optspec = """
-bup split [-tcb] [-n name] [--bench] [filenames...]
+bup split <-t|-c|-b|-n name|--copy|--noop> [--bench] [filenames...]
--
-r,remote= remote repository path
+ Modes:
b,blobs output a series of blob ids
t,tree output a tree id
c,commit output a commit id
-n,name= name of backup set to update (if any)
+n,name= save the result under the given name
+noop split the input, but throw away the result
+copy split the input, copy it to stdout, don't save to repo
+ Options:
+r,remote= remote repository path
d,date= date for the commit (seconds since the epoch)
q,quiet don't print progress messages
v,verbose increase log output (can be used more than once)
git-ids read a list of git object ids from stdin and split their contents
keep-boundaries don't let one chunk span two input files
-noop don't actually save the data anywhere
-copy just copy input to output, hashsplitting along the way
bench print benchmark timings to stderr
max-pack-size= maximum bytes in a single pack
max-pack-objects= maximum number of objects in a single pack
if (opt.noop or opt.copy) and (opt.blobs or opt.tree or
opt.commit or opt.name):
o.fatal('-N and --copy are incompatible with -b, -t, -c, -n')
+if opt.blobs and (opt.tree or opt.commit or opt.name):
+ o.fatal('-b is incompatible with -t, -c, -n')
if extra and opt.git_ids:
o.fatal("don't provide filenames when using --git-ids")
# the input either comes from a series of files or from stdin.
files = extra and (open(fn) for fn in extra) or [sys.stdin]
-if pack_writer:
+if pack_writer and opt.blobs:
+ shalist = hashsplit.split_to_blobs(pack_writer, files,
+ keep_boundaries=opt.keep_boundaries,
+ progress=prog)
+ for (sha, size, level) in shalist:
+ print sha.encode('hex')
+ reprogress()
+elif pack_writer: # tree or commit or name
shalist = hashsplit.split_to_shalist(pack_writer, files,
keep_boundaries=opt.keep_boundaries,
progress=prog)
sys.stdout.write(str(blob))
megs = hashsplit.total_split/1024/1024
if not opt.quiet and last != megs:
- progress('%d Mbytes read\r' % megs)
last = megs
- progress('%d Mbytes read, done.\n' % megs)
if opt.verbose:
log('\n')
-if opt.blobs:
- for (mode,name,bin) in shalist:
- print bin.encode('hex')
if opt.tree:
print tree.encode('hex')
if opt.commit or opt.name:
secs = time.time() - start_time
size = hashsplit.total_split
if opt.bench:
- log('\nbup: %.2fkbytes in %.2f secs = %.2f kbytes/sec\n'
+ log('bup: %.2fkbytes in %.2f secs = %.2f kbytes/sec\n'
% (size/1024., secs, size/1024./secs))
if saved_errors:
total_split = 0
-def _split_to_blobs(w, files, keep_boundaries, progress):
+def split_to_blobs(w, files, keep_boundaries, progress):
global total_split
for (blob, level) in hashsplit_iter(files, keep_boundaries, progress):
sha = w.new_blob(blob)
def split_to_shalist(w, files, keep_boundaries, progress=None):
- sl = _split_to_blobs(w, files, keep_boundaries, progress)
+ sl = split_to_blobs(w, files, keep_boundaries, progress)
+ assert(fanout != 0)
if not fanout:
shal = []
for (sha,size,level) in sl:
istty = os.isatty(2) or atoi(os.environ.get('BUP_FORCE_TTY'))
+_last_progress = ''
def progress(s):
"""Calls log() if stderr is a TTY. Does nothing otherwise."""
+ global _last_progress
if istty:
log(s)
+ _last_progress = s
def qprogress(s):
"""Calls progress() only if we haven't printed progress in a while.
- This avoids overloading the stderr buffer with excess junk."""
+ This avoids overloading the stderr buffer with excess junk.
+ """
global _last_prog
now = time.time()
if now - _last_prog > 0.1:
_last_prog = now
+def reprogress():
+ """Calls progress() to redisplay the most recent progress message.
+
+ Useful after you've printed some other message that wipes out the
+ progress line.
+ """
+ if _last_progress and _last_progress.endswith('\r'):
+ progress(_last_progress)
+
+
def mkdirp(d, mode=None):
"""Recursively create directories on path 'd'.
projects / bup.git / commitdiff