inst:
             if inst.errno != 0:
                 raise
             self.close()
             raise IOError(errno.EPIPE, 'Broken pipe')
 
     def flush(self):
         try:
             return self.fp.flush()
         except IOError as inst:
             if inst.errno != errno.EINVAL:
                 raise
             self.close()
             raise IOError(errno.EPIPE, 'Broken pipe')
 
 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 encoding.environ.get('comspec', '')
 
 def openhardlinks():
     return not _is_win_9x()
 
 def parsepatchoutput(output_line):
     """parses the output produced by patch and returns the filename"""
     pf = output_line[14:]
     if pf[0] == '`':
         pf = pf[1:-1] # Remove the quotes
     return pf
 
 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
     return port and ("%s %s %s" % (args, pflag, port)) or args
 
 def setflags(f, l, x):
     pass
 
 def copymode(src, dst, mode=None):
     pass
 
 def checkexec(path):
     return False
 
 def checklink(path):
     return False
 
 def setbinary(fd):
     # When run without console, pipes may expose invalid
     # fileno(), usually set to -1.
     fno = getattr(fd, 'fileno', None)
     if fno is not None and fno() >= 0:
         msvcrt.setmode(fno(), os.O_BINARY)
 
 def pconvert(path):
     return path.replace(pycompat.ossep, '/')
 
 def localpath(path):
     return path.replace('/', '\\')
 
 def normpath(path):
     return pconvert(os.path.normpath(path))
 
 def normcase(path):
     return encoding.upper(path) # NTFS compares via upper()
 
 # see posix.py for definitions
 normcasespec = encoding.normcasespecs.upper
 normcasefallback = encoding.upperfallback
 
 def samestat(s1, s2):
     return False
 
 # 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 precede double quotes and add another
 # backslash before every double quote (being careful with the double
 # quote we've appended to the end)
 _quotere = None
 _needsshellquote = None
 def shellquote(s):
     r"""
     >>> shellquote(r'C:\Users\xyz')
     '"C:\\Users\\xyz"'
     >>> shellquote(r'C:\Users\xyz/mixed')
     '"C:\\Users\\xyz/mixed"'
     >>> # Would be safe not to quote too, since it is all double backslashes
     >>> shellquote(r'C:\\Users\\xyz')
     '"C:\\\\Users\\\\xyz"'
     >>> # But this must be quoted
     >>> shellquote(r'C:\\Users\\xyz/abc')
     '"C:\\\\Users\\\\xyz/abc"'
     """
     global _quotere
     if _quotere is None:
         _quotere = re.compile(r'(\\*)("|\\$)')
     global _needsshellquote
     if _needsshellquote is None:
         # ":" is also treated as "safe character", because it is used as a part
         # of path name on Windows.  "\" is also part of a path name, but isn't
         # safe because shlex.split() (kind of) treats it as an escape char and
         # drops it.  It will leave the next character, even if it is another
         # "\".
         _needsshellquote = re.compile(r'[^a-zA-Z0-9._:/-]').search
     if s and not _needsshellquote(s) and not _quotere.search(s):
         # "s" shouldn't have to be quoted
         return s
     return '"%s"' % _quotere.sub(r'\1\1\\\2', s)
 
 def quotecommand(cmd):
     """Build a command string suitable for os.popen* calls."""
     if sys.version_info < (2, 7, 1):
         # Python versions since 2.7.1 do this extra quoting themselves
         return '"' + cmd + '"'
     return cmd
 
 def popen(command, mode='r'):
     # Work around "popen spawned process may not write to stdout
     # under windows"
     # http://bugs.python.org/issue1366
     command += " 2> %s" % os.devnull
     return os.popen(quotecommand(command), mode)
 
 def explainexit(code):
     return _("exited with status %d") % code, code
 
 # if you change this stub into a real check, please try to implement the
 # username and groupname functions above, too.
 def isowner(st):
     return True
 
 def findexe(command):
     '''Find executable for command searching like cmd.exe does.
     If command is a basename then PATH is searched for command.
     PATH isn't searched if command is an absolute or relative path.
     An extension from PATHEXT is found and added if not present.
     If command isn't found None is returned.'''
     pathext = encoding.environ.get('PATHEXT', '.COM;.EXE;.BAT;.CMD')
     pathexts = [ext for ext in pathext.lower().split(pycompat.ospathsep)]
     if os.path.splitext(command)[1].lower() in pathexts:
         pathexts = ['']
 
     def findexisting(pathcommand):
         'Will append extension (if needed) and return existing file'
         for ext in pathexts:
             executable = pathcommand + ext
             if os.path.exists(executable):
                 return executable
         return None
 
     if pycompat.ossep in command:
         return findexisting(command)
 
     for path in encoding.environ.get('PATH', '').split(pycompat.ospathsep):
         executable = findexisting(os.path.join(path, command))
         if executable is not None:
             return executable
     return findexisting(os.path.expanduser(os.path.expandvars(command)))
 
 _wantedkinds = {stat.S_IFREG, stat.S_IFLNK}
 
 def statfiles(files):
     '''Stat each file in files. Yield each stat, or None if a file
     does not exist or has a type we don't care about.
 
     Cluster and cache stat per directory to minimize number of OS stat calls.'''
     dircache = {} # dirname -> filename -> status | None if file does not exist
     getkind = stat.S_IFMT
     for nf in files:
         nf  = normcase(nf)
         dir, base = os.path.split(nf)
         if not dir:
             dir = '.'
         cache = dircache.get(dir, None)
         if cache is None:
             try:
                 dmap = dict([(normcase(n), s)
                              for n, k, s in listdir(dir, True)
                              if getkind(s.st_mode) in _wantedkinds])
             except OSError as err:
                 # Python >= 2.5 returns ENOENT and adds winerror field
                 # EINVAL is raised if dir is not a directory.
                 if err.errno not in (errno.ENOENT, errno.EINVAL,
                                      errno.ENOTDIR):
                     raise
                 dmap = {}
             cache = dircache.setdefault(dir, dmap)
         yield cache.get(base, None)
 
 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."""
     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."""
     return None
 
 def removedirs(name):
     """special version of os.removedirs that does not remove symlinked
     directories or junction points if they actually contain files"""
     if listdir(name):
         return
     os.rmdir(name)
     head, tail = os.path.split(name)
     if not tail:
         head, tail = os.path.split(head)
     while head and tail:
         try:
             if listdir(head):
                 return
             os.rmdir(head)
         except (ValueError, OSError):
             break
         head, tail = os.path.split(head)
 
 def rename(src, dst):
     '''atomically rename file src to dst, replacing dst if it exists'''
     try:
         os.rename(src, dst)
     except OSError as e:
         if e.errno != errno.EEXIST:
             raise
         unlink(dst)
         os.rename(src, dst)
 
 def gethgcmd():
     return [sys.executable] + sys.argv[:1]
 
 def groupmembers(name):
     # Don't support groups on Windows for now
     raise KeyError
 
 def isexec(f):
     return False
 
 class cachestat(object):
     def __init__(self, path):
         pass
 
     def cacheable(self):
         return False
 
 def lookupreg(key, valname=None, scope=None):
     ''' Look up a key/value name in the Windows registry.
 
     valname: value name. If unspecified, the default value for the key
     is used.
     scope: optionally specify scope for registry lookup, this can be
     a sequence of scopes to look up in order. Default (CURRENT_USER,
     LOCAL_MACHINE).
     '''
     if scope is None:
         scope = (winreg.HKEY_CURRENT_USER, winreg.HKEY_LOCAL_MACHINE)
     elif not isinstance(scope, (list, tuple)):
         scope = (scope,)
     for s in scope:
         try:
             val = winreg.QueryValueEx(winreg.OpenKey(s, key), valname)[0]
             # never let a Unicode string escape into the wild
             return encoding.unitolocal(val)
         except EnvironmentError:
             pass
 
 expandglobs = True
 
 def statislink(st):
     '''check whether a stat result is a symlink'''
     return False
 
 def statisexec(st):
     '''check whether a stat result is an executable file'''
     return False
 
 def poll(fds):
     # see posix.py for description
     raise NotImplementedError()
 
 def readpipe(pipe):
     """Read all available data from a pipe."""
     chunks = []
     while True:
         size = win32.peekpipe(pipe)
         if not size:
             break
 
         s = pipe.read(size)
         if not s:
             break
         chunks.append(s)
 
     return ''.join(chunks)
 
 def bindunixsocket(sock, path):
     raise NotImplementedError('unsupported platform')

setup.py

@@ -1,802 +1,802 b''
 #
 # This is the mercurial setup script.
 #
 # 'python setup.py install', or
 # 'python setup.py --help' for more options
 
 import sys, platform
 if sys.version_info < (2, 7, 0, 'final'):
     raise SystemExit('Mercurial requires Python 2.7 or later.')
 
 if sys.version_info[0] >= 3:
     printf = eval('print')
     libdir_escape = 'unicode_escape'
 else:
     libdir_escape = 'string_escape'
     def printf(*args, **kwargs):
         f = kwargs.get('file', sys.stdout)
         end = kwargs.get('end', '\n')
         f.write(b' '.join(args) + end)
 
 # Solaris Python packaging brain damage
 try:
     import hashlib
     sha = hashlib.sha1()
 except ImportError:
     try:
         import sha
         sha.sha # silence unused import warning
     except ImportError:
         raise SystemExit(
             "Couldn't import standard hashlib (incomplete Python install).")
 
 try:
     import zlib
     zlib.compressobj # silence unused import warning
 except ImportError:
     raise SystemExit(
         "Couldn't import standard zlib (incomplete Python install).")
 
 # The base IronPython distribution (as of 2.7.1) doesn't support bz2
 isironpython = False
 try:
     isironpython = (platform.python_implementation()
                     .lower().find("ironpython") != -1)
 except AttributeError:
     pass
 
 if isironpython:
     sys.stderr.write("warning: IronPython detected (no bz2 support)\n")
 else:
     try:
         import bz2
         bz2.BZ2Compressor # silence unused import warning
     except ImportError:
         raise SystemExit(
             "Couldn't import standard bz2 (incomplete Python install).")
 
 ispypy = "PyPy" in sys.version
 
 import ctypes
 import os, stat, subprocess, time
 import re
 import shutil
 import tempfile
 from distutils import log
 # We have issues with setuptools on some platforms and builders. Until
 # those are resolved, setuptools is opt-in except for platforms where
 # we don't have issues.
 if os.name == 'nt' or 'FORCE_SETUPTOOLS' in os.environ:
     from setuptools import setup
 else:
     from distutils.core import setup
 from distutils.ccompiler import new_compiler
 from distutils.core import Command, Extension
 from distutils.dist import Distribution
 from distutils.command.build import build
 from distutils.command.build_ext import build_ext
 from distutils.command.build_py import build_py
 from distutils.command.build_scripts import build_scripts
 from distutils.command.install_lib import install_lib
 from distutils.command.install_scripts import install_scripts
 from distutils.spawn import spawn, find_executable
 from distutils import file_util
 from distutils.errors import (
     CCompilerError,
     DistutilsError,
     DistutilsExecError,
 )
 from distutils.sysconfig import get_python_inc, get_config_var
 from distutils.version import StrictVersion
 
 scripts = ['hg']
 if os.name == 'nt':
     # We remove hg.bat if we are able to build hg.exe.
     scripts.append('contrib/win32/hg.bat')
 
 def cancompile(cc, code):
     tmpdir = tempfile.mkdtemp(prefix='hg-install-')
     devnull = oldstderr = None
     try:
         fname = os.path.join(tmpdir, 'testcomp.c')
         f = open(fname, 'w')
         f.write(code)
         f.close()
         # Redirect stderr to /dev/null to hide any error messages
         # from the compiler.
         # This will have to be changed if we ever have to check
         # for a function on Windows.
         devnull = open('/dev/null', 'w')
         oldstderr = os.dup(sys.stderr.fileno())
         os.dup2(devnull.fileno(), sys.stderr.fileno())
         objects = cc.compile([fname], output_dir=tmpdir)
         cc.link_executable(objects, os.path.join(tmpdir, "a.out"))
         return True
     except Exception:
         return False
     finally:
         if oldstderr is not None:
             os.dup2(oldstderr, sys.stderr.fileno())
         if devnull is not None:
             devnull.close()
         shutil.rmtree(tmpdir)
 
 # simplified version of distutils.ccompiler.CCompiler.has_function
 # that actually removes its temporary files.
 def hasfunction(cc, funcname):
     code = 'int main(void) { %s(); }\n' % funcname
     return cancompile(cc, code)
 
 def hasheader(cc, headername):
     code = '#include <%s>\nint main(void) { return 0; }\n' % headername
     return cancompile(cc, code)
 
 # py2exe needs to be installed to work
 try:
     import py2exe
     py2exe.Distribution # silence unused import warning
     py2exeloaded = True
     # import py2exe's patched Distribution class
     from distutils.core import Distribution
 except ImportError:
     py2exeloaded = False
 
 def runcmd(cmd, env):
     if (sys.platform == 'plan9'
        and (sys.version_info[0] == 2 and sys.version_info[1] < 7)):
         # subprocess kludge to work around issues in half-baked Python
         # ports, notably bichued/python:
         _, out, err = os.popen3(cmd)
         return str(out), str(err)
     else:
         p = subprocess.Popen(cmd, stdout=subprocess.PIPE,
                              stderr=subprocess.PIPE, env=env)
         out, err = p.communicate()
         return out, err
 
 def runhg(cmd, env):
     out, err = runcmd(cmd, env)
     # If root is executing setup.py, but the repository is owned by
     # another user (as in "sudo python setup.py install") we will get
     # trust warnings since the .hg/hgrc file is untrusted. That is
     # fine, we don't want to load it anyway.  Python may warn about
     # a missing __init__.py in mercurial/locale, we also ignore that.
     err = [e for e in err.splitlines()
            if not e.startswith(b'not trusting file') \
               and not e.startswith(b'warning: Not importing') \
               and not e.startswith(b'obsolete feature not enabled')]
     if err:
         printf("stderr from '%s':" % (' '.join(cmd)), file=sys.stderr)
         printf(b'\n'.join([b'  ' + e for e in err]), file=sys.stderr)
         return ''
     return out
 
 version = ''
 
 # Execute hg out of this directory with a custom environment which takes care
 # to not use any hgrc files and do no localization.
 env = {'HGMODULEPOLICY': 'py',
        'HGRCPATH': '',
        'LANGUAGE': 'C',
        'PATH': ''} # make pypi modules that use os.environ['PATH'] happy
 if 'LD_LIBRARY_PATH' in os.environ:
     env['LD_LIBRARY_PATH'] = os.environ['LD_LIBRARY_PATH']
 if 'SystemRoot' in os.environ:
     # Copy SystemRoot into the custom environment for Python 2.6
     # under Windows. Otherwise, the subprocess will fail with
     # error 0xc0150004. See: http://bugs.python.org/issue3440
     env['SystemRoot'] = os.environ['SystemRoot']
 
 if os.path.isdir('.hg'):
     cmd = [sys.executable, 'hg', 'log', '-r', '.', '--template', '{tags}\n']
     numerictags = [t for t in runhg(cmd, env).split() if t[0].isdigit()]
     hgid = runhg([sys.executable, 'hg', 'id', '-i'], env).strip()
     if numerictags: # tag(s) found
         version = numerictags[-1]
         if hgid.endswith('+'): # propagate the dirty status to the tag
             version += '+'
     else: # no tag found
         ltagcmd = [sys.executable, 'hg', 'parents', '--template',
                    '{latesttag}']
         ltag = runhg(ltagcmd, env)
         changessincecmd = [sys.executable, 'hg', 'log', '-T', 'x\n', '-r',
                            "only(.,'%s')" % ltag]
         changessince = len(runhg(changessincecmd, env).splitlines())
         version = '%s+%s-%s' % (ltag, changessince, hgid)
     if version.endswith('+'):
         version += time.strftime('%Y%m%d')
 elif os.path.exists('.hg_archival.txt'):
     kw = dict([[t.strip() for t in l.split(':', 1)]
                for l in open('.hg_archival.txt')])
     if 'tag' in kw:
         version = kw['tag']
     elif 'latesttag' in kw:
         if 'changessincelatesttag' in kw:
             version = '%(latesttag)s+%(changessincelatesttag)s-%(node).12s' % kw
         else:
             version = '%(latesttag)s+%(latesttagdistance)s-%(node).12s' % kw
     else:
         version = kw.get('node', '')[:12]
 
 if version:
     with open("mercurial/__version__.py", "w") as f:
         f.write('# this file is autogenerated by setup.py\n')
         f.write('version = "%s"\n' % version)
 
 try:
     oldpolicy = os.environ.get('HGMODULEPOLICY', None)
     os.environ['HGMODULEPOLICY'] = 'py'
     from mercurial import __version__
     version = __version__.version
 except ImportError:
     version = 'unknown'
 finally:
     if oldpolicy is None:
         del os.environ['HGMODULEPOLICY']
     else:
         os.environ['HGMODULEPOLICY'] = oldpolicy
 
 class hgbuild(build):
     # Insert hgbuildmo first so that files in mercurial/locale/ are found
     # when build_py is run next.
     sub_commands = [('build_mo', None)] + build.sub_commands
 
 class hgbuildmo(build):
 
     description = "build translations (.mo files)"
 
     def run(self):
         if not find_executable('msgfmt'):
             self.warn("could not find msgfmt executable, no translations "
                      "will be built")
             return
 
         podir = 'i18n'
         if not os.path.isdir(podir):
             self.warn("could not find %s/ directory" % podir)
             return
 
         join = os.path.join
         for po in os.listdir(podir):
             if not po.endswith('.po'):
                 continue
             pofile = join(podir, po)
             modir = join('locale', po[:-3], 'LC_MESSAGES')
             mofile = join(modir, 'hg.mo')
             mobuildfile = join('mercurial', mofile)
             cmd = ['msgfmt', '-v', '-o', mobuildfile, pofile]
             if sys.platform != 'sunos5':
                 # msgfmt on Solaris does not know about -c
                 cmd.append('-c')
             self.mkpath(join('mercurial', modir))
             self.make_file([pofile], mobuildfile, spawn, (cmd,))
 
 
 class hgdist(Distribution):
     pure = False
     cffi = ispypy
 
     global_options = Distribution.global_options + \
                      [('pure', None, "use pure (slow) Python "
                         "code instead of C extensions"),
                      ]
 
     def has_ext_modules(self):
         # self.ext_modules is emptied in hgbuildpy.finalize_options which is
         # too late for some cases
         return not self.pure and Distribution.has_ext_modules(self)
 
 # This is ugly as a one-liner. So use a variable.
 buildextnegops = dict(getattr(build_ext, 'negative_options', {}))
 buildextnegops['no-zstd'] = 'zstd'
 
 class hgbuildext(build_ext):
     user_options = build_ext.user_options + [
         ('zstd', None, 'compile zstd bindings [default]'),
         ('no-zstd', None, 'do not compile zstd bindings'),
     ]
 
     boolean_options = build_ext.boolean_options + ['zstd']
     negative_opt = buildextnegops
 
     def initialize_options(self):
         self.zstd = True
         return build_ext.initialize_options(self)
 
     def build_extensions(self):
         # Filter out zstd if disabled via argument.
         if not self.zstd:
             self.extensions = [e for e in self.extensions
                                if e.name != 'mercurial.zstd']
 
         return build_ext.build_extensions(self)
 
     def build_extension(self, ext):
         try:
             build_ext.build_extension(self, ext)
         except CCompilerError:
             if not getattr(ext, 'optional', False):
                 raise
             log.warn("Failed to build optional extension '%s' (skipping)",
                      ext.name)
 
 class hgbuildscripts(build_scripts):
     def run(self):
         if os.name != 'nt' or self.distribution.pure:
             return build_scripts.run(self)
 
         exebuilt = False
         try:
             self.run_command('build_hgexe')
             exebuilt = True
         except (DistutilsError, CCompilerError):
             log.warn('failed to build optional hg.exe')
 
         if exebuilt:
             # Copying hg.exe to the scripts build directory ensures it is
             # installed by the install_scripts command.
             hgexecommand = self.get_finalized_command('build_hgexe')
             dest = os.path.join(self.build_dir, 'hg.exe')
             self.mkpath(self.build_dir)
             self.copy_file(hgexecommand.hgexepath, dest)
 
             # Remove hg.bat because it is redundant with hg.exe.
             self.scripts.remove('contrib/win32/hg.bat')
 
         return build_scripts.run(self)
 
 class hgbuildpy(build_py):
     def finalize_options(self):
         build_py.finalize_options(self)
 
         if self.distribution.pure:
             self.distribution.ext_modules = []
         elif self.distribution.cffi:
             from mercurial.cffi import (
                 bdiff,
                 mpatch,
             )
             exts = [mpatch.ffi.distutils_extension(),
                     bdiff.ffi.distutils_extension()]
             # cffi modules go here
             if sys.platform == 'darwin':
                 from mercurial.cffi import osutil
                 exts.append(osutil.ffi.distutils_extension())
             self.distribution.ext_modules = exts
         else:
             h = os.path.join(get_python_inc(), 'Python.h')
             if not os.path.exists(h):
                 raise SystemExit('Python headers are required to build '
                                  'Mercurial but weren\'t found in %s' % h)
 
     def run(self):
         if self.distribution.pure:
             modulepolicy = 'py'
         elif self.build_lib == '.':
             # in-place build should run without rebuilding C extensions
             modulepolicy = 'allow'
         else:
             modulepolicy = 'c'
         with open("mercurial/__modulepolicy__.py", "w") as f:
             f.write('# this file is autogenerated by setup.py\n')
             f.write('modulepolicy = b"%s"\n' % modulepolicy)
 
         build_py.run(self)
 
 class buildhgextindex(Command):
     description = 'generate prebuilt index of hgext (for frozen package)'
     user_options = []
     _indexfilename = 'hgext/__index__.py'
 
     def initialize_options(self):
         pass
 
     def finalize_options(self):
         pass
 
     def run(self):
         if os.path.exists(self._indexfilename):
             with open(self._indexfilename, 'w') as f:
                 f.write('# empty\n')
 
         # here no extension enabled, disabled() lists up everything
         code = ('import pprint; from mercurial import extensions; '
                 'pprint.pprint(extensions.disabled())')
         out, err = runcmd([sys.executable, '-c', code], env)
         if err:
             raise DistutilsExecError(err)
 
         with open(self._indexfilename, 'w') as f:
             f.write('# this file is autogenerated by setup.py\n')
             f.write('docs = ')
             f.write(out)
 
 class buildhgexe(build_ext):
     description = 'compile hg.exe from mercurial/exewrapper.c'
 
     def build_extensions(self):
         if os.name != 'nt':
             return
         if isinstance(self.compiler, HackedMingw32CCompiler):
             self.compiler.compiler_so = self.compiler.compiler # no -mdll
             self.compiler.dll_libraries = [] # no -lmsrvc90
 
         # Different Python installs can have different Python library
         # names. e.g. the official CPython distribution uses pythonXY.dll
         # and MinGW uses libpythonX.Y.dll.
         _kernel32 = ctypes.windll.kernel32
         _kernel32.GetModuleFileNameA.argtypes = [ctypes.c_void_p,
                                                  ctypes.c_void_p,
                                                  ctypes.c_ulong]
         _kernel32.GetModuleFileNameA.restype = ctypes.c_ulong
         size = 1000
         buf = ctypes.create_string_buffer(size + 1)
         filelen = _kernel32.GetModuleFileNameA(sys.dllhandle, ctypes.byref(buf),
                                                size)
 
         if filelen > 0 and filelen != size:
             dllbasename = os.path.basename(buf.value)
             if not dllbasename.lower().endswith('.dll'):
                 raise SystemExit('Python DLL does not end with .dll: %s' %
                                  dllbasename)
             pythonlib = dllbasename[:-4]
         else:
             log.warn('could not determine Python DLL filename; '
                      'assuming pythonXY')
 
             hv = sys.hexversion
             pythonlib = 'python%d%d' % (hv >> 24, (hv >> 16) & 0xff)
 
         log.info('using %s as Python library name' % pythonlib)
         with open('mercurial/hgpythonlib.h', 'wb') as f:
             f.write('/* this file is autogenerated by setup.py */\n')
             f.write('#define HGPYTHONLIB "%s"\n' % pythonlib)
         objects = self.compiler.compile(['mercurial/exewrapper.c'],
                                          output_dir=self.build_temp)
         dir = os.path.dirname(self.get_ext_fullpath('dummy'))
         target = os.path.join(dir, 'hg')
         self.compiler.link_executable(objects, target,
                                       libraries=[],
                                       output_dir=self.build_temp)
 
     @property
     def hgexepath(self):
         dir = os.path.dirname(self.get_ext_fullpath('dummy'))
         return os.path.join(self.build_temp, dir, 'hg.exe')
 
 class hginstalllib(install_lib):
     '''
     This is a specialization of install_lib that replaces the copy_file used
     there so that it supports setting the mode of files after copying them,
     instead of just preserving the mode that the files originally had.  If your
     system has a umask of something like 027, preserving the permissions when
     copying will lead to a broken install.
 
     Note that just passing keep_permissions=False to copy_file would be
     insufficient, as it might still be applying a umask.
     '''
 
     def run(self):
         realcopyfile = file_util.copy_file
         def copyfileandsetmode(*args, **kwargs):
             src, dst = args[0], args[1]
             dst, copied = realcopyfile(*args, **kwargs)
             if copied:
                 st = os.stat(src)
                 # Persist executable bit (apply it to group and other if user
                 # has it)
                 if st[stat.ST_MODE] & stat.S_IXUSR:
                     setmode = int('0755', 8)
                 else:
                     setmode = int('0644', 8)
                 m = stat.S_IMODE(st[stat.ST_MODE])
                 m = (m & ~int('0777', 8)) | setmode
                 os.chmod(dst, m)
         file_util.copy_file = copyfileandsetmode
         try:
             install_lib.run(self)
         finally:
             file_util.copy_file = realcopyfile
 
 class hginstallscripts(install_scripts):
     '''
     This is a specialization of install_scripts that replaces the @LIBDIR@ with
     the configured directory for modules. If possible, the path is made relative
     to the directory for scripts.
     '''
 
     def initialize_options(self):
         install_scripts.initialize_options(self)
 
         self.install_lib = None
 
     def finalize_options(self):
         install_scripts.finalize_options(self)
         self.set_undefined_options('install',
                                    ('install_lib', 'install_lib'))
 
     def run(self):
         install_scripts.run(self)
 
         # It only makes sense to replace @LIBDIR@ with the install path if
         # the install path is known. For wheels, the logic below calculates
         # the libdir to be "../..". This is because the internal layout of a
         # wheel archive looks like:
         #
         #   mercurial-3.6.1.data/scripts/hg
         #   mercurial/__init__.py
         #
         # When installing wheels, the subdirectories of the "<pkg>.data"
         # directory are translated to system local paths and files therein
         # are copied in place. The mercurial/* files are installed into the
         # site-packages directory. However, the site-packages directory
         # isn't known until wheel install time. This means we have no clue
         # at wheel generation time what the installed site-packages directory
         # will be. And, wheels don't appear to provide the ability to register
         # custom code to run during wheel installation. This all means that
         # we can't reliably set the libdir in wheels: the default behavior
         # of looking in sys.path must do.
 
         if (os.path.splitdrive(self.install_dir)[0] !=
             os.path.splitdrive(self.install_lib)[0]):
             # can't make relative paths from one drive to another, so use an
             # absolute path instead
             libdir = self.install_lib
         else:
             common = os.path.commonprefix((self.install_dir, self.install_lib))
             rest = self.install_dir[len(common):]
             uplevel = len([n for n in os.path.split(rest) if n])
 
             libdir = uplevel * ('..' + os.sep) + self.install_lib[len(common):]
 
         for outfile in self.outfiles:
             with open(outfile, 'rb') as fp:
                 data = fp.read()
 
             # skip binary files
             if b'\0' in data:
                 continue
 
             # During local installs, the shebang will be rewritten to the final
             # install path. During wheel packaging, the shebang has a special
             # value.
             if data.startswith(b'#!python'):
                 log.info('not rewriting @LIBDIR@ in %s because install path '
                          'not known' % outfile)
                 continue
 
             data = data.replace(b'@LIBDIR@', libdir.encode(libdir_escape))
             with open(outfile, 'wb') as fp:
                 fp.write(data)
 
 cmdclass = {'build': hgbuild,
             'build_mo': hgbuildmo,
             'build_ext': hgbuildext,
             'build_py': hgbuildpy,
             'build_scripts': hgbuildscripts,
             'build_hgextindex': buildhgextindex,
             'install_lib': hginstalllib,
             'install_scripts': hginstallscripts,
             'build_hgexe': buildhgexe,
             }
 
 packages = ['mercurial',
             'mercurial.cext',
             'mercurial.hgweb',
             'mercurial.httpclient',
             'mercurial.pure',
             'hgext', 'hgext.convert', 'hgext.fsmonitor',
             'hgext.fsmonitor.pywatchman', 'hgext.highlight',
             'hgext.largefiles', 'hgext.zeroconf', 'hgext3rd']
 
 common_depends = ['mercurial/bitmanipulation.h',
                   'mercurial/compat.h',
                   'mercurial/util.h']
 common_include_dirs = ['mercurial']
 
 osutil_cflags = []
 osutil_ldflags = []
 
 # platform specific macros
 for plat, func in [('bsd', 'setproctitle')]:
     if re.search(plat, sys.platform) and hasfunction(new_compiler(), func):
         osutil_cflags.append('-DHAVE_%s' % func.upper())
 
 for plat, macro, code in [
     ('bsd|darwin', 'BSD_STATFS', '''
      #include <sys/param.h>
      #include <sys/mount.h>
      int main() { struct statfs s; return sizeof(s.f_fstypename); }
      '''),
     ('linux', 'LINUX_STATFS', '''
      #include <linux/magic.h>
      #include <sys/vfs.h>
      int main() { struct statfs s; return sizeof(s.f_type); }
      '''),
 ]:
     if re.search(plat, sys.platform) and cancompile(new_compiler(), code):
         osutil_cflags.append('-DHAVE_%s' % macro)
 
 if sys.platform == 'darwin':
     osutil_ldflags += ['-framework', 'ApplicationServices']
 
 extmodules = [
     Extension('mercurial.base85', ['mercurial/base85.c'],
               include_dirs=common_include_dirs,
               depends=common_depends),
     Extension('mercurial.bdiff', ['mercurial/bdiff.c',
                                   'mercurial/bdiff_module.c'],
               include_dirs=common_include_dirs,
               depends=common_depends + ['mercurial/bdiff.h']),
     Extension('mercurial.diffhelpers', ['mercurial/diffhelpers.c'],
               include_dirs=common_include_dirs,
               depends=common_depends),
     Extension('mercurial.mpatch', ['mercurial/mpatch.c',
                                    'mercurial/mpatch_module.c'],
               include_dirs=common_include_dirs,
               depends=common_depends),
     Extension('mercurial.parsers', ['mercurial/dirs.c',
                                     'mercurial/manifest.c',
                                     'mercurial/parsers.c',
                                     'mercurial/pathencode.c'],
               include_dirs=common_include_dirs,
               depends=common_depends),
-    Extension('mercurial.osutil', ['mercurial/osutil.c'],
+    Extension('mercurial.cext.osutil', ['mercurial/cext/osutil.c'],
               include_dirs=common_include_dirs,
               extra_compile_args=osutil_cflags,
               extra_link_args=osutil_ldflags,
               depends=common_depends),
     Extension('hgext.fsmonitor.pywatchman.bser',
               ['hgext/fsmonitor/pywatchman/bser.c']),
     ]
 
 sys.path.insert(0, 'contrib/python-zstandard')
 import setup_zstd
 extmodules.append(setup_zstd.get_c_extension(name='mercurial.zstd'))
 
 try:
     from distutils import cygwinccompiler
 
     # the -mno-cygwin option has been deprecated for years
     compiler = cygwinccompiler.Mingw32CCompiler
 
     class HackedMingw32CCompiler(cygwinccompiler.Mingw32CCompiler):
         def __init__(self, *args, **kwargs):
             compiler.__init__(self, *args, **kwargs)
             for i in 'compiler compiler_so linker_exe linker_so'.split():
                 try:
                     getattr(self, i).remove('-mno-cygwin')
                 except ValueError:
                     pass
 
     cygwinccompiler.Mingw32CCompiler = HackedMingw32CCompiler
 except ImportError:
     # the cygwinccompiler package is not available on some Python
     # distributions like the ones from the optware project for Synology
     # DiskStation boxes
     class HackedMingw32CCompiler(object):
         pass
 
 packagedata = {'mercurial': ['locale/*/LC_MESSAGES/hg.mo',
                              'help/*.txt',
                              'help/internals/*.txt',
                              'default.d/*.rc',
                              'dummycert.pem']}
 
 def ordinarypath(p):
     return p and p[0] != '.' and p[-1] != '~'
 
 for root in ('templates',):
     for curdir, dirs, files in os.walk(os.path.join('mercurial', root)):
         curdir = curdir.split(os.sep, 1)[1]
         dirs[:] = filter(ordinarypath, dirs)
         for f in filter(ordinarypath, files):
             f = os.path.join(curdir, f)
             packagedata['mercurial'].append(f)
 
 datafiles = []
 
 # distutils expects version to be str/unicode. Converting it to
 # unicode on Python 2 still works because it won't contain any
 # non-ascii bytes and will be implicitly converted back to bytes
 # when operated on.
 assert isinstance(version, bytes)
 setupversion = version.decode('ascii')
 
 extra = {}
 
 if py2exeloaded:
     extra['console'] = [
         {'script':'hg',
          'copyright':'Copyright (C) 2005-2017 Matt Mackall and others',
          'product_version':version}]
     # sub command of 'build' because 'py2exe' does not handle sub_commands
     build.sub_commands.insert(0, ('build_hgextindex', None))
     # put dlls in sub directory so that they won't pollute PATH
     extra['zipfile'] = 'lib/library.zip'
 
 if os.name == 'nt':
     # Windows binary file versions for exe/dll files must have the
     # form W.X.Y.Z, where W,X,Y,Z are numbers in the range 0..65535
     setupversion = version.split('+', 1)[0]
 
 if sys.platform == 'darwin' and os.path.exists('/usr/bin/xcodebuild'):
     version = runcmd(['/usr/bin/xcodebuild', '-version'], {})[0].splitlines()
     if version:
         version = version[0]
         if sys.version_info[0] == 3:
             version = version.decode('utf-8')
         xcode4 = (version.startswith('Xcode') and
                   StrictVersion(version.split()[1]) >= StrictVersion('4.0'))
         xcode51 = re.match(r'^Xcode\s+5\.1', version) is not None
     else:
         # xcodebuild returns empty on OS X Lion with XCode 4.3 not
         # installed, but instead with only command-line tools. Assume
         # that only happens on >= Lion, thus no PPC support.
         xcode4 = True
         xcode51 = False
 
     # XCode 4.0 dropped support for ppc architecture, which is hardcoded in
     # distutils.sysconfig
     if xcode4:
         os.environ['ARCHFLAGS'] = ''
 
     # XCode 5.1 changes clang such that it now fails to compile if the
     # -mno-fused-madd flag is passed, but the version of Python shipped with
     # OS X 10.9 Mavericks includes this flag. This causes problems in all
     # C extension modules, and a bug has been filed upstream at
     # http://bugs.python.org/issue21244. We also need to patch this here
     # so Mercurial can continue to compile in the meantime.
     if xcode51:
         cflags = get_config_var('CFLAGS')
         if cflags and re.search(r'-mno-fused-madd\b', cflags) is not None:
             os.environ['CFLAGS'] = (
                 os.environ.get('CFLAGS', '') + ' -Qunused-arguments')
 
 setup(name='mercurial',
       version=setupversion,
       author='Matt Mackall and many others',
       author_email='[e-mail]',
       url='https://mercurial-scm.org/',
       download_url='https://mercurial-scm.org/release/',
       description=('Fast scalable distributed SCM (revision control, version '
                    'control) system'),
       long_description=('Mercurial is a distributed SCM tool written in Python.'
                         ' It is used by a number of large projects that require'
                         ' fast, reliable distributed revision control, such as '
                         'Mozilla.'),
       license='GNU GPLv2 or any later version',
       classifiers=[
           'Development Status :: 6 - Mature',
           'Environment :: Console',
           'Intended Audience :: Developers',
           'Intended Audience :: System Administrators',
           'License :: OSI Approved :: GNU General Public License (GPL)',
           'Natural Language :: Danish',
           'Natural Language :: English',
           'Natural Language :: German',
           'Natural Language :: Italian',
           'Natural Language :: Japanese',
           'Natural Language :: Portuguese (Brazilian)',
           'Operating System :: Microsoft :: Windows',
           'Operating System :: OS Independent',
           'Operating System :: POSIX',
           'Programming Language :: C',
           'Programming Language :: Python',
           'Topic :: Software Development :: Version Control',
       ],
       scripts=scripts,
       packages=packages,
       ext_modules=extmodules,
       data_files=datafiles,
       package_data=packagedata,
       cmdclass=cmdclass,
       distclass=hgdist,
       options={'py2exe': {'packages': ['hgext', 'email']},
                'bdist_mpkg': {'zipdist': False,
                               'license': 'COPYING',
                               'readme': 'contrib/macosx/Readme.html',
                               'welcome': 'contrib/macosx/Welcome.html',
                               },
                },
       **extra)