When --exclude-rx <pattern> is provided to bup index, exclude any path
matching <pattern>, which must be a Python regular expression
(http://docs.python.org/library/re.html). The pattern will be
compared against the full path, without anchoring, so "x/y" will match
"ox/yard" or "box/yards". To exclude the contents of /tmp, but not
the directory itself, use "^/tmp/.".
You may check the behavior at runtime by setting BUP_DEBUG=2 in the
environment.
Thanks to Zoran Zaric <zz@zoranzaric.de> for reporting a bug in an
earlier version of this patch.
Signed-off-by: Rob Browning <rlb@defaultvalue.org>
Reviewed-by: Zoran Zaric <zz@zoranzaric.de>
bup index \<-p|-m|-s|-u\> [-H] [-l] [-x] [\--fake-valid] [\--no-check-device]
[\--fake-invalid] [\--check] [\--clear] [-f *indexfile*] [\--exclude *path*]
-[\--exclude-from *filename*] [-v] \<filenames...\>
+[\--exclude-from *filename*] [\--exclude-rx *pattern*] [-v] \<filenames...\>
# DESCRIPTION
: a file that contains exclude paths (can be used more
than once)
+\--exclude-rx=*pattern*
+: exclude any path matching *pattern*, which must be a Python regular
+ expression (http://docs.python.org/library/re.html). The pattern
+ will be compared against the full path, without anchoring, so
+ "x/y" will match "ox/yard" or "box/yards". To exclude the
+ contents of /tmp, but not the directory itself, use
+ "^/tmp/.". (can be specified more than once)
+
+ Examples:
+
+ * '/foo$' - exclude any file named foo
+ * '/foo/$' - exclude any directory named foo
+ * '/foo/.' - exclude the content of any directory named foo
+ * '^/tmp/.' - exclude root-level /tmp's content, but not /tmp itself
+
\--no-check-device
: don't mark a an entry invalid if the device number (stat(2)
st_dev) changes. This can be useful when indexing remote,
#!/usr/bin/env python
-import sys, stat, time, os, errno
+import sys, stat, time, os, errno, re
from bup import metadata, options, git, index, drecurse, hlinkdb
from bup.helpers import *
from bup.hashsplit import GIT_MODE_TREE, GIT_MODE_FILE
bup_dir = os.path.abspath(git.repo())
for (path,pst) in drecurse.recursive_dirlist([top], xdev=opt.xdev,
bup_dir=bup_dir,
- excluded_paths=excluded_paths):
+ excluded_paths=excluded_paths,
+ exclude_rxs=exclude_rxs):
if opt.verbose>=2 or (opt.verbose==1 and stat.S_ISDIR(pst.st_mode)):
sys.stdout.write('%s\n' % path)
sys.stdout.flush()
f,indexfile= the name of the index file (normally BUP_DIR/bupindex)
exclude= a path to exclude from the backup (can be used more than once)
exclude-from= a file that contains exclude paths (can be used more than once)
+exclude-rx= skip paths that match the unanchored regular expression
v,verbose increase log output (can be used more than once)
x,xdev,one-file-system don't cross filesystem boundaries
"""
clear_index(indexfile)
excluded_paths = parse_excludes(flags, o.fatal)
+exclude_rxs = parse_rx_excludes(flags, o.fatal)
paths = index.reduce_paths(extra)
if opt.update:
return l
-def _recursive_dirlist(prepend, xdev, bup_dir=None, excluded_paths=None):
+def _recursive_dirlist(prepend, xdev, bup_dir=None,
+ excluded_paths=None,
+ exclude_rxs=None):
for (name,pst) in _dirlist():
+ path = prepend + name
if excluded_paths:
- if os.path.normpath(prepend+name) in excluded_paths:
- debug1('Skipping %r: excluded.\n' % (prepend+name))
+ if os.path.normpath(path) in excluded_paths:
+ debug1('Skipping %r: excluded.\n' % path)
continue
+ if exclude_rxs and should_rx_exclude_path(path, exclude_rxs):
+ continue
if name.endswith('/'):
if xdev != None and pst.st_dev != xdev:
debug1('Skipping %r: different filesystem.\n' % (prepend+name))
else:
for i in _recursive_dirlist(prepend=prepend+name, xdev=xdev,
bup_dir=bup_dir,
- excluded_paths=excluded_paths):
+ excluded_paths=excluded_paths,
+ exclude_rxs=exclude_rxs):
yield i
os.chdir('..')
yield (prepend + name, pst)
-def recursive_dirlist(paths, xdev, bup_dir=None, excluded_paths=None):
+def recursive_dirlist(paths, xdev, bup_dir=None, excluded_paths=None,
+ exclude_rxs=None):
startdir = OsFile('.')
try:
assert(type(paths) != type(''))
prepend = os.path.join(path, '')
for i in _recursive_dirlist(prepend=prepend, xdev=xdev,
bup_dir=bup_dir,
- excluded_paths=excluded_paths):
+ excluded_paths=excluded_paths,
+ exclude_rxs=exclude_rxs):
yield i
startdir.fchdir()
else:
return excluded_paths
+def parse_rx_excludes(options, fatal):
+ """Traverse the options and extract all rx excludes, or call
+ Option.fatal()."""
+ rxs = [v for f, v in options if f == '--exclude-rx']
+ for i in range(len(rxs)):
+ try:
+ rxs[i] = re.compile(rxs[i])
+ except re.error, ex:
+ o.fatal('invalid --exclude-rx pattern (%s):' % (ex, rxs[i]))
+ return rxs
+
+
+def should_rx_exclude_path(path, exclude_rxs):
+ """Return True if path matches a regular expression in exclude_rxs."""
+ for rx in exclude_rxs:
+ if rx.search(path):
+ debug1('Skipping %r: excluded by rx pattern %r.\n'
+ % (path, rx.pattern))
+ return True
+ return False
+
+
# FIXME: Carefully consider the use of functions (os.path.*, etc.)
# that resolve against the current filesystem in the strip/graft
# functions for example, but elsewhere as well. I suspect bup's not
WVPASSEQ "$(bup ls -a src/latest)" "$(echo -e "$top_dir/\ntmp/" | sort)"
) || WVFAIL
-
WVSTART "clear-index"
D=clear-index.tmp
export BUP_DIR="$TOP/$D/.bup"
WVPASSEQ "$(bup index -p)" "$D/bar
$D/
./"
+
+# bup index --exclude-rx ...
+(
+ set -e
+ export BUP_DIR="$TOP/buptest.tmp"
+ D=bupdata.tmp
+
+ WVSTART "index --exclude-rx '^/foo' (root anchor)"
+ rm -rf "$D" "$BUP_DIR" buprestore.tmp
+ WVPASS bup init
+ mkdir $D
+ touch $D/a
+ touch $D/b
+ mkdir $D/sub1
+ mkdir $D/sub2
+ touch $D/sub1/a
+ touch $D/sub2/b
+ WVPASS bup index -u $D --exclude-rx "^$(pwd)/$D/sub1/"
+ bup save --strip -n bupdir $D
+ bup restore -C buprestore.tmp /bupdir/latest/
+ WVPASSEQ "$(cd buprestore.tmp && find . | sort)" ".
+./a
+./b
+./sub2
+./sub2/b"
+
+ WVSTART "index --exclude-rx '/foo$' (non-dir, tail anchor)"
+ rm -rf "$D" "$BUP_DIR" buprestore.tmp
+ WVPASS bup init
+ mkdir $D
+ touch $D/a
+ touch $D/b
+ touch $D/foo
+ mkdir $D/sub
+ mkdir $D/sub/foo
+ touch $D/sub/foo/a
+ WVPASS bup index -u $D --exclude-rx '/foo$'
+ bup save --strip -n bupdir $D
+ bup restore -C buprestore.tmp /bupdir/latest/
+ WVPASSEQ "$(cd buprestore.tmp && find . | sort)" ".
+./a
+./b
+./sub
+./sub/foo
+./sub/foo/a"
+
+ WVSTART "index --exclude-rx '/foo/$' (dir, tail anchor)"
+ rm -rf "$D" "$BUP_DIR" buprestore.tmp
+ WVPASS bup init
+ mkdir $D
+ touch $D/a
+ touch $D/b
+ touch $D/foo
+ mkdir $D/sub
+ mkdir $D/sub/foo
+ touch $D/sub/foo/a
+ WVPASS bup index -u $D --exclude-rx '/foo/$'
+ bup save --strip -n bupdir $D
+ bup restore -C buprestore.tmp /bupdir/latest/
+ WVPASSEQ "$(cd buprestore.tmp && find . | sort)" ".
+./a
+./b
+./foo
+./sub"
+
+ WVSTART "index --exclude-rx '/foo/.' (dir content)"
+ rm -rf "$D" "$BUP_DIR" buprestore.tmp
+ WVPASS bup init
+ mkdir $D
+ touch $D/a
+ touch $D/b
+ touch $D/foo
+ mkdir $D/sub
+ mkdir $D/sub/foo
+ touch $D/sub/foo/a
+ WVPASS bup index -u $D --exclude-rx '/foo/.'
+ bup save --strip -n bupdir $D
+ bup restore -C buprestore.tmp /bupdir/latest/
+ WVPASSEQ "$(cd buprestore.tmp && find . | sort)" ".
+./a
+./b
+./foo
+./sub
+./sub/foo"
+) || WVFAIL
projects / bup.git / commitdiff