viewvc-4intranet/bin/svndbadmin

361 lines
12 KiB
Python
Executable File

#!/usr/bin/env python
# -*-python-*-
#
# Copyright (C) 2004-2007 James Henstridge
#
# By using this file, you agree to the terms and conditions set forth in
# the LICENSE.html file which can be found at the top level of the ViewVC
# distribution or at http://viewvc.org/license-1.html.
#
# For more information, visit http://viewvc.org/
#
# -----------------------------------------------------------------------
#
# administrative program for loading Subversion revision information
# into the checkin database. It can be used to add a single revision
# to the database, or rebuild/update all revisions.
#
# To add all the checkins from a Subversion repository to the checkin
# database, run the following:
# /path/to/svndbadmin rebuild /path/to/repo
#
# This script can also be called from the Subversion post-commit hook,
# something like this:
# REPOS="$1"
# REV="$2"
# /path/to/svndbadmin update "$REPOS" "$REV"
#
# If you allow changes to revision properties in your repository, you
# might also want to set up something similar in the
# post-revprop-change hook using "update" with the --force option to
# keep the checkin database consistent with the repository.
#
# -----------------------------------------------------------------------
#
#########################################################################
#
# INSTALL-TIME CONFIGURATION
#
# These values will be set during the installation process. During
# development, they will remain None.
#
LIBRARY_DIR = None
CONF_PATHNAME = None
# Adjust sys.path to include our library directory
import sys
import os
if LIBRARY_DIR:
sys.path.insert(0, LIBRARY_DIR)
else:
sys.path.insert(0, os.path.abspath(os.path.join(sys.argv[0], "../../lib")))
#########################################################################
import os
import string
import re
import svn.core
import svn.repos
import svn.fs
import svn.delta
import cvsdb
import viewvc
import vclib
class SvnRepo:
"""Class used to manage a connection to a SVN repository."""
def __init__(self, path):
self.path = path
self.repo = svn.repos.svn_repos_open(path)
self.fs = svn.repos.svn_repos_fs(self.repo)
self.rev_max = svn.fs.youngest_rev(self.fs)
def __getitem__(self, rev):
if rev is None:
rev = self.rev_max
elif rev < 0:
rev = rev + self.rev_max + 1
assert 0 <= rev <= self.rev_max
rev = SvnRev(self, rev)
return rev
_re_diff_change_command = re.compile('(\d+)(?:,(\d+))?([acd])(\d+)(?:,(\d+))?')
def _get_diff_counts(diff_fp):
"""Calculate the plus/minus counts by parsing the output of a
normal diff. The reasons for choosing Normal diff format are:
- the output is short, so should be quicker to parse.
- only the change commands need be parsed to calculate the counts.
- All file data is prefixed, so won't be mistaken for a change
command.
This code is based on the description of the format found in the
GNU diff manual."""
plus, minus = 0, 0
line = diff_fp.readline()
while line:
match = re.match(_re_diff_change_command, line)
if match:
# size of first range
if match.group(2):
count1 = int(match.group(2)) - int(match.group(1)) + 1
else:
count1 = 1
cmd = match.group(3)
# size of second range
if match.group(5):
count2 = int(match.group(5)) - int(match.group(4)) + 1
else:
count2 = 1
if cmd == 'a':
# LaR - insert after line L of file1 range R of file2
plus = plus + count2
elif cmd == 'c':
# FcT - replace range F of file1 with range T of file2
minus = minus + count1
plus = plus + count2
elif cmd == 'd':
# RdL - remove range R of file1, which would have been
# at line L of file2
minus = minus + count1
line = diff_fp.readline()
return plus, minus
class SvnRev:
"""Class used to hold information about a particular revision of
the repository."""
def __init__(self, repo, rev):
self.repo = repo
self.rev = rev
self.rev_roots = {} # cache of revision roots
# revision properties ...
revprops = svn.fs.revision_proplist(repo.fs, rev)
self.author = str(revprops.get(svn.core.SVN_PROP_REVISION_AUTHOR,''))
self.date = str(revprops.get(svn.core.SVN_PROP_REVISION_DATE, ''))
self.log = str(revprops.get(svn.core.SVN_PROP_REVISION_LOG, ''))
# convert the date string to seconds since epoch ...
try:
self.date = svn.core.svn_time_from_cstring(self.date) / 1000000
except:
self.date = None
# get a root for the current revisions
fsroot = self._get_root_for_rev(rev)
# find changes in the revision
editor = svn.repos.RevisionChangeCollector(repo.fs, rev)
e_ptr, e_baton = svn.delta.make_editor(editor)
svn.repos.svn_repos_replay(fsroot, e_ptr, e_baton)
self.changes = []
for path, change in editor.changes.items():
# skip non-file changes
if change.item_kind != svn.core.svn_node_file:
continue
# deal with the change types we handle
base_root = None
if change.base_path:
base_root = self._get_root_for_rev(change.base_rev)
if not change.path:
action = 'remove'
elif change.added:
action = 'add'
else:
action = 'change'
diffobj = svn.fs.FileDiff(base_root and base_root or None,
base_root and change.base_path or None,
change.path and fsroot or None,
change.path and change.path or None)
diff_fp = diffobj.get_pipe()
plus, minus = _get_diff_counts(diff_fp)
self.changes.append((path, action, plus, minus))
def _get_root_for_rev(self, rev):
"""Fetch a revision root from a cache of such, or a fresh root
(which is then cached for later use."""
if not self.rev_roots.has_key(rev):
self.rev_roots[rev] = svn.fs.revision_root(self.repo.fs, rev)
return self.rev_roots[rev]
def handle_revision(db, command, repo, rev, verbose, force=0):
"""Adds a particular revision of the repository to the checkin database."""
revision = repo[rev]
committed = 0
if verbose: print "Building commit info for revision %d..." % (rev),
if not revision.changes:
if verbose: print "skipped (no changes)."
return
for (path, action, plus, minus) in revision.changes:
directory, file = os.path.split(path)
commit = cvsdb.CreateCommit()
commit.SetRepository(repo.path)
commit.SetDirectory(directory)
commit.SetFile(file)
commit.SetRevision(str(rev))
commit.SetAuthor(revision.author)
commit.SetDescription(revision.log)
commit.SetTime(revision.date)
commit.SetPlusCount(plus)
commit.SetMinusCount(minus)
commit.SetBranch(None)
if action == 'add':
commit.SetTypeAdd()
elif action == 'remove':
commit.SetTypeRemove()
elif action == 'change':
commit.SetTypeChange()
if command == 'update':
result = db.CheckCommit(commit)
if result and not force:
continue # already recorded
# commit to database
db.AddCommit(commit)
committed = 1
if verbose:
if committed:
print "done."
else:
print "skipped (already recorded)."
def main(command, repository, revs=[], verbose=0, force=0):
cfg = viewvc.load_config(CONF_PATHNAME)
db = cvsdb.ConnectDatabase(cfg)
if command in ('rebuild', 'purge'):
if verbose:
print "Purging commit info for repository root `%s'" % repository
db.PurgeRepository(repository)
repo = SvnRepo(repository)
if command == 'rebuild' or (command == 'update' and not revs):
for rev in range(repo.rev_max+1):
handle_revision(db, command, repo, rev, verbose)
elif command == 'update':
if revs[0] is None:
revs[0] = repo.rev_max
if revs[1] is None:
revs[1] = repo.rev_max
revs.sort()
for rev in range(revs[0], revs[1]+1):
handle_revision(db, command, repo, rev, verbose, force)
def _rev2int(r):
if r == 'HEAD':
r = None
else:
r = int(r)
if r < 0:
raise ValueError, "invalid revision '%d'" % (r)
return r
def usage():
cmd = os.path.basename(sys.argv[0])
sys.stderr.write(
"""Administer the ViewVC checkins database data for the Subversion repository
located at REPOS-PATH.
Usage: 1. %s [-v] rebuild REPOS-PATH
2. %s [-v] update REPOS-PATH [REV:[REV2]] [--force]
3. %s [-v] purge REPOS-PATH
1. Rebuild the commit database information for the repository located
at REPOS-PATH across all revisions, after first purging
information specific to that repository (if any).
2. Update the commit database information for the repository located
at REPOS-PATH across all revisions or, optionally, only for the
specified revision REV (or revision range REV:REV2). This is just
like rebuilding, except that, unless --force is specified, no
commit information will be stored for commits already present in
the database. If a range is specified, the revisions will be
processed in ascending order, and you may specify "HEAD" to
indicate "the youngest revision currently in the repository".
3. Purge information specific to the repository located at REPOS-PATH
from the database.
Use the -v flag to cause this script to give progress information as it works.
""" % (cmd, cmd, cmd))
sys.exit(1)
if __name__ == '__main__':
verbose = 0
force = 0
args = sys.argv
try:
index = args.index('-v')
verbose = 1
del args[index]
except ValueError:
pass
try:
index = args.index('--force')
force = 1
del args[index]
except ValueError:
pass
if len(args) < 3:
usage()
command = string.lower(args[1])
if command not in ('rebuild', 'update', 'purge'):
sys.stderr.write('ERROR: unknown command %s\n' % command)
usage()
repository = args[2]
if not os.path.exists(repository):
sys.stderr.write('ERROR: could not find repository %s\n' % args[2])
usage()
repository = vclib.svn.canonicalize_rootpath(repository)
revs = []
if len(sys.argv) > 3:
if command == 'rebuild':
sys.stderr.write('ERROR: rebuild no longer accepts a revision '
'number argument. Usage update --force.')
usage()
elif command != 'update':
usage()
try:
revs = map(lambda x: _rev2int(x), sys.argv[3].split(':'))
if len(revs) > 2:
raise ValueError, "too many revisions in range"
if len(revs) == 1:
revs.append(revs[0])
except ValueError:
sys.stderr.write('ERROR: invalid revision specification "%s"\n' \
% sys.argv[3])
usage()
else:
rev = None
try:
repository = cvsdb.CleanRepository(os.path.abspath(repository))
main(command, repository, revs, verbose, force)
except KeyboardInterrupt:
print
print '** break **'
sys.exit(0)