diff --git a/swh/loader/svn/loader.py b/swh/loader/svn/loader.py index bbb2a69..a3d61a5 100644 --- a/swh/loader/svn/loader.py +++ b/swh/loader/svn/loader.py @@ -1,524 +1,522 @@ # Copyright (C) 2015-2017 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information """Loader in charge of injecting either new or existing svn mirrors to swh-storage. """ import abc import os import shutil from swh.core import utils from swh.model import hashutil from swh.model.from_disk import Directory from swh.model.identifiers import identifier_to_bytes, revision_identifier from swh.loader.core.loader import SWHLoader from . import svn, converters from .utils import init_svn_repo_from_archive_dump def _revision_id(revision): return identifier_to_bytes(revision_identifier(revision)) class SvnLoaderEventful(ValueError): """A wrapper exception to transit the swh_revision onto which the loading failed. """ def __init__(self, e, swh_revision): super().__init__(e) self.swh_revision = swh_revision class SvnLoaderUneventful(ValueError): def __init__(self, e, *args): super().__init__(e, *args) class SvnLoaderHistoryAltered(ValueError): def __init__(self, e, *args): super().__init__(e, *args) class BaseSvnLoader(SWHLoader, metaclass=abc.ABCMeta): """Base Svn loader to load one svn repository according to specific policies (only swh one now). The main entry point of this is (no need to override it):: def load(self, origin_visit, last_known_swh_revision=None): pass Inherit this class and then override the following functions:: def build_swh_revision(self, rev, commit, dir_id, parents): # This is in charge of converting an svn revision to a compliant # swh revision def process_repository(self): # This is in charge of processing the actual svn repository and # store the result to swh storage. """ CONFIG_BASE_FILENAME = 'loader/svn' ADDITIONAL_CONFIG = { 'check_revision': ('int', 1000), } def __init__(self): super().__init__(logging_class='swh.loader.svn.SvnLoader') self.check_revision = self.config['check_revision'] self.origin_id = None @abc.abstractmethod def swh_revision_hash_tree_at_svn_revision(self, revision): """Compute and return the hash tree at a given svn revision. Args: rev (int): the svn revision we want to check Returns: The hash tree directory as bytes. """ pass @abc.abstractmethod def get_svn_repo(self, svn_url, destination_path, origin): """Instantiates the needed svnrepo collaborator to permit reading svn repository. Args: svn_url: the svn repository url to read from destination_path: the local path on disk to compute data origin: the corresponding origin Returns: Instance of :mod:`swh.loader.svn.svn` clients """ raise NotImplementedError @abc.abstractmethod def build_swh_revision(self, rev, commit, dir_id, parents): """Convert an svn revision to an swh one according to the loader's policy (git-svn or swh). Args: rev: the svn revision number commit: dictionary with keys: author\_name, author\_date, rev, message dir_id: the hash tree computation parents: the revision's parents Returns: The swh revision """ raise NotImplementedError @abc.abstractmethod def process_repository(self, origin_visit, last_known_swh_revision=None): """The main idea of this function is to: - iterate over the svn commit logs - extract the svn commit log metadata - compute the hashes from the current directory down to the file - compute the equivalent swh revision - send all those objects for storage - create an swh occurrence pointing to the last swh revision seen - send that occurrence for storage in swh-storage. """ raise NotImplementedError def process_svn_revisions(self, svnrepo, revision_start, revision_end, revision_parents): """Process revisions from revision_start to revision_end and send to swh for storage. At each svn revision, checkout the repository, compute the tree hash and blobs and send for swh storage to store. Then computes and yields the swh revision. Note that at every self.check_revision, an svn export is done and a hash tree is computed to check that no divergence occurred. Yields: swh revision as a dictionary with keys, sha1_git, sha1, etc... """ gen_revs = svnrepo.swh_hash_data_per_revision( revision_start, revision_end) swh_revision = None count = 0 for rev, nextrev, commit, new_objects, root_directory in gen_revs: count += 1 # Send the associated contents/directories self.maybe_load_contents(new_objects.get('content', {}).values()) self.maybe_load_directories( new_objects.get('directory', {}).values()) # compute the fs tree's checksums dir_id = root_directory.hash swh_revision = self.build_swh_revision( rev, commit, dir_id, revision_parents[rev]) swh_revision['id'] = _revision_id(swh_revision) self.log.debug('rev: %s, swhrev: %s, dir: %s' % ( rev, hashutil.hash_to_hex(swh_revision['id']), hashutil.hash_to_hex(dir_id))) if (count % self.check_revision) == 0: # hash computation check self.log.info('Checking hash computations on revision %s...' % rev) checked_dir_id = self.swh_revision_hash_tree_at_svn_revision( rev) if checked_dir_id != dir_id: err = 'Hash tree computation divergence detected (%s != %s), stopping!' % ( # noqa hashutil.hash_to_hex(dir_id), hashutil.hash_to_hex(checked_dir_id) ) raise ValueError(err) if nextrev: revision_parents[nextrev] = [swh_revision['id']] yield swh_revision def process_swh_revisions(self, svnrepo, revision_start, revision_end, revision_parents): """Process and store revision to swh (sent by blocks of revision_packet_size) Returns: The latest revision stored. """ try: swh_revision_gen = self.process_svn_revisions(svnrepo, revision_start, revision_end, revision_parents) revs = [] for revisions in utils.grouper( swh_revision_gen, self.config['revision_packet_size']): revs = list(revisions) self.log.info('Processed %s revisions: [%s, ...]' % ( len(revs), hashutil.hash_to_hex(revs[0]['id']))) self.maybe_load_revisions(revs) except Exception as e: if revs: # flush remaining revisions self.maybe_load_revisions(revs) # Take the last one as the last known revisions known_swh_rev = revs[-1] _id = known_swh_rev.get('id') if not _id: _id = _revision_id(known_swh_rev) # Then notify something is wrong, and we stopped at that rev. raise SvnLoaderEventful(e, swh_revision={ 'id': _id, }) raise e return revs[-1] def process_swh_occurrence(self, revision, origin_visit): """Process and load the occurrence pointing to the latest revision. """ occ = converters.build_swh_occurrence( revision['id'], origin_visit['origin'], origin_visit['visit']) self.log.debug('occ: %s' % occ) self.maybe_load_occurrences([occ]) def send_origin(self, origin): """Inhibit multiple calls to send_origin. """ if self.origin_id: return self.origin_id return super().send_origin(origin) def prepare(self, *args, **kwargs): self.args = args destination_path = kwargs['destination_path'] # local svn url svn_url = kwargs['svn_url'] origin_url = kwargs.get('origin_url') self.visit_date = kwargs.get('visit_date') origin = { 'url': origin_url if origin_url else svn_url, 'type': 'svn', } self.origin_id = self.send_origin(origin) origin['id'] = self.origin_id self.origin = origin if 'swh_revision' in kwargs: self.last_known_swh_revision = hashutil.hash_to_bytes( kwargs['swh_revision']) else: self.last_known_swh_revision = None self.svnrepo = self.get_svn_repo(svn_url, destination_path, origin) - self.fetch_history_id = self.open_fetch_history() - def get_origin(self): """Retrieve the origin we are working with. """ return self.origin # set in prepare method def fetch_data(self): """We need to fetch and stream the data to store directly. So fetch_data do actually nothing. The method ``store_data`` below is in charge to do everything, fetch and store. """ pass def store_data(self): """We need to fetch and stream the data to store directly because there is too much data and state changes. Everything is intertwined together (We receive patch and apply on disk and compute at the hashes at the same time) So every data to fetch and store is done here. Note: origin_visit and last_known_swh_revision must have been set in the prepare method. """ origin_visit = {'origin': self.origin_id, 'visit': self.visit} try: latest_rev = self.process_repository(origin_visit, self.last_known_swh_revision) except SvnLoaderUneventful as e: # Nothing needed to be done, the visit is full nonetheless self.log.info('Uneventful visit. Detail: %s' % e) except SvnLoaderEventful as e: self.log.error('Eventful partial visit. Detail: %s' % e) latest_rev = e.swh_revision self.process_swh_occurrence(latest_rev, origin_visit) raise except SvnLoaderHistoryAltered as e: self.log.error('History altered. Detail: %s' % e) raise else: self.process_swh_occurrence(latest_rev, origin_visit) class SWHSvnLoader(BaseSvnLoader): """Swh svn loader is the main implementation destined for production. This implementation is able to deal with update on known svn repository. Default policy: Keep data as close as possible from the original svn data. We only add information that are needed for update or continuing from last known revision (svn revision and svn repository's uuid). """ def cleanup(self): """Clean after oneself. This is in charge to flush the remaining data to write in swh storage. And to clean up the svn repository's working representation on disk. """ self.svnrepo.clean_fs() def swh_revision_hash_tree_at_svn_revision(self, revision): """Compute a given hash tree at specific revision. """ local_dirname, local_url = self.svnrepo.export_temporary(revision) h = Directory.from_disk(path=local_url).hash self.svnrepo.clean_fs(local_dirname) return h def get_svn_repo(self, svn_url, destination_path, origin): return svn.SWHSvnRepo( svn_url, origin['id'], self.storage, destination_path=destination_path) def swh_previous_revision(self, prev_swh_revision=None): """Retrieve swh's previous revision if any. """ return self.svnrepo.swh_previous_revision(prev_swh_revision) def check_history_not_altered(self, svnrepo, revision_start, swh_rev): """Given a svn repository, check if the history was not tampered with. """ revision_id = swh_rev['id'] parents = swh_rev['parents'] hash_data_per_revs = svnrepo.swh_hash_data_at_revision(revision_start) rev = revision_start rev, _, commit, _, root_dir = list(hash_data_per_revs)[0] dir_id = root_dir.hash swh_revision = self.build_swh_revision(rev, commit, dir_id, parents) swh_revision_id = _revision_id(swh_revision) return swh_revision_id == revision_id def build_swh_revision(self, rev, commit, dir_id, parents): """Build the swh revision dictionary. This adds: - the `'synthetic`' flag to true - the '`extra_headers`' containing the repository's uuid and the svn revision number. Args: rev: the svn revision commit: the commit metadata dir_id: the upper tree's hash identifier parents: the parents' identifiers Returns: The swh revision corresponding to the svn revision. """ return converters.build_swh_revision(rev, commit, self.svnrepo.uuid, dir_id, parents) def init_from(self, partial_swh_revision, previous_swh_revision): """Function to determine from where to start from. Args: partial_swh_revision: A known revision from which the previous loading did not finish. known_previous_revision: A known revision from which the previous loading did finish. Returns: The revision from which to start or None if nothing (fresh start). """ if partial_swh_revision and not previous_swh_revision: return partial_swh_revision if not partial_swh_revision and previous_swh_revision: return previous_swh_revision if partial_swh_revision and previous_swh_revision: # will determine from which to start from extra_headers1 = dict( partial_swh_revision['metadata']['extra_headers']) extra_headers2 = dict( previous_swh_revision['metadata']['extra_headers']) rev_start1 = int(extra_headers1['svn_revision']) rev_start2 = int(extra_headers2['svn_revision']) if rev_start1 <= rev_start2: return previous_swh_revision return partial_swh_revision return None def process_repository(self, origin_visit, last_known_swh_revision=None): svnrepo = self.svnrepo # default configuration revision_start = 1 revision_parents = { revision_start: [] } # Check if we already know a previous revision for that origin swh_rev = self.swh_previous_revision() # Determine from which known revision to start swh_rev = self.init_from(last_known_swh_revision, previous_swh_revision=swh_rev) if swh_rev: # Yes, we know a previous revision. Try and update it. extra_headers = dict(swh_rev['metadata']['extra_headers']) revision_start = int(extra_headers['svn_revision']) revision_parents = { revision_start: swh_rev['parents'], } self.log.debug('svn export --ignore-keywords %s@%s' % ( svnrepo.remote_url, revision_start)) if swh_rev and not self.check_history_not_altered( svnrepo, revision_start, swh_rev): msg = 'History of svn %s@%s history modified. Skipping...' % ( svnrepo.remote_url, revision_start) raise SvnLoaderHistoryAltered(msg, *self.args) # now we know history is ok, we start at next revision revision_start = revision_start + 1 # and the parent become the latest know revision for # that repository revision_parents[revision_start] = [swh_rev['id']] revision_end = svnrepo.head_revision() self.log.info('[revision_start-revision_end]: [%s-%s]' % ( revision_start, revision_end)) if revision_start > revision_end and revision_start is not 1: msg = '%s@%s already injected.' % (svnrepo.remote_url, revision_end) raise SvnLoaderUneventful(msg, *self.args) self.log.info('Processing %s.' % svnrepo) # process and store revision to swh (sent by by blocks of # 'revision_packet_size') return self.process_swh_revisions( svnrepo, revision_start, revision_end, revision_parents) class SWHSvnLoaderFromDumpArchive(SWHSvnLoader): """Load a svn repository from an archive (containing a dump). """ def __init__(self, archive_path): super().__init__() self.log.info('Archive to mount and load %s' % archive_path) self.temp_dir, self.repo_path = init_svn_repo_from_archive_dump( archive_path) def cleanup(self): super().cleanup() if self.temp_dir and os.path.exists(self.temp_dir): self.log.debug('Clean up temp directory %s for project %s' % ( self.temp_dir, os.path.basename(self.repo_path))) shutil.rmtree(self.temp_dir)