From: Brian Warner Date: Thu, 12 Mar 2009 20:56:06 +0000 (-0700) Subject: add 'tahoe debug consolidate' command, to merge directories created by repeated ... X-Git-Tag: allmydata-tahoe-1.4.0~65 X-Git-Url: https://git.rkrishnan.org/uri/%3C?a=commitdiff_plain;h=c4c6a629544ea9ad9c9e37f2f670e4905d2c3e11;p=tahoe-lafs%2Ftahoe-lafs.git add 'tahoe debug consolidate' command, to merge directories created by repeated 'tahoe cp -r' or the allmydata win32 backup tool, into the form that would have been created by 'tahoe backup'. --- diff --git a/src/allmydata/scripts/consolidate.py b/src/allmydata/scripts/consolidate.py new file mode 100644 index 00000000..5fccff3d --- /dev/null +++ b/src/allmydata/scripts/consolidate.py @@ -0,0 +1,364 @@ + +import os, pickle +import sqlite3 as sqlite + +import urllib +import simplejson +from allmydata.scripts.common_http import do_http +from allmydata.scripts.tahoe_backup import parse_old_timestamp, readonly, \ + raiseHTTPError, HTTPError +from allmydata.util import hashutil, base32 +from allmydata.util.netstring import netstring +from allmydata.scripts.common import get_alias, DEFAULT_ALIAS + +TAG = "consolidator_dirhash_v1" + +class CycleDetected(Exception): + pass + + +class Consolidator: + def __init__(self, options): + self.options = options + self.rootcap, path = get_alias(options.aliases, options.where, + DEFAULT_ALIAS) + assert path == "" + self.dbfile = options["dbfile"] + assert self.dbfile, "--dbfile is required" + self.backupfile = options["backupfile"] + assert self.backupfile, "--backupfile is required" + self.nodeurl = options["node-url"] + if not self.nodeurl.endswith("/"): + self.nodeurl += "/" + self.must_rescan_readonly_snapshots = not os.path.exists(self.dbfile) + self.db = sqlite.connect(self.dbfile) + self.cursor = self.db.cursor() + try: + self.cursor.execute("CREATE TABLE dirhashes" + "(" + " dirhash TEXT PRIMARY KEY," + " dircap TEXT" + ")") + except sqlite.OperationalError, e: + if "table dirhashes already exists" not in str(e): + raise + self.visited = set() + + def read_directory_json(self, dircap): + url = self.nodeurl + "uri/%s?t=json" % urllib.quote(dircap) + resp = do_http("GET", url) + if resp.status != 200: + raiseHTTPError("Error during directory GET", resp) + jd = simplejson.load(resp) + ntype, ndata = jd + if ntype != "dirnode": + return None + return ndata + + def msg(self, text): + print >>self.options.stdout, text + def err(self, text): + print >>self.options.stderr, text + + def consolidate(self): + try: + data = self.read_directory_json(self.rootcap + "/Backups") + except HTTPError: + self.err("Unable to list /Backups, maybe this account has none?") + return 1 + kids = data["children"] + potential_systems = {} + for (childname, (childtype, childdata)) in kids.items(): + if childtype != "dirnode": + continue + potential_systems[childname] = str(childdata["rw_uri"]) + backup_data = {"Backups": data, "systems": {}, "archives": {}} + systems = {} + for name, sdircap in potential_systems.items(): + sdata = self.read_directory_json(sdircap) + kids = sdata["children"] + if not u"Archives" in kids and not u"Latest Backup" in kids: + self.msg("%s: not a backupdir, no 'Archives' and 'Latest'" % name) + continue + self.msg("%s is a system" % name) + backup_data["systems"][name] = sdata + archives_dircap = kids[u"Archives"][1]["rw_uri"] + archives_data = self.read_directory_json(archives_dircap) + backup_data["archives"][name] = archives_data + systems[name] = archives_dircap + if not systems: + self.msg("No systems under /Backups, nothing to consolidate") + return 0 + if not os.path.exists(self.backupfile): + f = open(self.backupfile, "wb") + pickle.dump(backup_data, f) + f.close() + + for name, archives_dircap in sorted(systems.items()): + self.do_system(name, archives_dircap) + return 0 + + def do_system(self, system_name, archives_dircap): + # first we walk through the Archives list, looking for the existing + # snapshots. Each one will have a $NAME like "2008-11-16 10.34 PM" + # (in various forms: we use tahoe_backup.parse_old_timestamp to + # interpret it). At first, they'll all have $NAME and be writecaps. + # As we run, we will create $NAME-readonly (with a readcap) for each + # one (the first one will just be the readonly equivalent of the + # oldest snapshot: the others will be constructed out of shared + # directories). When we're done we'll have a $NAME-readonly for + # everything except the latest snapshot (to avoid any danger of + # modifying a backup that's already in progress). The very last step, + # which won't be enabled until we're sure that everything is working + # correctly, will replace each $NAME with $NAME-readonly. + + # We maintain a table that maps dirhash (hash of directory contents) + # to a directory readcap which contains those contents. We use this + # to decide if we can link to an existing directory, or if we must + # create a brand new one. Usually we add to this table in two places: + # when we scan the oldest snapshot (which we've just converted to + # readonly form), and when we must create a brand new one. If the + # table doesn't exist (probably because we've manually deleted it), + # we will scan *all* the existing readonly snapshots, and repopulate + # the table. We keep this table in a SQLite database (rather than a + # pickle) because we want to update it persistently after every + # directory creation, and writing out a 10k entry pickle takes about + # 250ms + + # 'snapshots' maps timestamp to [rwname, writecap, roname, readcap]. + # The possibilities are: + # [$NAME, writecap, None, None] : haven't touched it + # [$NAME, writecap, $NAME-readonly, readcap] : processed, not replaced + # [None, None, $NAME, readcap] : processed and replaced + + self.msg("consolidating system %s" % system_name) + self.directories_reused = 0 + self.directories_used_as_is = 0 + self.directories_created = 0 + data = self.read_directory_json(archives_dircap) + snapshots = {} + + for (childname, (childtype, childdata)) in data["children"].items(): + if childtype != "dirnode": + self.msg("non-dirnode %s in Archives/" % childname) + continue + timename = childname + if childname.endswith("-readonly"): + timename = childname[:-len("-readonly")] + timestamp = parse_old_timestamp(timename, self.options) + assert timestamp is not None, timename + snapshots.setdefault(timestamp, [None, None, None, None]) + # if the snapshot is readonly (i.e. it has no rw_uri), we might + # need to re-scan it + is_readonly = not childdata.has_key("rw_uri") + if is_readonly: + readcap = str(childdata["ro_uri"]) + if self.must_rescan_readonly_snapshots: + self.scan_old_directory(str(childdata["ro_uri"])) + snapshots[timestamp][2] = childname + snapshots[timestamp][3] = readcap + else: + writecap = str(childdata["rw_uri"]) + snapshots[timestamp][0] = childname + snapshots[timestamp][1] = writecap + else: + self.msg(" No snapshots in Backups/Archives") + snapshots = [ [timestamp] + values + for (timestamp, values) in snapshots.items() ] + # now 'snapshots' is [timestamp, rwname, writecap, roname, readcap], + # which makes it easier to process in temporal order + snapshots.sort() + self.msg(" %d snapshots" % len(snapshots)) + # we always ignore the last one, for safety + snapshots = snapshots[:-1] + + first_snapshot = True + for (timestamp, rwname, writecap, roname, readcap) in snapshots: + start_created = self.directories_created + start_used_as_is = self.directories_used_as_is + start_reused = self.directories_reused + + # [None, None, $NAME, readcap] : processed and replaced + # [$NAME, writecap, $NAME-readonly, readcap] : processed, not replaced + # [$NAME, writecap, None, None] : haven't touched it + + if readcap and not writecap: + # skip past anything we've already processed and replaced + assert roname + assert not rwname + first_snapshot = False + self.msg(" %s already readonly" % roname) + continue + if readcap and writecap: + # we've processed it, creating a -readonly version, but we + # haven't replaced it. + assert roname + assert rwname + first_snapshot = False + self.msg(" %s processed but not yet replaced" % roname) + if self.options["really"]: + self.msg(" replacing %s with %s" % (rwname, roname)) + self.put_child(archives_dircap, rwname, readcap) + self.delete_child(archives_dircap, roname) + continue + assert writecap + assert rwname + assert not readcap + assert not roname + roname = rwname + "-readonly" + # for the oldest snapshot, we can do a simple readonly conversion + if first_snapshot: + first_snapshot = False + readcap = readonly(writecap) + self.directories_used_as_is += 1 + self.msg(" %s: oldest snapshot, using as-is" % rwname) + self.scan_old_directory(readcap) + else: + # for the others, we must scan their contents and build up a new + # readonly directory (which shares common subdirs with previous + # backups) + self.msg(" %s: processing" % rwname) + readcap = self.process_directory(readonly(writecap)) + if self.options["really"]: + self.msg(" replaced %s" % rwname) + self.put_child(archives_dircap, rwname, readcap) + else: + self.msg(" created %s" % roname) + self.put_child(archives_dircap, roname, readcap) + + snapshot_created = self.directories_created - start_created + snapshot_used_as_is = self.directories_used_as_is - start_used_as_is + snapshot_reused = self.directories_reused - start_reused + self.msg(" %s: done: %d dirs created, %d used as-is, %d reused" + % (rwname, + snapshot_created, snapshot_used_as_is, snapshot_reused)) + # done! + self.msg(" system done, %d dirs created, %d used as-is, %d reused" \ + % (self.directories_created, self.directories_used_as_is, + self.directories_reused)) + + def process_directory(self, readcap): + # I walk all my children (recursing over any subdirectories), build + # up a table of my contents, then see if I can re-use an old + # directory with the same contents. If not, I create a new directory + # for my contents. In all cases I return a directory readcap that + # points to my contents. + + # build up contents to pass to mkdir() (which uses t=set_children) + contents = {} # childname -> (type, rocap, metadata) + data = self.read_directory_json(readcap) + assert data is not None + hashkids = [] + num_dirs = 0 + for (childname, (childtype, childdata)) in sorted(data["children"].items()): + if childtype == "dirnode": + num_dirs += 1 + childcap = self.process_directory(str(childdata["ro_uri"])) + contents[childname] = ("dirnode", childcap, None) + else: + childcap = str(childdata["ro_uri"]) + contents[childname] = (childtype, childcap, None) + hashkids.append( (childname, childcap) ) + + dirhash = self.hash_directory_contents(hashkids) + old_dircap = self.get_old_dirhash(dirhash) + if old_dircap: + assert isinstance(old_dircap, str) + self.directories_reused += 1 + return old_dircap + if num_dirs == 0: + # we're allowed to re-use this directory + new_dircap = readonly(readcap) + assert isinstance(new_dircap, str) + self.store_dirhash(dirhash, new_dircap) + self.directories_used_as_is += 1 + return new_dircap + # otherwise, we need to create a new directory + new_dircap = readonly(self.mkdir(contents)) + assert isinstance(new_dircap, str) + self.store_dirhash(dirhash, new_dircap) + self.directories_created += 1 + return new_dircap + + def put_child(self, dircap, childname, childcap): + url = self.nodeurl + "uri/%s/%s?t=uri" % (urllib.quote(dircap), + urllib.quote(childname)) + resp = do_http("PUT", url, childcap) + if resp.status not in (200, 201): + raiseHTTPError("error during put_child", resp) + + def delete_child(self, dircap, childname): + url = self.nodeurl + "uri/%s/%s" % (urllib.quote(dircap), + urllib.quote(childname)) + resp = do_http("DELETE", url) + if resp.status not in (200, 201): + raiseHTTPError("error during delete_child", resp) + + def mkdir(self, contents): + url = self.nodeurl + "uri?t=mkdir" + resp = do_http("POST", url) + if resp.status < 200 or resp.status >= 300: + raiseHTTPError("error during mkdir", resp) + dircap = str(resp.read().strip()) + url = self.nodeurl + "uri/%s?t=set_children" % urllib.quote(dircap) + body = dict([ (childname, (contents[childname][0], + {"ro_uri": contents[childname][1], + "metadata": contents[childname][2], + })) + for childname in contents + ]) + resp = do_http("POST", url, simplejson.dumps(body)) + if resp.status != 200: + raiseHTTPError("error during set_children", resp) + return dircap + + def scan_old_directory(self, dircap): + # scan this directory (recursively) and stash a hash of its contents + # in the DB. This assumes that all subdirs can be used as-is (i.e. + # they've already been declared immutable) + dircap = readonly(dircap) + if dircap in self.visited: + raise CycleDetected + self.visited.add(dircap) + data = self.read_directory_json(dircap) + kids = [] + for (childname, (childtype, childdata)) in data["children"].items(): + childcap = str(childdata["ro_uri"]) + if childtype == "dirnode": + self.scan_old_directory(childcap) + kids.append( (childname, childcap) ) + dirhash = self.hash_directory_contents(kids) + self.store_dirhash(dirhash, dircap) + return dirhash + + def hash_directory_contents(self, kids): + kids.sort() + s = "".join([netstring(childname.encode("utf-8"))+netstring(childcap) + for (childname, childcap) in kids]) + return hashutil.tagged_hash(TAG, s) + + def store_dirhash(self, dirhash, dircap): + assert isinstance(dircap, str) + # existing items should prevail + try: + c = self.cursor + c.execute("INSERT INTO dirhashes (dirhash, dircap) VALUES (?,?)", + (base32.b2a(dirhash), dircap)) + self.db.commit() + except sqlite.IntegrityError: + # already present + pass + + def get_old_dirhash(self, dirhash): + self.cursor.execute("SELECT dircap FROM dirhashes WHERE dirhash=?", + (base32.b2a(dirhash),)) + row = self.cursor.fetchone() + if not row: + return None + (dircap,) = row + return str(dircap) + + +def main(options): + c = Consolidator(options) + return c.consolidate() diff --git a/src/allmydata/scripts/debug.py b/src/allmydata/scripts/debug.py index 7be41694..75e5b595 100644 --- a/src/allmydata/scripts/debug.py +++ b/src/allmydata/scripts/debug.py @@ -4,6 +4,7 @@ import struct, time, os from twisted.python import usage, failure from twisted.internet import defer +from allmydata.scripts.cli import VDriveOptions class DumpOptions(usage.Options): def getSynopsis(self): @@ -760,6 +761,21 @@ def repl(options): return code.interact() +class ConsolidateOptions(VDriveOptions): + optParameters = [ + ("dbfile", None, None, "persistent file for reusable dirhashes"), + ("backupfile", "b", None, "file to store backup of Archives/ contents"), + ] + optFlags = [ + ("really", None, "Really remove old snapshot directories"), + ] + def parseArgs(self, where): + self.where = where + +def consolidate(options): + from consolidate import main; return main(options) + + class DebugCommand(usage.Options): subCommands = [ ["dump-share", None, DumpOptions, @@ -769,6 +785,7 @@ class DebugCommand(usage.Options): ["catalog-shares", None, CatalogSharesOptions, "Describe shares in node dirs"], ["corrupt-share", None, CorruptShareOptions, "Corrupt a share"], ["repl", None, ReplOptions, "Open a python interpreter"], + ["consolidate", None, ConsolidateOptions, "Consolidate non-shared backups"], ] def postOptions(self): if not hasattr(self, 'subOptions'): @@ -784,6 +801,7 @@ Subcommands: tahoe debug find-shares Locate sharefiles in node directories tahoe debug catalog-shares Describe all shares in node dirs tahoe debug corrupt-share Corrupt a share by flipping a bit. + tahoe debug consolidate Consolidate old non-shared backups into shared ones. Please run e.g. 'tahoe debug dump-share --help' for more details on each subcommand. @@ -797,6 +815,7 @@ subDispatch = { "catalog-shares": catalog_shares, "corrupt-share": corrupt_share, "repl": repl, + "consolidate": consolidate, } diff --git a/src/allmydata/test/test_consolidate.py b/src/allmydata/test/test_consolidate.py new file mode 100644 index 00000000..acc0f49f --- /dev/null +++ b/src/allmydata/test/test_consolidate.py @@ -0,0 +1,297 @@ + +import os +from cStringIO import StringIO +import pickle +from twisted.trial import unittest +from allmydata.test.no_network import GridTestMixin +from allmydata.util import fileutil +from allmydata.scripts import runner, debug +from allmydata.scripts.common import get_aliases +from twisted.internet import defer, threads # CLI tests use deferToThread +from allmydata.interfaces import IDirectoryNode + + +class CLITestMixin: + def do_cli(self, verb, *args, **kwargs): + nodeargs = [ + "--node-directory", self.get_clientdir(), + ] + if verb == "debug": + argv = [verb, args[0]] + nodeargs + list(args[1:]) + else: + argv = [verb] + nodeargs + list(args) + stdin = kwargs.get("stdin", "") + stdout, stderr = StringIO(), StringIO() + d = threads.deferToThread(runner.runner, argv, run_by_human=False, + stdin=StringIO(stdin), + stdout=stdout, stderr=stderr) + def _done(rc): + return rc, stdout.getvalue(), stderr.getvalue() + d.addCallback(_done) + return d + +class Consolidate(GridTestMixin, CLITestMixin, unittest.TestCase): + + def writeto(self, path, data): + d = os.path.dirname(os.path.join(self.basedir, "home", path)) + fileutil.make_dirs(d) + f = open(os.path.join(self.basedir, "home", path), "w") + f.write(data) + f.close() + + def writeto_snapshot(self, sn, path, data): + p = "Backups/fluxx/Archives/2009-03-%02d 01.01.01/%s" % (sn, path) + return self.writeto(p, data) + + def do_cli_good(self, verb, *args, **kwargs): + d = self.do_cli(verb, *args, **kwargs) + def _check((rc,out,err)): + self.failUnlessEqual(err, "", verb) + self.failUnlessEqual(rc, 0, verb) + return out + d.addCallback(_check) + return d + + def test_arg_parsing(self): + self.basedir = "consolidate/Consolidate/arg_parsing" + self.set_up_grid(num_clients=1, num_servers=1) + co = debug.ConsolidateOptions() + co.parseOptions(["--node-directory", self.get_clientdir(), + "--dbfile", "foo.db", "--backupfile", "backup", "--really", + "URI:DIR2:foo"]) + self.failUnlessEqual(co["dbfile"], "foo.db") + self.failUnlessEqual(co["backupfile"], "backup") + self.failUnless(co["really"]) + self.failUnlessEqual(co.where, "URI:DIR2:foo") + + def OFF_test_basic(self): + # rename this method to enable the test. I've disabled it because, in + # my opinion: + # + # 1: 'tahoe debug consolidate' is useful enough to include in trunk, + # but not useful enough justify a lot of compatibility effort or + # extra test time + # 2: it requires sqlite3; I did not bother to make it work with + # pysqlite, nor did I bother making it fail gracefully when + # sqlite3 is not available + # 3: this test takes 30 seconds to run on my workstation, and it likely + # to take several minutes on the old slow dapper buildslave + # 4: I don't want other folks to see a SkipTest and wonder "oh no, what + # did I do wrong to not allow this test to run" + # + # These may not be strong arguments: I welcome feedback. In particular, + # this command may be more suitable for a plugin of some sort, if we + # had plugins of some sort. -warner 12-Mar-09 + + self.basedir = "consolidate/Consolidate/basic" + self.set_up_grid(num_clients=1) + + fileutil.make_dirs(os.path.join(self.basedir, "home/Backups/nonsystem")) + fileutil.make_dirs(os.path.join(self.basedir, "home/Backups/fluxx/Latest")) + self.writeto(os.path.join(self.basedir, + "home/Backups/fluxx/Archives/nondir"), + "not a directory: ignore me") + + # set up a number of non-shared "snapshots" + for i in range(1,8): + self.writeto_snapshot(i, "parent/README", "README") + self.writeto_snapshot(i, "parent/foo.txt", "foo") + self.writeto_snapshot(i, "parent/subdir1/bar.txt", "bar") + self.writeto_snapshot(i, "parent/subdir1/baz.txt", "baz") + self.writeto_snapshot(i, "parent/subdir2/yoy.txt", "yoy") + self.writeto_snapshot(i, "parent/subdir2/hola.txt", "hola") + + if i >= 1: + pass # initial snapshot + if i >= 2: + pass # second snapshot: same as the first + if i >= 3: + # modify a file + self.writeto_snapshot(i, "parent/foo.txt", "FOOF!") + if i >= 4: + # foo.txt goes back to normal + self.writeto_snapshot(i, "parent/foo.txt", "foo") + if i >= 5: + # new file + self.writeto_snapshot(i, "parent/subdir1/new.txt", "new") + if i >= 6: + # copy parent/subdir1 to parent/subdir2/copy1 + self.writeto_snapshot(i, "parent/subdir2/copy1/bar.txt", "bar") + self.writeto_snapshot(i, "parent/subdir2/copy1/baz.txt", "baz") + self.writeto_snapshot(i, "parent/subdir2/copy1/new.txt", "new") + if i >= 7: + # the last snapshot shall remain untouched + pass + + # now copy the whole thing into tahoe + d = self.do_cli_good("create-alias", "tahoe") + d.addCallback(lambda ign: + self.do_cli_good("cp", "-r", + os.path.join(self.basedir, "home/Backups"), + "tahoe:Backups")) + def _copied(res): + rootcap = get_aliases(self.get_clientdir())["tahoe"] + # now scan the initial directory structure + n = self.g.clients[0].create_node_from_uri(rootcap) + return n.get_child_at_path([u"Backups", u"fluxx", u"Archives"]) + d.addCallback(_copied) + self.nodes = {} + self.caps = {} + def stash(node, name): + self.nodes[name] = node + self.caps[name] = node.get_uri() + return node + d.addCallback(stash, "Archives") + self.manifests = {} + def stash_manifest(manifest, which): + self.manifests[which] = dict(manifest) + d.addCallback(lambda ignored: self.build_manifest(self.nodes["Archives"])) + d.addCallback(stash_manifest, "start") + def c(n): + pieces = n.split("-") + which = "finish" + if len(pieces) == 3: + which = pieces[-1] + sn = int(pieces[0]) + name = pieces[1] + path = [u"2009-03-%02d 01.01.01" % sn] + path.extend( {"b": [], + "bp": [u"parent"], + "bps1": [u"parent", u"subdir1"], + "bps2": [u"parent", u"subdir2"], + "bps2c1": [u"parent", u"subdir2", u"copy1"], + }[name] ) + return self.manifests[which][tuple(path)] + + dbfile = os.path.join(self.basedir, "dirhash.db") + backupfile = os.path.join(self.basedir, "backup.pickle") + + d.addCallback(lambda ign: + self.do_cli_good("debug", "consolidate", + "--dbfile", dbfile, + "--backupfile", backupfile, + "tahoe:")) + def _check_consolidate_output1(out): + lines = out.splitlines() + last = lines[-1] + self.failUnlessEqual(last.strip(), + "system done, " + "7 dirs created, 2 used as-is, 13 reused") + self.failUnless(os.path.exists(dbfile)) + self.failUnless(os.path.exists(backupfile)) + backup = pickle.load(open(backupfile, "rb")) + self.failUnless(u"fluxx" in backup["systems"]) + self.failUnless(u"fluxx" in backup["archives"]) + adata = backup["archives"]["fluxx"] + kids = adata[u"children"] + self.failUnlessEqual(str(kids[u"2009-03-01 01.01.01"][1][u"rw_uri"]), + c("1-b-start")) + d.addCallback(_check_consolidate_output1) + d.addCallback(lambda ign: + self.do_cli_good("debug", "consolidate", + "--dbfile", dbfile, + "--backupfile", backupfile, + "--really", "tahoe:")) + def _check_consolidate_output2(out): + lines = out.splitlines() + last = lines[-1] + self.failUnlessEqual(last.strip(), + "system done, " + "0 dirs created, 0 used as-is, 0 reused") + d.addCallback(_check_consolidate_output2) + + d.addCallback(lambda ignored: self.build_manifest(self.nodes["Archives"])) + d.addCallback(stash_manifest, "finish") + + def check_consolidation(ignored): + #for which in ("finish",): + # for path in sorted(self.manifests[which].keys()): + # print "%s %s %s" % (which, "/".join(path), + # self.manifests[which][path]) + + # last snapshot should be untouched + self.failUnlessEqual(c("7-b"), c("7-b-start")) + + # first snapshot should be a readonly form of the original + from allmydata.scripts.tahoe_backup import readonly + self.failUnlessEqual(c("1-b-finish"), readonly(c("1-b-start"))) + self.failUnlessEqual(c("1-bp-finish"), readonly(c("1-bp-start"))) + self.failUnlessEqual(c("1-bps1-finish"), readonly(c("1-bps1-start"))) + self.failUnlessEqual(c("1-bps2-finish"), readonly(c("1-bps2-start"))) + + # new directories should be different than the old ones + self.failIfEqual(c("1-b"), c("1-b-start")) + self.failIfEqual(c("1-bp"), c("1-bp-start")) + self.failIfEqual(c("1-bps1"), c("1-bps1-start")) + self.failIfEqual(c("1-bps2"), c("1-bps2-start")) + self.failIfEqual(c("2-b"), c("2-b-start")) + self.failIfEqual(c("2-bp"), c("2-bp-start")) + self.failIfEqual(c("2-bps1"), c("2-bps1-start")) + self.failIfEqual(c("2-bps2"), c("2-bps2-start")) + self.failIfEqual(c("3-b"), c("3-b-start")) + self.failIfEqual(c("3-bp"), c("3-bp-start")) + self.failIfEqual(c("3-bps1"), c("3-bps1-start")) + self.failIfEqual(c("3-bps2"), c("3-bps2-start")) + self.failIfEqual(c("4-b"), c("4-b-start")) + self.failIfEqual(c("4-bp"), c("4-bp-start")) + self.failIfEqual(c("4-bps1"), c("4-bps1-start")) + self.failIfEqual(c("4-bps2"), c("4-bps2-start")) + self.failIfEqual(c("5-b"), c("5-b-start")) + self.failIfEqual(c("5-bp"), c("5-bp-start")) + self.failIfEqual(c("5-bps1"), c("5-bps1-start")) + self.failIfEqual(c("5-bps2"), c("5-bps2-start")) + + # snapshot 1 and snapshot 2 should be identical + self.failUnlessEqual(c("2-b"), c("1-b")) + + # snapshot 3 modified a file underneath parent/ + self.failIfEqual(c("3-b"), c("2-b")) # 3 modified a file + self.failIfEqual(c("3-bp"), c("2-bp")) + # but the subdirs are the same + self.failUnlessEqual(c("3-bps1"), c("2-bps1")) + self.failUnlessEqual(c("3-bps2"), c("2-bps2")) + + # snapshot 4 should be the same as 2 + self.failUnlessEqual(c("4-b"), c("2-b")) + self.failUnlessEqual(c("4-bp"), c("2-bp")) + self.failUnlessEqual(c("4-bps1"), c("2-bps1")) + self.failUnlessEqual(c("4-bps2"), c("2-bps2")) + + # snapshot 5 added a file under subdir1 + self.failIfEqual(c("5-b"), c("4-b")) + self.failIfEqual(c("5-bp"), c("4-bp")) + self.failIfEqual(c("5-bps1"), c("4-bps1")) + self.failUnlessEqual(c("5-bps2"), c("4-bps2")) + + # snapshot 6 copied a directory-it should be shared + self.failIfEqual(c("6-b"), c("5-b")) + self.failIfEqual(c("6-bp"), c("5-bp")) + self.failUnlessEqual(c("6-bps1"), c("5-bps1")) + self.failIfEqual(c("6-bps2"), c("5-bps2")) + self.failUnlessEqual(c("6-bps2c1"), c("6-bps1")) + + d.addCallback(check_consolidation) + + return d + + def build_manifest(self, root): + # like dirnode.build_manifest, but this one doesn't skip duplicate + # nodes (i.e. it is not cycle-resistant). + manifest = [] + manifest.append( ( (), root.get_uri() ) ) + d = self.manifest_of(None, root, manifest, () ) + d.addCallback(lambda ign: manifest) + return d + + def manifest_of(self, ignored, dirnode, manifest, path): + d = dirnode.list() + def _got_children(children): + d = defer.succeed(None) + for name, (child, metadata) in children.iteritems(): + childpath = path + (name,) + manifest.append( (childpath, child.get_uri()) ) + if IDirectoryNode.providedBy(child): + d.addCallback(self.manifest_of, child, manifest, childpath) + return d + d.addCallback(_got_children) + return d