diff --git a/swh/loader/svn/config.py b/swh/loader/svn/config.py new file mode 100644 index 0000000..365d576 --- /dev/null +++ b/swh/loader/svn/config.py @@ -0,0 +1,6 @@ +# Copyright (C) 2018 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 + +DEFAULT_BRANCH = b'master' diff --git a/swh/loader/svn/converters.py b/swh/loader/svn/converters.py index c0a3c64..8a06418 100644 --- a/swh/loader/svn/converters.py +++ b/swh/loader/svn/converters.py @@ -1,204 +1,187 @@ # Copyright (C) 2015-2018 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 from email import utils from .utils import strdate_to_timestamp def svn_date_to_gitsvn_date(strdate): """Convert a string date to an swh one. Args: strdate: A string formatted for .utils.strdate_to_timestamp to do its jobs Returns: An swh date format with an integer timestamp. """ ts = strdate_to_timestamp(strdate) return { 'timestamp': { 'seconds': ts['seconds'], 'microseconds': 0, }, 'offset': 0 } def svn_date_to_swh_date(strdate): """Convert a string date to an swh one. Args: strdate: A string formatted for .utils.strdate_to_timestamp to do its jobs Returns: An swh date format """ return { 'timestamp': strdate_to_timestamp(strdate), 'offset': 0 } def svn_author_to_swh_person(author): """Convert an svn author to an swh person. Default policy: No information is added. Args: author (string): the svn author (in bytes) Returns: a dictionary with keys: fullname: the author's associated fullname name: the author's associated name email: None (no email in svn) """ if not author: return {'fullname': b'', 'name': None, 'email': None} if isinstance(author, str): author = author.encode('utf-8') if b'<' in author and b'>' in author: name, email = utils.parseaddr(author.decode('utf-8')) return { 'fullname': author, 'name': name.encode('utf-8'), 'email': email.encode('utf-8') } return {'fullname': author, 'email': None, 'name': author} def svn_author_to_gitsvn_person(author, repo_uuid): """Convert an svn author to a person suitable for insertion. Default policy: If no email is found, the email is created using the author and the repo_uuid. Args: author (string): the svn author (in bytes) repo_uuid (bytes): the repository's uuid Returns: a dictionary with keys: fullname: the author's associated fullname name: the author's associated name email: None (no email in svn) """ if not author: author = '(no author)' author = author.encode('utf-8') if b'<' in author and b'>' in author: name, email = utils.parseaddr(author.decode('utf-8')) return { 'fullname': author, 'name': name.encode('utf-8'), 'email': email.encode('utf-8') } # we'll construct the author's fullname the same way git svn does # 'user ' email = b'@'.join([author, repo_uuid]) return { 'fullname': b''.join([author, b' ', b'<', email, b'>']), 'name': author, 'email': email, } def build_swh_revision(rev, commit, repo_uuid, dir_id, parents): """Given a svn revision, build a swh revision. This adds an ['metadata']['extra-headers'] entry with the repository's uuid and the svn revision. Args: - rev: the svn revision number - commit: the commit metadata - repo_uuid: The repository's uuid - dir_id: the tree's hash identifier - parents: the revision's parents identifier Returns: The swh revision dictionary. """ author = commit['author_name'] msg = commit['message'] date = commit['author_date'] metadata = { 'extra_headers': [ ['svn_repo_uuid', repo_uuid], ['svn_revision', str(rev).encode('utf-8')] ] } return { 'date': date, 'committer_date': date, 'type': 'svn', 'directory': dir_id, 'message': msg, 'author': author, 'committer': author, 'synthetic': True, 'metadata': metadata, 'parents': parents, } def build_gitsvn_swh_revision(rev, commit, dir_id, parents): """Given a svn revision, build a swh revision. Args: - rev: the svn revision number - commit: the commit metadata - dir_id: the tree's hash identifier - parents: the revision's parents identifier Returns: The swh revision dictionary. """ author = commit['author_name'] msg = commit['message'] date = commit['author_date'] return { 'date': date, 'committer_date': date, 'type': 'svn', 'directory': dir_id, 'message': msg, 'author': author, 'committer': author, 'synthetic': True, 'metadata': None, 'parents': parents, } - - -def build_swh_snapshot(revision_id, origin_id, visit): - """Build a swh snapshot from the revision id, origin id, and visit. - - """ - return { - 'id': None, - 'branches': { - b'master': { - 'target': revision_id, - 'target_type': 'revision', - 'origin': origin_id, - 'visit': visit, - } - } - } diff --git a/swh/loader/svn/loader.py b/swh/loader/svn/loader.py index b97fe91..bf9e148 100644 --- a/swh/loader/svn/loader.py +++ b/swh/loader/svn/loader.py @@ -1,522 +1,545 @@ # Copyright (C) 2015-2018 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.model.identifiers import snapshot_identifier from swh.loader.core.loader import SWHLoader + from . import svn, converters +from .config import DEFAULT_BRANCH from .utils import init_svn_repo_from_archive_dump def _revision_id(revision): return identifier_to_bytes(revision_identifier(revision)) +def build_swh_snapshot(revision_id, origin_id, visit): + """Build a swh snapshot from the revision id, origin id, and visit. + + """ + return { + 'id': None, + 'branches': { + DEFAULT_BRANCH: { + 'target': revision_id, + 'target_type': 'revision', + 'origin': origin_id, + 'visit': visit, + } + } + } + + 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, start_from_scratch=False): """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_snapshot(self, revision, origin_visit): """Process and load the occurrence pointing to the latest revision. """ - snap = converters.build_swh_snapshot( + snap = build_swh_snapshot( revision['id'], origin_visit['origin'], origin_visit['visit']) - self.log.debug('snap: %s' % snap) + snap_id = snapshot_identifier(snap) + snap['id'] = identifier_to_bytes(snap_id) + self.log.debug('snapshot: %s, id: %s' % (snap, snap_id)) + snap['id'] = identifier_to_bytes(snap_id) self.maybe_load_snapshot(snap) 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') self.start_from_scratch = kwargs.get('start_from_scratch', False) 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) def get_origin(self): """Retrieve the origin we are working with (setup-ed in the prepare method) """ 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, last_known_swh_revision=self.last_known_swh_revision, start_from_scratch=self.start_from_scratch) 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_snapshot(latest_rev, origin_visit) raise except SvnLoaderHistoryAltered as e: self.log.error('History altered. Detail: %s' % e) raise else: self.process_swh_snapshot(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. """ + self.log.debug('#####: %s' % prev_swh_revision) 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, start_from_scratch=False): svnrepo = self.svnrepo # default configuration revision_start = 1 revision_parents = { revision_start: [] } if not start_from_scratch: # 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...' % ( # noqa 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) diff --git a/swh/loader/svn/svn.py b/swh/loader/svn/svn.py index ccc269b..ab359fd 100644 --- a/swh/loader/svn/svn.py +++ b/swh/loader/svn/svn.py @@ -1,369 +1,378 @@ # Copyright (C) 2015-2018 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 """SVN client in charge of iterating over svn logs and yield commit representations including the hash tree/content computations per svn commit. """ import os import tempfile import shutil from subvertpy.ra import RemoteAccess, Auth, get_username_provider from subvertpy import client, properties from swh.model.from_disk import Directory +from .config import DEFAULT_BRANCH from . import ra, converters # When log message contains empty data DEFAULT_AUTHOR_MESSAGE = '' class SvnRepoException(ValueError): def __init__(self, svnrepo, e): super().__init__(e) self.svnrepo = svnrepo class BaseSvnRepo(): """Base svn repository representation for swh. To override some of the behavior regarding the message log properties, you can instantiate a subclass of this class and override:: def convert_commit_author(self, author) def convert_commit_message(self, msg) def convert_commit_date(self, date) see :class:`GitSvnSvnRepo`, :class:`SwhSvnRepo` for instanciation example. """ def __init__(self, remote_url, origin_id, storage, destination_path=None): self.remote_url = remote_url.rstrip('/') self.storage = storage self.origin_id = origin_id if destination_path: os.makedirs(destination_path, exist_ok=True) self.root_dir = destination_path else: self.root_dir = '/tmp' auth = Auth([get_username_provider()]) # one connection for log iteration self.conn_log = RemoteAccess(self.remote_url, auth=auth) # another for replay self.conn = RemoteAccess(self.remote_url, auth=auth) # one client for update operation self.client = client.Client(auth=auth) self.local_dirname = tempfile.mkdtemp(suffix='.tmp', prefix='swh.loader.svn.', dir=self.root_dir) local_name = os.path.basename(self.remote_url) self.local_url = os.path.join(self.local_dirname, local_name).encode( 'utf-8') self.uuid = self.conn.get_uuid().encode('utf-8') def __str__(self): return str({ 'remote_url': self.remote_url, 'local_url': self.local_url, 'uuid': self.uuid, 'swh-origin': self.origin_id }) def head_revision(self): """Retrieve current revision of the repository's working copy. """ return self.conn.get_latest_revnum() def initial_revision(self): """Retrieve the initial revision from which the remote url appeared. Note: This should always be 1 since we won't be dealing with in-depth url. """ return 1 def convert_commit_message(self, msg): """Do something with message (e.g add extra line, etc...) cf. SvnRepo for a simple implementation. Args: msg (str): the commit message to convert. Returns: The transformed message as bytes. """ raise NotImplementedError('Should be overridden by subclass.') def convert_commit_date(self, date): """Convert the message date (e.g, convert into timestamp or whatever makes sense to you.). Args: date (str): the commit date to convert. Returns: The transformed date. """ raise NotImplementedError('Should be overridden by subclass.') def convert_commit_author(self, author): """Convert the commit author (e.g, convert into dict or whatever makes sense to you.). Args: author (str): the commit author to convert. Returns: The transformed author as dict. """ raise NotImplementedError('Should be overridden by subclass.') def __to_entry(self, log_entry): changed_paths, rev, revprops, has_children = log_entry author_date = self.convert_commit_date( revprops.get(properties.PROP_REVISION_DATE)) author = self.convert_commit_author( revprops.get(properties.PROP_REVISION_AUTHOR)) message = self.convert_commit_message( revprops.get(properties.PROP_REVISION_LOG, DEFAULT_AUTHOR_MESSAGE)) return { 'rev': rev, 'author_date': author_date, 'author_name': author, 'message': message, } def logs(self, revision_start, revision_end): """Stream svn logs between revision_start and revision_end by chunks of block_size logs. Yields revision and associated revision information between the revision start and revision_end. Args: revision_start: the svn revision starting bound revision_end: the svn revision ending bound Yields: tuple: tuple of revisions and logs: - revisions: list of revisions in order - logs: Dictionary with key revision number and value the log entry. The log entry is a dictionary with the following keys: - author_date: date of the commit - author_name: name of the author - message: commit message """ for log_entry in self.conn_log.iter_log(paths=None, start=revision_start, end=revision_end, discover_changed_paths=False): yield self.__to_entry(log_entry) def export(self, revision): """Export the repository to a given version. """ self.client.export(self.remote_url, to=self.local_url.decode('utf-8'), rev=revision, ignore_keywords=True) def export_temporary(self, revision): """Export the repository to a given revision in a temporary location. This is up to the caller of this function to clean up the temporary location when done (cf. self.clean_fs method) Args: revision: Revision to export at Returns: The tuple local_dirname the temporary location root folder, local_url where the repository was exported. """ local_dirname = tempfile.mkdtemp( prefix='check-revision-%s.' % revision, dir=self.local_dirname) local_name = os.path.basename(self.remote_url) local_url = os.path.join(local_dirname, local_name) self.client.export( self.remote_url, to=local_url, rev=revision, ignore_keywords=True) return local_dirname, os.fsencode(local_url) def swh_previous_revision(self, previous_swh_revision=None): """Look for possible existing revision in swh. Args: previous_swh_revision: (optional) id of a possible previous swh revision Returns: If previous_swh_revision is not None and exists, returns the complete instance. Otherwise, check for a possible snapshot and returns the targeted revision if it exists. Otherwise, returns None. """ storage = self.storage if not previous_swh_revision: # check latest snapshot's revision latest_snap = storage.snapshot_get_latest(self.origin_id) - if not latest_snap or latest_snap['target_type'] != 'revision': - return None - previous_swh_revision = latest_snap['target'] + if latest_snap: + branches = latest_snap.get('branches') + if not branches: + return None + branch = branches.get(DEFAULT_BRANCH) + if not branch: + return None + target_type = branch['target_type'] + if target_type != 'revision': + return None + previous_swh_revision = branch['target'] revs = list(storage.revision_get([previous_swh_revision])) if revs: return revs[0] def swh_hash_data_per_revision(self, start_revision, end_revision): """Compute swh hash data per each revision between start_revision and end_revision. Args: start_revision: starting revision end_revision: ending revision Yields: tuple (rev, nextrev, commit, objects_per_path) - rev: current revision - nextrev: next revision - commit: commit data (author, date, message) for such revision - objects_per_path: dictionary of path, swh hash data with type """ for commit in self.logs(start_revision, end_revision): rev = commit['rev'] objects = self.swhreplay.compute_hashes(rev) if rev == end_revision: nextrev = None else: nextrev = rev + 1 yield rev, nextrev, commit, objects, self.swhreplay.directory def swh_hash_data_at_revision(self, revision): """Compute the hash data at revision. Expected to be used for update only. """ # Update the disk at revision self.export(revision) # Compute the current hashes on disk directory = Directory.from_disk(path=os.fsencode(self.local_url), save_path=True) # Update the replay collaborator with the right state self.swhreplay = ra.SWHReplay( conn=self.conn, rootpath=self.local_url, directory=directory) # Retrieve the commit information for revision commit = list(self.logs(revision, revision))[0] yield revision, revision + 1, commit, {}, directory def clean_fs(self, local_dirname=None): """Clean up the local working copy. Args: local_dirname (str): Path to remove recursively if provided. Otherwise, remove the temporary upper root tree used for svn repository loading. """ if local_dirname: shutil.rmtree(local_dirname) else: shutil.rmtree(self.local_dirname) class SWHSvnRepo(BaseSvnRepo): """Same as :class:`BaseSvnRepo` except for: - the commit message which is simply encoded - the commit author is left as is. - the commit timestamp is left as is. """ def __init__(self, remote_url, origin_id, storage, destination_path=None): super().__init__(remote_url, origin_id, storage, destination_path=destination_path) self.swhreplay = ra.SWHReplay( conn=self.conn, rootpath=self.local_url) def convert_commit_message(self, msg): """Simply encode the commit message. Args: msg (str): the commit message to convert. Returns: The transformed message as bytes. """ if isinstance(msg, bytes): return msg return msg.encode('utf-8') def convert_commit_date(self, date): """Convert the message commit date into a timestamp in swh format. The precision is kept. Args: date (str): the commit date to convert. Returns: The transformed date. """ return converters.svn_date_to_swh_date(date) def convert_commit_author(self, author): """Convert the commit author into an swh person. The user becomes a dictionary of the form:: { name: author, email: '', fullname: author } Args: author (str): the commit author to convert. Returns: The transformed author as dict. """ return converters.svn_author_to_swh_person(author) diff --git a/swh/loader/svn/tests/test_converters.py b/swh/loader/svn/tests/test_converters.py index 7e88fa8..fba7ae7 100644 --- a/swh/loader/svn/tests/test_converters.py +++ b/swh/loader/svn/tests/test_converters.py @@ -1,314 +1,294 @@ # Copyright (C) 2015-2018 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 import unittest from nose.tools import istest from swh.loader.svn import converters class TestAuthorGitSvnConverters(unittest.TestCase): @istest def svn_author_to_gitsvn_person(self): """The author should have name, email and fullname filled. """ actual_person = converters.svn_author_to_gitsvn_person( 'tony ', repo_uuid=None) self.assertEquals(actual_person, { 'fullname': b'tony ', 'name': b'tony', 'email': b'ynot@dagobah', }) @istest def svn_author_to_gitsvn_person_no_email(self): """The author should see his/her email filled with author@. """ actual_person = converters.svn_author_to_gitsvn_person( 'tony', repo_uuid=b'some-uuid') self.assertEquals(actual_person, { 'fullname': b'tony ', 'name': b'tony', 'email': b'tony@some-uuid', }) @istest def svn_author_to_gitsvn_person_empty_person(self): """The empty person should see name, fullname and email filled. """ actual_person = converters.svn_author_to_gitsvn_person( '', repo_uuid=b'some-uuid') self.assertEqual(actual_person, { 'fullname': b'(no author) <(no author)@some-uuid>', 'name': b'(no author)', 'email': b'(no author)@some-uuid' }) class TestAuthorSWHConverters(unittest.TestCase): @istest def svn_author_to_swh_person(self): """The author should have name, email and fullname filled. """ actual_person = converters.svn_author_to_swh_person( 'tony ') self.assertEquals(actual_person, { 'fullname': b'tony ', 'name': b'tony', 'email': b'ynot@dagobah', }) @istest def svn_author_to_swh_person_no_email(self): """The author and fullname should be the same as the input (author). """ actual_person = converters.svn_author_to_swh_person('tony') self.assertEquals(actual_person, { 'fullname': b'tony', 'name': b'tony', 'email': None, }) @istest def svn_author_to_swh_person_empty_person(self): """Empty person has only its fullname filled with the empty byte-string. """ actual_person = converters.svn_author_to_swh_person('') self.assertEqual(actual_person, { 'fullname': b'', 'name': None, 'email': None, }) class TestSWHRevisionConverters(unittest.TestCase): @istest def build_swh_revision_default(self): """This should build the swh revision with the swh revision's extra headers about the repository. """ actual_swh_revision = converters.build_swh_revision( repo_uuid=b'uuid', dir_id='dir-id', commit={ 'author_name': { 'name': b'theo', 'email': b'theo@uuid', 'fullname': b'theo ' }, 'message': b'commit message', 'author_date': { 'timestamp': { 'seconds': 1088108379, 'microseconds': 0, }, 'offset': 0 } }, rev=10, parents=['123']) date = { 'timestamp': { 'seconds': 1088108379, 'microseconds': 0, }, 'offset': 0, } self.assertEquals(actual_swh_revision, { 'date': date, 'committer_date': date, 'type': 'svn', 'directory': 'dir-id', 'message': b'commit message', 'author': { 'name': b'theo', 'email': b'theo@uuid', 'fullname': b'theo ' }, 'committer': { 'name': b'theo', 'email': b'theo@uuid', 'fullname': b'theo ' }, 'synthetic': True, 'metadata': { 'extra_headers': [ ['svn_repo_uuid', b'uuid'], ['svn_revision', b'10'], ] }, 'parents': ['123'], }) class TestGitSvnRevisionConverters(unittest.TestCase): @istest def build_gitsvn_swh_revision_default(self): """This should build the swh revision without the swh revision's extra headers about the repository. """ actual_swh_revision = converters.build_gitsvn_swh_revision( dir_id='dir-id', commit={ 'author_name': { 'name': b'theo', 'email': b'theo@uuid', 'fullname': b'theo ' }, 'message': b'commit message', 'author_date': { 'timestamp': { 'seconds': 1088108379, 'microseconds': 0, }, 'offset': 0 } }, rev=10, parents=['123']) date = { 'timestamp': { 'seconds': 1088108379, 'microseconds': 0, }, 'offset': 0, } self.assertEquals(actual_swh_revision, { 'date': date, 'committer_date': date, 'type': 'svn', 'directory': 'dir-id', 'message': b'commit message', 'author': { 'name': b'theo', 'email': b'theo@uuid', 'fullname': b'theo ' }, 'committer': { 'name': b'theo', 'email': b'theo@uuid', 'fullname': b'theo ' }, 'synthetic': True, 'metadata': None, 'parents': ['123'], }) -class TestSWHOccurrence(unittest.TestCase): - @istest - def build_swh_snapshot(self): - actual_snap = converters.build_swh_snapshot('revision-id', - 'origin-id', - visit=10) - - self.assertEquals(actual_snap, { - 'id': None, - 'branches': { - b'master': { - 'target': 'revision-id', - 'target_type': 'revision', - 'origin': 'origin-id', - 'visit': 10 - } - } - }) - - class ConvertSWHDate(unittest.TestCase): @istest def svn_date_to_swh_date(self): """The timestamp should not be tampered with and include the decimals. """ self.assertEquals( converters.svn_date_to_swh_date('2011-05-31T06:04:39.500900Z'), { 'timestamp': { 'seconds': 1306821879, 'microseconds': 500900, }, 'offset': 0 }) self.assertEquals( converters.svn_date_to_swh_date('2011-05-31T06:04:39.800722Z'), { 'timestamp': { 'seconds': 1306821879, 'microseconds': 800722, }, 'offset': 0 }) @istest def svn_date_to_swh_date_epoch(self): """Empty date should be EPOCH (timestamp and offset at 0).""" # It should return 0, epoch self.assertEquals({ 'timestamp': { 'seconds': 0, 'microseconds': 0, }, 'offset': 0, }, converters.svn_date_to_swh_date('')) self.assertEquals({ 'timestamp': { 'seconds': 0, 'microseconds': 0, }, 'offset': 0, }, converters.svn_date_to_swh_date(None)) class ConvertGitSvnDate(unittest.TestCase): @istest def svn_date_to_gitsvn_date(self): """The timestamp should be truncated to be an integer.""" actual_ts = converters.svn_date_to_gitsvn_date( '2011-05-31T06:04:39.800722Z') self.assertEquals(actual_ts, { 'timestamp': { 'seconds': 1306821879, 'microseconds': 0, }, 'offset': 0, }) @istest def svn_date_to_gitsvn_date_epoch(self): """Empty date should be EPOCH (timestamp and offset at 0).""" # It should return 0, epoch self.assertEquals({ 'timestamp': { 'seconds': 0, 'microseconds': 0, }, 'offset': 0, }, converters.svn_date_to_gitsvn_date('')) self.assertEquals({ 'timestamp': { 'seconds': 0, 'microseconds': 0, }, 'offset': 0, }, converters.svn_date_to_gitsvn_date(None)) diff --git a/swh/loader/svn/tests/test_loader.py b/swh/loader/svn/tests/test_loader.py index a1d56c0..8c1f58e 100644 --- a/swh/loader/svn/tests/test_loader.py +++ b/swh/loader/svn/tests/test_loader.py @@ -1,854 +1,876 @@ # Copyright (C) 2016-2018 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 from nose.tools import istest +from test_base import BaseTestSvnLoader +from unittest import TestCase from swh.model import hashutil + +from swh.loader.svn.config import DEFAULT_BRANCH +from swh.loader.svn.loader import build_swh_snapshot from swh.loader.svn.loader import SWHSvnLoader, SvnLoaderEventful from swh.loader.svn.loader import SvnLoaderHistoryAltered, SvnLoaderUneventful -from test_base import BaseTestSvnLoader + +class TestSWHSnapshot(TestCase): + @istest + def build_swh_snapshot(self): + actual_snap = build_swh_snapshot('revision-id', 'origin-id', visit=10) + + self.assertEquals(actual_snap, { + 'id': None, + 'branches': { + DEFAULT_BRANCH: { + 'target': 'revision-id', + 'target_type': 'revision', + 'origin': 'origin-id', + 'visit': 10 + } + } + }) + # Define loaders with no storage # They'll just accumulate the data in place # Only for testing purposes. class TestSvnLoader: """Mixin class to inhibit the persistence and keep in memory the data sent for storage. cf. SWHSvnLoaderNoStorage """ def __init__(self): super().__init__() self.all_contents = [] self.all_directories = [] self.all_revisions = [] self.all_releases = [] # Check at each svn revision that the hash tree computation # does not diverge self.check_revision = 10 # typed data self.objects = { 'content': self.all_contents, 'directory': self.all_directories, 'revision': self.all_revisions, 'release': self.all_releases, } def _add(self, type, l): """Add without duplicates and keeping the insertion order. Args: type (str): Type of objects concerned by the action l ([object]): List of 'type' object """ col = self.objects[type] for o in l: if o in col: continue col.extend([o]) def maybe_load_contents(self, all_contents): self._add('content', all_contents) def maybe_load_directories(self, all_directories): self._add('directory', all_directories) def maybe_load_revisions(self, all_revisions): self._add('revision', all_revisions) def maybe_load_releases(self, releases): raise ValueError('If called, the test must break.') # Override to do nothing at the end def close_failure(self): pass def close_success(self): pass # Override to only prepare the svn repository def prepare(self, *args, **kwargs): self.svnrepo = self.get_svn_repo(*args) class SWHSvnLoaderNoStorage(TestSvnLoader, SWHSvnLoader): """An SWHSVNLoader with no persistence. Context: Load a new svn repository using the swh policy (so no update). """ def swh_previous_revision(self, prev_swh_revision=None): """We do not know this repository so no revision. """ return None class SWHSvnLoaderUpdateNoStorage(TestSvnLoader, SWHSvnLoader): """An SWHSVNLoader with no persistence. Context: Load a known svn repository using the swh policy. We can either: - do nothing since it does not contain any new commit (so no change) - either check its history is not altered and update in consequence by loading the new revision """ def swh_previous_revision(self, prev_swh_revision=None): """Avoid the storage persistence call and return the expected previous revision for that repository. Check the following for explanation about the hashes: - test_loader.org for (swh policy). - cf. SWHSvnLoaderITTest """ return { 'id': hashutil.hash_to_bytes( '4876cb10aec6f708f7466dddf547567b65f6c39c'), 'parents': [hashutil.hash_to_bytes( 'a3a577948fdbda9d1061913b77a1588695eadb41')], 'directory': hashutil.hash_to_bytes( '0deab3023ac59398ae467fc4bff5583008af1ee2'), 'target_type': 'revision', 'metadata': { 'extra_headers': [ ['svn_repo_uuid', '3187e211-bb14-4c82-9596-0b59d67cd7f4'], ['svn_revision', '6'] ] } } class SWHSvnLoaderUpdateHistoryAlteredNoStorage(TestSvnLoader, SWHSvnLoader): """An SWHSVNLoader with no persistence. Context: Load a known svn repository using the swh policy with its history altered so we do not update it. """ def swh_previous_revision(self, prev_swh_revision=None): """Avoid the storage persistence call and return the expected previous revision for that repository. Check the following for explanation about the hashes: - test_loader.org for (swh policy). - cf. SWHSvnLoaderITTest """ return { # Changed the revision id's hash to simulate history altered 'id': hashutil.hash_to_bytes( 'badbadbadbadf708f7466dddf547567b65f6c39d'), 'parents': [hashutil.hash_to_bytes( 'a3a577948fdbda9d1061913b77a1588695eadb41')], 'directory': hashutil.hash_to_bytes( '0deab3023ac59398ae467fc4bff5583008af1ee2'), 'target_type': 'revision', 'metadata': { 'extra_headers': [ ['svn_repo_uuid', '3187e211-bb14-4c82-9596-0b59d67cd7f4'], ['svn_revision', b'6'] ] } } class SWHSvnLoaderNewRepositoryITTest(BaseTestSvnLoader): def setUp(self): super().setUp() self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 2, } self.loader = SWHSvnLoaderNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process a new repository with swh policy should be ok. """ # when self.loader.process_repository(self.origin_visit) # then self.assertEquals(len(self.loader.all_revisions), 6) self.assertEquals(len(self.loader.all_releases), 0) last_revision = '4876cb10aec6f708f7466dddf547567b65f6c39c' # cf. test_loader.org for explaining from where those hashes # come from expected_revisions = { # revision hash | directory hash '0d7dd5f751cef8fe17e8024f7d6b0e3aac2cfd71': '669a71cce6c424a81ba42b7dc5d560d32252f0ca', # noqa '95edacc8848369d6fb1608e887d6d2474fd5224f': '008ac97a1118560797c50e3392fa1443acdaa349', # noqa 'fef26ea45a520071711ba2b9d16a2985ee837021': '3780effbe846a26751a95a8c95c511fb72be15b4', # noqa '3f51abf3b3d466571be0855dfa67e094f9ceff1b': 'ffcca9b09c5827a6b8137322d4339c8055c3ee1e', # noqa 'a3a577948fdbda9d1061913b77a1588695eadb41': '7dc52cc04c3b8bd7c085900d60c159f7b846f866', # noqa last_revision: '0deab3023ac59398ae467fc4bff5583008af1ee2', # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderUpdateWithNoChangeITTest(BaseTestSvnLoader): def setUp(self): super().setUp() self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 3, } self.loader = SWHSvnLoaderUpdateNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process a known repository with swh policy and no new data should be ok. """ # when with self.assertRaises(SvnLoaderUneventful): self.loader.args = (self.origin_visit,) self.loader.process_repository(self.origin_visit) # then self.assertEquals(len(self.loader.all_revisions), 0) self.assertEquals(len(self.loader.all_releases), 0) class SWHSvnLoaderUpdateWithHistoryAlteredITTest(BaseTestSvnLoader): def setUp(self): # the svn repository pkg-gourmet has been updated with changes super().setUp(archive_name='pkg-gourmet-with-updates.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 4, } self.loader = SWHSvnLoaderUpdateHistoryAlteredNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process a known repository with swh policy and history altered should stop and do nothing. """ # when with self.assertRaises(SvnLoaderHistoryAltered): self.loader.args = (self.origin_visit,) self.loader.process_repository(self.origin_visit) # then # we got the previous run's last revision (rev 6) # so 2 news + 1 old self.assertEquals(len(self.loader.all_revisions), 0) self.assertEquals(len(self.loader.all_releases), 0) class SWHSvnLoaderUpdateWithChangesITTest(BaseTestSvnLoader): def setUp(self): # the svn repository pkg-gourmet has been updated with changes super().setUp(archive_name='pkg-gourmet-with-updates.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 5, } self.loader = SWHSvnLoaderUpdateNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process updated repository should yield new revisions """ # when self.loader.process_repository(self.origin_visit) # then # we got the previous run's last revision (rev 6) # so 2 new self.assertEquals(len(self.loader.all_revisions), 5) self.assertEquals(len(self.loader.all_releases), 0) last_revision = '171dc35522bfd17dda4e90a542a0377fb2fc707a' # cf. test_loader.org for explaining from where those hashes # come from expected_revisions = { # revision hash | directory hash '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa last_revision: 'fd24a76c87a3207428e06612b49860fc78e9f6dc' # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderUpdateWithChangesStartFromScratchITTest(BaseTestSvnLoader): def setUp(self): # the svn repository pkg-gourmet has been updated with changes super().setUp(archive_name='pkg-gourmet-with-updates.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 5, } self.loader = SWHSvnLoaderUpdateNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process known repository from scratch should yield revisions again """ # when self.loader.process_repository(self.origin_visit, start_from_scratch=True) # then # we got the previous run's last revision (rev 6) # but we do not inspect that as we start from from scratch so # we should have all revisions so 11 self.assertEquals(len(self.loader.all_revisions), 11) self.assertEquals(len(self.loader.all_releases), 0) expected_revisions = { '0d7dd5f751cef8fe17e8024f7d6b0e3aac2cfd71': '669a71cce6c424a81ba42b7dc5d560d32252f0ca', # noqa '95edacc8848369d6fb1608e887d6d2474fd5224f': '008ac97a1118560797c50e3392fa1443acdaa349', # noqa 'fef26ea45a520071711ba2b9d16a2985ee837021': '3780effbe846a26751a95a8c95c511fb72be15b4', # noqa '3f51abf3b3d466571be0855dfa67e094f9ceff1b': 'ffcca9b09c5827a6b8137322d4339c8055c3ee1e', # noqa 'a3a577948fdbda9d1061913b77a1588695eadb41': '7dc52cc04c3b8bd7c085900d60c159f7b846f866', # noqa '4876cb10aec6f708f7466dddf547567b65f6c39c': '0deab3023ac59398ae467fc4bff5583008af1ee2', # noqa '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa '171dc35522bfd17dda4e90a542a0377fb2fc707a': 'fd24a76c87a3207428e06612b49860fc78e9f6dc', # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderUpdateWithUnfinishedLoadingChangesITTest(BaseTestSvnLoader): def setUp(self): super().setUp(archive_name='pkg-gourmet-with-updates.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 6 } self.loader = SWHSvnLoaderNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process partially visited repository should finish loading """ previous_unfinished_revision = { 'id': hashutil.hash_to_bytes( '4876cb10aec6f708f7466dddf547567b65f6c39c'), 'parents': [hashutil.hash_to_bytes( 'a3a577948fdbda9d1061913b77a1588695eadb41')], 'directory': hashutil.hash_to_bytes( '0deab3023ac59398ae467fc4bff5583008af1ee2'), 'target_type': 'revision', 'metadata': { 'extra_headers': [ ['svn_repo_uuid', '3187e211-bb14-4c82-9596-0b59d67cd7f4'], ['svn_revision', '6'] ] } } # when self.loader.process_repository( self.origin_visit, last_known_swh_revision=previous_unfinished_revision) # then # we got the previous run's last revision (rev 6) # so 2 new self.assertEquals(len(self.loader.all_revisions), 5) self.assertEquals(len(self.loader.all_releases), 0) last_revision = '171dc35522bfd17dda4e90a542a0377fb2fc707a' # cf. test_loader.org for explaining from where those hashes # come from expected_revisions = { # revision hash | directory hash '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa last_revision: 'fd24a76c87a3207428e06612b49860fc78e9f6dc' # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderUpdateWithUnfinishedLoadingChangesButVisitDoneITTest( BaseTestSvnLoader): def setUp(self): super().setUp(archive_name='pkg-gourmet-with-updates.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 9, } self.loader = SWHSvnLoaderUpdateNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process known and partial repository should start from last visit """ previous_unfinished_revision = { 'id': hashutil.hash_to_bytes( 'a3a577948fdbda9d1061913b77a1588695eadb41'), 'parents': [hashutil.hash_to_bytes( '3f51abf3b3d466571be0855dfa67e094f9ceff1b')], 'directory': hashutil.hash_to_bytes( '7dc52cc04c3b8bd7c085900d60c159f7b846f866'), 'target_type': 'revision', 'metadata': { 'extra_headers': [ ['svn_repo_uuid', '3187e211-bb14-4c82-9596-0b59d67cd7f4'], ['svn_revision', '5'] ] } } # when self.loader.process_repository( self.origin_visit, last_known_swh_revision=previous_unfinished_revision) # then # we got the previous run's last revision (rev 6) # so 2 new self.assertEquals(len(self.loader.all_revisions), 5) self.assertEquals(len(self.loader.all_releases), 0) last_revision = '171dc35522bfd17dda4e90a542a0377fb2fc707a' # cf. test_loader.org for explaining from where those hashes # come from expected_revisions = { # revision hash | directory hash '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa last_revision: 'fd24a76c87a3207428e06612b49860fc78e9f6dc' # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderUpdateLessRecentNoStorage(TestSvnLoader, SWHSvnLoader): """An SWHSVNLoader with no persistence. Context: Load a known svn repository using the swh policy. The last visit seen is less recent than a previous unfinished crawl. """ def swh_previous_revision(self, prev_swh_revision=None): """Avoid the storage persistence call and return the expected previous revision for that repository. Check the following for explanation about the hashes: - test_loader.org for (swh policy). - cf. SWHSvnLoaderITTest """ return { 'id': hashutil.hash_to_bytes( 'a3a577948fdbda9d1061913b77a1588695eadb41'), 'parents': [hashutil.hash_to_bytes( '3f51abf3b3d466571be0855dfa67e094f9ceff1b')], 'directory': hashutil.hash_to_bytes( '7dc52cc04c3b8bd7c085900d60c159f7b846f866'), 'target_type': 'revision', 'metadata': { 'extra_headers': [ ['svn_repo_uuid', '3187e211-bb14-4c82-9596-0b59d67cd7f4'], ['svn_revision', '5'] ] } } class SWHSvnLoaderUnfinishedLoadingChangesSinceLastVisitITTest( BaseTestSvnLoader): def setUp(self): super().setUp(archive_name='pkg-gourmet-with-updates.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 1, } self.loader = SWHSvnLoaderUpdateLessRecentNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Process updated repository should yield revisions from last visit """ previous_unfinished_revision = { 'id': hashutil.hash_to_bytes( '4876cb10aec6f708f7466dddf547567b65f6c39c'), 'parents': [hashutil.hash_to_bytes( 'a3a577948fdbda9d1061913b77a1588695eadb41')], 'directory': hashutil.hash_to_bytes( '0deab3023ac59398ae467fc4bff5583008af1ee2'), 'target_type': 'revision', 'metadata': { 'extra_headers': [ ['svn_repo_uuid', '3187e211-bb14-4c82-9596-0b59d67cd7f4'], ['svn_revision', '6'] ] } } # when self.loader.process_repository( self.origin_visit, last_known_swh_revision=previous_unfinished_revision) # then # we got the previous run's last revision (rev 6) # so 2 new self.assertEquals(len(self.loader.all_revisions), 5) self.assertEquals(len(self.loader.all_releases), 0) last_revision = '171dc35522bfd17dda4e90a542a0377fb2fc707a' # cf. test_loader.org for explaining from where those hashes # come from expected_revisions = { # revision hash | directory hash '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa last_revision: 'fd24a76c87a3207428e06612b49860fc78e9f6dc' # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderUpdateAndTestCornerCasesAboutEolITTest(BaseTestSvnLoader): def setUp(self): super().setUp(archive_name='pkg-gourmet-with-eol-corner-cases.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 1, } self.loader = SWHSvnLoaderUpdateLessRecentNoStorage() self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """EOL corner cases and update. """ previous_unfinished_revision = { 'id': hashutil.hash_to_bytes( '171dc35522bfd17dda4e90a542a0377fb2fc707a'), 'parents': [hashutil.hash_to_bytes( '902f29b4323a9b9de3af6d28e72dd581e76d9397')], 'directory': hashutil.hash_to_bytes( 'fd24a76c87a3207428e06612b49860fc78e9f6dc'), 'target_type': 'revision', 'metadata': { 'extra_headers': [ ['svn_repo_uuid', '3187e211-bb14-4c82-9596-0b59d67cd7f4'], ['svn_revision', '11'] ] } } # when self.loader.process_repository( self.origin_visit, last_known_swh_revision=previous_unfinished_revision) # then # we got the previous run's last revision (rev 11) # so 8 new self.assertEquals(len(self.loader.all_revisions), 8) self.assertEquals(len(self.loader.all_releases), 0) last_revision = '0148ae3eaa520b73a50802c59f3f416b7a36cf8c' # cf. test_loader.org for explaining from where those hashes # come from expected_revisions = { # revision hash | directory hash '027e8769f4786597436ab94a91f85527d04a6cbb': '2d9ca72c6afec6284fb01e459588cbb007017c8c', # noqa '4474d96018877742d9697d5c76666c9693353bfc': 'ab111577e0ab39e4a157c476072af48f2641d93f', # noqa '97ad21eab92961e2a22ca0285f09c6d1e9a7ffbc': 'ab111577e0ab39e4a157c476072af48f2641d93f', # noqa 'd04ea8afcee6205cc8384c091bfc578931c169fd': 'b0a648b02e55a4dce356ac35187a058f89694ec7', # noqa 'ded78810401fd354ffe894aa4a1e5c7d30a645d1': 'b0a648b02e55a4dce356ac35187a058f89694ec7', # noqa '4ee95e39358712f53c4fc720da3fafee9249ed19': 'c3c98df624733fef4e592bef983f93e2ed02b179', # noqa 'ffa901b69ca0f46a2261f42948838d19709cb9f8': 'c3c98df624733fef4e592bef983f93e2ed02b179', # noqa last_revision: '844d4646d6c2b4f3a3b2b22ab0ee38c7df07bab2', # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderExternalIdCornerCaseITTest(BaseTestSvnLoader): def setUp(self): super().setUp(archive_name='pkg-gourmet-with-external-id.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 1, } self.loader = SWHSvnLoaderNoStorage() # override revision-block size self.loader.config['revision_packet_size'] = 3 self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Repository with svn:externals property, will stop raising an error """ previous_unfinished_revision = None # when with self.assertRaises(SvnLoaderEventful) as exc: self.loader.process_repository( self.origin_visit, last_known_swh_revision=previous_unfinished_revision) actual_raised_revision = exc.exception.swh_revision # then repositories holds 21 revisions, but the last commit # one holds an 'svn:externals' property which will make the # loader-svn stops. This will then stop at the 6th iterations # of 3-revision block size, so only 18 revisions will be # flushed self.assertEquals(len(self.loader.all_revisions), 18) self.assertEquals(len(self.loader.all_releases), 0) last_revision = 'ffa901b69ca0f46a2261f42948838d19709cb9f8' expected_revisions = { # revision hash | directory hash '0d7dd5f751cef8fe17e8024f7d6b0e3aac2cfd71': '669a71cce6c424a81ba42b7dc5d560d32252f0ca', # noqa '95edacc8848369d6fb1608e887d6d2474fd5224f': '008ac97a1118560797c50e3392fa1443acdaa349', # noqa 'fef26ea45a520071711ba2b9d16a2985ee837021': '3780effbe846a26751a95a8c95c511fb72be15b4', # noqa '3f51abf3b3d466571be0855dfa67e094f9ceff1b': 'ffcca9b09c5827a6b8137322d4339c8055c3ee1e', # noqa 'a3a577948fdbda9d1061913b77a1588695eadb41': '7dc52cc04c3b8bd7c085900d60c159f7b846f866', # noqa '4876cb10aec6f708f7466dddf547567b65f6c39c': '0deab3023ac59398ae467fc4bff5583008af1ee2', # noqa '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa '171dc35522bfd17dda4e90a542a0377fb2fc707a': 'fd24a76c87a3207428e06612b49860fc78e9f6dc', # noqa '027e8769f4786597436ab94a91f85527d04a6cbb': '2d9ca72c6afec6284fb01e459588cbb007017c8c', # noqa '4474d96018877742d9697d5c76666c9693353bfc': 'ab111577e0ab39e4a157c476072af48f2641d93f', # noqa '97ad21eab92961e2a22ca0285f09c6d1e9a7ffbc': 'ab111577e0ab39e4a157c476072af48f2641d93f', # noqa 'd04ea8afcee6205cc8384c091bfc578931c169fd': 'b0a648b02e55a4dce356ac35187a058f89694ec7', # noqa 'ded78810401fd354ffe894aa4a1e5c7d30a645d1': 'b0a648b02e55a4dce356ac35187a058f89694ec7', # noqa '4ee95e39358712f53c4fc720da3fafee9249ed19': 'c3c98df624733fef4e592bef983f93e2ed02b179', # noqa last_revision : 'c3c98df624733fef4e592bef983f93e2ed02b179', # noqa } # The last revision being the one used later to start back from self.assertEquals(hashutil.hash_to_hex(actual_raised_revision['id']), last_revision) self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderLinkFileAndFolderWithSameNameITTest(BaseTestSvnLoader): def setUp(self): # edge cases: # - first create a file and commit it. # Remove it, then add folder holding the same name, commit. # - do the same scenario with symbolic link (instead of file) super().setUp( archive_name='pkg-gourmet-with-edge-case-links-and-files.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 1, } self.loader = SWHSvnLoaderNoStorage() # override revision-block size self.loader.config['revision_packet_size'] = 3 self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """File/Link destroyed prior to folder with same name creation should be ok """ previous_unfinished_revision = None # when self.loader.process_repository( self.origin_visit, last_known_swh_revision=previous_unfinished_revision) # then repositories holds 14 revisions, but the last commit self.assertEquals(len(self.loader.all_revisions), 19) self.assertEquals(len(self.loader.all_releases), 0) last_revision = '3f43af2578fccf18b0d4198e48563da7929dc608' expected_revisions = { # revision hash | directory hash '0d7dd5f751cef8fe17e8024f7d6b0e3aac2cfd71': '669a71cce6c424a81ba42b7dc5d560d32252f0ca', # noqa '95edacc8848369d6fb1608e887d6d2474fd5224f': '008ac97a1118560797c50e3392fa1443acdaa349', # noqa 'fef26ea45a520071711ba2b9d16a2985ee837021': '3780effbe846a26751a95a8c95c511fb72be15b4', # noqa '3f51abf3b3d466571be0855dfa67e094f9ceff1b': 'ffcca9b09c5827a6b8137322d4339c8055c3ee1e', # noqa 'a3a577948fdbda9d1061913b77a1588695eadb41': '7dc52cc04c3b8bd7c085900d60c159f7b846f866', # noqa '4876cb10aec6f708f7466dddf547567b65f6c39c': '0deab3023ac59398ae467fc4bff5583008af1ee2', # noqa '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa '171dc35522bfd17dda4e90a542a0377fb2fc707a': 'fd24a76c87a3207428e06612b49860fc78e9f6dc', # noqa '9231f9a98a9051a0cd34231cddd4e11773f8348e': '6c07f4f4ac780eaf99a247fbfd0897533598dd36', # noqa 'c309bd3b57796696d6655ab3ab0b438fdd2d8201': 'fd24a76c87a3207428e06612b49860fc78e9f6dc', # noqa 'bb78300cc1ac9119eb6fffa9e9fa04a7f9340b11': 'ee995a0d85f6917c75bcee3aa448bea7726b265d', # noqa 'f2e01111329f84580dc3febb1fd45515692c5886': 'e2baec7b6a5543758e9c73695bc847db0a4f7941', # noqa '1a0f70c34e211f073e1be3435ecf6f0dd7700267': 'e7536e721fa806c19971b749c091c144b2f2b88e', # noqa '0c612a23d293cc3100496a54ae4ad13d750efe4c': '2123d12749294bbfb54e73f9d73fac658aabb266', # noqa '69a53d972e2f863acbbbda546d9da96287af6a88': '13896cb96ec004140ce5be8852fee8c29830d9c7', # noqa last_revision: '6b1e0243768ff9ac060064b2eeade77e764ffc82', # noqa } self.assertRevisionsOk(expected_revisions) class SWHSvnLoaderWrongLinkCasesITTest(BaseTestSvnLoader): def setUp(self): # edge cases: # - wrong symbolic link # - wrong symbolic link with empty space names super().setUp( archive_name='pkg-gourmet-with-wrong-link-cases.tgz') self.origin = {'id': 2, 'type': 'svn', 'url': 'file:///dev/null'} self.origin_visit = { 'origin': self.origin['id'], 'visit': 1, } self.loader = SWHSvnLoaderNoStorage() # override revision-block size self.loader.config['revision_packet_size'] = 3 self.loader.prepare( self.svn_mirror_url, self.destination_path, self.origin) @istest def process_repository(self): """Wrong link or empty space-named link should be ok """ previous_unfinished_revision = None # when self.loader.process_repository( self.origin_visit, last_known_swh_revision=previous_unfinished_revision) # then repositories holds 14 revisions, but the last commit self.assertEquals(len(self.loader.all_revisions), 21) self.assertEquals(len(self.loader.all_releases), 0) last_revision = 'cf30d3bb9d5967d0a2bbeacc405f10a5dd9b138a' expected_revisions = { # revision hash | directory hash '0d7dd5f751cef8fe17e8024f7d6b0e3aac2cfd71': '669a71cce6c424a81ba42b7dc5d560d32252f0ca', # noqa '95edacc8848369d6fb1608e887d6d2474fd5224f': '008ac97a1118560797c50e3392fa1443acdaa349', # noqa 'fef26ea45a520071711ba2b9d16a2985ee837021': '3780effbe846a26751a95a8c95c511fb72be15b4', # noqa '3f51abf3b3d466571be0855dfa67e094f9ceff1b': 'ffcca9b09c5827a6b8137322d4339c8055c3ee1e', # noqa 'a3a577948fdbda9d1061913b77a1588695eadb41': '7dc52cc04c3b8bd7c085900d60c159f7b846f866', # noqa '4876cb10aec6f708f7466dddf547567b65f6c39c': '0deab3023ac59398ae467fc4bff5583008af1ee2', # noqa '7f5bc909c29d4e93d8ccfdda516e51ed44930ee1': '752c52134dcbf2fff13c7be1ce4e9e5dbf428a59', # noqa '38d81702cb28db4f1a6821e64321e5825d1f7fd6': '39c813fb4717a4864bacefbd90b51a3241ae4140', # noqa '99c27ebbd43feca179ac0e895af131d8314cafe1': '3397ca7f709639cbd36b18a0d1b70bce80018c45', # noqa '902f29b4323a9b9de3af6d28e72dd581e76d9397': 'c4e12483f0a13e6851459295a4ae735eb4e4b5c4', # noqa '171dc35522bfd17dda4e90a542a0377fb2fc707a': 'fd24a76c87a3207428e06612b49860fc78e9f6dc', # noqa '9231f9a98a9051a0cd34231cddd4e11773f8348e': '6c07f4f4ac780eaf99a247fbfd0897533598dd36', # noqa 'c309bd3b57796696d6655ab3ab0b438fdd2d8201': 'fd24a76c87a3207428e06612b49860fc78e9f6dc', # noqa 'bb78300cc1ac9119eb6fffa9e9fa04a7f9340b11': 'ee995a0d85f6917c75bcee3aa448bea7726b265d', # noqa 'f2e01111329f84580dc3febb1fd45515692c5886': 'e2baec7b6a5543758e9c73695bc847db0a4f7941', # noqa '1a0f70c34e211f073e1be3435ecf6f0dd7700267': 'e7536e721fa806c19971b749c091c144b2f2b88e', # noqa '0c612a23d293cc3100496a54ae4ad13d750efe4c': '2123d12749294bbfb54e73f9d73fac658aabb266', # noqa '69a53d972e2f863acbbbda546d9da96287af6a88': '13896cb96ec004140ce5be8852fee8c29830d9c7', # noqa '3f43af2578fccf18b0d4198e48563da7929dc608': '6b1e0243768ff9ac060064b2eeade77e764ffc82', # noqa '4ab5fc264732cd474d2e695c5ac66e4933bdad74': '9a1f5e3961db89422250ce6c1441476f40d65205', # noqa last_revision: 'd853d9628f6f0008d324fed27dadad00ce48bc62', # noqa } self.assertRevisionsOk(expected_revisions)