def index_basic():
cd = os.path.realpath("../../../t")
WVPASS(cd)
sd = os.path.realpath(cd + "/sampledata")
WVPASSEQ(helpers.resolve_parent(cd + "/sampledata"), sd)
WVPASSEQ(os.path.realpath(cd + "/sampledata/x"), sd + "/x")
WVPASSEQ(os.path.realpath(cd + "/sampledata/var/abs-symlink"), sd + "/var/abs-symlink-target")
WVPASSEQ(helpers.resolve_parent(cd + "/sampledata/var/abs-symlink"), sd + "/var/abs-symlink")
def index_basic():
cd = os.path.realpath('../../../t')
WVPASS(cd)
sd = os.path.realpath(cd + '/sampledata')
WVPASSEQ(resolve_parent(cd + '/sampledata'), sd)
WVPASSEQ(os.path.realpath(cd + '/sampledata/x'), sd + '/x')
WVPASSEQ(os.path.realpath(cd + '/sampledata/var/abs-symlink'),
sd + '/var/abs-symlink-target')
WVPASSEQ(resolve_parent(cd + '/sampledata/var/abs-symlink'),
sd + '/var/abs-symlink')
def test_index_basic():
cd = os.path.realpath(os.path.join(lib_t_dir, b'../'))
WVPASS(cd)
sd = os.path.realpath(cd + b'/sampledata')
WVPASSEQ(resolve_parent(cd + b'/sampledata'), sd)
WVPASSEQ(os.path.realpath(cd + b'/sampledata/x'), sd + b'/x')
WVPASSEQ(os.path.realpath(cd + b'/sampledata/var/abs-symlink'),
sd + b'/var/abs-symlink-target')
WVPASSEQ(resolve_parent(cd + b'/sampledata/var/abs-symlink'),
sd + b'/var/abs-symlink')
def index_basic():
cd = os.path.realpath('../../../t')
WVPASS(cd)
sd = os.path.realpath(cd + '/sampledata')
WVPASSEQ(helpers.resolve_parent(cd + '/sampledata'), sd)
WVPASSEQ(os.path.realpath(cd + '/sampledata/x'), sd + '/x')
WVPASSEQ(os.path.realpath(cd + '/sampledata/var/abs-symlink'),
sd + '/var/abs-symlink-target')
WVPASSEQ(helpers.resolve_parent(cd + '/sampledata/var/abs-symlink'),
sd + '/var/abs-symlink')
def index_basic():
with no_lingering_errors():
cd = os.path.realpath('../../../t')
WVPASS(cd)
sd = os.path.realpath(cd + '/sampledata')
WVPASSEQ(resolve_parent(cd + '/sampledata'), sd)
WVPASSEQ(os.path.realpath(cd + '/sampledata/x'), sd + '/x')
WVPASSEQ(os.path.realpath(cd + '/sampledata/var/abs-symlink'),
sd + '/var/abs-symlink-target')
WVPASSEQ(resolve_parent(cd + '/sampledata/var/abs-symlink'),
sd + '/var/abs-symlink')
def test_metadata_method():
initial_failures = wvfailure_count()
tmpdir = tempfile.mkdtemp(dir=bup_tmp, prefix='bup-tmetadata-')
bup_dir = tmpdir + '/bup'
data_path = tmpdir + '/foo'
os.mkdir(data_path)
ex('touch', data_path + '/file')
ex('ln', '-s', 'file', data_path + '/symlink')
test_time1 = 13 * 1000000000
test_time2 = 42 * 1000000000
utime(data_path + '/file', (0, test_time1))
lutime(data_path + '/symlink', (0, 0))
utime(data_path, (0, test_time2))
ex(bup_path, '-d', bup_dir, 'init')
ex(bup_path, '-d', bup_dir, 'index', '-v', data_path)
ex(bup_path, '-d', bup_dir, 'save', '-tvvn', 'test', data_path)
git.check_repo_or_die(bup_dir)
top = vfs.RefList(None)
n = top.lresolve('/test/latest' + resolve_parent(data_path))
m = n.metadata()
WVPASS(m.mtime == test_time2)
WVPASS(len(n.subs()) == 2)
WVPASS(n.name == 'foo')
WVPASS(set([x.name for x in n.subs()]) == set(['file', 'symlink']))
for sub in n:
if sub.name == 'file':
m = sub.metadata()
WVPASS(m.mtime == test_time1)
elif sub.name == 'symlink':
m = sub.metadata()
WVPASS(m.mtime == 0)
if wvfailure_count() == initial_failures:
subprocess.call(['rm', '-rf', tmpdir])
def test_metadata_method():
with no_lingering_errors(), test_tempdir('bup-tmetadata-') as tmpdir:
bup_dir = tmpdir + '/bup'
data_path = tmpdir + '/foo'
os.mkdir(data_path)
ex('touch', data_path + '/file')
ex('ln', '-s', 'file', data_path + '/symlink')
test_time1 = 13 * 1000000000
test_time2 = 42 * 1000000000
utime(data_path + '/file', (0, test_time1))
lutime(data_path + '/symlink', (0, 0))
utime(data_path, (0, test_time2))
ex(bup_path, '-d', bup_dir, 'init')
ex(bup_path, '-d', bup_dir, 'index', '-v', data_path)
ex(bup_path, '-d', bup_dir, 'save', '-tvvn', 'test', data_path)
git.check_repo_or_die(bup_dir)
top = vfs.RefList(None)
n = top.lresolve('/test/latest' + resolve_parent(data_path))
m = n.metadata()
WVPASS(m.mtime == test_time2)
WVPASS(len(n.subs()) == 2)
WVPASS(n.name == 'foo')
WVPASS(set([x.name for x in n.subs()]) == set(['file', 'symlink']))
for sub in n:
if sub.name == 'file':
m = sub.metadata()
WVPASS(m.mtime == test_time1)
elif sub.name == 'symlink':
m = sub.metadata()
WVPASS(m.mtime == 0)
def test_metadata_method():
initial_failures = wvfailure_count()
tmpdir = tempfile.mkdtemp(dir=bup_tmp, prefix='bup-tmetadata-')
bup_dir = tmpdir + '/bup'
data_path = tmpdir + '/foo'
os.mkdir(data_path)
ex('touch', data_path + '/file')
ex('ln', '-s', 'file', data_path + '/symlink')
test_time1 = 13 * 1000000000
test_time2 = 42 * 1000000000
utime(data_path + '/file', (0, test_time1))
lutime(data_path + '/symlink', (0, 0))
utime(data_path, (0, test_time2))
ex(bup_path, '-d', bup_dir, 'init')
ex(bup_path, '-d', bup_dir, 'index', '-v', data_path)
ex(bup_path, '-d', bup_dir, 'save', '-tvvn', 'test', data_path)
git.check_repo_or_die(bup_dir)
top = vfs.RefList(None)
n = top.lresolve('/test/latest' + resolve_parent(data_path))
m = n.metadata()
WVPASS(m.mtime == test_time2)
WVPASS(len(n.subs()) == 2)
WVPASS(n.name == 'foo')
WVPASS(set([x.name for x in n.subs()]) == set(['file', 'symlink']))
for sub in n:
if sub.name == 'file':
m = sub.metadata()
WVPASS(m.mtime == test_time1)
elif sub.name == 'symlink':
m = sub.metadata()
WVPASS(m.mtime == 0)
if wvfailure_count() == initial_failures:
subprocess.call(['rm', '-rf', tmpdir])
def test_metadata_method(tmpdir):
bup_dir = tmpdir + b'/bup'
data_path = tmpdir + b'/foo'
os.mkdir(data_path)
ex(b'touch', data_path + b'/file')
ex(b'ln', b'-s', b'file', data_path + b'/symlink')
test_time1 = 13 * 1000000000
test_time2 = 42 * 1000000000
utime(data_path + b'/file', (0, test_time1))
lutime(data_path + b'/symlink', (0, 0))
utime(data_path, (0, test_time2))
ex(bup_path, b'-d', bup_dir, b'init')
ex(bup_path, b'-d', bup_dir, b'index', b'-v', data_path)
ex(bup_path, b'-d', bup_dir, b'save', b'-tvvn', b'test', data_path)
repo = LocalRepo(bup_dir)
resolved = vfs.resolve(repo,
b'/test/latest' + resolve_parent(data_path),
follow=False)
leaf_name, leaf_item = resolved[-1]
m = leaf_item.meta
WVPASS(m.mtime == test_time2)
WVPASS(leaf_name == b'foo')
contents = tuple(vfs.contents(repo, leaf_item))
WVPASS(len(contents) == 3)
WVPASSEQ(frozenset(name for name, item in contents),
frozenset((b'.', b'file', b'symlink')))
for name, item in contents:
if name == b'file':
m = item.meta
WVPASS(m.mtime == test_time1)
elif name == b'symlink':
m = item.meta
WVPASSEQ(m.symlink_target, b'file')
WVPASSEQ(m.size, 4)
WVPASSEQ(m.mtime, 0)
def __init__(self, filename, metastore, tmax):
self.rootlevel = self.level = Level([], None)
self.f = None
self.count = 0
self.lastfile = None
self.filename = None
self.filename = filename = resolve_parent(filename)
self.metastore = metastore
self.tmax = tmax
(dir, name) = os.path.split(filename)
ffd, self.tmpname = tempfile.mkstemp(b'.tmp', filename, dir)
self.f = os.fdopen(ffd, 'wb', 65536)
self.f.write(INDEX_HDR)
def _dirlist():
l = []
for n in os.listdir('.'):
try:
st = xstat.lstat(n)
except OSError as e:
add_error(Exception('%s: %s' % (resolve_parent(n), str(e))))
continue
if (st.st_mode & _IFMT) == stat.S_IFDIR:
n += '/'
l.append((n,st))
l.sort(reverse=True)
return l
def __init__(self, filename, metastore, tmax):
self.rootlevel = self.level = Level([], None)
self.f = None
self.count = 0
self.lastfile = None
self.filename = None
self.filename = filename = resolve_parent(filename)
self.metastore = metastore
self.tmax = tmax
(dir,name) = os.path.split(filename)
(ffd,self.tmpname) = tempfile.mkstemp('.tmp', filename, dir)
self.f = os.fdopen(ffd, 'wb', 65536)
self.f.write(INDEX_HDR)
def _dirlist():
l = []
for n in os.listdir(b'.'):
try:
st = xstat.lstat(n)
except OSError as e:
add_error(Exception('%s: %s' % (resolve_parent(n), str(e))))
continue
if stat.S_ISDIR(st.st_mode):
n += b'/'
l.append((n, st))
l.sort(reverse=True)
return l
def reduce_paths(paths):
xpaths = []
for p in paths:
rp = _slashappend_or_add_error(resolve_parent(p), "reduce_paths")
if rp:
xpaths.append((rp, slashappend(p) if rp.endswith("/") else p))
xpaths.sort()
paths = []
prev = None
for (rp, p) in xpaths:
if prev and (prev == rp or (prev.endswith("/") and rp.startswith(prev))):
continue # already superceded by previous path
paths.append((rp, p))
prev = rp
paths.sort(reverse=True)
return paths
def reduce_paths(paths):
xpaths = []
for p in paths:
rp = _slashappend_or_add_error(resolve_parent(p), 'reduce_paths')
if rp:
xpaths.append((rp, slashappend(p) if rp.endswith(b'/') else p))
xpaths.sort()
paths = []
prev = None
for (rp, p) in xpaths:
if prev and (prev == rp or
(prev.endswith(b'/') and rp.startswith(prev))):
continue # already superceded by previous path
paths.append((rp, p))
prev = rp
paths.sort(reverse=True)
return paths
def __init__(self, filename, metastore, tmax):
self.closed = False
self.rootlevel = self.level = Level([], None)
self.pending_index = None
self.f = None
self.count = 0
self.lastfile = None
self.filename = None
self.filename = filename = resolve_parent(filename)
self.metastore = metastore
self.tmax = tmax
(dir, name) = os.path.split(filename)
with ExitStack() as self.cleanup:
self.pending_index = atomically_replaced_file(self.filename,
mode='wb',
buffering=65536)
self.f = self.cleanup.enter_context(self.pending_index)
self.cleanup.enter_context(self.f)
self.f.write(INDEX_HDR)
self.cleanup = self.cleanup.pop_all()
def test_metadata_method():
with no_lingering_errors():
with test_tempdir('bup-tmetadata-') as tmpdir:
bup_dir = tmpdir + '/bup'
data_path = tmpdir + '/foo'
os.mkdir(data_path)
ex('touch', data_path + '/file')
ex('ln', '-s', 'file', data_path + '/symlink')
test_time1 = 13 * 1000000000
test_time2 = 42 * 1000000000
utime(data_path + '/file', (0, test_time1))
lutime(data_path + '/symlink', (0, 0))
utime(data_path, (0, test_time2))
ex(bup_path, '-d', bup_dir, 'init')
ex(bup_path, '-d', bup_dir, 'index', '-v', data_path)
ex(bup_path, '-d', bup_dir, 'save', '-tvvn', 'test', data_path)
git.check_repo_or_die(bup_dir)
repo = LocalRepo()
resolved = vfs.resolve(repo,
'/test/latest' + resolve_parent(data_path),
follow=False)
leaf_name, leaf_item = resolved[-1]
m = leaf_item.meta
WVPASS(m.mtime == test_time2)
WVPASS(leaf_name == 'foo')
contents = tuple(vfs.contents(repo, leaf_item))
WVPASS(len(contents) == 3)
WVPASSEQ(frozenset(name for name, item in contents),
frozenset(('.', 'file', 'symlink')))
for name, item in contents:
if name == 'file':
m = item.meta
WVPASS(m.mtime == test_time1)
elif name == 'symlink':
m = item.meta
WVPASSEQ(m.symlink_target, 'file')
WVPASSEQ(m.size, 4)
WVPASSEQ(m.mtime, 0)
def test_metadata_method():
with no_lingering_errors():
with test_tempdir('bup-tmetadata-') as tmpdir:
bup_dir = tmpdir + '/bup'
data_path = tmpdir + '/foo'
os.mkdir(data_path)
ex('touch', data_path + '/file')
ex('ln', '-s', 'file', data_path + '/symlink')
test_time1 = 13 * 1000000000
test_time2 = 42 * 1000000000
utime(data_path + '/file', (0, test_time1))
lutime(data_path + '/symlink', (0, 0))
utime(data_path, (0, test_time2))
ex(bup_path, '-d', bup_dir, 'init')
ex(bup_path, '-d', bup_dir, 'index', '-v', data_path)
ex(bup_path, '-d', bup_dir, 'save', '-tvvn', 'test', data_path)
git.check_repo_or_die(bup_dir)
repo = LocalRepo()
resolved = vfs.resolve(repo,
'/test/latest' + resolve_parent(data_path),
follow=False)
leaf_name, leaf_item = resolved[-1]
m = leaf_item.meta
WVPASS(m.mtime == test_time2)
WVPASS(leaf_name == 'foo')
contents = tuple(vfs.contents(repo, leaf_item))
WVPASS(len(contents) == 3)
WVPASSEQ(frozenset(name for name, item in contents),
frozenset(('.', 'file', 'symlink')))
for name, item in contents:
if name == 'file':
m = item.meta
WVPASS(m.mtime == test_time1)
elif name == 'symlink':
m = item.meta
WVPASSEQ(m.symlink_target, 'file')
WVPASSEQ(m.size, 4)
WVPASSEQ(m.mtime, 0)
def reduce_paths(paths):
xpaths = []
for p in paths:
rp = resolve_parent(p)
try:
st = os.lstat(rp)
if stat.S_ISDIR(st.st_mode):
rp = slashappend(rp)
p = slashappend(p)
xpaths.append((rp, p))
except OSError as e:
add_error('reduce_paths: %s' % e)
xpaths.sort()
paths = []
prev = None
for (rp, p) in xpaths:
if prev and (prev == rp or
(prev.endswith('/') and rp.startswith(prev))):
continue # already superceded by previous path
paths.append((rp, p))
prev = rp
paths.sort(reverse=True)
return paths
def reduce_paths(paths):
xpaths = []
for p in paths:
rp = resolve_parent(p)
try:
st = os.lstat(rp)
if stat.S_ISDIR(st.st_mode):
rp = slashappend(rp)
p = slashappend(p)
xpaths.append((rp, p))
except OSError as e:
add_error('reduce_paths: %s' % e)
xpaths.sort()
paths = []
prev = None
for (rp, p) in xpaths:
if prev and (prev == rp
or (prev.endswith('/') and rp.startswith(prev))):
continue # already superceded by previous path
paths.append((rp, p))
prev = rp
paths.sort(reverse=True)
return paths
if opt.strip:
o.fatal("--strip is incompatible with --graft")
if opt.strip_path:
o.fatal("--strip-path is incompatible with --graft")
for (option, parameter) in flags:
if option == "--graft":
parameter = argv_bytes(parameter)
splitted_parameter = parameter.split(b'=')
if len(splitted_parameter) != 2:
o.fatal("a graft point must be of the form old_path=new_path")
old_path, new_path = splitted_parameter
if not (old_path and new_path):
o.fatal("a graft point cannot be empty")
graft_points.append((resolve_parent(old_path),
resolve_parent(new_path)))
is_reverse = environ.get(b'BUP_SERVER_REVERSE')
if is_reverse and opt.remote:
o.fatal("don't use -r in reverse mode; it's automatic")
name = opt.name
if name and not valid_save_name(name):
o.fatal("'%s' is not a valid branch name" % path_msg(name))
refname = name and b'refs/heads/%s' % name or None
if opt.remote or is_reverse:
try:
cli = client.Client(opt.remote)
except client.ClientError as e:
log('error: %s' % e)
if opt.graft:
if opt.strip:
o.fatal("--strip is incompatible with --graft")
if opt.strip_path:
o.fatal("--strip-path is incompatible with --graft")
for (option, parameter) in flags:
if option == "--graft":
splitted_parameter = parameter.split('=')
if len(splitted_parameter) != 2:
o.fatal("a graft point must be of the form old_path=new_path")
old_path, new_path = splitted_parameter
if not (old_path and new_path):
o.fatal("a graft point cannot be empty")
graft_points.append((resolve_parent(old_path),
resolve_parent(new_path)))
is_reverse = os.environ.get('BUP_SERVER_REVERSE')
if is_reverse and opt.remote:
o.fatal("don't use -r in reverse mode; it's automatic")
if opt.name and opt.name.startswith('.'):
o.fatal("'%s' is not a valid branch name" % opt.name)
refname = opt.name and 'refs/heads/%s' % opt.name or None
if opt.remote or is_reverse:
try:
cli = client.Client(opt.remote)
except client.ClientError as e:
log('error: %s' % e)
sys.exit(1)
o.fatal("--strip is incompatible with --graft")
if opt.strip_path:
o.fatal("--strip-path is incompatible with --graft")
for (option, parameter) in flags:
if option == "--graft":
parameter = argv_bytes(parameter)
splitted_parameter = parameter.split(b'=')
if len(splitted_parameter) != 2:
o.fatal("a graft point must be of the form old_path=new_path")
old_path, new_path = splitted_parameter
if not (old_path and new_path):
o.fatal("a graft point cannot be empty")
graft_points.append(
(resolve_parent(old_path), resolve_parent(new_path)))
name = opt.name
if name and not valid_save_name(name):
o.fatal("'%s' is not a valid branch name" % path_msg(name))
refname = name and b'refs/heads/%s' % name or None
repo = repo.from_opts(opt)
use_treesplit = repo.config(b'bup.treesplit', opttype='bool')
blobbits = repo.config(b'bup.blobbits', opttype='int')
oldref = refname and repo.read_ref(refname) or None
handle_ctrl_c()
# Metadata is stored in a file named .bupm in each directory. The
def opts_from_cmdline(argv):
o = options.Options(optspec)
opt, flags, extra = o.parse_bytes(argv[1:])
if opt.indexfile:
opt.indexfile = argv_bytes(opt.indexfile)
if opt.name:
opt.name = argv_bytes(opt.name)
if opt.remote:
opt.remote = argv_bytes(opt.remote)
if opt.strip_path:
opt.strip_path = argv_bytes(opt.strip_path)
if not (opt.tree or opt.commit or opt.name):
o.fatal("use one or more of -t, -c, -n")
if not extra:
o.fatal("no filenames given")
if opt.date:
opt.date = parse_date_or_fatal(opt.date, o.fatal)
else:
opt.date = time.time()
opt.progress = (istty2 and not opt.quiet)
opt.smaller = parse_num(opt.smaller or 0)
if opt.bwlimit:
opt.bwlimit = parse_num(opt.bwlimit)
if opt.strip and opt.strip_path:
o.fatal("--strip is incompatible with --strip-path")
opt.sources = [argv_bytes(x) for x in extra]
grafts = []
if opt.graft:
if opt.strip:
o.fatal("--strip is incompatible with --graft")
if opt.strip_path:
o.fatal("--strip-path is incompatible with --graft")
for (option, parameter) in flags:
if option == "--graft":
parameter = argv_bytes(parameter)
splitted_parameter = parameter.split(b'=')
if len(splitted_parameter) != 2:
o.fatal("a graft point must be of the form old_path=new_path")
old_path, new_path = splitted_parameter
if not (old_path and new_path):
o.fatal("a graft point cannot be empty")
grafts.append((resolve_parent(old_path),
resolve_parent(new_path)))
opt.grafts = grafts
opt.is_reverse = environ.get(b'BUP_SERVER_REVERSE')
if opt.is_reverse and opt.remote:
o.fatal("don't use -r in reverse mode; it's automatic")
if opt.name and not valid_save_name(opt.name):
o.fatal("'%s' is not a valid branch name" % path_msg(opt.name))
return opt
def unique_resolved_paths(paths):
"Return a collection of unique resolved paths."
rps = (_slashappend_or_add_error(resolve_parent(p),
'unique_resolved_paths') for p in paths)
return frozenset((x for x in rps if x is not None))
def unique_resolved_paths(paths):
"Return a collection of unique resolved paths."
rps = (_slashappend_or_add_error(resolve_parent(p), 'unique_resolved_paths')
for p in paths)
return frozenset((x for x in rps if x is not None))
def main(argv):
# Hack around lack of nonlocal vars in python 2
_nonlocal = {}
o = options.Options(optspec)
opt, flags, extra = o.parse_bytes(argv[1:])
if opt.indexfile:
opt.indexfile = argv_bytes(opt.indexfile)
if opt.name:
opt.name = argv_bytes(opt.name)
if opt.remote:
opt.remote = argv_bytes(opt.remote)
if opt.strip_path:
opt.strip_path = argv_bytes(opt.strip_path)
git.check_repo_or_die()
if not (opt.tree or opt.commit or opt.name):
o.fatal("use one or more of -t, -c, -n")
if not extra:
o.fatal("no filenames given")
extra = [argv_bytes(x) for x in extra]
opt.progress = (istty2 and not opt.quiet)
opt.smaller = parse_num(opt.smaller or 0)
if opt.bwlimit:
client.bwlimit = parse_num(opt.bwlimit)
if opt.date:
date = parse_date_or_fatal(opt.date, o.fatal)
else:
date = time.time()
if opt.strip and opt.strip_path:
o.fatal("--strip is incompatible with --strip-path")
graft_points = []
if opt.graft:
if opt.strip:
o.fatal("--strip is incompatible with --graft")
if opt.strip_path:
o.fatal("--strip-path is incompatible with --graft")
for (option, parameter) in flags:
if option == "--graft":
parameter = argv_bytes(parameter)
splitted_parameter = parameter.split(b'=')
if len(splitted_parameter) != 2:
o.fatal(
"a graft point must be of the form old_path=new_path")
old_path, new_path = splitted_parameter
if not (old_path and new_path):
o.fatal("a graft point cannot be empty")
graft_points.append(
(resolve_parent(old_path), resolve_parent(new_path)))
is_reverse = environ.get(b'BUP_SERVER_REVERSE')
if is_reverse and opt.remote:
o.fatal("don't use -r in reverse mode; it's automatic")
name = opt.name
if name and not valid_save_name(name):
o.fatal("'%s' is not a valid branch name" % path_msg(name))
refname = name and b'refs/heads/%s' % name or None
if opt.remote or is_reverse:
try:
cli = client.Client(opt.remote)
except client.ClientError as e:
log('error: %s' % e)
sys.exit(1)
oldref = refname and cli.read_ref(refname) or None
w = cli.new_packwriter(compression_level=opt.compress)
else:
cli = None
oldref = refname and git.read_ref(refname) or None
w = git.PackWriter(compression_level=opt.compress)
handle_ctrl_c()
# Metadata is stored in a file named .bupm in each directory. The
# first metadata entry will be the metadata for the current directory.
# The remaining entries will be for each of the other directory
# elements, in the order they're listed in the index.
#
# Since the git tree elements are sorted according to
# git.shalist_item_sort_key, the metalist items are accumulated as
# (sort_key, metadata) tuples, and then sorted when the .bupm file is
# created. The sort_key should have been computed using the element's
# mangled name and git mode (after hashsplitting), but the code isn't
# actually doing that but rather uses the element's real name and mode.
# This makes things a bit more difficult when reading it back, see
# vfs.ordered_tree_entries().
# Maintain a stack of information representing the current location in
# the archive being constructed. The current path is recorded in
# parts, which will be something like ['', 'home', 'someuser'], and
# the accumulated content and metadata for of the dirs in parts is
# stored in parallel stacks in shalists and metalists.
parts = [] # Current archive position (stack of dir names).
shalists = [] # Hashes for each dir in paths.
metalists = [] # Metadata for each dir in paths.
def _push(part, metadata):
# Enter a new archive directory -- make it the current directory.
parts.append(part)
shalists.append([])
metalists.append([(b'', metadata)]) # This dir's metadata (no name).
def _pop(force_tree, dir_metadata=None):
# Leave the current archive directory and add its tree to its parent.
assert (len(parts) >= 1)
part = parts.pop()
shalist = shalists.pop()
metalist = metalists.pop()
# FIXME: only test if collision is possible (i.e. given --strip, etc.)?
if force_tree:
tree = force_tree
else:
names_seen = set()
clean_list = []
metaidx = 1 # entry at 0 is for the dir
for x in shalist:
name = x[1]
if name in names_seen:
parent_path = b'/'.join(parts) + b'/'
add_error('error: ignoring duplicate path %s in %s' %
(path_msg(name), path_msg(parent_path)))
if not stat.S_ISDIR(x[0]):
del metalist[metaidx]
else:
names_seen.add(name)
clean_list.append(x)
if not stat.S_ISDIR(x[0]):
metaidx += 1
if dir_metadata: # Override the original metadata pushed for this dir.
metalist = [(b'', dir_metadata)] + metalist[1:]
sorted_metalist = sorted(metalist, key=lambda x: x[0])
metadata = b''.join([m[1].encode() for m in sorted_metalist])
metadata_f = BytesIO(metadata)
mode, id = hashsplit.split_to_blob_or_tree(w.new_blob,
w.new_tree,
[metadata_f],
keep_boundaries=False)
clean_list.append((mode, b'.bupm', id))
tree = w.new_tree(clean_list)
if shalists:
shalists[-1].append((GIT_MODE_TREE,
git.mangle_name(part, GIT_MODE_TREE,
GIT_MODE_TREE), tree))
return tree
_nonlocal['count'] = 0
_nonlocal['subcount'] = 0
_nonlocal['lastremain'] = None
def progress_report(n):
_nonlocal['subcount'] += n
cc = _nonlocal['count'] + _nonlocal['subcount']
pct = total and (cc * 100.0 / total) or 0
now = time.time()
elapsed = now - tstart
kps = elapsed and int(cc / 1024. / elapsed)
kps_frac = 10**int(math.log(kps + 1, 10) - 1)
kps = int(kps / kps_frac) * kps_frac
if cc:
remain = elapsed * 1.0 / cc * (total - cc)
else:
remain = 0.0
if (_nonlocal['lastremain'] and (remain > _nonlocal['lastremain']) and
((remain - _nonlocal['lastremain']) / _nonlocal['lastremain'] <
0.05)):
remain = _nonlocal['lastremain']
else:
_nonlocal['lastremain'] = remain
hours = int(remain / 60 / 60)
mins = int(remain / 60 - hours * 60)
secs = int(remain - hours * 60 * 60 - mins * 60)
if elapsed < 30:
remainstr = ''
kpsstr = ''
else:
kpsstr = '%dk/s' % kps
if hours:
remainstr = '%dh%dm' % (hours, mins)
elif mins:
remainstr = '%dm%d' % (mins, secs)
else:
remainstr = '%ds' % secs
qprogress(
'Saving: %.2f%% (%d/%dk, %d/%d files) %s %s\r' %
(pct, cc / 1024, total / 1024, fcount, ftotal, remainstr, kpsstr))
indexfile = opt.indexfile or git.repo(b'bupindex')
r = index.Reader(indexfile)
try:
msr = index.MetaStoreReader(indexfile + b'.meta')
except IOError as ex:
if ex.errno != EACCES:
raise
log('error: cannot access %r; have you run bup index?' %
path_msg(indexfile))
sys.exit(1)
hlink_db = hlinkdb.HLinkDB(indexfile + b'.hlink')
def already_saved(ent):
return ent.is_valid() and w.exists(ent.sha) and ent.sha
def wantrecurse_pre(ent):
return not already_saved(ent)
def wantrecurse_during(ent):
return not already_saved(ent) or ent.sha_missing()
def find_hardlink_target(hlink_db, ent):
if hlink_db and not stat.S_ISDIR(ent.mode) and ent.nlink > 1:
link_paths = hlink_db.node_paths(ent.dev, ent.ino)
if link_paths:
return link_paths[0]
total = ftotal = 0
if opt.progress:
for (transname, ent) in r.filter(extra, wantrecurse=wantrecurse_pre):
if not (ftotal % 10024):
qprogress('Reading index: %d\r' % ftotal)
exists = ent.exists()
hashvalid = already_saved(ent)
ent.set_sha_missing(not hashvalid)
if not opt.smaller or ent.size < opt.smaller:
if exists and not hashvalid:
total += ent.size
ftotal += 1
progress('Reading index: %d, done.\n' % ftotal)
hashsplit.progress_callback = progress_report
# Root collisions occur when strip or graft options map more than one
# path to the same directory (paths which originally had separate
# parents). When that situation is detected, use empty metadata for
# the parent. Otherwise, use the metadata for the common parent.
# Collision example: "bup save ... --strip /foo /foo/bar /bar".
# FIXME: Add collision tests, or handle collisions some other way.
# FIXME: Detect/handle strip/graft name collisions (other than root),
# i.e. if '/foo/bar' and '/bar' both map to '/'.
first_root = None
root_collision = None
tstart = time.time()
fcount = 0
lastskip_name = None
lastdir = b''
for (transname, ent) in r.filter(extra, wantrecurse=wantrecurse_during):
(dir, file) = os.path.split(ent.name)
exists = (ent.flags & index.IX_EXISTS)
hashvalid = already_saved(ent)
wasmissing = ent.sha_missing()
oldsize = ent.size
if opt.verbose:
if not exists:
status = 'D'
elif not hashvalid:
if ent.sha == index.EMPTY_SHA:
status = 'A'
else:
status = 'M'
else:
status = ' '
if opt.verbose >= 2:
log('%s %-70s\n' % (status, path_msg(ent.name)))
elif not stat.S_ISDIR(ent.mode) and lastdir != dir:
if not lastdir.startswith(dir):
log('%s %-70s\n' %
(status, path_msg(os.path.join(dir, b''))))
lastdir = dir
if opt.progress:
progress_report(0)
fcount += 1
if not exists:
continue
if opt.smaller and ent.size >= opt.smaller:
if exists and not hashvalid:
if opt.verbose:
log('skipping large file "%s"\n' % path_msg(ent.name))
lastskip_name = ent.name
continue
assert (dir.startswith(b'/'))
if opt.strip:
dirp = stripped_path_components(dir, extra)
elif opt.strip_path:
dirp = stripped_path_components(dir, [opt.strip_path])
elif graft_points:
dirp = grafted_path_components(graft_points, dir)
else:
dirp = path_components(dir)
# At this point, dirp contains a representation of the archive
# path that looks like [(archive_dir_name, real_fs_path), ...].
# So given "bup save ... --strip /foo/bar /foo/bar/baz", dirp
# might look like this at some point:
# [('', '/foo/bar'), ('baz', '/foo/bar/baz'), ...].
# This dual representation supports stripping/grafting, where the
# archive path may not have a direct correspondence with the
# filesystem. The root directory is represented by an initial
# component named '', and any component that doesn't have a
# corresponding filesystem directory (due to grafting, for
# example) will have a real_fs_path of None, i.e. [('', None),
# ...].
if first_root == None:
first_root = dirp[0]
elif first_root != dirp[0]:
root_collision = True
# If switching to a new sub-tree, finish the current sub-tree.
while parts > [x[0] for x in dirp]:
_pop(force_tree=None)
# If switching to a new sub-tree, start a new sub-tree.
for path_component in dirp[len(parts):]:
dir_name, fs_path = path_component
# Not indexed, so just grab the FS metadata or use empty metadata.
try:
meta = metadata.from_path(fs_path, normalized=True) \
if fs_path else metadata.Metadata()
except (OSError, IOError) as e:
add_error(e)
lastskip_name = dir_name
meta = metadata.Metadata()
_push(dir_name, meta)
if not file:
if len(parts) == 1:
continue # We're at the top level -- keep the current root dir
# Since there's no filename, this is a subdir -- finish it.
oldtree = already_saved(ent) # may be None
newtree = _pop(force_tree=oldtree)
if not oldtree:
if lastskip_name and lastskip_name.startswith(ent.name):
ent.invalidate()
else:
ent.validate(GIT_MODE_TREE, newtree)
ent.repack()
if exists and wasmissing:
_nonlocal['count'] += oldsize
continue
# it's not a directory
if hashvalid:
id = ent.sha
git_name = git.mangle_name(file, ent.mode, ent.gitmode)
git_info = (ent.gitmode, git_name, id)
shalists[-1].append(git_info)
sort_key = git.shalist_item_sort_key((ent.mode, file, id))
meta = msr.metadata_at(ent.meta_ofs)
meta.hardlink_target = find_hardlink_target(hlink_db, ent)
# Restore the times that were cleared to 0 in the metastore.
(meta.atime, meta.mtime, meta.ctime) = (ent.atime, ent.mtime,
ent.ctime)
metalists[-1].append((sort_key, meta))
else:
id = None
hlink = find_hardlink_target(hlink_db, ent)
try:
meta = metadata.from_path(
ent.name,
hardlink_target=hlink,
normalized=True,
after_stat=after_nondir_metadata_stat)
except (OSError, IOError) as e:
add_error(e)
lastskip_name = ent.name
continue
if stat.S_IFMT(ent.mode) != stat.S_IFMT(meta.mode):
# The mode changed since we indexed the file, this is bad.
# This can cause two issues:
# 1) We e.g. think the file is a regular file, but now it's
# something else (a device, socket, FIFO or symlink, etc.)
# and _read_ from it when we shouldn't.
# 2) We then record it as valid, but don't update the index
# metadata, and on a subsequent save it has 'hashvalid'
# but is recorded as the file type from the index, when
# the content is something else ...
# Avoid all of these consistency issues by just skipping such
# things - it really ought to not happen anyway.
add_error("%s: mode changed since indexing, skipping." %
path_msg(ent.name))
lastskip_name = ent.name
continue
if stat.S_ISREG(ent.mode):
try:
# If the file changes while we're reading it, then our reading
# may stop at some point, but the stat() above may have gotten
# a different size already. Recalculate the meta size so that
# the repository records the accurate size in the metadata, even
# if the other stat() data might be slightly older than the file
# content (which we can't fix, this is inherently racy, but we
# can prevent the size mismatch.)
meta.size = 0
def new_blob(data):
meta.size += len(data)
return w.new_blob(data)
before_saving_regular_file(ent.name)
with hashsplit.open_noatime(ent.name) as f:
(mode, id) = hashsplit.split_to_blob_or_tree(
new_blob, w.new_tree, [f], keep_boundaries=False)
except (IOError, OSError) as e:
add_error('%s: %s' % (ent.name, e))
lastskip_name = ent.name
elif stat.S_ISDIR(ent.mode):
assert (0) # handled above
elif stat.S_ISLNK(ent.mode):
mode, id = (GIT_MODE_SYMLINK, w.new_blob(meta.symlink_target))
else:
# Everything else should be fully described by its
# metadata, so just record an empty blob, so the paths
# in the tree and .bupm will match up.
(mode, id) = (GIT_MODE_FILE, w.new_blob(b''))
if id:
ent.validate(mode, id)
ent.repack()
git_name = git.mangle_name(file, ent.mode, ent.gitmode)
git_info = (mode, git_name, id)
shalists[-1].append(git_info)
sort_key = git.shalist_item_sort_key((ent.mode, file, id))
metalists[-1].append((sort_key, meta))
if exists and wasmissing:
_nonlocal['count'] += oldsize
_nonlocal['subcount'] = 0
if opt.progress:
pct = total and _nonlocal['count'] * 100.0 / total or 100
progress(
'Saving: %.2f%% (%d/%dk, %d/%d files), done. \n' %
(pct, _nonlocal['count'] / 1024, total / 1024, fcount, ftotal))
while len(parts) > 1: # _pop() all the parts above the root
_pop(force_tree=None)
assert (len(shalists) == 1)
assert (len(metalists) == 1)
# Finish the root directory.
tree = _pop(
force_tree=None,
# When there's a collision, use empty metadata for the root.
dir_metadata=metadata.Metadata() if root_collision else None)
sys.stdout.flush()
out = byte_stream(sys.stdout)
if opt.tree:
out.write(hexlify(tree))
out.write(b'\n')
if opt.commit or name:
if compat.py_maj > 2:
# Strip b prefix from python 3 bytes reprs to preserve previous format
msgcmd = b'[%s]' % b', '.join(
[repr(argv_bytes(x))[1:].encode('ascii') for x in argv])
else:
msgcmd = repr(argv)
msg = b'bup save\n\nGenerated by command:\n%s\n' % msgcmd
userline = (b'%s <%s@%s>' % (userfullname(), username(), hostname()))
commit = w.new_commit(tree, oldref, userline, date, None, userline,
date, None, msg)
if opt.commit:
out.write(hexlify(commit))
out.write(b'\n')
msr.close()
w.close() # must close before we can update the ref
if opt.name:
if cli:
cli.update_ref(refname, commit, oldref)
else:
git.update_ref(refname, commit, oldref)
if cli:
cli.close()
if saved_errors:
log('WARNING: %d errors encountered while saving.\n' %
len(saved_errors))
sys.exit(1)
