diff --git a/swh/web/api/views/snapshot.py b/swh/web/api/views/snapshot.py index cfeb3286..5b2e0035 100644 --- a/swh/web/api/views/snapshot.py +++ b/swh/web/api/views/snapshot.py @@ -1,117 +1,118 @@ # Copyright (C) 2018 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from swh.web.common import service from swh.web.common.utils import reverse from swh.web.config import get_config from swh.web.api.apidoc import api_doc from swh.web.api import utils from swh.web.api.apiurls import api_route from swh.web.api.views.utils import api_lookup @api_route(r'/snapshot/(?P[0-9a-f]+)/', 'api-snapshot', checksum_args=['snapshot_id']) @api_doc('/snapshot/') def api_snapshot(request, snapshot_id): """ .. http:get:: /api/1/snapshot/(snapshot_id)/ Get information about a snapshot in the archive. A snapshot is a set of named branches, which are pointers to objects at any level of the Software Heritage DAG. It represents a full picture of an origin at a given time. As well as pointing to other objects in the Software Heritage DAG, branches can also be aliases, in which case their target is the name of another branch in the same snapshot, or dangling, in which case the target is unknown. A snapshot identifier is a salted sha1. See :func:`swh.model.identifiers.snapshot_identifier` in our data model module for details about how they are computed. :param sha1 snapshot_id: a snapshot identifier :query str branches_from: optional parameter used to skip branches whose name is lesser than it before returning them :query int branches_count: optional parameter used to restrain the amount of returned branches (default to 1000) :query str target_types: optional comma separated list parameter used to filter the target types of branch to return (possible values that can be contained in that list are ``content``, ``directory``, ``revision``, ``release``, ``snapshot`` or ``alias``) :reqheader Accept: the requested response content type, either ``application/json`` (default) or ``application/yaml`` :resheader Content-Type: this depends on :http:header:`Accept` header of request :resheader Link: indicates that a subsequent result page is available and contains the url pointing to it :>json object branches: object containing all branches associated to the snapshot, for each of them the associated target type and id are given but also a link to get information about that target :>json string id: the unique identifier of the snapshot **Allowed HTTP Methods:** :http:method:`get`, :http:method:`head`, :http:method:`options` :statuscode 200: no error :statuscode 400: an invalid snapshot identifier has been provided :statuscode 404: requested snapshot can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`snapshot/6a3a2cf0b2b90ce7ae1cf0a221ed68035b686f5a/` """ # noqa def _enrich_snapshot(snapshot): s = snapshot.copy() if 'branches' in s: s['branches'] = { k: utils.enrich_object(v) if v else None for k, v in s['branches'].items() } for k, v in s['branches'].items(): if v and v['target_type'] == 'alias': if v['target'] in s['branches']: branch_alias = s['branches'][v['target']] - v['target_url'] = branch_alias['target_url'] + if branch_alias: + v['target_url'] = branch_alias['target_url'] else: snp = \ service.lookup_snapshot(s['id'], branches_from=v['target'], branches_count=1) if snp and v['target'] in snp['branches']: branch = snp['branches'][v['target']] branch = utils.enrich_object(branch) v['target_url'] = branch['target_url'] return s snapshot_content_max_size = get_config()['snapshot_content_max_size'] branches_from = request.GET.get('branches_from', '') branches_count = int(request.GET.get('branches_count', snapshot_content_max_size)) target_types = request.GET.get('target_types', None) target_types = target_types.split(',') if target_types else None results = api_lookup( service.lookup_snapshot, snapshot_id, branches_from, branches_count, target_types, notfound_msg='Snapshot with id {} not found.'.format(snapshot_id), enrich_fn=_enrich_snapshot) response = {'results': results, 'headers': {}} if results['next_branch'] is not None: response['headers']['link-next'] = \ reverse('api-snapshot', url_args={'snapshot_id': snapshot_id}, query_params={'branches_from': results['next_branch'], 'branches_count': branches_count, 'target_types': target_types}) return response diff --git a/swh/web/browse/utils.py b/swh/web/browse/utils.py index c7a2a32e..52368a0b 100644 --- a/swh/web/browse/utils.py +++ b/swh/web/browse/utils.py @@ -1,1113 +1,1116 @@ # Copyright (C) 2017-2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import base64 from collections import defaultdict import magic import pypandoc import stat import textwrap from django.core.cache import cache from django.utils.safestring import mark_safe from django.utils.html import escape from importlib import reload from swh.model.identifiers import persistent_identifier from swh.web.common import highlightjs, service from swh.web.common.exc import NotFoundExc, http_status_code_message from swh.web.common.origin_visits import get_origin_visit from swh.web.common.utils import ( reverse, format_utc_iso_date, get_swh_persistent_id, swh_object_icons ) from swh.web.config import get_config def get_directory_entries(sha1_git): """Function that retrieves the content of a directory from the archive. The directories entries are first sorted in lexicographical order. Sub-directories and regular files are then extracted. Args: sha1_git: sha1_git identifier of the directory Returns: A tuple whose first member corresponds to the sub-directories list and second member the regular files list Raises: NotFoundExc if the directory is not found """ cache_entry_id = 'directory_entries_%s' % sha1_git cache_entry = cache.get(cache_entry_id) if cache_entry: return cache_entry entries = list(service.lookup_directory(sha1_git)) for e in entries: e['perms'] = stat.filemode(e['perms']) if e['type'] == 'rev': # modify dir entry name to explicitly show it points # to a revision e['name'] = '%s @ %s' % (e['name'], e['target'][:7]) dirs = [e for e in entries if e['type'] in ('dir', 'rev')] files = [e for e in entries if e['type'] == 'file'] dirs = sorted(dirs, key=lambda d: d['name']) files = sorted(files, key=lambda f: f['name']) cache.set(cache_entry_id, (dirs, files)) return dirs, files def get_mimetype_and_encoding_for_content(content): """Function that returns the mime type and the encoding associated to a content buffer using the magic module under the hood. Args: content (bytes): a content buffer Returns: A tuple (mimetype, encoding), for instance ('text/plain', 'us-ascii'), associated to the provided content. """ while True: try: magic_result = magic.detect_from_content(content) mime_type = magic_result.mime_type encoding = magic_result.encoding break except Exception: # workaround an issue with the magic module who can fail # if detect_from_content is called multiple times in # a short amount of time reload(magic) return mime_type, encoding # maximum authorized content size in bytes for HTML display # with code highlighting content_display_max_size = get_config()['content_display_max_size'] snapshot_content_max_size = get_config()['snapshot_content_max_size'] def _reencode_content(mimetype, encoding, content_data): # encode textual content to utf-8 if needed if mimetype.startswith('text/'): # probably a malformed UTF-8 content, re-encode it # by replacing invalid chars with a substitution one if encoding == 'unknown-8bit': content_data = content_data.decode('utf-8', 'replace')\ .encode('utf-8') elif 'ascii' not in encoding and encoding not in ['utf-8', 'binary']: content_data = content_data.decode(encoding, 'replace')\ .encode('utf-8') elif mimetype.startswith('application/octet-stream'): # file may detect a text content as binary # so try to decode it for display encodings = ['us-ascii'] encodings += ['iso-8859-%s' % i for i in range(1, 17)] for encoding in encodings: try: content_data = content_data.decode(encoding)\ .encode('utf-8') except Exception: pass else: # ensure display in content view mimetype = 'text/plain' break return mimetype, content_data def request_content(query_string, max_size=content_display_max_size, raise_if_unavailable=True, reencode=True): """Function that retrieves a content from the archive. Raw bytes content is first retrieved, then the content mime type. If the mime type is not stored in the archive, it will be computed using Python magic module. Args: query_string: a string of the form "[ALGO_HASH:]HASH" where optional ALGO_HASH can be either ``sha1``, ``sha1_git``, ``sha256``, or ``blake2s256`` (default to ``sha1``) and HASH the hexadecimal representation of the hash value max_size: the maximum size for a content to retrieve (default to 1MB, no size limit if None) Returns: A tuple whose first member corresponds to the content raw bytes and second member the content mime type Raises: NotFoundExc if the content is not found """ content_data = service.lookup_content(query_string) filetype = None language = None license = None # requests to the indexer db may fail so properly handle # those cases in order to avoid content display errors try: filetype = service.lookup_content_filetype(query_string) language = service.lookup_content_language(query_string) license = service.lookup_content_license(query_string) except Exception: pass mimetype = 'unknown' encoding = 'unknown' if filetype: mimetype = filetype['mimetype'] encoding = filetype['encoding'] # workaround when encountering corrupted data due to implicit # conversion from bytea to text in the indexer db (see T818) # TODO: Remove that code when all data have been correctly converted if mimetype.startswith('\\'): filetype = None content_data['error_code'] = 200 content_data['error_message'] = '' content_data['error_description'] = '' if not max_size or content_data['length'] < max_size: try: content_raw = service.lookup_content_raw(query_string) except Exception as e: if raise_if_unavailable: raise e else: content_data['raw_data'] = None content_data['error_code'] = 404 content_data['error_description'] = \ 'The bytes of the content are currently not available in the archive.' # noqa content_data['error_message'] = \ http_status_code_message[content_data['error_code']] else: content_data['raw_data'] = content_raw['data'] if not filetype: mimetype, encoding = \ get_mimetype_and_encoding_for_content(content_data['raw_data']) # noqa if reencode: mimetype, raw_data = _reencode_content( mimetype, encoding, content_data['raw_data']) content_data['raw_data'] = raw_data else: content_data['raw_data'] = None content_data['mimetype'] = mimetype content_data['encoding'] = encoding if language: content_data['language'] = language['lang'] else: content_data['language'] = 'not detected' if license: content_data['licenses'] = ', '.join(license['facts'][0]['licenses']) else: content_data['licenses'] = 'not detected' return content_data _browsers_supported_image_mimes = set(['image/gif', 'image/png', 'image/jpeg', 'image/bmp', 'image/webp', 'image/svg', 'image/svg+xml']) def prepare_content_for_display(content_data, mime_type, path): """Function that prepares a content for HTML display. The function tries to associate a programming language to a content in order to perform syntax highlighting client-side using highlightjs. The language is determined using either the content filename or its mime type. If the mime type corresponds to an image format supported by web browsers, the content will be encoded in base64 for displaying the image. Args: content_data (bytes): raw bytes of the content mime_type (string): mime type of the content path (string): path of the content including filename Returns: A dict containing the content bytes (possibly different from the one provided as parameter if it is an image) under the key 'content_data and the corresponding highlightjs language class under the key 'language'. """ language = highlightjs.get_hljs_language_from_filename(path) if not language: language = highlightjs.get_hljs_language_from_mime_type(mime_type) if not language: language = 'nohighlight' elif mime_type.startswith('application/'): mime_type = mime_type.replace('application/', 'text/') if mime_type.startswith('image/'): if mime_type in _browsers_supported_image_mimes: content_data = base64.b64encode(content_data) else: content_data = None if mime_type.startswith('image/svg'): mime_type = 'image/svg+xml' return {'content_data': content_data, 'language': language, 'mimetype': mime_type} def process_snapshot_branches(snapshot): """ Process a dictionary describing snapshot branches: extract those targeting revisions and releases, put them in two different lists, then sort those lists in lexicographical order of the branches' names. Args: snapshot_branches (dict): A dict describing the branches of a snapshot as returned for instance by :func:`swh.web.common.service.lookup_snapshot` Returns: tuple: A tuple whose first member is the sorted list of branches targeting revisions and second member the sorted list of branches targeting releases """ # noqa snapshot_branches = snapshot['branches'] branches = {} branch_aliases = {} releases = {} revision_to_branch = defaultdict(set) revision_to_release = defaultdict(set) release_to_branch = defaultdict(set) for branch_name, target in snapshot_branches.items(): if not target: # FIXME: display branches with an unknown target anyway continue target_id = target['target'] target_type = target['target_type'] if target_type == 'revision': branches[branch_name] = { 'name': branch_name, 'revision': target_id, } revision_to_branch[target_id].add(branch_name) elif target_type == 'release': release_to_branch[target_id].add(branch_name) elif target_type == 'alias': branch_aliases[branch_name] = target_id # FIXME: handle pointers to other object types def _enrich_release_branch(branch, release): releases[branch] = { 'name': release['name'], 'branch_name': branch, 'date': format_utc_iso_date(release['date']), 'id': release['id'], 'message': release['message'], 'target_type': release['target_type'], 'target': release['target'], } def _enrich_revision_branch(branch, revision): branches[branch].update({ 'revision': revision['id'], 'directory': revision['directory'], 'date': format_utc_iso_date(revision['date']), 'message': revision['message'] }) releases_info = service.lookup_release_multiple( release_to_branch.keys() ) for release in releases_info: branches_to_update = release_to_branch[release['id']] for branch in branches_to_update: _enrich_release_branch(branch, release) if release['target_type'] == 'revision': revision_to_release[release['target']].update( branches_to_update ) revisions = service.lookup_revision_multiple( set(revision_to_branch.keys()) | set(revision_to_release.keys()) ) for revision in revisions: if not revision: continue for branch in revision_to_branch[revision['id']]: _enrich_revision_branch(branch, revision) for release in revision_to_release[revision['id']]: releases[release]['directory'] = revision['directory'] for branch_alias, branch_target in branch_aliases.items(): if branch_target in branches: branches[branch_alias] = dict(branches[branch_target]) else: snp = service.lookup_snapshot(snapshot['id'], branches_from=branch_target, branches_count=1) if snp and branch_target in snp['branches']: + if snp['branches'][branch_target] is None: + continue + target_type = snp['branches'][branch_target]['target_type'] target = snp['branches'][branch_target]['target'] if target_type == 'revision': branches[branch_alias] = snp['branches'][branch_target] revision = service.lookup_revision(target) _enrich_revision_branch(branch_alias, revision) elif target_type == 'release': release = service.lookup_release(target) _enrich_release_branch(branch_alias, release) if branch_alias in branches: branches[branch_alias]['name'] = branch_alias ret_branches = list(sorted(branches.values(), key=lambda b: b['name'])) ret_releases = list(sorted(releases.values(), key=lambda b: b['name'])) return ret_branches, ret_releases def get_snapshot_content(snapshot_id): """Returns the lists of branches and releases associated to a swh snapshot. That list is put in cache in order to speedup the navigation in the swh-web/browse ui. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: snapshot_id (str): hexadecimal representation of the snapshot identifier Returns: A tuple with two members. The first one is a list of dict describing the snapshot branches. The second one is a list of dict describing the snapshot releases. Raises: NotFoundExc if the snapshot does not exist """ cache_entry_id = 'swh_snapshot_%s' % snapshot_id cache_entry = cache.get(cache_entry_id) if cache_entry: return cache_entry['branches'], cache_entry['releases'] branches = [] releases = [] if snapshot_id: snapshot = service.lookup_snapshot( snapshot_id, branches_count=snapshot_content_max_size) branches, releases = process_snapshot_branches(snapshot) cache.set(cache_entry_id, { 'branches': branches, 'releases': releases, }) return branches, releases def get_origin_visit_snapshot(origin_info, visit_ts=None, visit_id=None, snapshot_id=None): """Returns the lists of branches and releases associated to a swh origin for a given visit. The visit is expressed by a timestamp. In the latter case, the closest visit from the provided timestamp will be used. If no visit parameter is provided, it returns the list of branches found for the latest visit. That list is put in cache in order to speedup the navigation in the swh-web/browse ui. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: origin_info (dict): a dict filled with origin information (id, url, type) visit_ts (int or str): an ISO date string or Unix timestamp to parse visit_id (int): optional visit id for disambiguation in case several visits have the same timestamp Returns: A tuple with two members. The first one is a list of dict describing the origin branches for the given visit. The second one is a list of dict describing the origin releases for the given visit. Raises: NotFoundExc if the origin or its visit are not found """ visit_info = get_origin_visit(origin_info, visit_ts, visit_id, snapshot_id) return get_snapshot_content(visit_info['snapshot']) def gen_link(url, link_text=None, link_attrs=None): """ Utility function for generating an HTML link to insert in Django templates. Args: url (str): an url link_text (str): optional text for the produced link, if not provided the url will be used link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ attrs = ' ' if link_attrs: for k, v in link_attrs.items(): attrs += '%s="%s" ' % (k, v) if not link_text: link_text = url link = '%s' \ % (attrs, escape(url), escape(link_text)) return mark_safe(link) def _snapshot_context_query_params(snapshot_context): query_params = None if snapshot_context and snapshot_context['origin_info']: origin_info = snapshot_context['origin_info'] query_params = {'origin': origin_info['url']} if 'timestamp' in snapshot_context['url_args']: query_params['timestamp'] = \ snapshot_context['url_args']['timestamp'] if 'visit_id' in snapshot_context['query_params']: query_params['visit_id'] = \ snapshot_context['query_params']['visit_id'] elif snapshot_context: query_params = {'snapshot_id': snapshot_context['snapshot_id']} return query_params def gen_person_link(person_id, person_name, snapshot_context=None, link_attrs=None): """ Utility function for generating a link to a person HTML view to insert in Django templates. Args: person_id (int): a person id person_name (str): the associated person name link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'person_name' """ query_params = _snapshot_context_query_params(snapshot_context) person_url = reverse('browse-person', url_args={'person_id': person_id}, query_params=query_params) return gen_link(person_url, person_name or 'None', link_attrs) def gen_revision_url(revision_id, snapshot_context=None): """ Utility function for generating an url to a revision. Args: revision_id (str): a revision id snapshot_context (dict): if provided, generate snapshot-dependent browsing url Returns: str: The url to browse the revision """ query_params = _snapshot_context_query_params(snapshot_context) return reverse('browse-revision', url_args={'sha1_git': revision_id}, query_params=query_params) def gen_revision_link(revision_id, shorten_id=False, snapshot_context=None, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}): """ Utility function for generating a link to a revision HTML view to insert in Django templates. Args: revision_id (str): a revision id shorten_id (boolean): whether to shorten the revision id to 7 characters for the link text snapshot_context (dict): if provided, generate snapshot-dependent browsing link link_text (str): optional text for the generated link (the revision id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: str: An HTML link in the form 'revision_id' """ if not revision_id: return None revision_url = gen_revision_url(revision_id, snapshot_context) if shorten_id: return gen_link(revision_url, revision_id[:7], link_attrs) else: if not link_text: link_text = revision_id return gen_link(revision_url, link_text, link_attrs) def gen_directory_link(sha1_git, snapshot_context=None, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}): """ Utility function for generating a link to a directory HTML view to insert in Django templates. Args: sha1_git (str): directory identifier link_text (str): optional text for the generated link (the directory id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ if not sha1_git: return None query_params = _snapshot_context_query_params(snapshot_context) directory_url = reverse('browse-directory', url_args={'sha1_git': sha1_git}, query_params=query_params) if not link_text: link_text = sha1_git return gen_link(directory_url, link_text, link_attrs) def gen_snapshot_link(snapshot_id, snapshot_context=None, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}): """ Utility function for generating a link to a snapshot HTML view to insert in Django templates. Args: snapshot_id (str): snapshot identifier link_text (str): optional text for the generated link (the snapshot id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ query_params = _snapshot_context_query_params(snapshot_context) snapshot_url = reverse('browse-snapshot', url_args={'snapshot_id': snapshot_id}, query_params=query_params) if not link_text: link_text = snapshot_id return gen_link(snapshot_url, link_text, link_attrs) def gen_content_link(sha1_git, snapshot_context=None, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}): """ Utility function for generating a link to a content HTML view to insert in Django templates. Args: sha1_git (str): content identifier link_text (str): optional text for the generated link (the content sha1_git will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ if not sha1_git: return None query_params = _snapshot_context_query_params(snapshot_context) content_url = reverse('browse-content', url_args={'query_string': 'sha1_git:' + sha1_git}, query_params=query_params) if not link_text: link_text = sha1_git return gen_link(content_url, link_text, link_attrs) def get_revision_log_url(revision_id, snapshot_context=None): """ Utility function for getting the URL for a revision log HTML view (possibly in the context of an origin). Args: revision_id (str): revision identifier the history heads to snapshot_context (dict): if provided, generate snapshot-dependent browsing link Returns: The revision log view URL """ query_params = {'revision': revision_id} if snapshot_context and snapshot_context['origin_info']: origin_info = snapshot_context['origin_info'] url_args = {'origin_url': origin_info['url']} if 'timestamp' in snapshot_context['url_args']: url_args['timestamp'] = \ snapshot_context['url_args']['timestamp'] if 'visit_id' in snapshot_context['query_params']: query_params['visit_id'] = \ snapshot_context['query_params']['visit_id'] revision_log_url = reverse('browse-origin-log', url_args=url_args, query_params=query_params) elif snapshot_context: url_args = {'snapshot_id': snapshot_context['snapshot_id']} revision_log_url = reverse('browse-snapshot-log', url_args=url_args, query_params=query_params) else: revision_log_url = reverse('browse-revision-log', url_args={'sha1_git': revision_id}) return revision_log_url def gen_revision_log_link(revision_id, snapshot_context=None, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}): """ Utility function for generating a link to a revision log HTML view (possibly in the context of an origin) to insert in Django templates. Args: revision_id (str): revision identifier the history heads to snapshot_context (dict): if provided, generate snapshot-dependent browsing link link_text (str): optional text to use for the generated link (the revision id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ if not revision_id: return None revision_log_url = get_revision_log_url(revision_id, snapshot_context) if not link_text: link_text = revision_id return gen_link(revision_log_url, link_text, link_attrs) def gen_release_link(sha1_git, snapshot_context=None, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}): """ Utility function for generating a link to a release HTML view to insert in Django templates. Args: sha1_git (str): release identifier link_text (str): optional text for the generated link (the release id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ query_params = _snapshot_context_query_params(snapshot_context) release_url = reverse('browse-release', url_args={'sha1_git': sha1_git}, query_params=query_params) if not link_text: link_text = sha1_git return gen_link(release_url, link_text, link_attrs) def format_log_entries(revision_log, per_page, snapshot_context=None): """ Utility functions that process raw revision log data for HTML display. Its purpose is to: * add links to relevant browse views * format date in human readable format * truncate the message log Args: revision_log (list): raw revision log as returned by the swh-web api per_page (int): number of log entries per page snapshot_context (dict): if provided, generate snapshot-dependent browsing link """ revision_log_data = [] for i, rev in enumerate(revision_log): if i == per_page: break author_name = 'None' author_fullname = 'None' committer_fullname = 'None' if rev['author']: author_name = rev['author']['name'] or rev['author']['fullname'] author_fullname = rev['author']['fullname'] if rev['committer']: committer_fullname = rev['committer']['fullname'] author_date = format_utc_iso_date(rev['date']) committer_date = format_utc_iso_date(rev['committer_date']) tooltip = 'revision %s\n' % rev['id'] tooltip += 'author: %s\n' % author_fullname tooltip += 'author date: %s\n' % author_date tooltip += 'committer: %s\n' % committer_fullname tooltip += 'committer date: %s\n\n' % committer_date if rev['message']: tooltip += textwrap.indent(rev['message'], ' '*4) revision_log_data.append({ 'author': author_name, 'id': rev['id'][:7], 'message': rev['message'], 'date': author_date, 'commit_date': committer_date, 'url': gen_revision_url(rev['id'], snapshot_context), 'tooltip': tooltip }) return revision_log_data # list of origin types that can be found in the swh archive # TODO: retrieve it dynamically in an efficient way instead # of hardcoding it _swh_origin_types = ['git', 'svn', 'deb', 'hg', 'ftp', 'deposit', 'pypi', 'npm'] def get_origin_info(origin_url, origin_type=None): """ Get info about a software origin. Its main purpose is to automatically find an origin type when it is not provided as parameter. Args: origin_url (str): complete url of a software origin origin_type (str): optional origin type Returns: A dict with the following entries: * type: the origin type * url: the origin url * id: the internal id of the origin """ if origin_type: return service.lookup_origin({'type': origin_type, 'url': origin_url}) else: for origin_type in _swh_origin_types: try: origin_info = service.lookup_origin({'type': origin_type, 'url': origin_url}) return origin_info except Exception: pass raise NotFoundExc('Origin with url %s not found!' % escape(origin_url)) def get_snapshot_context(snapshot_id=None, origin_type=None, origin_url=None, timestamp=None, visit_id=None): """ Utility function to compute relevant information when navigating the archive in a snapshot context. The snapshot is either referenced by its id or it will be retrieved from an origin visit. Args: snapshot_id (str): hexadecimal representation of a snapshot identifier, all other parameters will be ignored if it is provided origin_type (str): the origin type (git, svn, deposit, ...) origin_url (str): the origin_url (e.g. https://github.com/(user)/(repo)/) timestamp (str): a datetime string for retrieving the closest visit of the origin visit_id (int): optional visit id for disambiguation in case of several visits with the same timestamp Returns: A dict with the following entries: * origin_info: dict containing origin information * visit_info: dict containing visit information * branches: the list of branches for the origin found during the visit * releases: the list of releases for the origin found during the visit * origin_browse_url: the url to browse the origin * origin_branches_url: the url to browse the origin branches * origin_releases_url': the url to browse the origin releases * origin_visit_url: the url to browse the snapshot of the origin found during the visit * url_args: dict containing url arguments to use when browsing in the context of the origin and its visit Raises: NotFoundExc: if no snapshot is found for the visit of an origin. """ # noqa origin_info = None visit_info = None url_args = None query_params = {} branches = [] releases = [] browse_url = None visit_url = None branches_url = None releases_url = None swh_type = 'snapshot' if origin_url: swh_type = 'origin' origin_info = get_origin_info(origin_url, origin_type) visit_info = get_origin_visit(origin_info, timestamp, visit_id, snapshot_id) fmt_date = format_utc_iso_date(visit_info['date']) visit_info['fmt_date'] = fmt_date snapshot_id = visit_info['snapshot'] if not snapshot_id: raise NotFoundExc('No snapshot associated to the visit of origin ' '%s on %s' % (escape(origin_url), fmt_date)) # provided timestamp is not necessarily equals to the one # of the retrieved visit, so get the exact one in order # use it in the urls generated below if timestamp: timestamp = visit_info['date'] branches, releases = \ get_origin_visit_snapshot(origin_info, timestamp, visit_id, snapshot_id) url_args = {'origin_type': origin_type, 'origin_url': origin_info['url']} query_params = {'visit_id': visit_id} browse_url = reverse('browse-origin-visits', url_args=url_args) if timestamp: url_args['timestamp'] = format_utc_iso_date(timestamp, '%Y-%m-%dT%H:%M:%S') visit_url = reverse('browse-origin-directory', url_args=url_args, query_params=query_params) visit_info['url'] = visit_url branches_url = reverse('browse-origin-branches', url_args=url_args, query_params=query_params) releases_url = reverse('browse-origin-releases', url_args=url_args, query_params=query_params) elif snapshot_id: branches, releases = get_snapshot_content(snapshot_id) url_args = {'snapshot_id': snapshot_id} browse_url = reverse('browse-snapshot', url_args=url_args) branches_url = reverse('browse-snapshot-branches', url_args=url_args) releases_url = reverse('browse-snapshot-releases', url_args=url_args) releases = list(reversed(releases)) snapshot_size = service.lookup_snapshot_size(snapshot_id) is_empty = sum(snapshot_size.values()) == 0 swh_snp_id = persistent_identifier('snapshot', snapshot_id) return { 'swh_type': swh_type, 'swh_object_id': swh_snp_id, 'snapshot_id': snapshot_id, 'snapshot_size': snapshot_size, 'is_empty': is_empty, 'origin_info': origin_info, # keep track if the origin type was provided as url argument 'origin_type': origin_type, 'visit_info': visit_info, 'branches': branches, 'releases': releases, 'branch': None, 'release': None, 'browse_url': browse_url, 'branches_url': branches_url, 'releases_url': releases_url, 'url_args': url_args, 'query_params': query_params } # list of common readme names ordered by preference # (lower indices have higher priority) _common_readme_names = [ "readme.markdown", "readme.md", "readme.rst", "readme.txt", "readme" ] def get_readme_to_display(readmes): """ Process a list of readme files found in a directory in order to find the adequate one to display. Args: readmes: a list of dict where keys are readme file names and values are readme sha1s Returns: A tuple (readme_name, readme_sha1) """ readme_name = None readme_url = None readme_sha1 = None readme_html = None lc_readmes = {k.lower(): {'orig_name': k, 'sha1': v} for k, v in readmes.items()} # look for readme names according to the preference order # defined by the _common_readme_names list for common_readme_name in _common_readme_names: if common_readme_name in lc_readmes: readme_name = lc_readmes[common_readme_name]['orig_name'] readme_sha1 = lc_readmes[common_readme_name]['sha1'] readme_url = reverse('browse-content-raw', url_args={'query_string': readme_sha1}, query_params={'reencode': 'true'}) break # otherwise pick the first readme like file if any if not readme_name and len(readmes.items()) > 0: readme_name = next(iter(readmes)) readme_sha1 = readmes[readme_name] readme_url = reverse('browse-content-raw', url_args={'query_string': readme_sha1}, query_params={'reencode': 'true'}) # convert rst README to html server side as there is # no viable solution to perform that task client side if readme_name and readme_name.endswith('.rst'): cache_entry_id = 'readme_%s' % readme_sha1 cache_entry = cache.get(cache_entry_id) if cache_entry: readme_html = cache_entry else: try: rst_doc = request_content(readme_sha1) readme_html = pypandoc.convert_text(rst_doc['raw_data'], 'html', format='rst') cache.set(cache_entry_id, readme_html) except Exception: readme_html = 'Readme bytes are not available' return readme_name, readme_url, readme_html def get_swh_persistent_ids(swh_objects, snapshot_context=None): """ Returns a list of dict containing info related to persistent identifiers of swh objects. Args: swh_objects (list): a list of dict with the following keys: * type: swh object type (content/directory/release/revision/snapshot) * id: swh object id snapshot_context (dict): optional parameter describing the snapshot in which the object has been found Returns: list: a list of dict with the following keys: * object_type: the swh object type (content/directory/release/revision/snapshot) * object_icon: the swh object icon to use in HTML views * swh_id: the computed swh object persistent identifier * swh_id_url: the url resolving the persistent identifier * show_options: boolean indicating if the persistent id options must be displayed in persistent ids HTML view """ # noqa swh_ids = [] for swh_object in swh_objects: if not swh_object['id']: continue swh_id = get_swh_persistent_id(swh_object['type'], swh_object['id']) show_options = swh_object['type'] == 'content' or \ (snapshot_context and snapshot_context['origin_info'] is not None) object_icon = swh_object_icons[swh_object['type']] swh_ids.append({ 'object_type': swh_object['type'], 'object_icon': object_icon, 'swh_id': swh_id, 'swh_id_url': reverse('browse-swh-id', url_args={'swh_id': swh_id}), 'show_options': show_options }) return swh_ids diff --git a/swh/web/tests/api/views/test_snapshot.py b/swh/web/tests/api/views/test_snapshot.py index 4934900c..c755239e 100644 --- a/swh/web/tests/api/views/test_snapshot.py +++ b/swh/web/tests/api/views/test_snapshot.py @@ -1,173 +1,189 @@ # Copyright (C) 2018 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random from hypothesis import given from rest_framework.test import APITestCase +from swh.model.hashutil import hash_to_hex from swh.web.common.utils import reverse -from swh.web.tests.strategies import snapshot, unknown_snapshot +from swh.web.tests.strategies import ( + snapshot, unknown_snapshot, new_snapshot +) from swh.web.tests.testcase import WebTestCase class SnapshotApiTestCase(WebTestCase, APITestCase): @given(snapshot()) def test_api_snapshot(self, snapshot): url = reverse('api-snapshot', url_args={'snapshot_id': snapshot}) rv = self.client.get(url) self.assertEqual(rv.status_code, 200) self.assertEqual(rv['Content-Type'], 'application/json') expected_data = self.snapshot_get(snapshot) expected_data = self._enrich_snapshot(expected_data) self.assertEqual(rv.data, expected_data) @given(snapshot()) def test_api_snapshot_paginated(self, snapshot): branches_offset = 0 branches_count = 2 snapshot_branches = [] for k, v in sorted(self.snapshot_get(snapshot)['branches'].items()): snapshot_branches.append({ 'name': k, 'target_type': v['target_type'], 'target': v['target'] }) whole_snapshot = {'id': snapshot, 'branches': {}, 'next_branch': None} while branches_offset < len(snapshot_branches): branches_from = snapshot_branches[branches_offset]['name'] url = reverse('api-snapshot', url_args={'snapshot_id': snapshot}, query_params={'branches_from': branches_from, 'branches_count': branches_count}) rv = self.client.get(url) self.assertEqual(rv.status_code, 200) self.assertEqual(rv['Content-Type'], 'application/json') expected_data = self.snapshot_get_branches(snapshot, branches_from, branches_count) expected_data = self._enrich_snapshot(expected_data) branches_offset += branches_count if branches_offset < len(snapshot_branches): next_branch = snapshot_branches[branches_offset]['name'] expected_data['next_branch'] = next_branch else: expected_data['next_branch'] = None self.assertEqual(rv.data, expected_data) whole_snapshot['branches'].update(expected_data['branches']) if branches_offset < len(snapshot_branches): next_url = reverse( 'api-snapshot', url_args={'snapshot_id': snapshot}, query_params={'branches_from': next_branch, 'branches_count': branches_count}) self.assertEqual(rv['Link'], '<%s>; rel="next"' % next_url) else: self.assertFalse(rv.has_header('Link')) url = reverse('api-snapshot', url_args={'snapshot_id': snapshot}) rv = self.client.get(url) self.assertEqual(rv.status_code, 200) self.assertEqual(rv['Content-Type'], 'application/json') self.assertEqual(rv.data, whole_snapshot) @given(snapshot()) def test_api_snapshot_filtered(self, snapshot): snapshot_branches = [] for k, v in sorted(self.snapshot_get(snapshot)['branches'].items()): snapshot_branches.append({ 'name': k, 'target_type': v['target_type'], 'target': v['target'] }) target_type = random.choice(snapshot_branches)['target_type'] url = reverse('api-snapshot', url_args={'snapshot_id': snapshot}, query_params={'target_types': target_type}) rv = self.client.get(url) expected_data = self.snapshot_get_branches( snapshot, target_types=target_type) expected_data = self._enrich_snapshot(expected_data) self.assertEqual(rv.status_code, 200) self.assertEqual(rv['Content-Type'], 'application/json') self.assertEqual(rv.data, expected_data) @given(unknown_snapshot()) def test_api_snapshot_errors(self, unknown_snapshot): url = reverse('api-snapshot', url_args={'snapshot_id': '63ce369'}) rv = self.client.get(url) self.assertEqual(rv.status_code, 400) url = reverse('api-snapshot', url_args={'snapshot_id': unknown_snapshot}) rv = self.client.get(url) self.assertEqual(rv.status_code, 404) def _enrich_snapshot(self, snapshot): def _get_branch_url(target_type, target): url = None if target_type == 'revision': url = reverse('api-revision', url_args={'sha1_git': target}) if target_type == 'release': url = reverse('api-release', url_args={'sha1_git': target}) return url for branch in snapshot['branches'].keys(): target = snapshot['branches'][branch]['target'] target_type = snapshot['branches'][branch]['target_type'] snapshot['branches'][branch]['target_url'] = \ _get_branch_url(target_type, target) for branch in snapshot['branches'].keys(): target = snapshot['branches'][branch]['target'] target_type = snapshot['branches'][branch]['target_type'] if target_type == 'alias': if target in snapshot['branches']: snapshot['branches'][branch]['target_url'] = \ snapshot['branches'][target]['target_url'] else: snp = self.snapshot_get_branches(snapshot['id'], branches_from=target, branches_count=1) alias_target = snp['branches'][target]['target'] alias_target_type = snp['branches'][target]['target_type'] snapshot['branches'][branch]['target_url'] = \ _get_branch_url(alias_target_type, alias_target) return snapshot @given(snapshot()) def test_api_snapshot_uppercase(self, snapshot): url = reverse('api-snapshot-uppercase-checksum', url_args={'snapshot_id': snapshot.upper()}) resp = self.client.get(url) self.assertEqual(resp.status_code, 302) redirect_url = reverse('api-snapshot-uppercase-checksum', url_args={'snapshot_id': snapshot}) self.assertEqual(resp['location'], redirect_url) + + @given(new_snapshot(min_size=4)) + def test_api_snapshot_null_branch(self, new_snapshot): + snp_dict = new_snapshot.to_dict() + snp_id = hash_to_hex(snp_dict['id']) + for branch in snp_dict['branches'].keys(): + snp_dict['branches'][branch] = None + break + self.storage.snapshot_add([snp_dict]) + url = reverse('api-snapshot', + url_args={'snapshot_id': snp_id}) + rv = self.client.get(url) + self.assertEqual(rv.status_code, 200) diff --git a/swh/web/tests/browse/views/test_origin.py b/swh/web/tests/browse/views/test_origin.py index 0f327760..a4b65b95 100644 --- a/swh/web/tests/browse/views/test_origin.py +++ b/swh/web/tests/browse/views/test_origin.py @@ -1,870 +1,896 @@ # Copyright (C) 2017-2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random from unittest.mock import patch from django.utils.html import escape from hypothesis import given +from swh.model.hashutil import hash_to_bytes from swh.web.browse.utils import process_snapshot_branches from swh.web.common.exc import NotFoundExc from swh.web.common.utils import ( reverse, gen_path_info, format_utc_iso_date, parse_timestamp, get_swh_persistent_id ) from swh.web.tests.data import get_content from swh.web.tests.strategies import ( - origin, origin_with_multiple_visits + origin, origin_with_multiple_visits, new_origin, + new_snapshot, visit_dates, revisions ) from swh.web.tests.testcase import WebTestCase class SwhBrowseOriginTest(WebTestCase): @given(origin_with_multiple_visits()) def test_origin_visits_browse(self, origin): url = reverse('browse-origin-visits', url_args={'origin_type': origin['type'], 'origin_url': origin['url']}) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('origin-visits.html') url = reverse('browse-origin-visits', url_args={'origin_url': origin['url']}) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('origin-visits.html') visits = self.origin_visit_get(origin['id']) for v in visits: vdate = format_utc_iso_date(v['date'], '%Y-%m-%dT%H:%M:%SZ') browse_dir_url = reverse('browse-origin-directory', url_args={'origin_url': origin['url'], 'timestamp': vdate}) self.assertContains(resp, browse_dir_url) def origin_content_view_helper(self, origin_info, origin_visits, origin_branches, origin_releases, root_dir_sha1, content, visit_id=None, timestamp=None): content_path = '/'.join(content['path'].split('/')[1:]) url_args = {'origin_type': origin_info['type'], 'origin_url': origin_info['url'], 'path': content_path} if not visit_id: visit_id = origin_visits[-1]['visit'] query_params = {} if timestamp: url_args['timestamp'] = timestamp if visit_id: query_params['visit_id'] = visit_id url = reverse('browse-origin-content', url_args=url_args, query_params=query_params) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('content.html') self.assertContains(resp, '' % content['hljs_language']) self.assertContains(resp, escape(content['data'])) split_path = content_path.split('/') filename = split_path[-1] path = content_path.replace(filename, '')[:-1] path_info = gen_path_info(path) del url_args['path'] if timestamp: url_args['timestamp'] = \ format_utc_iso_date(parse_timestamp(timestamp).isoformat(), '%Y-%m-%dT%H:%M:%S') root_dir_url = reverse('browse-origin-directory', url_args=url_args, query_params=query_params) self.assertContains(resp, '
  • ', count=len(path_info)+1) self.assertContains(resp, '%s' % (root_dir_url, root_dir_sha1[:7])) for p in path_info: url_args['path'] = p['path'] dir_url = reverse('browse-origin-directory', url_args=url_args, query_params=query_params) self.assertContains(resp, '%s' % (dir_url, p['name'])) self.assertContains(resp, '
  • %s
  • ' % filename) query_string = 'sha1_git:' + content['sha1_git'] url_raw = reverse('browse-content-raw', url_args={'query_string': query_string}, query_params={'filename': filename}) self.assertContains(resp, url_raw) if 'args' in url_args: del url_args['path'] origin_branches_url = reverse('browse-origin-branches', url_args=url_args, query_params=query_params) self.assertContains(resp, 'Branches (%s)' % (origin_branches_url, len(origin_branches))) origin_releases_url = reverse('browse-origin-releases', url_args=url_args, query_params=query_params) self.assertContains(resp, 'Releases (%s)' % (origin_releases_url, len(origin_releases))) self.assertContains(resp, '
  • ', count=len(origin_branches)) url_args['path'] = content_path for branch in origin_branches: query_params['branch'] = branch['name'] root_dir_branch_url = reverse('browse-origin-content', url_args=url_args, query_params=query_params) self.assertContains(resp, '' % root_dir_branch_url) self.assertContains(resp, '
  • ', count=len(origin_releases)) query_params['branch'] = None for release in origin_releases: query_params['release'] = release['name'] root_dir_release_url = reverse('browse-origin-content', url_args=url_args, query_params=query_params) self.assertContains(resp, '' % root_dir_release_url) del url_args['origin_type'] url = reverse('browse-origin-content', url_args=url_args, query_params=query_params) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('content.html') swh_cnt_id = get_swh_persistent_id('content', content['sha1_git']) swh_cnt_id_url = reverse('browse-swh-id', url_args={'swh_id': swh_cnt_id}) self.assertContains(resp, swh_cnt_id) self.assertContains(resp, swh_cnt_id_url) self.assertContains(resp, 'swh-take-new-snapshot') @given(origin_with_multiple_visits()) def test_origin_content_view(self, origin): origin_visits = self.origin_visit_get(origin['id']) def _get_test_data(visit_idx): snapshot = self.snapshot_get(origin_visits[visit_idx]['snapshot']) head_rev_id = snapshot['branches']['HEAD']['target'] head_rev = self.revision_get(head_rev_id) dir_content = self.directory_ls(head_rev['directory']) dir_files = [e for e in dir_content if e['type'] == 'file'] dir_file = random.choice(dir_files) branches, releases = process_snapshot_branches(snapshot) return { 'branches': branches, 'releases': releases, 'root_dir_sha1': head_rev['directory'], 'content': get_content(dir_file['checksums']['sha1']), 'visit': origin_visits[visit_idx] } test_data = _get_test_data(-1) self.origin_content_view_helper(origin, origin_visits, test_data['branches'], test_data['releases'], test_data['root_dir_sha1'], test_data['content']) self.origin_content_view_helper(origin, origin_visits, test_data['branches'], test_data['releases'], test_data['root_dir_sha1'], test_data['content'], timestamp=test_data['visit']['date']) visit_unix_ts = parse_timestamp(test_data['visit']['date']).timestamp() visit_unix_ts = int(visit_unix_ts) self.origin_content_view_helper(origin, origin_visits, test_data['branches'], test_data['releases'], test_data['root_dir_sha1'], test_data['content'], timestamp=visit_unix_ts) test_data = _get_test_data(0) self.origin_content_view_helper(origin, origin_visits, test_data['branches'], test_data['releases'], test_data['root_dir_sha1'], test_data['content'], visit_id=test_data['visit']['visit']) def origin_directory_view_helper(self, origin_info, origin_visits, origin_branches, origin_releases, root_directory_sha1, directory_entries, visit_id=None, timestamp=None, path=None): dirs = [e for e in directory_entries if e['type'] in ('dir', 'rev')] files = [e for e in directory_entries if e['type'] == 'file'] if not visit_id: visit_id = origin_visits[-1]['visit'] url_args = {'origin_url': origin_info['url']} query_params = {} if timestamp: url_args['timestamp'] = timestamp else: query_params['visit_id'] = visit_id if path: url_args['path'] = path url = reverse('browse-origin-directory', url_args=url_args, query_params=query_params) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('directory.html') self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('directory.html') self.assertContains(resp, '', count=len(dirs)) self.assertContains(resp, '', count=len(files)) if timestamp: url_args['timestamp'] = \ format_utc_iso_date(parse_timestamp(timestamp).isoformat(), '%Y-%m-%dT%H:%M:%S') for d in dirs: if d['type'] == 'rev': dir_url = reverse('browse-revision', url_args={'sha1_git': d['target']}) else: dir_path = d['name'] if path: dir_path = "%s/%s" % (path, d['name']) dir_url_args = dict(url_args) dir_url_args['path'] = dir_path dir_url = reverse('browse-origin-directory', url_args=dir_url_args, query_params=query_params) self.assertContains(resp, dir_url) for f in files: file_path = f['name'] if path: file_path = "%s/%s" % (path, f['name']) file_url_args = dict(url_args) file_url_args['path'] = file_path file_url = reverse('browse-origin-content', url_args=file_url_args, query_params=query_params) self.assertContains(resp, file_url) if 'path' in url_args: del url_args['path'] root_dir_branch_url = \ reverse('browse-origin-directory', url_args=url_args, query_params=query_params) nb_bc_paths = 1 if path: nb_bc_paths = len(path.split('/')) + 1 self.assertContains(resp, '
  • ', count=nb_bc_paths) self.assertContains(resp, '%s' % (root_dir_branch_url, root_directory_sha1[:7])) origin_branches_url = reverse('browse-origin-branches', url_args=url_args, query_params=query_params) self.assertContains(resp, 'Branches (%s)' % (origin_branches_url, len(origin_branches))) origin_releases_url = reverse('browse-origin-releases', url_args=url_args, query_params=query_params) nb_releases = len(origin_releases) if nb_releases > 0: self.assertContains(resp, 'Releases (%s)' % (origin_releases_url, nb_releases)) if path: url_args['path'] = path self.assertContains(resp, '
  • ', count=len(origin_branches)) for branch in origin_branches: query_params['branch'] = branch['name'] root_dir_branch_url = \ reverse('browse-origin-directory', url_args=url_args, query_params=query_params) self.assertContains(resp, '' % root_dir_branch_url) self.assertContains(resp, '
  • ', count=len(origin_releases)) query_params['branch'] = None for release in origin_releases: query_params['release'] = release['name'] root_dir_release_url = \ reverse('browse-origin-directory', url_args=url_args, query_params=query_params) self.assertContains(resp, '' % root_dir_release_url) self.assertContains(resp, 'vault-cook-directory') self.assertContains(resp, 'vault-cook-revision') swh_dir_id = get_swh_persistent_id('directory', directory_entries[0]['dir_id']) # noqa swh_dir_id_url = reverse('browse-swh-id', url_args={'swh_id': swh_dir_id}) self.assertContains(resp, swh_dir_id) self.assertContains(resp, swh_dir_id_url) self.assertContains(resp, 'swh-take-new-snapshot') @given(origin()) def test_origin_root_directory_view(self, origin): origin_visits = self.origin_visit_get(origin['id']) visit = origin_visits[-1] snapshot = self.snapshot_get(visit['snapshot']) head_rev_id = snapshot['branches']['HEAD']['target'] head_rev = self.revision_get(head_rev_id) root_dir_sha1 = head_rev['directory'] dir_content = self.directory_ls(root_dir_sha1) branches, releases = process_snapshot_branches(snapshot) visit_unix_ts = parse_timestamp(visit['date']).timestamp() visit_unix_ts = int(visit_unix_ts) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content, visit_id=visit['visit']) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content, timestamp=visit_unix_ts) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content, timestamp=visit['date']) origin = dict(origin) del origin['type'] self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content, visit_id=visit['visit']) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content, timestamp=visit_unix_ts) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, dir_content, timestamp=visit['date']) @given(origin()) def test_origin_sub_directory_view(self, origin): origin_visits = self.origin_visit_get(origin['id']) visit = origin_visits[-1] snapshot = self.snapshot_get(visit['snapshot']) head_rev_id = snapshot['branches']['HEAD']['target'] head_rev = self.revision_get(head_rev_id) root_dir_sha1 = head_rev['directory'] subdirs = [e for e in self.directory_ls(root_dir_sha1) if e['type'] == 'dir'] branches, releases = process_snapshot_branches(snapshot) visit_unix_ts = parse_timestamp(visit['date']).timestamp() visit_unix_ts = int(visit_unix_ts) if len(subdirs) == 0: return subdir = random.choice(subdirs) subdir_content = self.directory_ls(subdir['target']) subdir_path = subdir['name'] self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, visit_id=visit['visit']) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, timestamp=visit_unix_ts) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, timestamp=visit['date']) origin = dict(origin) del origin['type'] self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, visit_id=visit['visit']) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, timestamp=visit_unix_ts) self.origin_directory_view_helper(origin, origin_visits, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, timestamp=visit['date']) def origin_branches_helper(self, origin_info, origin_snapshot): url_args = {'origin_type': origin_info['type'], 'origin_url': origin_info['url']} url = reverse('browse-origin-branches', url_args=url_args) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('branches.html') origin_branches = origin_snapshot[0] origin_releases = origin_snapshot[1] origin_branches_url = reverse('browse-origin-branches', url_args=url_args) self.assertContains(resp, 'Branches (%s)' % (origin_branches_url, len(origin_branches))) origin_releases_url = reverse('browse-origin-releases', url_args=url_args) nb_releases = len(origin_releases) if nb_releases > 0: self.assertContains(resp, 'Releases (%s)' % (origin_releases_url, nb_releases)) self.assertContains(resp, '' % escape(browse_branch_url)) browse_revision_url = reverse( 'browse-revision', url_args={'sha1_git': branch['revision']}, query_params={'origin_type': origin_info['type'], 'origin': origin_info['url']}) self.assertContains(resp, '' % escape(browse_revision_url)) @given(origin()) def test_origin_branches(self, origin): origin_visits = self.origin_visit_get(origin['id']) visit = origin_visits[-1] snapshot = self.snapshot_get(visit['snapshot']) snapshot_content = process_snapshot_branches(snapshot) self.origin_branches_helper(origin, snapshot_content) origin = dict(origin) origin['type'] = None self.origin_branches_helper(origin, snapshot_content) def origin_releases_helper(self, origin_info, origin_snapshot): url_args = {'origin_type': origin_info['type'], 'origin_url': origin_info['url']} url = reverse('browse-origin-releases', url_args=url_args) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('releases.html') origin_branches = origin_snapshot[0] origin_releases = origin_snapshot[1] origin_branches_url = reverse('browse-origin-branches', url_args=url_args) self.assertContains(resp, 'Branches (%s)' % (origin_branches_url, len(origin_branches))) origin_releases_url = reverse('browse-origin-releases', url_args=url_args) nb_releases = len(origin_releases) if nb_releases > 0: self.assertContains(resp, 'Releases (%s)' % (origin_releases_url, nb_releases)) self.assertContains(resp, '' % escape(browse_release_url)) self.assertContains(resp, '' % escape(browse_revision_url)) @given(origin()) def test_origin_releases(self, origin): origin_visits = self.origin_visit_get(origin['id']) visit = origin_visits[-1] snapshot = self.snapshot_get(visit['snapshot']) snapshot_content = process_snapshot_branches(snapshot) self.origin_releases_helper(origin, snapshot_content) origin = dict(origin) origin['type'] = None self.origin_releases_helper(origin, snapshot_content) + @given(new_origin(), new_snapshot(min_size=4, max_size=4), visit_dates(), + revisions(min_size=3, max_size=3)) + def test_origin_snapshot_null_branch(self, new_origin, new_snapshot, + visit_dates, revisions): + snp_dict = new_snapshot.to_dict() + new_origin = self.storage.origin_add([new_origin])[0] + for i, branch in enumerate(snp_dict['branches'].keys()): + if i == 0: + snp_dict['branches'][branch] = None + else: + snp_dict['branches'][branch]['target_type'] = 'revision' + snp_dict['branches'][branch]['target'] = hash_to_bytes( + revisions[i-1]) + self.storage.snapshot_add([snp_dict]) + visit = self.storage.origin_visit_add(new_origin['id'], visit_dates[0]) + self.storage.origin_visit_update(new_origin['id'], visit['visit'], + status='partial', + snapshot=snp_dict['id']) + + url = reverse('browse-origin-directory', + url_args={'origin_url': new_origin['url']}) + rv = self.client.get(url) + self.assertEqual(rv.status_code, 200) + @patch('swh.web.browse.views.utils.snapshot_context.request_content') @patch('swh.web.common.origin_visits.get_origin_visits') @patch('swh.web.browse.utils.get_origin_visit_snapshot') @patch('swh.web.browse.utils.service') @patch('swh.web.browse.views.origin.service') @patch('swh.web.browse.views.utils.snapshot_context.service') @patch('swh.web.browse.views.origin.get_origin_info') def test_origin_request_errors(self, mock_get_origin_info, mock_snapshot_service, mock_origin_service, mock_utils_service, mock_get_origin_visit_snapshot, mock_get_origin_visits, mock_request_content): mock_get_origin_info.side_effect = \ NotFoundExc('origin not found') url = reverse('browse-origin-visits', url_args={'origin_type': 'foo', 'origin_url': 'bar'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertContains(resp, 'origin not found', status_code=404) mock_utils_service.lookup_origin.side_effect = None mock_utils_service.lookup_origin.return_value = {'type': 'foo', 'url': 'bar', 'id': 457} mock_get_origin_visits.return_value = [] url = reverse('browse-origin-directory', url_args={'origin_type': 'foo', 'origin_url': 'bar'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertContains(resp, "No visit", status_code=404) mock_get_origin_visits.return_value = [{'visit': 1}] mock_get_origin_visit_snapshot.side_effect = \ NotFoundExc('visit not found') url = reverse('browse-origin-directory', url_args={'origin_type': 'foo', 'origin_url': 'bar'}, query_params={'visit_id': 2}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertRegex(resp.content.decode('utf-8'), 'Visit.*not found') mock_get_origin_visits.return_value = [{ 'date': '2015-09-26T09:30:52.373449+00:00', 'metadata': {}, 'origin': 457, 'snapshot': 'bdaf9ac436488a8c6cda927a0f44e172934d3f65', 'status': 'full', 'visit': 1 }] mock_get_origin_visit_snapshot.side_effect = None mock_get_origin_visit_snapshot.return_value = ( [{'directory': 'ae59ceecf46367e8e4ad800e231fc76adc3afffb', 'name': 'HEAD', 'revision': '7bc08e1aa0b08cb23e18715a32aa38517ad34672', 'date': '04 May 2017, 13:27 UTC', 'message': ''}], [] ) mock_utils_service.lookup_snapshot_size.return_value = { 'revision': 1, 'release': 0 } mock_utils_service.lookup_directory.side_effect = \ NotFoundExc('Directory not found') url = reverse('browse-origin-directory', url_args={'origin_type': 'foo', 'origin_url': 'bar'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertContains(resp, 'Directory not found', status_code=404) with patch('swh.web.browse.views.utils.snapshot_context.' 'get_snapshot_context') as mock_get_snapshot_context: mock_get_snapshot_context.side_effect = \ NotFoundExc('Snapshot not found') url = reverse('browse-origin-directory', url_args={'origin_type': 'foo', 'origin_url': 'bar'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertContains(resp, 'Snapshot not found', status_code=404) mock_origin_service.lookup_origin.side_effect = None mock_origin_service.lookup_origin.return_value = {'type': 'foo', 'url': 'bar', 'id': 457} mock_get_origin_visits.return_value = [] url = reverse('browse-origin-content', url_args={'origin_type': 'foo', 'origin_url': 'bar', 'path': 'foo'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertContains(resp, "No visit", status_code=404) mock_get_origin_visits.return_value = [{'visit': 1}] mock_get_origin_visit_snapshot.side_effect = \ NotFoundExc('visit not found') url = reverse('browse-origin-content', url_args={'origin_type': 'foo', 'origin_url': 'bar', 'path': 'foo'}, query_params={'visit_id': 2}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertRegex(resp.content.decode('utf-8'), 'Visit.*not found') mock_get_origin_visits.return_value = [{ 'date': '2015-09-26T09:30:52.373449+00:00', 'metadata': {}, 'origin': 457, 'snapshot': 'bdaf9ac436488a8c6cda927a0f44e172934d3f65', 'status': 'full', 'visit': 1 }] mock_get_origin_visit_snapshot.side_effect = None mock_get_origin_visit_snapshot.return_value = ([], []) url = reverse('browse-origin-content', url_args={'origin_type': 'foo', 'origin_url': 'bar', 'path': 'baz'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertRegex(resp.content.decode('utf-8'), 'Origin.*has an empty list of branches') mock_get_origin_visit_snapshot.return_value = ( [{'directory': 'ae59ceecf46367e8e4ad800e231fc76adc3afffb', 'name': 'HEAD', 'revision': '7bc08e1aa0b08cb23e18715a32aa38517ad34672', 'date': '04 May 2017, 13:27 UTC', 'message': ''}], [] ) mock_snapshot_service.lookup_directory_with_path.return_value = \ {'target': '5ecd9f37b7a2d2e9980d201acd6286116f2ba1f1'} mock_request_content.side_effect = \ NotFoundExc('Content not found') url = reverse('browse-origin-content', url_args={'origin_type': 'foo', 'origin_url': 'bar', 'path': 'baz'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 404) self.assertTemplateUsed('error.html') self.assertContains(resp, 'Content not found', status_code=404) @patch('swh.web.common.origin_visits.get_origin_visits') @patch('swh.web.browse.utils.get_origin_visit_snapshot') @patch('swh.web.browse.utils.service') def test_origin_empty_snapshot(self, mock_utils_service, mock_get_origin_visit_snapshot, mock_get_origin_visits): mock_get_origin_visits.return_value = [{ 'date': '2015-09-26T09:30:52.373449+00:00', 'metadata': {}, 'origin': 457, 'snapshot': 'bdaf9ac436488a8c6cda927a0f44e172934d3f65', 'status': 'full', 'visit': 1 }] mock_get_origin_visit_snapshot.return_value = ([], []) mock_utils_service.lookup_snapshot_size.return_value = { 'revision': 0, 'release': 0 } url = reverse('browse-origin-directory', url_args={'origin_type': 'foo', 'origin_url': 'bar'}) resp = self.client.get(url) self.assertEqual(resp.status_code, 200) self.assertTemplateUsed('content.html') self.assertRegex(resp.content.decode('utf-8'), 'snapshot.*is empty') diff --git a/swh/web/tests/strategies.py b/swh/web/tests/strategies.py index eff83f94..53dd4b84 100644 --- a/swh/web/tests/strategies.py +++ b/swh/web/tests/strategies.py @@ -1,536 +1,536 @@ # Copyright (C) 2018-2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random from collections import defaultdict from datetime import datetime from hypothesis import settings, assume from hypothesis.strategies import ( just, sampled_from, lists, composite, datetimes, integers, binary, text, characters ) from swh.model.hashutil import hash_to_hex, hash_to_bytes from swh.model.identifiers import directory_identifier from swh.storage.algos.revisions_walker import get_revisions_walker from swh.model.hypothesis_strategies import ( origins as new_origin_strategy, snapshots as new_snapshot ) from swh.web.tests.data import get_tests_data # Module dedicated to the generation of input data for tests through # the use of hypothesis. # Some of these data are sampled from a test archive created and populated # in the swh.web.tests.data module. # Set the swh-web hypothesis profile if none has been explicitly set hypothesis_default_settings = settings.get_profile('default') if repr(settings()) == repr(hypothesis_default_settings): settings.load_profile('swh-web') # The following strategies exploit the hypothesis capabilities def _filter_checksum(cs): generated_checksums = get_tests_data()['generated_checksums'] if not int.from_bytes(cs, byteorder='little') or \ cs in generated_checksums: return False generated_checksums.add(cs) return True def _known_swh_object(object_type): return sampled_from(get_tests_data()[object_type]) def sha1(): """ Hypothesis strategy returning a valid hexadecimal sha1 value. """ return binary( min_size=20, max_size=20).filter(_filter_checksum).map(hash_to_hex) def invalid_sha1(): """ Hypothesis strategy returning an invalid sha1 representation. """ return binary( min_size=50, max_size=50).filter(_filter_checksum).map(hash_to_hex) def sha256(): """ Hypothesis strategy returning a valid hexadecimal sha256 value. """ return binary( min_size=32, max_size=32).filter(_filter_checksum).map(hash_to_hex) def content(): """ Hypothesis strategy returning a random content ingested into the test archive. """ return _known_swh_object('contents') def contents(): """ Hypothesis strategy returning random contents ingested into the test archive. """ return lists(content(), min_size=2, max_size=8) def content_text(): """ Hypothesis strategy returning random textual contents ingested into the test archive. """ return content().filter(lambda c: c['mimetype'].startswith('text/')) def content_text_non_utf8(): """ Hypothesis strategy returning random textual contents not encoded to UTF-8 ingested into the test archive. """ return content().filter(lambda c: c['mimetype'].startswith('text/') and c['encoding'] not in ('utf-8', 'us-ascii')) def content_text_no_highlight(): """ Hypothesis strategy returning random textual contents with no detected programming language to highlight ingested into the test archive. """ return content().filter(lambda c: c['mimetype'].startswith('text/') and c['hljs_language'] == 'nohighlight') def content_image_type(): """ Hypothesis strategy returning random image contents ingested into the test archive. """ return content().filter(lambda c: c['mimetype'].startswith('image/')) @composite def new_content(draw): blake2s256_hex = draw(sha256()) sha1_hex = draw(sha1()) sha1_git_hex = draw(sha1()) sha256_hex = draw(sha256()) assume(sha1_hex != sha1_git_hex) assume(blake2s256_hex != sha256_hex) return { 'blake2S256': blake2s256_hex, 'sha1': sha1_hex, 'sha1_git': sha1_git_hex, 'sha256': sha256_hex } def unknown_content(): """ Hypothesis strategy returning a random content not ingested into the test archive. """ return new_content().filter( lambda c: next(get_tests_data()['storage'].content_get( [hash_to_bytes(c['sha1'])])) is None) def unknown_contents(): """ Hypothesis strategy returning random contents not ingested into the test archive. """ return lists(unknown_content(), min_size=2, max_size=8) def directory(): """ Hypothesis strategy returning a random directory ingested into the test archive. """ return _known_swh_object('directories') def directory_with_subdirs(): """ Hypothesis strategy returning a random directory containing sub directories ingested into the test archive. """ storage = get_tests_data()['storage'] return directory().filter( lambda d: any([e['type'] == 'dir' for e in list(storage.directory_ls(hash_to_bytes(d)))])) def empty_directory(): """ Hypothesis strategy returning the empty directory ingested into the test archive. """ return just(directory_identifier({'entries': []})) def unknown_directory(): """ Hypothesis strategy returning a random directory not ingested into the test archive. """ storage = get_tests_data()['storage'] return sha1().filter( lambda s: len(list(storage.directory_missing([hash_to_bytes(s)]))) > 0) def origin(): """ Hypothesis strategy returning a random origin ingested into the test archive. """ return _known_swh_object('origins') def origin_with_multiple_visits(): """ Hypothesis strategy returning a random origin ingested into the test archive. """ ret = [] tests_data = get_tests_data() for origin in tests_data['origins']: visits = list(tests_data['storage'].origin_visit_get(origin['id'])) if len(visits) > 1: ret.append(origin) return sampled_from(ret) def origin_with_release(): """ Hypothesis strategy returning a random origin ingested into the test archive. """ ret = [] tests_data = get_tests_data() for origin in tests_data['origins']: snapshot = tests_data['storage'].snapshot_get_latest(origin['id']) if any([b['target_type'] == 'release' for b in snapshot['branches'].values()]): ret.append(origin) return sampled_from(ret) def unknown_origin_id(): """ Hypothesis strategy returning a random origin id not ingested into the test archive. """ return integers(min_value=1000000) def new_origin(): """ Hypothesis strategy returning a random origin not ingested into the test archive. """ storage = get_tests_data()['storage'] return new_origin_strategy().map(lambda origin: origin.to_dict()).filter( lambda origin: storage.origin_get([origin])[0] is None) def new_origins(nb_origins=None): """ Hypothesis strategy returning random origins not ingested into the test archive. """ min_size = nb_origins if nb_origins is not None else 2 max_size = nb_origins if nb_origins is not None else 8 size = random.randint(min_size, max_size) return lists(new_origin(), min_size=size, max_size=size, unique_by=lambda o: tuple(sorted(o.items()))) def visit_dates(nb_dates=None): """ Hypothesis strategy returning a list of visit dates. """ min_size = nb_dates if nb_dates else 2 max_size = nb_dates if nb_dates else 8 return lists(datetimes(min_value=datetime(2015, 1, 1, 0, 0), max_value=datetime(2018, 12, 31, 0, 0)), min_size=min_size, max_size=max_size, unique=True).map(sorted) def release(): """ Hypothesis strategy returning a random release ingested into the test archive. """ return _known_swh_object('releases') def unknown_release(): """ Hypothesis strategy returning a random revision not ingested into the test archive. """ return sha1().filter( lambda s: next(get_tests_data()['storage'].release_get([s])) is None) def revision(): """ Hypothesis strategy returning a random revision ingested into the test archive. """ return _known_swh_object('revisions') def unknown_revision(): """ Hypothesis strategy returning a random revision not ingested into the test archive. """ storage = get_tests_data()['storage'] return sha1().filter( lambda s: next(storage.revision_get([hash_to_bytes(s)])) is None) @composite def new_person(draw): """ Hypothesis strategy returning random raw swh person data. """ name = draw(text(min_size=5, max_size=30, alphabet=characters(min_codepoint=0, max_codepoint=255))) email = '%s@company.org' % name return { 'name': name.encode(), 'email': email.encode(), 'fullname': ('%s <%s>' % (name, email)).encode() } @composite def new_swh_date(draw): """ Hypothesis strategy returning random raw swh date data. """ timestamp = draw( datetimes(min_value=datetime(2015, 1, 1, 0, 0), max_value=datetime(2018, 12, 31, 0, 0)).map( lambda d: int(d.timestamp()))) return { 'timestamp': timestamp, 'offset': 0, 'negative_utc': False, } @composite def new_revision(draw): """ Hypothesis strategy returning random raw swh revision data not ingested into the test archive. """ return { 'id': draw(unknown_revision().map(hash_to_bytes)), 'directory': draw(sha1().map(hash_to_bytes)), 'author': draw(new_person()), 'committer': draw(new_person()), 'message': draw( text(min_size=20, max_size=100).map(lambda t: t.encode())), 'date': draw(new_swh_date()), 'committer_date': draw(new_swh_date()), 'synthetic': False, 'type': 'git', 'parents': [], 'metadata': [], } -def revisions(): +def revisions(min_size=2, max_size=8): """ Hypothesis strategy returning random revisions ingested into the test archive. """ - return lists(revision(), min_size=2, max_size=8) + return lists(revision(), min_size=min_size, max_size=max_size) -def unknown_revisions(): +def unknown_revisions(min_size=2, max_size=8): """ Hypothesis strategy returning random revisions not ingested into the test archive. """ - return lists(unknown_revision(), min_size=2, max_size=8) + return lists(unknown_revision(), min_size=min_size, max_size=max_size) def snapshot(): """ Hypothesis strategy returning a random snapshot ingested into the test archive. """ return _known_swh_object('snapshots') def new_snapshots(nb_snapshots=None): min_size = nb_snapshots if nb_snapshots else 2 max_size = nb_snapshots if nb_snapshots else 8 return lists(new_snapshot(min_size=2, max_size=10, only_objects=True) .map(lambda snp: snp.to_dict()), min_size=min_size, max_size=max_size) def unknown_snapshot(): """ Hypothesis strategy returning a random revision not ingested into the test archive. """ storage = get_tests_data()['storage'] return sha1().filter( lambda s: storage.snapshot_get(hash_to_bytes(s)) is None) def person(): """ Hypothesis strategy returning a random person ingested into the test archive. """ return _known_swh_object('persons') def unknown_person(): """ Hypothesis strategy returning a random person not ingested into the test archive. """ return integers(min_value=1000000) def _get_origin_dfs_revisions_walker(): tests_data = get_tests_data() storage = tests_data['storage'] origin = random.choice(tests_data['origins'][:-1]) snapshot = storage.snapshot_get_latest(origin['id']) head = snapshot['branches'][b'HEAD']['target'] return get_revisions_walker('dfs', storage, head) def ancestor_revisions(): """ Hypothesis strategy returning a pair of revisions ingested into the test archive with an ancestor relation. """ # get a dfs revisions walker for one of the origins # loaded into the test archive revisions_walker = _get_origin_dfs_revisions_walker() master_revisions = [] children = defaultdict(list) init_rev_found = False # get revisions only authored in the master branch for rev in revisions_walker: for rev_p in rev['parents']: children[rev_p].append(rev['id']) if not init_rev_found: master_revisions.append(rev) if not rev['parents']: init_rev_found = True # head revision root_rev = master_revisions[0] # pick a random revision, different from head, only authored # in the master branch ancestor_rev_idx = random.choice(list(range(1, len(master_revisions)-1))) ancestor_rev = master_revisions[ancestor_rev_idx] ancestor_child_revs = children[ancestor_rev['id']] return just({ 'sha1_git_root': hash_to_hex(root_rev['id']), 'sha1_git': hash_to_hex(ancestor_rev['id']), 'children': [hash_to_hex(r) for r in ancestor_child_revs] }) def non_ancestor_revisions(): """ Hypothesis strategy returning a pair of revisions ingested into the test archive with no ancestor relation. """ # get a dfs revisions walker for one of the origins # loaded into the test archive revisions_walker = _get_origin_dfs_revisions_walker() merge_revs = [] children = defaultdict(list) # get all merge revisions for rev in revisions_walker: if len(rev['parents']) > 1: merge_revs.append(rev) for rev_p in rev['parents']: children[rev_p].append(rev['id']) # find a merge revisions whose parents have a unique child revision random.shuffle(merge_revs) selected_revs = None for merge_rev in merge_revs: if all(len(children[rev_p]) == 1 for rev_p in merge_rev['parents']): selected_revs = merge_rev['parents'] return just({ 'sha1_git_root': hash_to_hex(selected_revs[0]), 'sha1_git': hash_to_hex(selected_revs[1]) }) # The following strategies returns data specific to some tests # that can not be generated and thus are hardcoded. def contents_with_ctags(): """ Hypothesis strategy returning contents ingested into the test archive. Those contents are ctags compatible, that is running ctags on those lay results. """ return just({ 'sha1s': ['0ab37c02043ebff946c1937523f60aadd0844351', '15554cf7608dde6bfefac7e3d525596343a85b6f', '2ce837f1489bdfb8faf3ebcc7e72421b5bea83bd', '30acd0b47fc25e159e27a980102ddb1c4bea0b95', '4f81f05aaea3efb981f9d90144f746d6b682285b', '5153aa4b6e4455a62525bc4de38ed0ff6e7dd682', '59d08bafa6a749110dfb65ba43a61963d5a5bf9f', '7568285b2d7f31ae483ae71617bd3db873deaa2c', '7ed3ee8e94ac52ba983dd7690bdc9ab7618247b4', '8ed7ef2e7ff9ed845e10259d08e4145f1b3b5b03', '9b3557f1ab4111c8607a4f2ea3c1e53c6992916c', '9c20da07ed14dc4fcd3ca2b055af99b2598d8bdd', 'c20ceebd6ec6f7a19b5c3aebc512a12fbdc9234b', 'e89e55a12def4cd54d5bff58378a3b5119878eb7', 'e8c0654fe2d75ecd7e0b01bee8a8fc60a130097e', 'eb6595e559a1d34a2b41e8d4835e0e4f98a5d2b5'], 'symbol_name': 'ABS' }) def revision_with_submodules(): """ Hypothesis strategy returning a revision that is known to point to a directory with revision entries (aka git submodule) """ return just({ 'rev_sha1_git': 'ffcb69001f3f6745dfd5b48f72ab6addb560e234', 'rev_dir_sha1_git': 'd92a21446387fa28410e5a74379c934298f39ae2', 'rev_dir_rev_path': 'libtess2' })