-#!/usr/bin/env python
-import sys, math, struct, glob, resource
-from bup import options, git
-from bup.helpers import *
+#!/bin/sh
+"""": # -*-python-*-
+bup_python="$(dirname "$0")/bup-python" || exit $?
+exec "$bup_python" "$0" ${1+"$@"}
+"""
+# end of bup preamble
+
+from __future__ import absolute_import
+import glob, math, os, resource, struct, sys, tempfile
+
+from bup import options, git, midx, _helpers, xstat
+from bup.compat import range
+from bup.helpers import (Sha1, add_error, atomically_replaced_file, debug1, fdatasync,
+ handle_ctrl_c, log, mmap_readwrite, qprogress,
+ saved_errors, unlink)
+
PAGE_SIZE=4096
SHA_PER_PAGE=PAGE_SIZE/20.
bup midx [options...] <idxnames...>
--
o,output= output midx filename (default: auto-generated)
-a,auto automatically create .midx from any unindexed .idx files
-f,force automatically create .midx from *all* .idx files
+a,auto automatically use all existing .midx/.idx files as input
+f,force merge produce exactly one .midx containing all objects
p,print print names of generated midx files
+check validate contents of the given midx files (with -a, all midx files)
max-files= maximum number of idx files to open at once [-1]
-dir= directory containing idx/midx files
+d,dir= directory containing idx/midx files
"""
+merge_into = _helpers.merge_into
+
+
def _group(l, count):
- for i in xrange(0, len(l), count):
+ for i in range(0, len(l), count):
yield l[i:i+count]
-
-
+
+
def max_files():
mf = min(resource.getrlimit(resource.RLIMIT_NOFILE))
if mf > 32:
return mf
-def merge(idxlist, bits, table):
- count = 0
- for e in git.idxmerge(idxlist, final_progress=False):
- count += 1
- prefix = git.extract_bits(e, bits)
- table[prefix] = count
- yield e
+def check_midx(name):
+ nicename = git.repo_rel(name)
+ log('Checking %s.\n' % nicename)
+ try:
+ ix = git.open_idx(name)
+ except git.GitError as e:
+ add_error('%s: %s' % (name, e))
+ return
+ for count,subname in enumerate(ix.idxnames):
+ sub = git.open_idx(os.path.join(os.path.dirname(name), subname))
+ for ecount,e in enumerate(sub):
+ if not (ecount % 1234):
+ qprogress(' %d/%d: %s %d/%d\r'
+ % (count, len(ix.idxnames),
+ git.shorten_hash(subname), ecount, len(sub)))
+ if not sub.exists(e):
+ add_error("%s: %s: %s missing from idx"
+ % (nicename, git.shorten_hash(subname),
+ str(e).encode('hex')))
+ if not ix.exists(e):
+ add_error("%s: %s: %s missing from midx"
+ % (nicename, git.shorten_hash(subname),
+ str(e).encode('hex')))
+ prev = None
+ for ecount,e in enumerate(ix):
+ if not (ecount % 1234):
+ qprogress(' Ordering: %d/%d\r' % (ecount, len(ix)))
+ if not e >= prev:
+ add_error('%s: ordering error: %s < %s'
+ % (nicename,
+ str(e).encode('hex'), str(prev).encode('hex')))
+ prev = e
+_first = None
def _do_midx(outdir, outfilename, infilenames, prefixstr):
+ global _first
if not outfilename:
assert(outdir)
sum = Sha1('\0'.join(infilenames)).hexdigest()
inp = []
total = 0
- allfilenames = {}
- for name in infilenames:
- ix = git.open_idx(name)
- for n in ix.idxnames:
- allfilenames[n] = 1
- inp.append(ix)
- total += len(ix)
-
- log('midx: %screating from %d files (%d objects).\n'
- % (prefixstr, len(infilenames), total))
- if (not opt.force and (total < 1024 and len(infilenames) < 3)) \
- or len(infilenames) < 2 \
- or (opt.force and not total):
- debug1('midx: nothing to do.\n')
- return
+ allfilenames = []
+ midxs = []
+ try:
+ for name in infilenames:
+ ix = git.open_idx(name)
+ midxs.append(ix)
+ inp.append((
+ ix.map,
+ len(ix),
+ ix.sha_ofs,
+ isinstance(ix, midx.PackMidx) and ix.which_ofs or 0,
+ len(allfilenames),
+ ))
+ for n in ix.idxnames:
+ allfilenames.append(os.path.basename(n))
+ total += len(ix)
+ inp.sort(reverse=True, key=lambda x: str(x[0][x[2]:x[2]+20]))
- pages = int(total/SHA_PER_PAGE) or 1
- bits = int(math.ceil(math.log(pages, 2)))
- entries = 2**bits
- debug1('midx: table size: %d (%d bits)\n' % (entries*4, bits))
-
- table = [0]*entries
+ if not _first: _first = outdir
+ dirprefix = (_first != outdir) and git.repo_rel(outdir)+': ' or ''
+ debug1('midx: %s%screating from %d files (%d objects).\n'
+ % (dirprefix, prefixstr, len(infilenames), total))
+ if (opt.auto and (total < 1024 and len(infilenames) < 3)) \
+ or ((opt.auto or opt.force) and len(infilenames) < 2) \
+ or (opt.force and not total):
+ debug1('midx: nothing to do.\n')
+ return
- try:
- os.unlink(outfilename)
- except OSError:
- pass
- f = open(outfilename + '.tmp', 'w+')
- f.write('MIDX\0\0\0\2')
- f.write(struct.pack('!I', bits))
- assert(f.tell() == 12)
- f.write('\0'*4*entries)
-
- for e in merge(inp, bits, table):
- f.write(e)
-
- f.write('\0'.join(os.path.basename(p) for p in allfilenames.keys()))
+ pages = int(total/SHA_PER_PAGE) or 1
+ bits = int(math.ceil(math.log(pages, 2)))
+ entries = 2**bits
+ debug1('midx: table size: %d (%d bits)\n' % (entries*4, bits))
- f.seek(12)
- f.write(struct.pack('!%dI' % entries, *table))
- f.close()
- os.rename(outfilename + '.tmp', outfilename)
+ unlink(outfilename)
+ with atomically_replaced_file(outfilename, 'wb') as f:
+ f.write('MIDX')
+ f.write(struct.pack('!II', midx.MIDX_VERSION, bits))
+ assert(f.tell() == 12)
- # this is just for testing
+ f.truncate(12 + 4*entries + 20*total + 4*total)
+ f.flush()
+ fdatasync(f.fileno())
+
+ fmap = mmap_readwrite(f, close=False)
+
+ count = merge_into(fmap, bits, total, inp)
+ del fmap # Assume this calls msync() now.
+ f.seek(0, os.SEEK_END)
+ f.write('\0'.join(allfilenames))
+ finally:
+ for ix in midxs:
+ if isinstance(ix, midx.PackMidx):
+ ix.close()
+ midxs = None
+ inp = None
+
+
+ # This is just for testing (if you enable this, don't clear inp above)
if 0:
- p = git.PackMidx(outfilename)
+ p = midx.PackMidx(outfilename)
assert(len(p.idxnames) == len(infilenames))
print p.idxnames
assert(len(p) == total)
- pi = iter(p)
- for i in merge(inp, total, bits, table):
- assert(i == pi.next())
+ for pe, e in p, git.idxmerge(inp, final_progress=False):
+ pin = next(pi)
+ assert(i == pin)
assert(p.exists(i))
- return total,outfilename
+ return total, outfilename
def do_midx(outdir, outfilename, infilenames, prefixstr):
print rv[1]
-def do_midx_dir(path):
+def do_midx_dir(path, outfilename):
already = {}
sizes = {}
if opt.force and not opt.auto:
contents[mname] = [('%s/%s' % (path,i)) for i in m.idxnames]
sizes[mname] = len(m)
- # sort the biggest midxes first, so that we can eliminate smaller
- # redundant ones that come later in the list
- midxs.sort(lambda x,y: -cmp(sizes[x], sizes[y]))
+ # sort the biggest+newest midxes first, so that we can eliminate
+ # smaller (or older) redundant ones that come later in the list
+ midxs.sort(key=lambda ix: (-sizes[ix], -xstat.stat(ix).st_mtime))
for mname in midxs:
any = 0
all.sort()
part1 = [name for sz,name in all[:len(all)-DESIRED_LWM+1]]
part2 = all[len(all)-DESIRED_LWM+1:]
- all = list(do_midx_group(path, part1)) + part2
+ all = list(do_midx_group(path, outfilename, part1)) + part2
if len(all) > DESIRED_HWM:
debug1('\nStill too many indexes (%d > %d). Merging again.\n'
% (len(all), DESIRED_HWM))
print name
-def do_midx_group(outdir, infiles):
+def do_midx_group(outdir, outfilename, infiles):
groups = list(_group(infiles, opt.max_files))
gprefix = ''
for n,sublist in enumerate(groups):
if len(groups) != 1:
gprefix = 'Group %d: ' % (n+1)
- rv = _do_midx(path, None, sublist, gprefix)
+ rv = _do_midx(outdir, outfilename, sublist, gprefix)
if rv:
yield rv
if extra and (opt.auto or opt.force):
o.fatal("you can't use -f/-a and also provide filenames")
+if opt.check and (not extra and not opt.auto):
+ o.fatal("if using --check, you must provide filenames or -a")
git.check_repo_or_die()
opt.max_files = max_files()
assert(opt.max_files >= 5)
-if extra:
- do_midx(git.repo('objects/pack'), opt.output, extra, '')
-elif opt.auto or opt.force:
- if opt.dir:
- paths = [opt.dir]
+if opt.check:
+ # check existing midx files
+ if extra:
+ midxes = extra
else:
- paths = [git.repo('objects/pack')]
- paths += glob.glob(git.repo('index-cache/*/.'))
- for path in paths:
- debug1('midx: scanning %s\n' % path)
- do_midx_dir(path)
- debug1('\n')
+ midxes = []
+ paths = opt.dir and [opt.dir] or git.all_packdirs()
+ for path in paths:
+ debug1('midx: scanning %s\n' % path)
+ midxes += glob.glob(os.path.join(path, '*.midx'))
+ for name in midxes:
+ check_midx(name)
+ if not saved_errors:
+ log('All tests passed.\n')
else:
- o.fatal("you must use -f or -a or provide input filenames")
+ if extra:
+ do_midx(git.repo('objects/pack'), opt.output, extra, '')
+ elif opt.auto or opt.force:
+ paths = opt.dir and [opt.dir] or git.all_packdirs()
+ for path in paths:
+ debug1('midx: scanning %s\n' % path)
+ do_midx_dir(path, opt.output)
+ else:
+ o.fatal("you must use -f or -a or provide input filenames")
+
+if saved_errors:
+ log('WARNING: %d errors encountered.\n' % len(saved_errors))
+ sys.exit(1)