diff --git a/swh/web/api/views/snapshot.py b/swh/web/api/views/snapshot.py index 43d56197..fc39631e 100644 --- a/swh/web/api/views/snapshot.py +++ b/swh/web/api/views/snapshot.py @@ -1,107 +1,121 @@ # 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') @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'] + 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+1, target_types, notfound_msg='Snapshot with id {} not found.'.format(snapshot_id), enrich_fn=_enrich_snapshot) next_branch = None if len(results['branches']) > branches_count: next_branch = sorted(results['branches'].keys())[-1] del results['branches'][next_branch] response = {'results': results, 'headers': {}} if next_branch: response['headers']['link-next'] = \ reverse('api-snapshot', url_args={'snapshot_id': snapshot_id}, query_params={'branches_from': 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 7e753153..2d37f35c 100644 --- a/swh/web/browse/utils.py +++ b/swh/web/browse/utils.py @@ -1,1178 +1,1209 @@ # Copyright (C) 2017-2018 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 base64 from collections import defaultdict import magic import math import pypandoc import stat import textwrap from django.core.cache import cache from django.utils.safestring import mark_safe 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.utils import ( reverse, format_utc_iso_date, parse_timestamp, get_origin_visits, 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 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 # encode textual content to utf-8 if needed if reencode and 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['raw_data'] = \ content_data['raw_data'].decode('utf-8', 'replace')\ .encode('utf-8') elif 'ascii' not in encoding and encoding not in ['utf-8', 'binary']: # noqa content_data['raw_data'] = \ content_data['raw_data'].decode(encoding, 'replace')\ .encode('utf-8') elif reencode and 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['raw_data'] = \ content_data['raw_data'].decode(encoding)\ .encode('utf-8') except Exception: pass else: # ensure display in content view mimetype = 'text/plain' break 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 get_origin_visit(origin_info, visit_ts=None, visit_id=None, snapshot_id=None): """Function that returns information about a visit for a given origin. The visit is retrieved from a provided timestamp. The closest visit from that timestamp is selected. 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 Returns: A dict containing the visit info as described below:: {'origin': 2, 'date': '2017-10-08T11:54:25.582463+00:00', 'metadata': {}, 'visit': 25, 'status': 'full'} """ visits = get_origin_visits(origin_info) if not visits: raise NotFoundExc('No visit associated to origin with' ' type %s and url %s!' % (origin_info['type'], origin_info['url'])) if snapshot_id: visit = [v for v in visits if v['snapshot'] == snapshot_id] if len(visit) == 0: raise NotFoundExc( 'Visit for snapshot with id %s for origin with type %s' ' and url %s not found!' % (snapshot_id, origin_info['type'], origin_info['url'])) return visit[0] if visit_id: visit = [v for v in visits if v['visit'] == int(visit_id)] if len(visit) == 0: raise NotFoundExc( 'Visit with id %s for origin with type %s' ' and url %s not found!' % (visit_id, origin_info['type'], origin_info['url'])) return visit[0] if not visit_ts: # returns the latest full visit when no timestamp is provided for v in reversed(visits): if v['status'] == 'full': return v return visits[-1] parsed_visit_ts = math.floor(parse_timestamp(visit_ts).timestamp()) visit_idx = None for i, visit in enumerate(visits): ts = math.floor(parse_timestamp(visit['date']).timestamp()) if i == 0 and parsed_visit_ts <= ts: return visit elif i == len(visits) - 1: if parsed_visit_ts >= ts: return visit else: next_ts = math.floor( parse_timestamp(visits[i+1]['date']).timestamp()) if parsed_visit_ts >= ts and parsed_visit_ts < next_ts: if (parsed_visit_ts - ts) < (next_ts - parsed_visit_ts): visit_idx = i break else: visit_idx = i+1 break if visit_idx is not None: visit = visits[visit_idx] while visit_idx < len(visits) - 1 and \ visit['date'] == visits[visit_idx+1]['date']: visit_idx = visit_idx + 1 visit = visits[visit_idx] return visit else: raise NotFoundExc( 'Visit with timestamp %s for origin with type %s and url %s not found!' % # noqa (visit_ts, origin_info['type'], origin_info['url'])) -def process_snapshot_branches(snapshot_branches): +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 - # FIXME: handle branch aliases + + 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: - 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'], - } + _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 - revision_data = { - 'directory': revision['directory'], - 'date': format_utc_iso_date(revision['date']), - 'message': revision['message'], - } for branch in revision_to_branch[revision['id']]: - branches[branch].update(revision_data) + _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']: + + 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['branches']) + 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={}): """ 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 = ' ' for k, v in link_attrs.items(): attrs += '%s="%s" ' % (k, v) if not link_text: link_text = url link = '%s' % (attrs, url, link_text) return mark_safe(link) def gen_person_link(person_id, person_name, snapshot_context=None, link_attrs={}): """ 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 = None if snapshot_context and snapshot_context['origin_info']: origin_info = snapshot_context['origin_info'] query_params = {'origin_type': origin_info['type'], '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']} 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 = None if snapshot_context and snapshot_context['origin_info']: origin_info = snapshot_context['origin_info'] origin_type = snapshot_context['origin_type'] query_params = {'origin_type': origin_type, '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 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=None, link_attrs={}): """ 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_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_origin_link(origin_info, link_attrs={}): """ Utility function for generating a link to a software origin HTML view to insert in Django templates. Args: origin_info (dict): a dict filled with origin information (id, type, url) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'Origin: origin_url' """ # noqa origin_browse_url = reverse('browse-origin', url_args={'origin_type': origin_info['type'], 'origin_url': origin_info['url']}) return gen_link(origin_browse_url, 'Origin: ' + origin_info['url'], link_attrs) def gen_directory_link(sha1_git, link_text=None, link_attrs={}): """ 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 generated url 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 directory_url = reverse('browse-directory', url_args={'sha1_git': sha1_git}) if not link_text: link_text = directory_url return gen_link(directory_url, link_text, link_attrs) def gen_snapshot_link(snapshot_id, link_text=None, link_attrs={}): """ 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 generated url 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' """ snapshot_url = reverse('browse-snapshot', url_args={'snapshot_id': snapshot_id}) if not link_text: link_text = snapshot_url return gen_link(snapshot_url, link_text, link_attrs) def gen_snapshot_directory_link(snapshot_context, revision_id=None, link_text=None, link_attrs={}): """ Utility function for generating a link to a directory HTML view in the context of a snapshot to insert in Django templates. Args: snapshot_context (dict): the snapshot information revision_id (str): optional revision identifier in order to use the associated directory link_text (str): optional text to use for the generated link link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'origin_directory_view_url' """ query_params = {'revision': revision_id} if 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'] directory_url = reverse('browse-origin-directory', url_args=url_args, query_params=query_params) else: url_args = {'snapshot_id': snapshot_context['snapshot_id']} directory_url = reverse('browse-snapshot-directory', url_args=url_args, query_params=query_params) if not link_text: link_text = directory_url return gen_link(directory_url, link_text, link_attrs) def gen_content_link(sha1_git, link_text=None, link_attrs={}): """ 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 generated url 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 content_url = reverse('browse-content', url_args={'query_string': 'sha1_git:' + sha1_git}) if not link_text: link_text = content_url 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=None, link_attrs={}): """ 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 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_log_url return gen_link(revision_log_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 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'] 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!' % 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' % (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}) 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}) # 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/browse/views/utils/snapshot_context.py b/swh/web/browse/views/utils/snapshot_context.py index cef89601..bdfe2c84 100644 --- a/swh/web/browse/views/utils/snapshot_context.py +++ b/swh/web/browse/views/utils/snapshot_context.py @@ -1,945 +1,947 @@ # 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 # Utility module implementing Django views for browsing the archive # in a snapshot context. # Its purpose is to factorize code for the views reachable from the # /origin/.* and /snapshot/.* endpoints. from django.shortcuts import render, redirect from django.template.defaultfilters import filesizeformat from swh.model.identifiers import snapshot_identifier from swh.web.browse.utils import ( get_snapshot_context, get_directory_entries, gen_directory_link, gen_revision_link, request_content, gen_content_link, prepare_content_for_display, content_display_max_size, format_log_entries, gen_revision_log_link, gen_link, get_readme_to_display, get_swh_persistent_ids, process_snapshot_branches ) from swh.web.common import service from swh.web.common.exc import ( handle_view_exception, NotFoundExc ) from swh.web.common.utils import ( reverse, gen_path_info, format_utc_iso_date, swh_object_icons ) _empty_snapshot_id = snapshot_identifier({'branches': {}}) def _get_branch(branches, branch_name, snapshot_id): """ Utility function to get a specific branch from a branches list. Its purpose is to get the default HEAD branch as some software origin (e.g those with svn type) does not have it. In that latter case, check if there is a master branch instead and returns it. """ filtered_branches = \ [b for b in branches if b['name'].endswith(branch_name)] if len(filtered_branches) > 0: return filtered_branches[0] elif branch_name == 'HEAD': filtered_branches = \ [b for b in branches if b['name'].endswith('master')] if len(filtered_branches) > 0: return filtered_branches[0] elif len(branches) > 0: return branches[0] else: # case where a large branches list has been truncated - snp_branch = service.lookup_snapshot(snapshot_id, - branches_from=branch_name, - branches_count=1, - target_types=['revision']) - snp_branch, _ = process_snapshot_branches(snp_branch['branches']) + snp = service.lookup_snapshot(snapshot_id, + branches_from=branch_name, + branches_count=1, + target_types=['revision', 'alias']) + snp_branch, _ = process_snapshot_branches(snp) if snp_branch: branches.append(snp_branch[0]) return snp_branch[0] return None def _get_release(releases, release_name): """ Utility function to get a specific release from a releases list. Returns None if the release can not be found in the list. """ filtered_releases = \ [r for r in releases if r['name'] == release_name] if len(filtered_releases) > 0: return filtered_releases[0] else: return None def _branch_not_found(branch_type, branch, branches, snapshot_id=None, origin_info=None, timestamp=None, visit_id=None): """ Utility function to raise an exception when a specified branch/release can not be found. """ if branch_type == 'branch': branch_type = 'Branch' branch_type_plural = 'branches' else: branch_type = 'Release' branch_type_plural = 'releases' if snapshot_id and len(branches) == 0: msg = 'Snapshot with id %s has an empty list' \ ' of %s!' % (snapshot_id, branch_type_plural) elif snapshot_id: msg = '%s %s for snapshot with id %s' \ ' not found!' % (branch_type, branch, snapshot_id) elif visit_id and len(branches) == 0: msg = 'Origin with type %s and url %s' \ ' for visit with id %s has an empty list' \ ' of %s!' % (origin_info['type'], origin_info['url'], visit_id, branch_type_plural) elif visit_id: msg = '%s %s associated to visit with' \ ' id %s for origin with type %s and url %s' \ ' not found!' % (branch_type, branch, visit_id, origin_info['type'], origin_info['url']) elif len(branches) == 0: msg = 'Origin with type %s and url %s' \ ' for visit with timestamp %s has an empty list' \ ' of %s!' % (origin_info['type'], origin_info['url'], timestamp, branch_type_plural) else: msg = '%s %s associated to visit with' \ ' timestamp %s for origin with type %s' \ ' and url %s not found!' % (branch_type, branch, timestamp, origin_info['type'], origin_info['url']) raise NotFoundExc(msg) def _process_snapshot_request(request, snapshot_id=None, origin_type=None, origin_url=None, timestamp=None, path=None, browse_context='directory'): """ Utility function to perform common input request processing for snapshot context views. """ visit_id = request.GET.get('visit_id', None) snapshot_context = get_snapshot_context(snapshot_id, origin_type, origin_url, timestamp, visit_id) swh_type = snapshot_context['swh_type'] origin_info = snapshot_context['origin_info'] branches = snapshot_context['branches'] releases = snapshot_context['releases'] url_args = snapshot_context['url_args'] query_params = snapshot_context['query_params'] if snapshot_context['visit_info']: timestamp = format_utc_iso_date(snapshot_context['visit_info']['date'], '%Y-%m-%dT%H:%M:%SZ') snapshot_context['timestamp'] = \ format_utc_iso_date(snapshot_context['visit_info']['date']) browse_view_name = 'browse-' + swh_type + '-' + browse_context root_sha1_git = None revision_id = request.GET.get('revision', None) release_name = request.GET.get('release', None) release_id = None branch_name = None snapshot_total_size = sum(snapshot_context['snapshot_size'].values()) if snapshot_total_size and revision_id: revision = service.lookup_revision(revision_id) root_sha1_git = revision['directory'] branches.append({'name': revision_id, 'revision': revision_id, 'directory': root_sha1_git, 'url': None}) branch_name = revision_id query_params['revision'] = revision_id elif snapshot_total_size and release_name: release = _get_release(releases, release_name) try: root_sha1_git = release['directory'] revision_id = release['target'] release_id = release['id'] query_params['release'] = release_name except Exception: _branch_not_found("release", release_name, releases, snapshot_id, origin_info, timestamp, visit_id) elif snapshot_total_size: branch_name = request.GET.get('branch', None) if branch_name: query_params['branch'] = branch_name branch = _get_branch(branches, branch_name or 'HEAD', snapshot_context['snapshot_id']) try: branch_name = branch['name'] revision_id = branch['revision'] root_sha1_git = branch['directory'] except Exception: _branch_not_found("branch", branch_name, branches, snapshot_id, origin_info, timestamp, visit_id) for b in branches: branch_url_args = dict(url_args) branch_query_params = dict(query_params) if 'release' in branch_query_params: del branch_query_params['release'] branch_query_params['branch'] = b['name'] if path: b['path'] = path branch_url_args['path'] = path b['url'] = reverse(browse_view_name, url_args=branch_url_args, query_params=branch_query_params) for r in releases: release_url_args = dict(url_args) release_query_params = dict(query_params) if 'branch' in release_query_params: del release_query_params['branch'] release_query_params['release'] = r['name'] if path: r['path'] = path release_url_args['path'] = path r['url'] = reverse(browse_view_name, url_args=release_url_args, query_params=release_query_params) snapshot_context['query_params'] = query_params snapshot_context['root_sha1_git'] = root_sha1_git snapshot_context['revision_id'] = revision_id snapshot_context['branch'] = branch_name snapshot_context['release'] = release_name snapshot_context['release_id'] = release_id return snapshot_context def browse_snapshot_directory(request, snapshot_id=None, origin_type=None, origin_url=None, timestamp=None, path=None): """ Django view implementation for browsing a directory in a snapshot context. """ try: snapshot_context = _process_snapshot_request(request, snapshot_id, origin_type, origin_url, timestamp, path, browse_context='directory') # noqa root_sha1_git = snapshot_context['root_sha1_git'] sha1_git = root_sha1_git if root_sha1_git and path: dir_info = service.lookup_directory_with_path(root_sha1_git, path) # some readme files can reference assets reachable from the # browsed directory, handle that special case in order to # correctly displayed them if dir_info and dir_info['type'] == 'file': file_raw_url = reverse( 'browse-content-raw', url_args={'query_string': dir_info['checksums']['sha1']}) return redirect(file_raw_url) sha1_git = dir_info['target'] dirs = [] files = [] if sha1_git: dirs, files = get_directory_entries(sha1_git) except Exception as exc: return handle_view_exception(request, exc) swh_type = snapshot_context['swh_type'] origin_info = snapshot_context['origin_info'] visit_info = snapshot_context['visit_info'] url_args = snapshot_context['url_args'] query_params = snapshot_context['query_params'] revision_id = snapshot_context['revision_id'] snapshot_id = snapshot_context['snapshot_id'] path_info = gen_path_info(path) browse_view_name = 'browse-' + swh_type + '-directory' breadcrumbs = [] if root_sha1_git: breadcrumbs.append({'name': root_sha1_git[:7], 'url': reverse(browse_view_name, url_args=url_args, query_params=query_params)}) for pi in path_info: bc_url_args = dict(url_args) bc_url_args['path'] = pi['path'] breadcrumbs.append({'name': pi['name'], 'url': reverse(browse_view_name, url_args=bc_url_args, query_params=query_params)}) path = '' if path is None else (path + '/') for d in dirs: if d['type'] == 'rev': d['url'] = reverse('browse-revision', url_args={'sha1_git': d['target']}) else: bc_url_args = dict(url_args) bc_url_args['path'] = path + d['name'] d['url'] = reverse(browse_view_name, url_args=bc_url_args, query_params=query_params) sum_file_sizes = 0 readmes = {} browse_view_name = 'browse-' + swh_type + '-content' for f in files: bc_url_args = dict(url_args) bc_url_args['path'] = path + f['name'] f['url'] = reverse(browse_view_name, url_args=bc_url_args, query_params=query_params) if f['length'] is not None: sum_file_sizes += f['length'] f['length'] = filesizeformat(f['length']) if f['name'].lower().startswith('readme'): readmes[f['name']] = f['checksums']['sha1'] readme_name, readme_url, readme_html = get_readme_to_display(readmes) browse_view_name = 'browse-' + swh_type + '-log' history_url = None if snapshot_id != _empty_snapshot_id: history_url = reverse(browse_view_name, url_args=url_args, query_params=query_params) nb_files = None nb_dirs = None sum_file_sizes = None dir_path = None if root_sha1_git: nb_files = len(files) nb_dirs = len(dirs) sum_file_sizes = filesizeformat(sum_file_sizes) dir_path = '/' + path browse_dir_link = \ gen_directory_link(sha1_git, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) browse_rev_link = \ gen_revision_link(revision_id, snapshot_context=snapshot_context, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) dir_metadata = {'id': sha1_git, 'context-independent directory': browse_dir_link, 'number of regular files': nb_files, 'number of subdirectories': nb_dirs, 'sum of regular file sizes': sum_file_sizes, 'path': dir_path, 'revision id': revision_id, 'revision': browse_rev_link, 'snapshot id': snapshot_id} if origin_info: dir_metadata['origin id'] = origin_info['id'] dir_metadata['origin type'] = origin_info['type'] dir_metadata['origin url'] = origin_info['url'] dir_metadata['origin visit date'] = format_utc_iso_date(visit_info['date']) # noqa dir_metadata['origin visit id'] = visit_info['visit'] snapshot_context_url = reverse('browse-snapshot-directory', url_args={'snapshot_id': snapshot_id}, query_params=request.GET) browse_snapshot_link = \ gen_link(snapshot_context_url, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) dir_metadata['snapshot context'] = browse_snapshot_link vault_cooking = { 'directory_context': True, 'directory_id': sha1_git, 'revision_context': True, 'revision_id': revision_id } swh_objects = [{'type': 'directory', 'id': sha1_git}, {'type': 'revision', 'id': revision_id}, {'type': 'snapshot', 'id': snapshot_id}] release_id = snapshot_context['release_id'] if release_id: swh_objects.append({'type': 'release', 'id': release_id}) swh_ids = get_swh_persistent_ids(swh_objects, snapshot_context) dir_path = '/'.join([bc['name'] for bc in breadcrumbs]) + '/' context_found = 'snapshot: %s' % snapshot_context['snapshot_id'] if origin_info: context_found = 'origin: %s' % origin_info['url'] heading = 'Directory - %s - %s - %s' %\ (dir_path, snapshot_context['branch'], context_found) return render(request, 'browse/directory.html', {'heading': heading, 'swh_object_name': 'Directory', 'swh_object_metadata': dir_metadata, 'dirs': dirs, 'files': files, 'breadcrumbs': breadcrumbs if root_sha1_git else [], 'top_right_link': { 'url': history_url, 'icon': swh_object_icons['revisions history'], 'text': 'History' }, 'readme_name': readme_name, 'readme_url': readme_url, 'readme_html': readme_html, 'snapshot_context': snapshot_context, 'vault_cooking': vault_cooking, 'show_actions_menu': True, 'swh_ids': swh_ids}) def browse_snapshot_content(request, snapshot_id=None, origin_type=None, origin_url=None, timestamp=None, path=None): """ Django view implementation for browsing a content in a snapshot context. """ try: snapshot_context = _process_snapshot_request(request, snapshot_id, origin_type, origin_url, timestamp, path, browse_context='content') root_sha1_git = snapshot_context['root_sha1_git'] sha1_git = None query_string = None content_data = None if root_sha1_git: content_info = service.lookup_directory_with_path(root_sha1_git, path) sha1_git = content_info['target'] query_string = 'sha1_git:' + sha1_git content_data = request_content(query_string, raise_if_unavailable=False) except Exception as exc: return handle_view_exception(request, exc) swh_type = snapshot_context['swh_type'] url_args = snapshot_context['url_args'] query_params = snapshot_context['query_params'] revision_id = snapshot_context['revision_id'] origin_info = snapshot_context['origin_info'] visit_info = snapshot_context['visit_info'] snapshot_id = snapshot_context['snapshot_id'] content = None language = None mimetype = None if content_data and content_data['raw_data'] is not None: content_display_data = prepare_content_for_display( content_data['raw_data'], content_data['mimetype'], path) content = content_display_data['content_data'] language = content_display_data['language'] mimetype = content_display_data['mimetype'] filename = None path_info = None browse_view_name = 'browse-' + swh_type + '-directory' breadcrumbs = [] split_path = path.split('/') filename = split_path[-1] path_info = gen_path_info(path[:-len(filename)]) if root_sha1_git: breadcrumbs.append({'name': root_sha1_git[:7], 'url': reverse(browse_view_name, url_args=url_args, query_params=query_params)}) for pi in path_info: bc_url_args = dict(url_args) bc_url_args['path'] = pi['path'] breadcrumbs.append({'name': pi['name'], 'url': reverse(browse_view_name, url_args=bc_url_args, query_params=query_params)}) breadcrumbs.append({'name': filename, 'url': None}) browse_content_link = \ gen_content_link(sha1_git, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) content_raw_url = None if query_string: content_raw_url = reverse('browse-content-raw', url_args={'query_string': query_string}, query_params={'filename': filename}) browse_rev_link = \ gen_revision_link(revision_id, snapshot_context=snapshot_context, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) content_metadata = { 'context-independent content': browse_content_link, 'path': None, 'filename': None, 'revision id': revision_id, 'revision': browse_rev_link, 'snapshot id': snapshot_id } cnt_sha1_git = None content_size = None error_code = 200 error_description = '' error_message = '' if content_data: content_metadata['sha1 checksum'] = \ content_data['checksums']['sha1'] content_metadata['sha1_git checksum'] = \ content_data['checksums']['sha1_git'] content_metadata['sha256 checksum'] = \ content_data['checksums']['sha256'] content_metadata['blake2s256 checksum'] = \ content_data['checksums']['blake2s256'] content_metadata['mime type'] = content_data['mimetype'] content_metadata['encoding'] = content_data['encoding'] content_metadata['size'] = filesizeformat(content_data['length']) content_metadata['language'] = content_data['language'] content_metadata['licenses'] = content_data['licenses'] content_metadata['path'] = '/' + path[:-len(filename)] content_metadata['filename'] = filename cnt_sha1_git = content_data['checksums']['sha1_git'] content_size = content_data['length'] error_code = content_data['error_code'] error_message = content_data['error_message'] error_description = content_data['error_description'] if origin_info: content_metadata['origin id'] = origin_info['id'] content_metadata['origin type'] = origin_info['type'] content_metadata['origin url'] = origin_info['url'] content_metadata['origin visit date'] = format_utc_iso_date(visit_info['date']) # noqa content_metadata['origin visit id'] = visit_info['visit'] browse_snapshot_url = reverse('browse-snapshot-content', url_args={'snapshot_id': snapshot_id, 'path': path}, query_params=request.GET) browse_snapshot_link = \ gen_link(browse_snapshot_url, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) content_metadata['snapshot context'] = browse_snapshot_link swh_objects = [{'type': 'content', 'id': cnt_sha1_git}, {'type': 'revision', 'id': revision_id}, {'type': 'snapshot', 'id': snapshot_id}] release_id = snapshot_context['release_id'] if release_id: swh_objects.append({'type': 'release', 'id': release_id}) swh_ids = get_swh_persistent_ids(swh_objects, snapshot_context) content_path = '/'.join([bc['name'] for bc in breadcrumbs]) context_found = 'snapshot: %s' % snapshot_context['snapshot_id'] if origin_info: context_found = 'origin: %s' % origin_info['url'] heading = 'Content - %s - %s - %s' %\ (content_path, snapshot_context['branch'], context_found) return render(request, 'browse/content.html', {'heading': heading, 'swh_object_name': 'Content', 'swh_object_metadata': content_metadata, 'content': content, 'content_size': content_size, 'max_content_size': content_display_max_size, 'mimetype': mimetype, 'language': language, 'breadcrumbs': breadcrumbs if root_sha1_git else [], 'top_right_link': { 'url': content_raw_url, 'icon': swh_object_icons['content'], 'text': 'Raw File' }, 'snapshot_context': snapshot_context, 'vault_cooking': None, 'show_actions_menu': True, 'swh_ids': swh_ids, 'error_code': error_code, 'error_message': error_message, 'error_description': error_description}, status=error_code) PER_PAGE = 100 def browse_snapshot_log(request, snapshot_id=None, origin_type=None, origin_url=None, timestamp=None): """ Django view implementation for browsing a revision history in a snapshot context. """ try: snapshot_context = _process_snapshot_request(request, snapshot_id, origin_type, origin_url, timestamp, browse_context='log') # noqa revision_id = snapshot_context['revision_id'] per_page = int(request.GET.get('per_page', PER_PAGE)) offset = int(request.GET.get('offset', 0)) revs_ordering = request.GET.get('revs_ordering', 'committer_date') session_key = 'rev_%s_log_ordering_%s' % (revision_id, revs_ordering) rev_log_session = request.session.get(session_key, None) rev_log = [] revs_walker_state = None if rev_log_session: rev_log = rev_log_session['rev_log'] revs_walker_state = rev_log_session['revs_walker_state'] if len(rev_log) < offset+per_page: revs_walker = \ service.get_revisions_walker(revs_ordering, revision_id, max_revs=offset+per_page+1, state=revs_walker_state) rev_log += list(revs_walker) revs_walker_state = revs_walker.export_state() revision_log = rev_log[offset:offset+per_page] request.session[session_key] = { 'rev_log': rev_log, 'revs_walker_state': revs_walker_state } except Exception as exc: return handle_view_exception(request, exc) swh_type = snapshot_context['swh_type'] origin_info = snapshot_context['origin_info'] visit_info = snapshot_context['visit_info'] url_args = snapshot_context['url_args'] query_params = snapshot_context['query_params'] snapshot_id = snapshot_context['snapshot_id'] query_params['per_page'] = per_page revs_ordering = request.GET.get('revs_ordering', '') query_params['revs_ordering'] = revs_ordering browse_view_name = 'browse-' + swh_type + '-log' prev_log_url = None if len(rev_log) > offset + per_page: query_params['offset'] = offset + per_page prev_log_url = reverse(browse_view_name, url_args=url_args, query_params=query_params) next_log_url = None if offset != 0: query_params['offset'] = offset - per_page next_log_url = reverse(browse_view_name, url_args=url_args, query_params=query_params) revision_log_data = format_log_entries(revision_log, per_page, snapshot_context) browse_log_link = \ gen_revision_log_link(revision_id, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) revision_metadata = { 'context-independent revision history': browse_log_link, 'snapshot id': snapshot_id } if origin_info: revision_metadata['origin id'] = origin_info['id'] revision_metadata['origin type'] = origin_info['type'] revision_metadata['origin url'] = origin_info['url'] revision_metadata['origin visit date'] = format_utc_iso_date(visit_info['date']) # noqa revision_metadata['origin visit id'] = visit_info['visit'] browse_snapshot_url = reverse('browse-snapshot-log', url_args={'snapshot_id': snapshot_id}, query_params=request.GET) browse_snapshot_link = \ gen_link(browse_snapshot_url, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) revision_metadata['snapshot context'] = browse_snapshot_link swh_objects = [{'type': 'revision', 'id': revision_id}, {'type': 'snapshot', 'id': snapshot_id}] release_id = snapshot_context['release_id'] if release_id: swh_objects.append({'type': 'release', 'id': release_id}) swh_ids = get_swh_persistent_ids(swh_objects, snapshot_context) context_found = 'snapshot: %s' % snapshot_context['snapshot_id'] if origin_info: context_found = 'origin: %s' % origin_info['url'] heading = 'Revision history - %s - %s' %\ (snapshot_context['branch'], context_found) return render(request, 'browse/revision-log.html', {'heading': heading, 'swh_object_name': 'Revisions history', 'swh_object_metadata': revision_metadata, 'revision_log': revision_log_data, 'revs_ordering': revs_ordering, 'next_log_url': next_log_url, 'prev_log_url': prev_log_url, 'breadcrumbs': None, 'top_right_link': None, 'snapshot_context': snapshot_context, 'vault_cooking': None, 'show_actions_menu': True, 'swh_ids': swh_ids}) def browse_snapshot_branches(request, snapshot_id=None, origin_type=None, origin_url=None, timestamp=None): """ Django view implementation for browsing a list of branches in a snapshot context. """ try: snapshot_context = _process_snapshot_request(request, snapshot_id, origin_type, origin_url, timestamp) branches_bc = request.GET.get('branches_breadcrumbs', '') branches_bc = \ branches_bc.split(',') if branches_bc else [] branches_from = branches_bc[-1] if branches_bc else '' swh_type = snapshot_context['swh_type'] origin_info = snapshot_context['origin_info'] url_args = snapshot_context['url_args'] query_params = snapshot_context['query_params'] browse_view_name = 'browse-' + swh_type + '-directory' - displayed_branches = \ + snapshot = \ service.lookup_snapshot(snapshot_context['snapshot_id'], branches_from, PER_PAGE+1, - target_types=['revision'])['branches'] + target_types=['revision', 'alias']) + + displayed_branches, _ = process_snapshot_branches(snapshot) + except Exception as exc: return handle_view_exception(request, exc) - displayed_branches, _ = process_snapshot_branches(displayed_branches) - for branch in displayed_branches: if snapshot_id: revision_url = reverse('browse-revision', url_args={'sha1_git': branch['revision']}, query_params={'snapshot_id': snapshot_id}) else: revision_url = reverse('browse-revision', url_args={'sha1_git': branch['revision']}, query_params={'origin_type': origin_type, 'origin': origin_info['url']}) query_params['branch'] = branch['name'] directory_url = reverse(browse_view_name, url_args=url_args, query_params=query_params) del query_params['branch'] branch['revision_url'] = revision_url branch['directory_url'] = directory_url browse_view_name = 'browse-' + swh_type + '-branches' prev_branches_url = None next_branches_url = None if branches_bc: query_params_prev = dict(query_params) query_params_prev['branches_breadcrumbs'] = \ ','.join(branches_bc[:-1]) prev_branches_url = reverse(browse_view_name, url_args=url_args, query_params=query_params_prev) elif branches_from: prev_branches_url = reverse(browse_view_name, url_args=url_args, query_params=query_params) if len(displayed_branches) > PER_PAGE: query_params_next = dict(query_params) next_branch = displayed_branches[-1]['name'] del displayed_branches[-1] branches_bc.append(next_branch) query_params_next['branches_breadcrumbs'] = \ ','.join(branches_bc) next_branches_url = reverse(browse_view_name, url_args=url_args, query_params=query_params_next) heading = 'Branches - ' if origin_info: heading += 'origin: %s' % origin_info['url'] else: heading += 'snapshot: %s' % snapshot_id return render(request, 'browse/branches.html', {'heading': heading, 'swh_object_name': 'Branches', 'swh_object_metadata': {}, 'top_right_link': None, 'displayed_branches': displayed_branches, 'prev_branches_url': prev_branches_url, 'next_branches_url': next_branches_url, 'snapshot_context': snapshot_context}) def browse_snapshot_releases(request, snapshot_id=None, origin_type=None, origin_url=None, timestamp=None): """ Django view implementation for browsing a list of releases in a snapshot context. """ try: snapshot_context = _process_snapshot_request(request, snapshot_id, origin_type, origin_url, timestamp) rel_bc = request.GET.get('releases_breadcrumbs', '') rel_bc = \ rel_bc.split(',') if rel_bc else [] rel_from = rel_bc[-1] if rel_bc else '' swh_type = snapshot_context['swh_type'] origin_info = snapshot_context['origin_info'] url_args = snapshot_context['url_args'] query_params = snapshot_context['query_params'] - displayed_releases = \ + snapshot = \ service.lookup_snapshot(snapshot_context['snapshot_id'], rel_from, PER_PAGE+1, - target_types=['release'])['branches'] + target_types=['release', 'alias']) + + _, displayed_releases = process_snapshot_branches(snapshot) + except Exception as exc: return handle_view_exception(request, exc) - _, displayed_releases = process_snapshot_branches(displayed_releases) - for release in displayed_releases: if snapshot_id: query_params_tgt = {'snapshot_id': snapshot_id} else: query_params_tgt = {'origin': origin_info['url']} release_url = reverse('browse-release', url_args={'sha1_git': release['id']}, query_params=query_params_tgt) target_url = '' if release['target_type'] == 'revision': target_url = reverse('browse-revision', url_args={'sha1_git': release['target']}, query_params=query_params_tgt) elif release['target_type'] == 'directory': target_url = reverse('browse-directory', url_args={'sha1_git': release['target']}, query_params=query_params_tgt) elif release['target_type'] == 'content': target_url = reverse('browse-content', url_args={'sha1_git': release['target']}, query_params=query_params_tgt) elif release['target_type'] == 'release': target_url = reverse('browse-release', url_args={'sha1_git': release['target']}, query_params=query_params_tgt) release['release_url'] = release_url release['target_url'] = target_url browse_view_name = 'browse-' + swh_type + '-releases' prev_releases_url = None next_releases_url = None if rel_bc: query_params_prev = dict(query_params) query_params_prev['releases_breadcrumbs'] = \ ','.join(rel_bc[:-1]) prev_releases_url = reverse(browse_view_name, url_args=url_args, query_params=query_params_prev) elif rel_from: prev_releases_url = reverse(browse_view_name, url_args=url_args, query_params=query_params) if len(displayed_releases) > PER_PAGE: query_params_next = dict(query_params) next_rel = displayed_releases[-1]['branch_name'] del displayed_releases[-1] rel_bc.append(next_rel) query_params_next['releases_breadcrumbs'] = \ ','.join(rel_bc) next_releases_url = reverse(browse_view_name, url_args=url_args, query_params=query_params_next) heading = 'Releases - ' if origin_info: heading += 'origin: %s' % origin_info['url'] else: heading += 'snapshot: %s' % snapshot_id return render(request, 'browse/releases.html', {'heading': heading, 'top_panel_visible': False, 'top_panel_collapsible': False, 'swh_object_name': 'Releases', 'swh_object_metadata': {}, 'top_right_link': None, 'displayed_releases': displayed_releases, 'prev_releases_url': prev_releases_url, 'next_releases_url': next_releases_url, 'snapshot_context': snapshot_context, 'vault_cooking': None, 'show_actions_menu': False}) diff --git a/swh/web/common/converters.py b/swh/web/common/converters.py index e4aded53..608a34f7 100644 --- a/swh/web/common/converters.py +++ b/swh/web/common/converters.py @@ -1,381 +1,387 @@ # Copyright (C) 2015-2018 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 datetime import json from swh.model import hashutil from swh.core.utils import decode_with_escape def _group_checksums(data): """Groups checksums values computed from hash functions used in swh and stored in data dict under a single entry 'checksums' """ if data: checksums = {} for hash in hashutil.ALGORITHMS: if hash in data and data[hash]: checksums[hash] = data[hash] del data[hash] if len(checksums) > 0: data['checksums'] = checksums def fmap(f, data): """Map f to data at each level. This must keep the origin data structure type: - map -> map - dict -> dict - list -> list - None -> None Args: f: function that expects one argument. data: data to traverse to apply the f function. list, map, dict or bare value. Returns: The same data-structure with modified values by the f function. """ if data is None: return data if isinstance(data, map): return map(lambda y: fmap(f, y), (x for x in data)) if isinstance(data, list): return [fmap(f, x) for x in data] if isinstance(data, dict): return {k: fmap(f, v) for (k, v) in data.items()} return f(data) def from_swh(dict_swh, hashess={}, bytess={}, dates={}, blacklist={}, removables_if_empty={}, empty_dict={}, empty_list={}, convert={}, convert_fn=lambda x: x): """Convert from a swh dictionary to something reasonably json serializable. Args: dict_swh: the origin dictionary needed to be transformed hashess: list/set of keys representing hashes values (sha1, sha256, sha1_git, etc...) as bytes. Those need to be transformed in hexadecimal string bytess: list/set of keys representing bytes values which needs to be decoded blacklist: set of keys to filter out from the conversion convert: set of keys whose associated values need to be converted using convert_fn convert_fn: the conversion function to apply on the value of key in 'convert' The remaining keys are copied as is in the output. Returns: dictionary equivalent as dict_swh only with its keys converted. """ def convert_hashes_bytes(v): """v is supposedly a hash as bytes, returns it converted in hex. """ if isinstance(v, bytes): return hashutil.hash_to_hex(v) return v def convert_bytes(v): """v is supposedly a bytes string, decode as utf-8. FIXME: Improve decoding policy. If not utf-8, break! """ if isinstance(v, bytes): return v.decode('utf-8') return v def convert_date(v): """ Args: v (dict or datatime): either: - a dict with three keys: - timestamp (dict or integer timestamp) - offset - negative_utc - or, a datetime We convert it to a human-readable string """ if not v: return v if isinstance(v, datetime.datetime): return v.isoformat() tz = datetime.timezone(datetime.timedelta(minutes=v['offset'])) swh_timestamp = v['timestamp'] if isinstance(swh_timestamp, dict): date = datetime.datetime.fromtimestamp( swh_timestamp['seconds'], tz=tz) else: date = datetime.datetime.fromtimestamp( swh_timestamp, tz=tz) datestr = date.isoformat() if v['offset'] == 0 and v['negative_utc']: # remove the rightmost + and replace it with a - return '-'.join(datestr.rsplit('+', 1)) return datestr if not dict_swh: return dict_swh new_dict = {} for key, value in dict_swh.items(): if key in blacklist or (key in removables_if_empty and not value): continue if key in dates: new_dict[key] = convert_date(value) elif key in convert: new_dict[key] = convert_fn(value) elif isinstance(value, dict): new_dict[key] = from_swh(value, hashess=hashess, bytess=bytess, dates=dates, blacklist=blacklist, removables_if_empty=removables_if_empty, empty_dict=empty_dict, empty_list=empty_list, convert=convert, convert_fn=convert_fn) elif key in hashess: new_dict[key] = fmap(convert_hashes_bytes, value) elif key in bytess: try: new_dict[key] = fmap(convert_bytes, value) except UnicodeDecodeError: if 'decoding_failures' not in new_dict: new_dict['decoding_failures'] = [key] else: new_dict['decoding_failures'].append(key) new_dict[key] = fmap(decode_with_escape, value) elif key in empty_dict and not value: new_dict[key] = {} elif key in empty_list and not value: new_dict[key] = [] else: new_dict[key] = value _group_checksums(new_dict) return new_dict def from_provenance(provenance): """Convert from a swh provenance information to a provenance dictionary. Args: provenance (dict): Dictionary with the following keys: - content (sha1_git): the content's identifier - revision (sha1_git): the revision the content was seen - origin (int): the origin the content was seen - visit (int): the visit it occurred - path (bytes): the path the content was seen at """ return from_swh(provenance, hashess={'content', 'revision'}, bytess={'path'}) def from_origin(origin): """Convert from a swh origin to an origin dictionary. """ return from_swh(origin) def from_release(release): """Convert from a swh release to a json serializable release dictionary. Args: release (dict): dictionary with keys: - id: identifier of the revision (sha1 in bytes) - revision: identifier of the revision the release points to (sha1 in bytes) comment: release's comment message (bytes) name: release's name (string) author: release's author identifier (swh's id) synthetic: the synthetic property (boolean) Returns: dict: Release dictionary with the following keys: - id: hexadecimal sha1 (string) - revision: hexadecimal sha1 (string) - comment: release's comment message (string) - name: release's name (string) - author: release's author identifier (swh's id) - synthetic: the synthetic property (boolean) """ return from_swh( release, hashess={'id', 'target'}, bytess={'message', 'name', 'fullname', 'email'}, dates={'date'}, ) class SWHMetadataEncoder(json.JSONEncoder): """Special json encoder for metadata field which can contain bytes encoded value. """ def default(self, obj): if isinstance(obj, bytes): return obj.decode('utf-8') # Let the base class default method raise the TypeError return json.JSONEncoder.default(self, obj) def convert_revision_metadata(metadata): """Convert json specific dict to a json serializable one. """ if not metadata: return {} return json.loads(json.dumps(metadata, cls=SWHMetadataEncoder)) def from_revision(revision): """Convert from a swh revision to a json serializable revision dictionary. Args: revision (dict): dict with keys: - id: identifier of the revision (sha1 in bytes) - directory: identifier of the directory the revision points to (sha1 in bytes) - author_name, author_email: author's revision name and email - committer_name, committer_email: committer's revision name and email - message: revision's message - date, date_offset: revision's author date - committer_date, committer_date_offset: revision's commit date - parents: list of parents for such revision - synthetic: revision's property nature - type: revision's type (git, tar or dsc at the moment) - metadata: if the revision is synthetic, this can reference dynamic properties. Returns: dict: Revision dictionary with the same keys as inputs, except: - sha1s are in hexadecimal strings (id, directory) - bytes are decoded in string (author_name, committer_name, author_email, committer_email) Remaining keys are left as is """ revision = from_swh(revision, hashess={'id', 'directory', 'parents', 'children'}, bytess={'name', 'fullname', 'email'}, convert={'metadata'}, convert_fn=convert_revision_metadata, dates={'date', 'committer_date'}) if revision: if 'parents' in revision: revision['merge'] = len(revision['parents']) > 1 if 'message' in revision: try: revision['message'] = revision['message'].decode('utf-8') except UnicodeDecodeError: revision['message_decoding_failed'] = True revision['message'] = None return revision def from_content(content): """Convert swh content to serializable content dictionary. """ return from_swh(content, hashess={'sha1', 'sha1_git', 'sha256', 'blake2s256'}, blacklist={'ctime'}, convert={'status'}, convert_fn=lambda v: 'absent' if v == 'hidden' else v) def from_person(person): """Convert swh person to serializable person dictionary. """ return from_swh(person, bytess={'name', 'fullname', 'email'}) def from_origin_visit(visit): """Convert swh origin_visit to serializable origin_visit dictionary. """ ov = from_swh(visit, hashess={'target', 'snapshot'}, bytess={'branch'}, dates={'date'}, empty_dict={'metadata'}) return ov def from_snapshot(snapshot): """Convert swh snapshot to serializable snapshot dictionary. """ sv = from_swh(snapshot, hashess={'id', 'target'}, bytess={'next_branch'}) if sv and 'branches' in sv: sv['branches'] = { decode_with_escape(k): v for k, v in sv['branches'].items() } + for k, v in snapshot['branches'].items(): + # alias target existing branch names, not a sha1 + if v and v['target_type'] == 'alias': + branch = decode_with_escape(k) + target = v['target'].decode('utf-8') + sv['branches'][branch]['target'] = target return sv def from_directory_entry(dir_entry): """Convert swh person to serializable person dictionary. """ return from_swh(dir_entry, hashess={'dir_id', 'sha1_git', 'sha1', 'sha256', 'blake2s256', 'target'}, bytess={'name'}, removables_if_empty={ 'sha1', 'sha1_git', 'sha256', 'blake2s256', 'status'}, convert={'status'}, convert_fn=lambda v: 'absent' if v == 'hidden' else v) def from_filetype(content_entry): """Convert swh person to serializable person dictionary. """ return from_swh(content_entry, hashess={'id'}) diff --git a/swh/web/tests/api/views/test_snapshot.py b/swh/web/tests/api/views/test_snapshot.py index abcaa4b6..62b74e78 100644 --- a/swh/web/tests/api/views/test_snapshot.py +++ b/swh/web/tests/api/views/test_snapshot.py @@ -1,186 +1,218 @@ # 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 rest_framework.test import APITestCase from unittest.mock import patch from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.common.utils import reverse from swh.web.tests.testcase import SWHWebTestCase _snapshot_id = '36ce36946fcd0f39bdfc40727af4acfce81f67af' _snapshot_branches = [ + { + 'name': 'gh-pages', + 'target': 'refs/heads/gh-pages', + 'target_type': 'alias', + }, + { + 'name': 'latest', + 'target': 'refs/tags/v1.3.0', + 'target_type': 'alias', + }, { 'name': 'refs/heads/andresgalante-dismissable-badges', 'target': '3af57e6db525015a25b4f3abb29263432e4af7b1', 'target_type': 'revision', }, { 'name': 'refs/heads/boom-toasted', 'target': 'e3f4957facfbdc25fdc4c6d3f7fcf1c488f06cea', 'target_type': 'revision', }, { 'name': 'refs/heads/flex-checks', 'target': '1b08ea1630a987d6f56f0c99e869896485bf230b', 'target_type': 'revision', }, { 'name': 'refs/heads/gh-pages', 'target': 'ea4129886adec992483aa592db717f908b4be355', 'target_type': 'revision', }, { 'name': 'refs/heads/grid-vertical-align-height', 'target': 'ba23eea651f483b88b78bee1084d7d0f61161e8d', 'target_type': 'revision', }, { 'name': 'refs/tags/v1.0.0', 'target': 'd28343dc3ad53a411ae3685e7d6a7866c8c22d6b', 'target_type': 'release', }, { 'name': 'refs/tags/v1.1.0', 'target': '0f11410b77140852f835ad456e5fbcf28760846c', 'target_type': 'release', }, { 'name': 'refs/tags/v1.1.1', 'target': '4ca26e76ee4867bfcd65ecf81039f183fc1d3b4d', 'target_type': 'release', }, { 'name': 'refs/tags/v1.2.0', 'target': 'cabae13db21e0e1db686f5628d1725c6191062ef', 'target_type': 'release', }, { 'name': 'refs/tags/v1.3.0', 'target': '3c3d596d94501509bec1959a4cfb44b20bfa8606', 'target_type': 'release', } ] def _lookup_snapshot(snapshot_id, branches_from='', branches_count=None, target_types=None): ret = {'id': snapshot_id, 'branches': {}} count = 0 for branch in _snapshot_branches: if branches_count and count == branches_count: break if branch['name'] >= branches_from: if not target_types or branch['target_type'] in target_types: # noqa ret['branches'][branch['name']] = { 'target': branch['target'], 'target_type': branch['target_type'] } count += 1 return ret +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 + + def _enrich_snapshot_data(snapshot_data): for branch in snapshot_data['branches'].keys(): target = snapshot_data['branches'][branch]['target'] - if snapshot_data['branches'][branch]['target_type'] == 'revision': # noqa - snapshot_data['branches'][branch]['target_url'] = \ - reverse('api-revision', url_args={'sha1_git': target}) - if snapshot_data['branches'][branch]['target_type'] == 'release': # noqa - snapshot_data['branches'][branch]['target_url'] = \ - reverse('api-release', url_args={'sha1_git': target}) + target_type = snapshot_data['branches'][branch]['target_type'] + snapshot_data['branches'][branch]['target_url'] = \ + _get_branch_url(target_type, target) + for branch in snapshot_data['branches'].keys(): + target = snapshot_data['branches'][branch]['target'] + target_type = snapshot_data['branches'][branch]['target_type'] + if target_type == 'alias': + if target in snapshot_data['branches']: + snapshot_data['branches'][branch]['target_url'] = \ + snapshot_data['branches'][target]['target_url'] + else: + snp = _lookup_snapshot(snapshot_data['id'], + branches_from=target, + branches_count=1) + alias_target = snp['branches'][target]['target'] + alias_target_type = snp['branches'][target]['target_type'] + snapshot_data['branches'][branch]['target_url'] = \ + _get_branch_url(alias_target_type, alias_target) + return snapshot_data @patch('swh.web.api.views.snapshot.service') class SnapshotApiTestCase(SWHWebTestCase, APITestCase): def test_api_snapshot(self, mock_service): mock_service.lookup_snapshot.side_effect = _lookup_snapshot url = reverse('api-snapshot', url_args={'snapshot_id': _snapshot_id}) rv = self.client.get(url) self.assertEqual(rv.status_code, 200) self.assertEqual(rv['Content-Type'], 'application/json') expected_data = _lookup_snapshot(_snapshot_id) expected_data = _enrich_snapshot_data(expected_data) self.assertEqual(rv.data, expected_data) def test_api_snapshot_paginated(self, mock_service): mock_service.lookup_snapshot.side_effect = _lookup_snapshot branches_offset = 0 branches_count = 2 whole_snapshot = {'id': _snapshot_id, 'branches': {}} while branches_offset < len(_snapshot_branches): branches_from = _snapshot_branches[branches_offset]['name'] url = reverse('api-snapshot', url_args={'snapshot_id': _snapshot_id}, 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 = _lookup_snapshot(_snapshot_id, branches_from, branches_count) expected_data = _enrich_snapshot_data(expected_data) self.assertEqual(rv.data, expected_data) whole_snapshot['branches'].update(expected_data['branches']) branches_offset += branches_count if branches_offset < len(_snapshot_branches): next_branch = _snapshot_branches[branches_offset]['name'] # noqa next_url = reverse('api-snapshot', url_args={'snapshot_id': _snapshot_id}, query_params={'branches_from': next_branch, 'branches_count': branches_count}) # noqa 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_id}) rv = self.client.get(url) self.assertEqual(rv.status_code, 200) self.assertEqual(rv['Content-Type'], 'application/json') self.assertEqual(rv.data, whole_snapshot) def test_api_snapshot_filtered(self, mock_service): mock_service.lookup_snapshot.side_effect = _lookup_snapshot target_types = 'release' url = reverse('api-snapshot', url_args={'snapshot_id': _snapshot_id}, query_params={'target_types': target_types}) rv = self.client.get(url) self.assertEqual(rv.status_code, 200) self.assertEqual(rv['Content-Type'], 'application/json') expected_data = _lookup_snapshot(_snapshot_id, target_types=target_types.split(',')) expected_data = _enrich_snapshot_data(expected_data) self.assertEqual(rv.data, expected_data) def test_api_snapshot_errors(self, mock_service): mock_service.lookup_snapshot.side_effect = \ BadInputExc('Invalid snapshot id!') url = reverse('api-snapshot', url_args={'snapshot_id': '63ce369'}) rv = self.client.get(url) self.assertEqual(rv.status_code, 400) mock_service.lookup_snapshot.side_effect = \ NotFoundExc('Snapshot not found!') snapshot_inexistent = '63ce36946fcd0f79bdfc40727af4acfce81f67fa' url = reverse('api-snapshot', url_args={'snapshot_id': snapshot_inexistent}) rv = self.client.get(url) self.assertEqual(rv.status_code, 404)