diff --git a/requirements-swh.txt b/requirements-swh.txt index 40c3624..fe77c47 100644 --- a/requirements-swh.txt +++ b/requirements-swh.txt @@ -1,4 +1,4 @@ swh.storage >= 0.0.143 swh.model >= 0.0.18 swh.scheduler >= 0.0.39 -swh.loader.core >= 0.0.37 +swh.loader.core >= 0.0.67 diff --git a/swh/loader/svn/loader.py b/swh/loader/svn/loader.py index e7c58ca..8e5c9f7 100644 --- a/swh/loader/svn/loader.py +++ b/swh/loader/svn/loader.py @@ -1,765 +1,765 @@ # Copyright (C) 2015-2019 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 os import pty import re import shutil import tempfile from mmap import mmap, ACCESS_WRITE from subprocess import Popen 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 BufferedLoader from swh.loader.core.utils import clean_dangling_folders from swh.storage.algos.snapshot import snapshot_get_all_branches from . import svn, converters from .utils import ( init_svn_repo_from_dump, init_svn_repo_from_archive_dump, OutputStream ) from .exception import SvnLoaderUneventful from .exception import SvnLoaderHistoryAltered DEFAULT_BRANCH = b'HEAD' def _revision_id(revision): return identifier_to_bytes(revision_identifier(revision)) def build_swh_snapshot(revision_id, branch=DEFAULT_BRANCH): """Build a swh snapshot from the revision id, origin url, and visit. """ return { 'id': None, 'branches': { branch: { 'target': revision_id, 'target_type': 'revision', } } } TEMPORARY_DIR_PREFIX_PATTERN = 'swh.loader.svn.' class SvnLoader(BufferedLoader): """Swh svn loader. The repository is either remote or local. The loader deals with update on an already previously loaded repository. """ CONFIG_BASE_FILENAME = 'loader/svn' ADDITIONAL_CONFIG = { 'temp_directory': ('str', '/tmp'), 'debug': ('bool', False), # NOT FOR PRODUCTION, False for production 'check_revision': ('dict', { 'status': False, # do we check the revision? 'limit': 1000, # at which pace do we check it? }), } visit_type = 'svn' def __init__(self, url, origin_url=None, visit_date=None, destination_path=None, swh_revision=None, start_from_scratch=False): super().__init__(logging_class='swh.loader.svn.SvnLoader') # technical svn uri to act on svn repository self.svn_url = url # origin url as unique identifier for origin in swh archive self.origin_url = origin_url if origin_url else self.svn_url self.debug = self.config['debug'] self.last_seen_revision = None self.temp_directory = self.config['temp_directory'] self.done = False self.svnrepo = None # Revision check is configurable check_revision = self.config['check_revision'] if check_revision['status']: self.check_revision = check_revision['limit'] else: self.check_revision = None # internal state used to store swh objects self._contents = [] self._directories = [] self._revisions = [] self._snapshot = None self._last_revision = None self._visit_status = 'full' self._load_status = 'uneventful' self.visit_date = visit_date self.destination_path = destination_path self.start_from_scratch = start_from_scratch self.swh_revision = swh_revision def pre_cleanup(self): """Cleanup potential dangling files from prior runs (e.g. OOM killed tasks) """ clean_dangling_folders(self.temp_directory, pattern_check=TEMPORARY_DIR_PREFIX_PATTERN, log=self.log) def cleanup(self): """Clean up the svn repository's working representation on disk. """ if not self.svnrepo: # could happen if `prepare` fails return if self.debug: self.log.error('''NOT FOR PRODUCTION - debug flag activated Local repository not cleaned up for investigation: %s''' % ( self.svnrepo.local_url.decode('utf-8'), )) return self.svnrepo.clean_fs() 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. """ 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, local_dirname, origin_url): """Instantiates the needed svnrepo collaborator to permit reading svn repository. Args: svn_url (str): the svn repository url to read from local_dirname (str): the local path on disk to compute data origin_url (str): the corresponding origin url Returns: Instance of :mod:`swh.loader.svn.svn` clients """ return svn.SvnRepo(svn_url, local_dirname=local_dirname, origin_url=origin_url) def swh_latest_snapshot_revision(self, origin_url, previous_swh_revision=None): """Look for latest snapshot revision and returns it if any. Args: origin_url (str): Origin identifier previous_swh_revision: (optional) id of a possible previous swh revision Returns: dict: The latest known point in time. Dict with keys: 'revision': latest visited revision 'snapshot': latest snapshot If None is found, return an empty dict. """ storage = self.storage if not previous_swh_revision: # check latest snapshot's revision visit = storage.origin_visit_get_latest( origin_url, require_snapshot=True) if visit: latest_snap = snapshot_get_all_branches( storage, visit['snapshot']) if latest_snap: branches = latest_snap.get('branches') if not branches: return {} branch = branches.get(DEFAULT_BRANCH) if not branch: return {} target_type = branch['target_type'] if target_type != 'revision': return {} previous_swh_revision = branch['target'] else: return {} else: return {} if isinstance(previous_swh_revision, dict): swh_id = previous_swh_revision['id'] else: swh_id = previous_swh_revision revs = list(storage.revision_get([swh_id])) if revs: return { 'snapshot': latest_snap, 'revision': revs[0] } return {} 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 (dict): the svn revision commit (dict): the commit metadata dir_id (bytes): the upper tree's hash identifier parents ([bytes]): 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 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 _init_from(self, partial_swh_revision, previous_swh_revision): """Function to determine from where to start from. Args: partial_swh_revision (dict): A known revision from which the previous loading did not finish. known_previous_revision (dict): 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 start_from(self, last_known_swh_revision=None, start_from_scratch=False): """Determine from where to start the loading. Args: last_known_swh_revision (dict): Last know swh revision or None start_from_scratch (bool): To start loading from scratch or not Returns: tuple (revision_start, revision_end, revision_parents) Raises: SvnLoaderHistoryAltered: When a hash divergence has been detected (should not happen) SvnLoaderUneventful: Nothing changed since last visit """ revision_head = self.svnrepo.head_revision() if revision_head == 0: # empty repository case revision_start = 0 revision_end = 0 else: # default configuration revision_start = self.svnrepo.initial_revision() revision_end = revision_head revision_parents = { revision_start: [] } if not start_from_scratch: # Check if we already know a previous revision for that origin if self.latest_snapshot: swh_rev = self.latest_snapshot['revision'] else: swh_rev = None # 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' % ( self.svnrepo.remote_url, revision_start)) if swh_rev and not self.check_history_not_altered( self.svnrepo, revision_start, swh_rev): msg = 'History of svn %s@%s altered. ' \ 'Skipping...' % ( self.svnrepo.remote_url, revision_start) raise SvnLoaderHistoryAltered(msg) # 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']] if revision_start > revision_end and revision_start != 1: msg = '%s@%s already injected.' % (self.svnrepo.remote_url, revision_end) raise SvnLoaderUneventful(msg) self.log.info('Processing revisions [%s-%s] for %s' % ( revision_start, revision_end, self.svnrepo)) return revision_start, revision_end, revision_parents def _check_revision_divergence(self, count, rev, dir_id): """Check for hash revision computation divergence. The Rationale behind this is that svn can trigger unknown edge cases (mixed CRLF, svn properties, etc...). Those are not always easy to spot. Adding a check will help in spotting missing edge cases. Args: count (int): The number of revisions done so far rev (dict): The actual revision we are computing from dir_id (bytes): The actual directory for the given revision Returns: False if no hash divergence detected Raises ValueError if a hash divergence is detected """ if (count % self.check_revision) == 0: # hash computation check self.log.debug('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!' % ( hashutil.hash_to_hex(dir_id), hashutil.hash_to_hex(checked_dir_id)) raise ValueError(err) def process_svn_revisions(self, svnrepo, revision_start, revision_end, revision_parents): """Process svn revisions from revision_start to revision_end. At each svn revision, apply new diffs and simultaneously compute swh hashes. This yields those computed swh hashes as a tuple (contents, directories, revision). Note that at every `self.check_revision`, a supplementary check takes place to check for hash-tree divergence (related T570). Yields: tuple (contents, directories, revision) of dict as a dictionary with keys, sha1_git, sha1, etc... Raises: ValueError in case of a hash divergence detection """ 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 _contents = new_objects.get('content', {}).values() _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 self.check_revision: self._check_revision_divergence(count, rev, dir_id) if nextrev: revision_parents[nextrev] = [swh_revision['id']] yield _contents, _directories, swh_revision def prepare_origin_visit(self, *args, **kwargs): self.origin = { 'url': self.origin_url if self.origin_url else self.svn_url, } def prepare(self, *args, **kwargs): if self.swh_revision: self.last_known_swh_revision = self.swh_revision else: self.last_known_swh_revision = None self.latest_snapshot = self.swh_latest_snapshot_revision( self.origin_url, self.last_known_swh_revision) if self.destination_path: local_dirname = self.destination_path else: local_dirname = tempfile.mkdtemp( suffix='-%s' % os.getpid(), prefix=TEMPORARY_DIR_PREFIX_PATTERN, dir=self.temp_directory) self.svnrepo = self.get_svn_repo( self.svn_url, local_dirname, self.origin_url) try: revision_start, revision_end, revision_parents = self.start_from( self.last_known_swh_revision, self.start_from_scratch) self.swh_revision_gen = self.process_svn_revisions( self.svnrepo, revision_start, revision_end, revision_parents) except SvnLoaderUneventful as e: self.log.warning(e) if self.latest_snapshot and 'snapshot' in self.latest_snapshot: self._snapshot = self.latest_snapshot['snapshot'] self.done = True except SvnLoaderHistoryAltered as e: self.log.error(e) self.done = True self._visit_status = 'partial' def fetch_data(self): """Fetching svn revision information. This will apply svn revision as patch on disk, and at the same time, compute the swh hashes. In effect, fetch_data fetches those data and compute the necessary swh objects. It's then stored in the internal state instance variables (initialized in `_prepare_state`). This is up to `store_data` to actually discuss with the storage to store those objects. Returns: bool: True to continue fetching data (next svn revision), False to stop. """ data = None if self.done: return False try: data = next(self.swh_revision_gen) self._load_status = 'eventful' except StopIteration: self.done = True self._visit_status = 'full' return False # Stopping iteration except Exception as e: # svn:external, hash divergence, i/o error... self.log.exception(e) self.done = True self._visit_status = 'partial' return False # Stopping iteration self._contents, self._directories, revision = data if revision: self._last_revision = revision self._revisions.append(revision) return True # next svn revision def store_data(self): """We store the data accumulated in internal instance variable. If the iteration over the svn revisions is done, we create the snapshot and flush to storage the data. This also resets the internal instance variable state. """ - self.maybe_load_contents(self._contents) - self.maybe_load_directories(self._directories) - self.maybe_load_revisions(self._revisions) + self.send_contents(self._contents) + self.send_directories(self._directories) + self.send_revisions(self._revisions) if self.done: # finish line, snapshot! self.generate_and_load_snapshot(revision=self._last_revision, snapshot=self._snapshot) self.flush() self._contents = [] self._directories = [] self._revisions = [] def generate_and_load_snapshot(self, revision=None, snapshot=None): """Create the snapshot either from existing revision or snapshot. Revision (supposedly new) has priority over the snapshot (supposedly existing one). Args: revision (dict): Last revision seen if any (None by default) snapshot (dict): Snapshot to use if any (None by default) """ if revision: # Priority to the revision snap = build_swh_snapshot(revision['id']) snap['id'] = identifier_to_bytes(snapshot_identifier(snap)) elif snapshot: # Fallback to prior snapshot snap = snapshot else: return None self.log.debug('snapshot: %s' % snap) - self.maybe_load_snapshot(snap) + self.send_snapshot(snap) def load_status(self): return { 'status': self._load_status, } def visit_status(self): return self._visit_status class SvnLoaderFromDumpArchive(SvnLoader): """Uncompress an archive containing an svn dump, mount the svn dump as an svn repository and load said repository. """ def __init__(self, url, archive_path, origin_url=None, destination_path=None, swh_revision=None, start_from_scratch=None, visit_date=None): super().__init__(url, origin_url=origin_url, destination_path=destination_path, swh_revision=swh_revision, start_from_scratch=start_from_scratch, visit_date=visit_date) self.archive_path = archive_path self.temp_dir = None self.repo_path = None def prepare(self, *args, **kwargs): self.log.info('Archive to mount and load %s' % self.archive_path) self.temp_dir, self.repo_path = init_svn_repo_from_archive_dump( self.archive_path, prefix=TEMPORARY_DIR_PREFIX_PATTERN, suffix='-%s' % os.getpid(), root_dir=self.temp_directory) super().prepare(*args, **kwargs) def cleanup(self): super().cleanup() if self.temp_dir and os.path.exists(self.temp_dir): msg = 'Clean up temporary directory dump %s for project %s' % ( self.temp_dir, os.path.basename(self.repo_path)) self.log.debug(msg) shutil.rmtree(self.temp_dir) class SvnLoaderFromRemoteDump(SvnLoader): """ Create a subversion repository dump using the svnrdump utility, mount it locally and load the repository from it. """ def __init__(self, url, origin_url=None, destination_path=None, swh_revision=None, start_from_scratch=False, visit_date=None): super().__init__(url, origin_url=origin_url, destination_path=destination_path, swh_revision=swh_revision, start_from_scratch=start_from_scratch, visit_date=visit_date) self.temp_dir = tempfile.mkdtemp(dir=self.temp_directory) self.repo_path = None self.truncated_dump = False def get_last_loaded_svn_rev(self, svn_url): """ Check if the svn repository has already been visited and return the last loaded svn revision number or -1 otherwise. """ last_loaded_svn_rev = -1 try: origin = \ self.storage.origin_get({'url': svn_url}) last_swh_rev = \ self.swh_latest_snapshot_revision(origin['url'])['revision'] last_swh_rev_headers = \ dict(last_swh_rev['metadata']['extra_headers']) last_loaded_svn_rev = int(last_swh_rev_headers['svn_revision']) except Exception: pass return last_loaded_svn_rev def dump_svn_revisions(self, svn_url, last_loaded_svn_rev=-1): """ Generate a subversion dump file using the svnrdump tool. If the svnrdump command failed somehow, the produced dump file is analyzed to determine if a partial loading is still feasible. """ # Build the svnrdump command line svnrdump_cmd = ['svnrdump', 'dump', svn_url] # Launch the svnrdump command while capturing stderr as # successfully dumped revision numbers are printed to it dump_temp_dir = tempfile.mkdtemp(dir=self.temp_dir) dump_name = ''.join(c for c in svn_url if c.isalnum()) dump_path = '%s/%s.svndump' % (dump_temp_dir, dump_name) stderr_lines = [] self.log.debug('Executing %s' % ' '.join(svnrdump_cmd)) with open(dump_path, 'wb') as dump_file: stderr_r, stderr_w = pty.openpty() svnrdump = Popen(svnrdump_cmd, stdout=dump_file, stderr=stderr_w) os.close(stderr_w) stderr_stream = OutputStream(stderr_r) readable = True while readable: lines, readable = stderr_stream.read_lines() stderr_lines += lines for line in lines: self.log.debug(line) svnrdump.wait() os.close(stderr_r) if svnrdump.returncode == 0: return dump_path # There was an error but it does not mean that no revisions # can be loaded. # Get the stderr line with latest dumped revision last_dumped_rev = None if len(stderr_lines) > 1: last_dumped_rev = stderr_lines[-2] if last_dumped_rev: # Get the latest dumped revision number matched_rev = re.search('.*revision ([0-9]+)', last_dumped_rev) last_dumped_rev = int(matched_rev.group(1)) if matched_rev else -1 # Check if revisions inside the dump file can be loaded anyway if last_dumped_rev > last_loaded_svn_rev: self.log.debug(('svnrdump did not dump all expected revisions ' 'but revisions range %s:%s are available in ' 'the generated dump file and will be loaded ' 'into the archive.') % (last_loaded_svn_rev+1, last_dumped_rev)) # Truncate the dump file after the last successfully dumped # revision to avoid the loading of corrupted data self.log.debug(('Truncating dump file after the last ' 'successfully dumped revision (%s) to avoid ' 'the loading of corrupted data') % last_dumped_rev) with open(dump_path, 'r+b') as f: with mmap(f.fileno(), 0, access=ACCESS_WRITE) as s: pattern = ('Revision-number: %s' % (last_dumped_rev+1)).encode() n = s.rfind(pattern) if n != -1: s.resize(n) self.truncated_dump = True return dump_path elif last_dumped_rev != -1: raise Exception(('Last dumped subversion revision (%s) is ' 'lesser than the last one loaded into the ' 'archive (%s).') % (last_dumped_rev, last_loaded_svn_rev)) raise Exception('An error occurred when running svnrdump and ' 'no exploitable dump file has been generated.') def prepare(self, *args, **kwargs): # First, check if previous revisions have been loaded for the # subversion origin and get the number of the last one last_loaded_svn_rev = self.get_last_loaded_svn_rev(self.svn_url) # Then try to generate a dump file containing relevant svn revisions # to load, an exception will be thrown if something wrong happened dump_path = self.dump_svn_revisions(self.svn_url, last_loaded_svn_rev) # Finally, mount the dump and load the repository self.log.debug('Mounting dump file with "svnadmin load".') _, self.repo_path = \ init_svn_repo_from_dump(dump_path, prefix=TEMPORARY_DIR_PREFIX_PATTERN, suffix='-%s' % os.getpid(), root_dir=self.temp_dir) self.svn_url = 'file://%s' % self.repo_path super().prepare(*args, **kwargs) def cleanup(self): super().cleanup() if self.temp_dir and os.path.exists(self.temp_dir): shutil.rmtree(self.temp_dir) def visit_status(self): if self.truncated_dump: return 'partial' else: return super().visit_status() diff --git a/swh/loader/svn/tests/conftest.py b/swh/loader/svn/tests/conftest.py index 4832a76..5af0b6c 100644 --- a/swh/loader/svn/tests/conftest.py +++ b/swh/loader/svn/tests/conftest.py @@ -1,57 +1,61 @@ # Copyright (C) 2019 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 os import pytest import yaml from typing import Any, Dict from swh.scheduler.tests.conftest import swh_app # noqa @pytest.fixture def swh_loader_config() -> Dict[str, Any]: return { 'storage': { - 'cls': 'memory', + 'cls': 'pipeline', + 'steps': [ + { + 'cls': 'filter' + }, + { + 'cls': 'buffer', + 'min_batch_size': { + 'content': 10000, + 'content_bytes': 1073741824, + 'directory': 2500, + 'revision': 10, + 'release': 100, + }, + }, + { + 'cls': 'memory' + }, + ] }, 'check_revision': {'limit': 100, 'status': False}, - 'content_packet_block_size_bytes': 104857600, - 'content_packet_size': 10000, - 'content_packet_size_bytes': 1073741824, - 'content_size_limit': 104857600, 'debug': False, - 'directory_packet_size': 2500, 'log_db': 'dbname=softwareheritage-log', - 'occurrence_packet_size': 1000, - 'release_packet_size': 1000, - 'revision_packet_size': 10, 'save_data': False, 'save_data_path': '', - 'send_contents': True, - 'send_directories': True, - 'send_occurrences': True, - 'send_releases': True, - 'send_revisions': True, - 'send_snapshot': True, 'temp_directory': '/tmp', } @pytest.fixture def swh_config(swh_loader_config, monkeypatch, tmp_path): conffile = os.path.join(str(tmp_path), 'loader.yml') with open(conffile, 'w') as f: f.write(yaml.dump(swh_loader_config)) monkeypatch.setenv('SWH_CONFIG_FILENAME', conffile) return conffile @pytest.fixture(scope='session') def celery_includes(): return [ 'swh.loader.svn.tasks', ] diff --git a/swh/loader/svn/tests/test_loader.py b/swh/loader/svn/tests/test_loader.py index a9ed474..ac3c15b 100644 --- a/swh/loader/svn/tests/test_loader.py +++ b/swh/loader/svn/tests/test_loader.py @@ -1,863 +1,870 @@ # 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 import copy import os from unittest import TestCase from swh.loader.core.tests import BaseLoaderTest from swh.loader.svn.loader import (DEFAULT_BRANCH, SvnLoader, SvnLoaderFromRemoteDump, build_swh_snapshot) from swh.model import hashutil class TestSnapshot(TestCase): def test_build_swh_snapshot(self): actual_snap = build_swh_snapshot('revision-id') self.assertEqual(actual_snap, { 'id': None, 'branches': { DEFAULT_BRANCH: { 'target': 'revision-id', 'target_type': 'revision', } } }) _LOADER_TEST_CONFIG = { 'check_revision': {'limit': 100, 'status': False}, - 'content_packet_block_size_bytes': 104857600, - 'content_packet_size': 10000, - 'content_packet_size_bytes': 1073741824, - 'content_size_limit': 104857600, 'debug': False, - 'directory_packet_size': 2500, 'log_db': 'dbname=softwareheritage-log', - 'occurrence_packet_size': 1000, - 'release_packet_size': 1000, - 'revision_packet_size': 10, 'save_data': False, 'save_data_path': '', - 'send_contents': True, - 'send_directories': True, - 'send_occurrences': True, - 'send_releases': True, - 'send_revisions': True, - 'send_snapshot': True, - 'storage': {'args': {}, 'cls': 'memory'}, - 'temp_directory': '/tmp' + 'temp_directory': '/tmp', + 'max_content_size': 100 * 1024 * 1024, + 'storage': { + 'cls': 'pipeline', + 'steps': [ + { + 'cls': 'filter' + }, + { + 'cls': 'buffer', + 'min_batch_size': { + 'content': 10000, + 'content_bytes': 1073741824, + 'directory': 2500, + 'revision': 10, + 'release': 100, + }, + }, + { + 'cls': 'memory' + }, + ] + }, } class SvnLoaderTest(SvnLoader): """An SVNLoader with no persistence. Context: Load a new svn repository using the swh policy (so no update). """ def __init__(self, url, last_snp_rev={}, destination_path=None, start_from_scratch=False, swh_revision=None): super().__init__(url, destination_path=destination_path, start_from_scratch=start_from_scratch, swh_revision=swh_revision) self.origin = { 'id': 1, 'url': url, } self.visit = { 'origin': 1, 'visit': 1, } self.last_snp_rev = last_snp_rev def parse_config_file(self, *args, **kwargs): return _LOADER_TEST_CONFIG def swh_latest_snapshot_revision(self, origin_url, 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. SvnLoaderTest """ return self.last_snp_rev class BaseSvnLoaderTest(BaseLoaderTest): """Base test loader class. In its setup, it's uncompressing a local svn mirror to /tmp. """ def setUp(self, archive_name='pkg-gourmet.tgz', filename='pkg-gourmet', loader=None, snapshot=None, type='default', start_from_scratch=False, swh_revision=None): super().setUp(archive_name=archive_name, filename=filename, prefix_tmp_folder_name='swh.loader.svn.', start_path=os.path.dirname(__file__)) self.svn_mirror_url = self.repo_url if type == 'default': loader_test_class = SvnLoaderTest else: loader_test_class = SvnLoaderTestFromRemoteDump if loader: self.loader = loader elif snapshot: self.loader = loader_test_class( self.svn_mirror_url, destination_path=self.destination_path, start_from_scratch=start_from_scratch, swh_revision=swh_revision, last_snp_rev=snapshot, ) else: self.loader = loader_test_class( self.svn_mirror_url, destination_path=self.destination_path, start_from_scratch=start_from_scratch, swh_revision=swh_revision ) self.storage = self.loader.storage class SvnLoaderTest1(BaseSvnLoaderTest): """Load an unknown svn repository results in new data. """ def test_load(self): """Load a new repository results in new swh object and snapshot """ # when self.loader.load() # then self.assertCountRevisions(6) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') _LAST_SNP_REV = { 'snapshot': { 'id': 'something', 'branches': {} }, # need a snapshot of sort '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'] ] } } } class SvnLoaderTest2(BaseSvnLoaderTest): """Load a visited repository with no new change results in no data change. """ def setUp(self): super().setUp(snapshot=_LAST_SNP_REV) def test_load(self): """Load a repository without new changes results in same snapshot """ # when self.loader.load() # then self.assertCountContents(0) self.assertCountDirectories(0) self.assertCountRevisions(0) self.assertCountReleases(0) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'uneventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest3(BaseSvnLoaderTest): """In this scenario, the dump has been tampered with to modify the commit log. This results in a hash divergence which is detected at startup. In effect, that stops the loading and do nothing. """ def setUp(self): last_snp_rev = copy.deepcopy(_LAST_SNP_REV) last_snp_rev['snapshot'] = None # Changed the revision id's hash to simulate history altered last_snp_rev['revision']['id'] = \ hashutil.hash_to_bytes('badbadbadbadf708f7466dddf547567b65f6c39d') # the svn repository pkg-gourmet has been updated with changes super().setUp(archive_name='pkg-gourmet-with-updates.tgz', snapshot=last_snp_rev) def test_load(self): """Load known repository with history altered should do nothing """ # when self.loader.load() # then # we got the previous run's last revision (rev 6) # so 2 news + 1 old self.assertCountContents(0) self.assertCountDirectories(0) self.assertCountRevisions(0) self.assertCountReleases(0) self.assertCountSnapshots(0) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'uneventful'}) self.assertEqual(self.loader.visit_status(), 'partial') class SvnLoaderTest4(BaseSvnLoaderTest): """In this scenario, the repository has been updated with new changes. The loading visit should result in new objects stored and 1 new snapshot. """ def setUp(self): # the svn repository pkg-gourmet has been updated with changes super().setUp(archive_name='pkg-gourmet-with-updates.tgz', snapshot=_LAST_SNP_REV) def test_process_repository(self): """Process updated repository should yield new objects """ # when self.loader.load() # then # we got the previous run's last revision (rev 6) # so 2 new self.assertCountRevisions(5) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest5(BaseSvnLoaderTest): """Context: - Repository already injected with successful data - New visit from scratch done with successful load """ def setUp(self): # the svn repository pkg-gourmet has been updated with changes super().setUp(archive_name='pkg-gourmet-with-updates.tgz', snapshot=_LAST_SNP_REV, start_from_scratch=True) def test_load(self): """Load an existing repository from scratch yields same swh objects """ # when self.loader.load() # 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.assertCountRevisions(11) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest6(BaseSvnLoaderTest): """Context: - repository already visited with load successful - Changes on existing repository - New Visit done with successful new data """ def setUp(self): last_snp_rev = { 'snapshot': None, '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'], # noqa ['svn_revision', '6'] ] } } } super().setUp(archive_name='pkg-gourmet-with-updates.tgz', snapshot=last_snp_rev) def test_load(self): """Load from partial previous visit result in new changes """ # when self.loader.load() # then # we got the previous run's last revision (rev 6) # so 2 new self.assertCountRevisions(5) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest7(BaseSvnLoaderTest): """Context: - repository already visited with load successful - Changes on existing repository - New Visit done with successful new data """ def setUp(self): 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'] ] } } super().setUp(archive_name='pkg-gourmet-with-updates.tgz', snapshot=_LAST_SNP_REV, swh_revision=previous_unfinished_revision) def test_load(self): """Load known and partial repository should start from last visit """ # when self.loader.load() # then # we got the previous run's last revision (rev 6) # so 2 new self.assertCountRevisions(5) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest8(BaseSvnLoaderTest): """Context: - Previous visit on existing repository done - Starting the loading from the last unfinished visit - New objects are created (1 new snapshot) """ def setUp(self): last_snp_rev = { 'snapshot': None, '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'] ] } } } 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'] ] } } super().setUp(archive_name='pkg-gourmet-with-updates.tgz', snapshot=last_snp_rev, swh_revision=previous_unfinished_revision) def test_load(self): """Load repository should yield revisions starting from last visit """ # when self.loader.load() # then # we got the previous run's last revision (rev 6) # so 2 new self.assertCountRevisions(5) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTTest9(BaseSvnLoaderTest): """Check that a svn repo containing a versioned file with CRLF line endings with svn:eol-style property set to 'native' (this is a violation of svn specification as the file should have been stored with LF line endings) can be loaded anyway. """ def setUp(self): super().setUp(archive_name='mediawiki-repo-r407-eol-native-crlf.tgz', filename='mediawiki-repo-r407-eol-native-crlf') def test_process_repository(self): """Load repository with CRLF endings (svn:eol-style: native) is ok """ # when self.loader.load() expected_revisions = { '7da4975c363101b819756d33459f30a866d01b1b': 'f63637223ee0f7d4951ffd2d4d9547a4882c5d8b' # noqa } self.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest10(BaseSvnLoaderTest): # noqa """Check that a svn repo containing a versioned file with mixed CRLF/LF line endings with svn:eol-style property set to 'native' (this is a violation of svn specification as mixed line endings for textual content should not be stored when the svn:eol-style property is set) can be loaded anyway. """ def setUp(self): super().setUp( archive_name='pyang-repo-r343-eol-native-mixed-lf-crlf.tgz', filename='pyang-repo-r343-eol-native-mixed-lf-crlf') def test_load(self): """Load repo with mixed CRLF/LF endings (svn:eol-style:native) is ok """ self.loader.load() expected_revisions = { '9c6962eeb9164a636c374be700672355e34a98a7': '16aa6b6271f3456d4643999d234cf39fe3d0cc5a' # noqa } self.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state # self.assertEqual(self.loader.all_snapshots[0], {}) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest11(BaseSvnLoaderTest): """Context: - Repository with svn:external (which is not deal with for now) - Visit is partial with as much data loaded as possible """ def setUp(self): previous_unfinished_revision = None super().setUp(archive_name='pkg-gourmet-with-external-id.tgz', swh_revision=previous_unfinished_revision) def test_load(self): """Repository with svn:externals property, will stop raising an error """ # when self.loader.load() # then repositories holds 21 revisions, but the last commit # one holds an 'svn:externals' property which will make the # loader-svn stops at the last revision prior to the bad one self.assertCountRevisions(20) self.assertCountReleases(0) last_revision = '82a7a4a09f9549223429143ba36ad77375e33c5c' 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 'ffa901b69ca0f46a2261f42948838d19709cb9f8': 'c3c98df624733fef4e592bef983f93e2ed02b179', # noqa '0148ae3eaa520b73a50802c59f3f416b7a36cf8c': '844d4646d6c2b4f3a3b2b22ab0ee38c7df07bab2', # noqa last_revision: '0de6e75d2b79ec90d00a3a7611aa3861b2e4aa5e', # noqa } # The last revision being the one used later to start back from self.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'partial') class SvnLoaderTest12(BaseSvnLoaderTest): """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) """ def setUp(self): previous_unfinished_revision = None super().setUp( archive_name='pkg-gourmet-with-edge-case-links-and-files.tgz', swh_revision=previous_unfinished_revision) def test_load(self): """File/Link removed prior to folder with same name creation is ok """ # when self.loader.load() # then repositories holds 14 revisions, but the last commit self.assertCountRevisions(19) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTest13(BaseSvnLoaderTest): """Edge cases: - wrong symbolic link - wrong symbolic link with empty space names """ def setUp(self): super().setUp( archive_name='pkg-gourmet-with-wrong-link-cases.tgz') def test_load(self): """Wrong link or empty space-named link should be ok """ # when self.loader.load() # then repositories holds 14 revisions, but the last commit self.assertCountRevisions(21) self.assertCountReleases(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.assertRevisionsContain(expected_revisions) self.assertCountSnapshots(1) # FIXME: Check the snapshot's state self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full') class SvnLoaderTestFromRemoteDump(SvnLoaderTest, SvnLoaderFromRemoteDump): pass class SvnLoaderFromRemoteDumpTest(BaseSvnLoaderTest): """ Check that the results obtained with the remote svn dump loader and the base svn loader are the same. """ def setUp(self): _LOADER_TEST_CONFIG['debug'] = True # to avoid cleanup in between load super().setUp(archive_name='pkg-gourmet.tgz', type='remote') def test_load(self): """ Compare results of remote dump loader and base loader """ dump_loader = self.loader dump_loader.load() self.assertCountContents(19) self.assertCountDirectories(17) self.assertCountRevisions(6) self.assertCountSnapshots(1) base_loader = SvnLoaderTest(self.svn_mirror_url) base_loader.load() dump_storage_stat = dump_loader.storage.stat_counters() base_storage_stat = base_loader.storage.stat_counters() self.assertEqual(dump_storage_stat, base_storage_stat) class SvnLoaderTest14(BaseSvnLoaderTest): """Edge cases: The repository held some user defined svn-properties with special encodings, this prevented the repository from being loaded even though we do not ingest those information. """ def setUp(self): super().setUp(archive_name='httthttt.tgz', filename='httthttt') def test_load(self): """Decoding user defined svn properties error should not fail loading """ # when self.loader.load() self.assertCountRevisions(7, '7 svn commits') self.assertCountReleases(0) last_revision = '604a17dbb15e8d7ecb3e9f3768d09bf493667a93' expected_revisions = { 'e6ae8487c6d14df9e6cb7196c6aac045798fd5be': '75ed58f260bfa4102d0e09657803511f5f0ab372', # noqa 'e1e3314e0e9c9d17e6a3f60d6662f48f0e3c2fa3': '7bfb95cef68c1affe8d7f786353213d92abbb2b7', # noqa '1632fd38a8653e9b607c00feb93a41faddfb544c': 'cd6de65c84d9405e7ca45fead02aa10162e30727', # noqa '0ad1ebbb92d00721644b0a46d6322d18dbcba848': 'cd6de65c84d9405e7ca45fead02aa10162e30727', # noqa '94b87c97697d178a9311b018daa5179f7d4ba31e': 'c2128108adecb59a0144339c2e701cd8118cff5a', # noqa 'bd741cf22f0642d88cd0d8b545e8896b898c692d': 'c2128108adecb59a0144339c2e701cd8118cff5a', # noqa last_revision: 'f051d60256b2d89a0ca2704d6f91ad1b0ab44e02', } self.assertRevisionsContain(expected_revisions) expected_snapshot_id = '70487267f682c07e52a2371061369b6cf5bffa47' expected_branches = { 'HEAD': { 'target': last_revision, 'target_type': 'revision' } } self.assertSnapshotEqual(expected_snapshot_id, expected_branches) self.assertEqual(self.loader.load_status(), {'status': 'eventful'}) self.assertEqual(self.loader.visit_status(), 'full')